AI Snake Oil—A New Book by 2 Princeton University Computer Scientists

24 sept. · Ground Truths

Arvind Narayanan and Sayash Kapoor are well regarded computer scientists at Princeton University and have just published a book with a provocative title, AI Snake Oil. Here I’ve interviewed Sayash and challenged him on this dismal title, for which he provides solid examples of predictive AI’s failures. Then we get into the promise of generative AI.

Full videos of all Ground Truths podcasts can be seen on YouTube here. The audios are also available on Apple and Spotify.

Transcript with links to audio and external links to key publications

Eric Topol (00:06):

Hello, it's Eric Topol with Ground Truths, and I'm delighted to welcome the co-author of a new book AI SNAKE OIL and it's Sayash Kapoor who has written this book with Arvind Narayanan of Princeton. And so welcome, Sayash. It's wonderful to have you on Ground Truths.

Sayash Kapoor (00:28):

Thank you so much. It's a pleasure to be here.

Eric Topol (00:31):

Well, congratulations on this book. What's interesting is how much you've achieved at such a young age. Here you are named in TIME100 AI’s inaugural edition as one of those eminent contributors to the field. And you're currently a PhD candidate at Princeton, is that right?

Sayash Kapoor (00:54):

That's correct, yes. I work at the Center for Information Technology Policy, which is a joint program between the computer science department and the school of public and international affairs.

Eric Topol (01:05):

So before you started working on your PhD in computer science, you already were doing this stuff, I guess, right?

Sayash Kapoor (01:14):

That's right. So before I started my PhD, I used to work at Facebook as a machine learning engineer.

Eric Topol (01:20):

Yeah, well you're taking it to a more formal level here. Before I get into the book itself, what was the background? I mean you did describe it in the book why you decided to write a book, especially one that was entitled AI Snake Oil: What Artificial Intelligence Can Do, What It Can't, and How to Tell the Difference.

Background to Writing the Book

Sayash Kapoor (01:44):

Yeah, absolutely. So I think for the longest time both Arvind and I had been sort of looking at how AI works and how it doesn't work, what are cases where people are somewhat fooled by the potential for this technology and fail to apply it in meaningful ways in their life. As an engineer at Facebook, I had seen how easy it is to slip up or make mistakes when deploying machine learning and AI tools in the real world. And had also seen that, especially when it comes to research, it's really easy to make mistakes even unknowingly that inflate the accuracy of a machine learning model. So as an example, one of the first research projects I did when I started my PhD was to look at the field of political science in the subfield of civil war prediction. This is a field which tries to predict where the next civil war will happen and in order to better be prepared for civil conflict.

(02:39):

And what we found was that there were a number of papers that claimed almost perfect accuracy at predicting when a civil war will take place. At first this seemed sort of astounding. If AI can really help us predict when a civil war will start like years in advance sometimes, it could be game changing, but when we dug in, it turned out that every single one of these claims where people claim that AI was better than two decades old logistic regression models, every single one of these claims was not reproducible. And so, that sort of set the alarm bells ringing for the both of us and we sort of dug in a little bit deeper and we found that this is pervasive. So this was a pervasive issue across fields that were quickly adopting AI and machine learning. We found, I think over 300 papers and the last time I compiled this list, I think it was over 600 papers that suffer from data leakage. That is when you can sort of train on the sets that you're evaluating your models on. It's sort of like teaching to the test. And so, machine learning model seems like it does much better when you evaluate it on your data compared to how it would really work out in the real world.

Eric Topol (03:48):

Right. You say in the book, “the goal of this book is to identify AI snake oil - and to distinguish it from AI that can work well if used in the right ways.” Now I have to tell you, it's kind of a downer book if you're an AI enthusiast because there's not a whole lot of positive here. We'll get to that in a minute. But you break down the types of AI, which I'm going to challenge a bit into three discrete areas, the predictive AI, which you take a really harsh stance on, say it will never work. Then there's generative AI, obviously the large language models that took the world by storm, although they were incubating for several years when ChatGPT came along and then content moderation AI. So maybe you could tell us about your breakdown to these three different domains of AI.

Three Types of AI: Predictive, Generative, Content Moderation

Sayash Kapoor (04:49):

Absolutely. I think one of our main messages across the book is that when we are talking about AI, often what we are really interested in are deeper questions about society. And so, our breakdown of predictive, generative, and content moderation AI sort of reflects how these tools are being used in the real world today. So for predictive AI, one of the motivations for including this in the book as a separate category was that we found that it often has nothing to do with modern machine learning methods. In some cases it can be as simple as decades old linear regression tools or logistic regression tools. And yet these tools are sold under the package of AI. Advances that are being made in generative AI are sold as if they apply to predictive AI as well. Perhaps as a result, what we are seeing is across dozens of different domains, including insurance, healthcare, education, criminal justice, you name it, companies have been selling predictive AI with the promise that we can use it to replace human decision making.

(05:51):

And I think that last part is where a lot of our issues really come down to because these tools are being sold as far more than they're actually capable of. These tools are being sold as if they can enable better decision making for criminal justice. And at the same time, when people have tried to interrogate these tools, what we found is these tools essentially often work no better than random, especially when it comes to some consequential decisions such as job automation. So basically deciding who gets to be called on the next level of like a job interview or who is rejected, right as soon as they submit the CV. And so, these are very, very consequential decisions and we felt like there is a lot of snake oil in part because people don't distinguish between applications that have worked really well or where we have seen tremendous advances such as generative AI and applications where essentially we've stalled for a number of decades and these tools don't really work as claimed by the developers.

Eric Topol (06:55):

I mean the way you partition that, the snake oil, which is a tough metaphor, and you even show the ad from 1905 of snake oil in the book. You're really getting at predictive AI and how it is using old tools and selling itself as some kind of breakthrough. Before I challenge that, are we going to be able to predict things? By the way, using generative AI, not as you described, but I would like to go through a few examples of how bad this has been and since a lot of our listeners and readers are in the medical world or biomedical world, I'll try to get to those. So one of the first ones you mentioned, which I completely agree, is how prediction of Covid from the chest x-ray and there were thousands of these studies that came throughout the pandemic. Maybe you could comment about that one.

Some Flagrant Examples

Sayash Kapoor (08:04):

Absolutely. Yeah, so this is one of my favorite examples as well. So essentially Michael Roberts and his team at the University of Cambridge a year or so after the pandemic looked back at what had happened. I think at the time there were around 500 studies that they included in the sample. And they looked back to see how many of these would be useful in a clinical setting beyond just the scope of writing a research paper. And they started out by using a simple checklist to see, okay, are these tools well validated? Does the training and the testing data, is it separate? And so on. So they ran through the simple checklist and that excluded all but 60 of these studies from consideration. So apart from 60 studies, none of these other studies even passed a very, very basic criteria for being included in the analysis. Now for these 60, it turns out that if you take a guess about how many were useful, I'm pretty confident most cases would be wrong.

(09:03):

There were exactly zero studies that were useful in a clinically relevant setting. And the reasons for this, I mean in some cases the reasons were as bizarre as training a machine learning model to predict Covid where all of the positive samples of people who had Covid were from adults. But all of the negative samples of people who didn't have Covid were from children. And so, essentially claiming that the resulting classifier can predict who has Covid is bizarre because all the classifier is doing is looking at the checks history and basically predicting which x-ray belongs to a child versus an adult. And so, this is the sort of error in some cases we saw duplicates in the training and test set. So you have the same person that is being used for training the model and that it is also used for evaluating the model. So simply memorizing a given sample of x-rays would be enough to achieve a very high performance. And so, for issues like these, I think all 60 of these studies prove to be not useful in a clinically relevant setting. And I think this is sort of the type of pattern that we've seen over and over again.

Eric Topol (10:14):

Yeah, and I agree with you on that point. I mean that was really a flagrant example and that would fulfill your title of your book, which as I said is a very tough title. But on page 29, and we'll have this in the post. You have a figure, the landscape of AI snake oil, hype, and harm.

And the problem is there is nothing good in this landscape. So on the y-axis you have works, hype, snake oil going up on the y-axis. And on the x-axis, you have benign and harmful. So the only thing you have that works and that's benign is autocomplete. I wouldn't say that works. And then you have works facial recognition for surveillance is harmful. This is a pretty sobering view of AI. Obviously, there's many things that are working that aren't on this landscape. So I just would like to challenge, are you a bit skewed here and only fixating on bad things? Because this diagram is really rough. I mean, there's so much progress in AI and you have in here you mentioned the predicting civil wars, and obviously we have these cheating detection, criminal risk prediction. I mean a lot of problems, video interviews that are deep fakes, but you don't present any good things.

Optimism on Generative AI

Sayash Kapoor (11:51):

So to be clear, I think both Arvind and are somewhat paradoxically optimistic about the future of generative AI. And so, the decision to focus on snake oil was a very intentional one from our end. So in particular, I think at various places in the book we outline why we're optimistic, what types of applications we think we're optimistic about as well. And the reason we don't focus on them is that it basically comes down to the fact that no one wants to read a book that has 300 pages about the virtues of spellcheck or AI for code generation or something like that. But I think I completely agree and acknowledge that there are lots of positive applications that didn't make the cut for the book as well. That was because we wanted people to come to this from a place of skepticism so that they're not fooled by the hype.

(12:43):

Because essentially we see even these positive uses of AI being lost out if people have unrealistic expectations from what an AI tool should do. And so, pointing out snake oil is almost a prerequisite for being able to use AI productively in your work environment. I can give a couple of examples of where or how we've sort of manifested this optimism. One is AI for coding. I think writing code is an application that I do, at least I use AI a lot. I think almost half of the code I write these days is generated, at least the first draft is generated using AI. And yet if I did not know how to program, it would be a completely different question, right? Because for me pointing out that, oh, this syntax looks incorrect or this is not handling the data in the correct way is as simple as looking at a piece of code because I've done this a few times. But if I weren't an expert on programming, it would be completely disastrous because even if the error rate is like 5%, I would have dozens of errors in my code if I'm using AI to generate it.

(13:51):

Another example of how we've been using it in our daily lives is Arvind has two little kids and he's built a number of applications for his kids using AI. So I think he's a big proponent of incorporating AI into children's lives as a force for good rather than having a completely hands-off approach. And I think both of these are just two examples, but I would say a large amount of our work these days occurs with the assistance of AI. So we are very much optimistic. And at the same time, I think one of the biggest hindrances to actually adopting AI in the real world is not understanding its limitations.

Eric Topol (14:31):

Right. Yeah, you say in the book quote, “the two of us are enthusiastic users of generative AI, both in our work and our personal lives.” It just doesn't come through as far as the examples. But before I leave the troubles of predictive AI, I liked to get into a few more examples because that's where your book shines in convincing that we got some trouble here and we need to be completely aware. So one of the most famous, well, there's a couple we're going to get into, but one I'd like to review with you, it's in the book, is the prediction of sepsis in the Epic model. So as you know very well, Epic is the most used IT and health systems electronic health records, and they launched never having published an algorithm that would tell when the patient was hospitalized if they actually had sepsis or risk of sepsis. Maybe you could take us through that, what you do in the book, and it truly was a fiasco.

The Sepsis Debacle

Sayash Kapoor (15:43):

Absolutely. So I think back in 2016/2017, Epic came up with a system that would help healthcare providers predict which patients are most at risk of sepsis. And I think, again, this is a very important problem. I think sepsis is one of the leading causes of death worldwide and even in the US. And so, if we could fix that, I think it would be a game changer. The problem was that there were no external validations of this algorithm for the next four years. So for four years, between 2017 to 2021, the algorithm wasn't used by hundreds of hospitals in the US. And in 2021, a team from University of Michigan did this study in their own hospital to see what the efficacy of the sepsis prediction model is. They found out that Epic had claimed an AUC of between 0.76 and 0.83, and the actual AUC was closer to 0.6, and AUC of 0.5 is making guesses at random.

(16:42):

So this was much, much worse than the company's claims. And I think even after that, it still took a year for sepsis to roll back this algorithm. So at first, Epic's claims were that this model works well and that's why hospitals are adopting it. But then it turned out that Epic was actually incentivizing hospitals to adopt sepsis prediction models. I think they were giving credits of hundreds of thousands of dollars in some cases. If a hospital satisfied a certain set of conditions, one of these conditions was using a sepsis prediction model. And so, we couldn't really take their claims at face value. And finally in October 2022, Epic essentially rolled back this algorithm. So they went from this one size fits all sepsis prediction model to a model that each hospital has to train on its own data, an approach which I think is more likely to work because each hospital's data is different. But it's also more time consuming and expensive for the hospitals because all of a sudden you now need your own data analysts to be able to roll out this model to be able to monitor it.

(17:47):

I think this study also highlights many of the more general issues with predictive AI. These tools are often sold as if they're replacements for an existing system, but then when things go bad, essentially they're replaced with tools that do far less. And companies often go back to the fine print saying that, oh, we should always deploy it with the human in the loop, or oh, it needs to have these extra protections that are not our responsibility, by the way. And I think that gap between what developers claim and how the tool actually works is what is most problematic.

Eric Topol (18:21):

Yeah, no, I mean it's an egregious example, and again, it fulfills like what we discussed with statistics, but even worse because it was marketed and it was incentivized financially and there's no doubt that some patients were completely miscategorized and potentially hurt. The other one, that's a classic example that went south is the Optum UnitedHealth algorithm. Maybe you could take us through that one as well, because that is yet another just horrible case of how people were discriminated against.

The Infamous Optum Algorithm

Sayash Kapoor (18:59):

Absolutely. So Optum, another health tech company created an algorithm to prioritize high risk patients for preemptive care. So I think it was around when Obamacare was being introduced that insurance networks started looking into how they could reduce costs. And one of the main ways they identified to reduce costs is basically preemptively caring for patients who are extremely high risk. So in this case, they decided to keep 3% of the patients in the high risk category and they built a classifier to decide who's the highest risk, because potentially once you have these patients, you can proactively treat them. There might be fewer emergency room visits, there might be fewer hospitalizations and so on. So that's all fine and good. But what happened when they implemented the algorithm was that every machine learning model needs like the target variable, what is being predicted at the end of the day. What they decided to predict was how much patient would pay, how much would they charge, what cost the hospital would incur if they admitted this patient.

(20:07):

And they essentially use that to predict who should be prioritized for healthcare. Now unsurprisingly, it turned out that white patients often pay a lot more or are able to pay a lot more when it comes to hospital visits. Maybe it's because of better insurance or better conditions at work that allow them to take leave and so on. But whatever the mechanism is, what ended up happening with this algorithm was I think black patients with the same level of healthcare prognosis were half as likely or about much less likely compared to white ones of getting enrolled in this high risk program. So they were much less likely to get this proactive care. And this was a fantastic study by Obermeyer, et al. It was published in Science in 2019. Now, what I think is the most disappointing part of this is that Optum did not stop using this algorithm after this study was released. And that was because in some sense the algorithm was working precisely as expected. It was an algorithm that was meant to lower healthcare costs. It wasn't an algorithm that was meant to provide better care for patients who need it most. And so, even after this study was rolled out, I think Optum continued using this algorithm as is. And I think as far as I know, even today this is or some version of this algorithm is still in use across the network of hospitals that Optum serves.

Eric Topol (21:31):

No, it's horrible the fact that it was exposed by Ziad Obermeyer’s paper in Science and that nothing has been done to change it, it's extraordinary. I mean, it's just hard to imagine. Now you do summarize the five reasons predictive AI fails in a nice table, we'll put that up on the post as well. And I think you've kind of reviewed that as these case examples. So now I get to challenge you about predictive AI because I don't know that such a fine line between that and generative AI are large language models. So as you know, the group at DeepMind and now others have done weather forecasting with multimodal large language models and have come up with some of the most accurate weather forecasting we've ever seen. And I've written a piece in Science about medical forecasting. Again, taking all the layers of a person's data and trying to predict if they're high risk for a particular condition, including not just their electronic record, but their genomics, proteomics, their scans and labs and on and on and on exposures, environmental.

Multimodal A.I. in Medicine

(22:44):

So I want to get your sense about that because this is now a coalescence of where you took down predictive AI for good reasons, and then now these much more sophisticated models that are integrating not just large data sets, but truly multimodal. Now, some people think multimodal means only text, audio, speech and video images, but here we're talking about multimodal layers of data as for the weather forecasting model or earthquake prediction or other things. So let's get your views on that because they weren't really presented in the book. I think they're a positive step, but I want to see what you think.

Sayash Kapoor (23:37):

No, absolutely. I think maybe the two questions are sort of slightly separate in my view. So for things like weather forecasting, I think weather forecasting is a problem that's extremely tenable for generative AI or for making predictions about the future. And I think one of the key differences there is that we don't have the problem of feedback loops with humans. We are not making predictions about individual human beings. We are rather making predictions about what happens with geological outcomes. We have good differential equations that we've used to predict them in the past, and those are already pretty good. But I do think deep learning has taken us one step further. So in that sense, I think that's an extremely good example of what doesn't really fit within the context of the chapter because we are thinking about decisions thinking about individual human beings. And you rightly point out that that's not really covered within the chapter.

(24:36):

For the second part about incorporating multimodal data, genomics data, everything about an individual, I think that approach is promising. What I will say though is that so far we haven't seen it used for making individual decisions and especially consequential decisions about human beings because oftentimes what ends up happening is we can make very good predictions. That's not in question at all. But even with these good predictions about what will happen to a person, sometimes intervening on the decision is hard because oftentimes we treat prediction as a problem of correlations, but making decisions is a problem of causal estimation. And that's where those two sort of approaches disentangle a little bit. So one of my examples, favorite examples of this is this model that was used to predict who should be released before screening when someone comes in with symptoms of pneumonia. So let's say a patient comes in with symptoms of pneumonia, should you release them on the day of?

(25:39):

Should you keep them in the hospital or should you transfer them to the ICU? And these ML researchers were basically trying to solve this problem. They found out that the neural network model they developed, this was two decades ago, by the way. The neural network model they developed was extremely accurate at predicting who would basically have a high risk of having complications once they get pneumonia. But it turned out that the model was saying essentially that anyone who comes in who has asthma and who comes in with symptoms of pneumonia is the lowest risk patient. Now, why was this? This was because when in the past training data, when some such patients would come into the hospital, these patients would be transferred directly to the ICU because the healthcare professionals realized that could be a serious condition. And so, it turned out that actually patients who had asthma who came in with symptoms of pneumonia were actually the lowest risk amongst the population because they were taken such good care of.

(26:38):

But now if you use this prediction that a patient comes in with symptoms of pneumonia and they have asthma, and so they're low risk, if you use this to make a decision to send them back home, that could be catastrophic. And I think that's the danger with using predictive models to make decisions about people. Now, again, I think the scope and consequences of decisions also vary. So you could think of using this to surface interesting patterns in the data, especially at a slightly larger statistical level to see how certain subpopulations behave or how certain groups of people are likely to develop symptoms or whatever. But I think when as soon as it comes to making decisions about people, the paradigm of problem solving changes because as long as we are using correlational models, I think it's very hard to say what will happen if we change the conditions, what will happen if the decision making mechanism is very different from one where the data was collected.

Eric Topol (27:37):

Right. No, I mean where we agree on this is that at the individual level, using multimodal AI with all these layers of data that have now recently become available or should be available, that has to be compared ideally in a randomized trial with standard of care today, which doesn't use any of that. And to see whether or not that decision's made, does it change the natural history and is it an advantage, that's yet to be done. And I agree, it's a very promising pathway for the future. Now, I think you have done what is a very comprehensive sweep on the predictive AI failures. You've mentioned here in our discussion, your enthusiasm and in the book about generative AI positive features and hope and excitement perhaps even. You didn't really yet, we haven't discussed much on the content moderation AI that you have discreetly categorized. Maybe you could just give us the skinny on your sense of that.

Content Moderation AI

Sayash Kapoor (28:46):

Absolutely. So content moderation AI is AI that's used to sort of clean up social media feeds. Social media platforms have a number of policies about what's allowed and not allowed on the platforms. Simple things such as spam are obviously not allowed because let's say people start spamming the platform, it becomes useless for everyone. But then there are other things like hate speech or nudity or pornography and things like that, which are also disallowed on most if not all social media platforms today. And I think a lot of the ways in which these policies are enforced today is using AI. So you might have an AI model that runs every single time you upload a photo to Facebook, for instance. And not just one perhaps hundreds of such models to detect if it has nudity or hate speech or any of these other things that might violate the platform's terms of service.

(29:40):

So content moderation AI is AI that's used to make these decisions. And very often in the last few years we've seen that when something gets taken down, for instance, Facebook deletes a post, people often blame the AI for having a poor understanding. Let's say of satire or not understanding what's in the image to basically say that their post was taken down because of bad AI. Now, there have been many claims that content moderation AI will solve social media's problems. In particular, we've heard claims from Mark Zuckerberg who in a senate testimony I think back in 2018, said that AI is going to solve most if not all of their content moderation problems. So our take on content moderation AI is basically this. AI is very, very useful for solving the simple parts of content moderation. What is a simple part? So basically the simple parts of content moderation are, let's say you have a large training data of the same type of policy violation on a platform like Facebook.

(30:44):

If you have large data sets, and if these data sets have a clear line in the sand, for instance, with nudity or pornography, it's very easy to create classifiers that will automate this. On the other hand, the hard part of content moderation is not actually just creating these AI models. The hard part is drawing the line. So when it comes to what is allowed and not allowed on platforms, these platforms are essentially making decisions about speech. And that is a topic that's extremely fraught. It's fraught in the US, it's also fraught globally. And essentially these platforms are trying to solve this really hard problem at scale. So they're trying to come up with rules that apply to every single user of the platform, like over 3 billion users in the case of Facebook. And this inevitably has these trade-offs about what speech is allowed versus disallowed that are hard to say one way or the other.

(31:42):

They're not black and white. And what we think is that AI has no place in this hard part of content moderation, which is essentially human. It's essentially about adjudicating between competing interests. And so, when people claim that AI will solve these many problems of content moderation, I think what they're often missing is that there's this extremely large number of things you need to do to get content moderation right. AI solves one of these dozen or so things, which is detecting and taking down content automatically, but all of the rest of it involves essentially human decisions. And so, this is sort of the brief gist of it. There are also other problems. For example, AI doesn't really work so well for low resource languages. It doesn't really work so well when it comes to nuances and so on that we discussed in the book. But we think some of these challenges are solvable in the medium to long term. But these questions around competing interests of power, I think are beyond the domain of AI even in the medium to long term.

Age 28! and Career Advice

Eric Topol (32:50):

No, I think you nailed that. I think this is an area that you've really aptly characterized and shown the shortcomings of AI and how the human factor is so critically important. So what's extraordinary here is you're just 28 and you are rocking it here with publications all over the place on reproducibility, transparency, evaluating generative AI, AI safety. You have a website on AI snake oil that you're collecting more things, writing more things, and of course you have the experience of having worked in the IT world with Facebook and also I guess also Columbia. So you're kind of off to the races here as one of the really young leaders in the field. And I am struck by that, and maybe you could comment about the inspiration you might provide to other young people. You're the youngest person I've interviewed for Ground Truths, by the way, by a pretty substantial margin, I would say. And this is a field where it attracts so many young people. So maybe you could just talk a bit about your career path and your advice for people. They may be the kids of some of our listeners, but they also may be some of the people listening as well.

Sayash Kapoor (34:16):

Absolutely. First, thank you so much for the kind words. I think a lot of this work is with collaborators without whom of course, I would never be able to do this. I think Arvind is a great co-author and supporter. I think in terms of my career parts, it was sort of like a zigzag, I would say. It wasn't clear to me when I was an undergrad if I wanted to do grad school or go into the industry, and I sort of on a whim went to work at Facebook, and it was because I'd been working on machine learning for a little bit of time, and I just thought, it's worth seeing what the other side has to offer beyond academia. And I think that experience was very, very helpful. One of the things, I talked to a lot of undergrads here at Princeton, and one of the things I've seen people be very concerned about is, what is the grad school they're going to get into right after undergrad?

(35:04):

And I think it's not really a question you need to answer now. I mean, in some cases I would say it's even very helpful to have a few years of industry experience before getting into grad school. That has definitely, at least that has been my experience. Beyond that, I think working in a field like AI, I think it's very easy to be caught up with all of the new things that are happening each day. So I'm not sure if you know, but AI has I think over 500-1,000 new archive papers every single day. And with this rush, I think there's this expectation that you might put on yourself on being successful requires a certain number of publications or a certain threshold of things. And I think more often than not, that is counterproductive. So it has been very helpful for me, for example, to have collaborators who are thinking long term, so this book, for instance, is not something that would be very valued within the CS community, I would say. I think the CS community values peer-reviewed papers a lot more than they do books, and yet we chose to write it because I think the staying power of a book or the longevity of a book is much more than any single paper could do. So the other concrete thing I found very helpful is optimizing for a different metric compared to what the rest of the community seems to be doing, especially when it comes to fast moving fields like AI.

Eric Topol (36:29):

Well, that last piece of advice is important because I think too often people, whether it's computer scientists, life scientists, whoever, they don't realize that their audience is much broader. And that reaching the public with things like a book or op-eds or essays, varied ways that are intended for public consumption, not for, in this case, computer scientists. So that's why I think the book is a nice contribution. I don't like the title because it's so skewed. And also the content is really trying to hammer it at home. I hope you write a sequel book on the positive sides of AI. I did want to ask you, when I read the book, I thought I heard your voice. I thought you had written the book, and Arvind maybe did some editing. You wrote about Arvind this and Arvind that. Did you write the first draft of the book and then he kind of came along?

Sayash Kapoor (37:28):

No, absolutely not. So the way we wrote the book was we basically started writing it in parallel, and I wrote the first draft of half the chapters and he wrote the first draft of the other half, and that was essentially all the way through. So we would sort of write a draft, pass it to the other person, and then keep doing this until we sent it to our publishers.

Eric Topol (37:51):

Okay. So I guess I was thinking of the chapters you wrote where it came through. I'm glad that it was a shared piece of work because that's good, because that’s what co-authoring is all about, right? Well, Sayash, it's really been a joy to meet you and congratulations on this book. I obviously have expressed my objections and my disagreements, but that's okay because this book will feed the skeptics of AI. They'll love this. And I hope that the positive side, which I think is under expressed, will not be lost and that you'll continue to work on this and be a conscience. You may know I've interviewed a few other people in the AI space that are similarly like you, trying to assure its safety, its transparency, the ethical issues. And I think we need folks like you. I mean, this is what helps get it on track, keeping it from getting off the rails or what it shouldn't be doing. So keep up the great work and thanks so much for joining.

Sayash Kapoor (39:09):

Thank you so much. It was a real pleasure.

************************************************

Thanks for listening, reading or watching!

The Ground Truths newsletters and podcasts are all free, open-access, without ads.

Please share this post/podcast with your friends and network if you found it informative!

Voluntary paid subscriptions all go to support Scripps Research. Many thanks for that—they greatly helped fund our summer internship programs for 2023 and 2024.

Thanks to my producer Jessica Nguyen and Sinjun Balabanoff for audio and video support at Scripps Research.

Note: you can select preferences to receive emails about newsletters, podcasts, or all I don’t want to bother you with an email for content that you’re not interested in.

Get full access to Ground Truths at erictopol.substack.com/subscribe

Écouter Écoute encore Continuer Écoutez...
S’abonner Se désabonner
Partager

Episodes

The Glaucomfleckens: The Best in Medical Comedy
16 nov.· Ground Truths
Below is a brief video snippet from our conversation. Full videos of all Ground Truths podcasts can be seen on YouTube here. The current one is here. If you like the YouTube format, please subscribe! This one has embedded one of my favorite TikTok’s from Will. There are several links to others in the transcript. The audios are also available on Apple and Spotify.
Transcript with links to both audio and videos, commencement addresses, NEJM article coverage
Eric Topol (00:06):
Hi, it's Eric Topol from Ground Truths, and I've got an amazing couple with me today. It's Will Flanary and Kristin Flanary, the Glaucomfleckens. I've had the chance to get to know them a bit through Knock Knock, Hi! which is their podcast. And of course, everyone knows Dr. Glaucomflecken from his TikTok world and his other about 4 million followers on Instagram and Twitter and all these other social media, and YouTube. So welcome.
Will Flanary (00:43):
Thanks for having us.
Kristin Flanary (00:44):
Thank you. Happy to be here.
By Way of Background
Eric Topol (00:45):
Yeah. Well, this is going to be fun because I'm going to go a quick background so we can go fast forward because we did an interview back in early 2022.
Kristin Flanary (00:56):
Yes.
Eric Topol (00:57):
And what you've been doing since then is rocking it. You're like a meteoric, right. And it was predictable, like rarefied talent and who couldn't love humor, medical humor, but by way of background, just for those who are not up to speed. I guess you got your start, Will, as a class clown when your mother was a teacher in the sixth grade.
Will Flanary (01:22):
Yep, yep. I misbehaved a little bit. It helped that I still made good grades, but I cut up a bit in class.
Eric Topol (01:32):
And then you were already in the comedy club circuits doing standup in Houston as an 18-year-old.
Will Flanary (01:40):
It was all amateur stuff, nothing, just dabble in it and trying to get better. I was always kind of naturally funny just with my friend group and everything. I loved making people laugh, but doing standups is a whole different ball game. And so, I started doing that around Houston as a high school senior and kept that going through college and a little bit into med school.
Kristin Flanary (02:02):
Houston was a good training ground, right? That where Harris Wittels was also coming up.
Will Flanary (02:07):
Yeah. A lot of famous comedians have come through Houston. Even going back to Bill Hicks back in the, was that the 80s, I think? Or 90s?
Eric Topol (02:17):
Well, and then of course, it was I think in 2020 when you launched Dr. Glaucomflecken, I think. Is that right?
Will Flanary (02:28):
That's when it really started to take off. I was on Twitter telling jokes back in 2016.
Kristin Flanary (02:39):
GomerBlog before that, that's actually where it was born.
Will Flanary (02:41):
I was doing satire writing. I basically do what I'm doing now, but in article form, trying to be The Onion of medicine. And then the pandemic hit, started doing video content and that's really with lockdown. That's when, because everybody was on social media, nobody had anything else to do. So it was right place, right time for me and branching out into video content.
On to Medical School Commencement Addresses
Eric Topol (03:11):
Alright, so that's the background of some incredible foundation for humor. But since we last got together, I'll link the Medicine and the Machine interview we did back then. What has been happening with you two is nothing short of incredible. I saw your graduation speeches, Will. Yale in 2022, I watched the UCSF in 2023 and then the University of Michigan in 2024. Maybe there's other ones I don't even know.
Kristin Flanary (03:45):
There’s a few others.
Will Flanary (03:45):
There's a few. But I feel like you've done, I'm sure your fair share of commencement addresses as well. It's kind of hard to come up with different ways to be inspirational to the next generation. So fortunately, we have together, we have some life experiences and learned a thing or two by doing all of this social media stuff and just the things we've been through that I guess I have enough things to say to entertain an interest.
Eric Topol (04:18):
Well, you're being humble as usual, but having watched those commencement addresses, they were the best medical commencement addresses I've ever seen. And even though you might have told us some of the same jokes, they were so great that it was all right. Yeah, and you know what is great about it is you've got these, not the students, they all love you of course, because they're probably addicted to when's your next video going to get posted.
(04:44):
But even the old professors, all the family members, it's great. But one of the things I wanted to get at. Well, I'll start with the graduation speeches, because you were such an inspiration, not just with humor, but your message. And this gets back to you as a couple and the tragedies you've been through. So you really, I think, got into this co-survivor story and maybe Kristin, since you are the co-survivor of two bouts of Will’s testicular cancer, and then the sudden cardiac death. I mean, people don't talk about this much, so maybe you could help enlighten us.
Tragedies and Being a Co-Survivor
Kristin Flanary (05:26):
Yeah, it’s funny because the experience of being a co-survivor is nothing new. It’s as long as we've had human beings, we've had co-survivors. But the concept around it and giving it a name and a label, a framework to be able to think about it, that is what I think is new and what people haven't talked about before. So co-survivor is just this idea that when a medical trauma happens to a patient, the patient has their experience and if they survive it, they are a survivor and they have a survivor experience. And also, most people are closely attached to at least one other person, if not many. And those people are co-surviving the medical event along with the survivor. That event is happening in their lives as was happening to them too. If someone comes in with a patient to the hospital, that person, you can just assume by default that their lives are pretty intimately or profoundly intertwined or else why would that person be there? And so, thinking of it as there's the patient and then there's also a co-patient, that family members in the past have only been thought of as caregivers if they've been thought of at all. And that is certainly one aspect of the role, but it's important to remember that whatever it is that's happening to the patient is also affecting the family members' lives in a really deep and profound way.
Eric Topol (07:04):
That's really helpful. Now, the fact that you recognize that in your graduation speech, Will, I think is somewhat unique. And of course, some of the other things that you touched on like playing to your creativity and the human factors, I mean, these are so important messages.
Will Flanary (07:23):
Well, in the discussion about co-survivorship and because I talk about that whenever I do my keynotes and when I do the commencement addresses, but all credit goes to Kristin for really being the driving force of this idea for me and for many others because as a physician, we take care of patients. Our focus is always on the patient. And it really wasn't until this happened to me and my family and Kristin in particular that I started to understand exactly what she's talking about and this idea. And so, Kristin gets a lot of credit for just really bringing that term and that idea to the forefront.
Eric Topol (08:09):
Yeah, well, you saved his life. It's just not many have that bond. And then the other thing I just want to mention now, you've been recognized by the American Heart Association and a whole bunch of other organizations awarded because of your advocacy for CPR. And you even mentioned that I think in one of your commencement addresses.
Will Flanary (08:31):
Yeah, I tried to get the crowd to do CPR. Like team up, partner up, and it kind of fell flat. It wasn't quite the right time, I think, to try to do a mass class on CPR. So maybe next time.
Eric Topol (08:47):
Right. Well, so you had this foundation with the Glaucomflecken General Hospital and taking on 37 specialties and all these incredible people that became part of the family, if you will, of spoof on medicine and your alter ego and these videos that you would do. And sometimes you have three or four different alter egos in there playing out, but now you've branched into new things. So one which is an outgrowth of what we were just talking about. You've been on this country tour, Wife & Death.
“Wife and Death,” A Nationwide Tour
Kristin Flanary (09:28):
Yes.
Eric Topol (09:29):
Wife and death. I mean, yeah, I guess we can make the connect of how you named it that, but what is it you've been selling out in cities all over the country, and by the way, I'm really upset you haven't come to San Diego, but tell us about wife and death.
Will Flanary (09:44):
Yeah. Well, we have this amazing story and all these medical challenges we've been through, and then developing the Glaucomflecken brand and universe, and we've done keynotes together for years, and then we thought, let's have more fun with it. Let's do keynotes. They're great. We can get our message out, but sometimes they're just a bit stuffy. It's an academic environment.
Kristin Flanary (10:15):
They're usually at seven in the morning also, so that's the downside.
Will Flanary (10:21):
So we thought, let's just put together our own live show. Let's put together something that we could just creatively, we can do whatever we want with it. I could dress up as characters, Kristin, who has these beautiful writing and monologues that she's put together around her experience and just to create something that people can come into a theater and just experience this wide range of emotions from just laughter to tears of all kinds, and just have them feel the story and enjoy this story. Fortunately, it has a happy ending because I'm still alive and it's been so much fun. The audiences have been incredible. Mostly healthcare, but even some non-healthcare people show up, and we've been blown away by the response. Honestly, we should have done bigger theaters. That's our lesson for the first go round.
Eric Topol (11:21):
I saw you had to do a second show in Pittsburgh.
Will Flanary (11:24):
We did.
Kristin Flanary (11:26):
That one sold out too. Something about Pittsburgh, that was a good crowd, and there was a lot of them.
Will Flanary (11:33):
It was almost like in Pittsburgh, they rarely ever get any internet comedian ophthalmologists that come through. I don't know.
Eric Topol (11:41):
Well, I see you got some still to come in Denver and Chicago. This is amazing. And I wondered who was coming and I mean, it's not at all surprising that there'd be this phenomenal popularity. So that's one thing you've done that's new, which is amazing. And of course, it's a multidimensional story. The one that shocked me, I have to tell you, shocked me, was the New England Journal partnership. The New England Journal is the most stodgy, arrogant, I mean so difficult. And not only that.
Kristin Flanary (12:17):
You said that. Not us.
Partnering with the New England Journal of Medicine!
Eric Topol (12:19):
Yeah, yeah. They'll get this too. They know we don't get along that well, but that's okay. You even made fun of journals. And now you're partnering with the New England Journal, God's greatest medical journal, or whatever. Tell us about that.
Will Flanary (12:39):
Well, so one thing that I really enjoy doing, and I've done it with my US healthcare system content is almost like tricking people into learning things. And so, if you make something funny, then people will actually sit there and listen to what you have to say about deductibles and physician-owned hospitals and all these inner workings. DIR fees and pharmacy, all these things that are really dry topics. But if you can make them funny, all of a sudden people will actually learn and listen to it. And the New England Journal of Medicine, they approached me with an idea. Basically just to take one or two of their trials per month. And I just make a skit out of that trial with the idea being to help disseminate some of the research findings that are out there, because I guess it's getting harder and harder for people to actually read, to sit down and read a journal article.
(13:43):
And so, I have to credit them for having this idea and thinking outside the box of a different way to get medical information and knowledge out to the masses. And you're absolutely right, that I have been critical of journals, and particularly I've been critical of the predatory nature of some of the larger journals out there, like Elsevier. I've specifically named Elsevier, Springer, these journals that have a 40% profit margin. And I certainly thought about that whenever I was looking into this partnership. And the reason I was okay with doing it with the New England Journal is because they're a nonprofit, first of all, so they're run by the Massachusetts Medical Society. That's the publisher for that journal. And so, I feel okay partnering with them because I feel like they're doing it in a much better way than some of the bigger journal corporations out there.
Kristin Flanary (14:54):
Well, and also part of the deal that we negotiated was that those articles that you make skits about those will be available open access.
Will Flanary (15:03):
Oh yeah. That was a prerequisite. Yes. It was like, if I'm going to do this, the articles that I'm talking about need to be free and readily available. That's part of it.
Eric Topol (15:14):
I think you've done about five already, something like that. And I watched them, and I just was blown away. I mean, the one that got me where I was just rolling on the floor, this one, the Belantamab Mafodotin for Multiple Myeloma. And when you were going on about the Bortezomib, Dexamethasone. We'll link to this. I said, oh my God.
Will Flanary (15:40):
Yeah. The joke there is, you don't have any idea how long it took me to say those things that quickly. And so, I was writing this skit and I'm like, wouldn't it be funny if somehow that triggered a code stroke in the hospital because this person is saying all these random words that don't have any meaning to anybody. Man, I tell you, I am learning. Why would I ever need to know any of this information as an ophthalmologist? So it's great. I know all this random stuff about multiple myeloma that I probably would never have learned otherwise.
Kristin Flanary (16:21):
It's the only way, you won't read a journal either.
Eric Topol (16:23):
Well, and if you read the comments on the post. These doctors saying, this is the only way they want to get journal information from now on.
Will Flanary (16:33):
Which is double-edged sword, maybe a little bit. Obviously, in a 90 second skit, there's no way I'm going to cover the ins and outs of a major trial. So it's really, in a lot of ways, it's basically like, I call it a comedy abstract. I’m not going much further than an abstract, but hopefully people that are actually interested in the topic can have their interest piqued and want to read more about it. That's kind of the idea.
Eric Topol (17:06):
Yeah. Well, they're phenomenal. We'll link to them. People will enjoy them. I know, because I sure did. And tenecteplase for stroke and all that you've done. Oh, they're just phenomenal.
Will Flanary (17:20):
Every two weeks we come out with a new one.
Eric Topol (17:24):
And that is basically between the fact that you are now on the commencement circuit of the top medical schools and doing New England Journal videos on their articles. You've crossed a line from just making fun of insurance companies and doctors of specialties.
Kristin Flanary (17:44):
Oh, he has crossed many lines, Dr. Topol.
Eric Topol (17:46):
Yeah. Oh yeah. Now you've done it, really. Back two years ago when we convened, actually it's almost three, but you said, when's it going to be your Netflix special?
Will Flanary (18:02):
Oh, gosh.
Eric Topol (18:02):
Is that in the works now?
Will Flanary (18:04):
Well, I'll tell you what's in the works now.
Kristin Flanary (18:06):
Do you know anyone at Netflix?
Will Flanary (18:09):
A New Animated Series
No. We're working on an animated series.
Eric Topol (18:12):
Oh, wow. Wow.
Will Flanary (18:13):
Yeah. All these characters. It's basically just this fictional hospital and all these characters are very cartoonish, the emergency physician that wears the bike helmet and everything. So it's like, well, what do we have together? What do we, Kristin and I have time for? And it wasn't like moving to LA and trying to make a live action with actors and do all, which is something we probably could have tried to do. So instead, we were like, let's just do an animated series.
Kristin Flanary (18:48):
Let's have someone else do the work and draw us.
Will Flanary (18:51):
So we've worked with a writer for the first time, which was a fun process, and putting together a few scripts and then also an animator. We learned a lot about that process. Kristin and I are doing the voiceovers. And yeah, it's in process.
Kristin Flanary (19:10):
We're the only actors we could afford.
Will Flanary (19:12):
Right, exactly.
Eric Topol (19:13):
I can't wait to see it. Now when will it get out there?
Will Flanary (19:17):
Well, we're hoping to be able to put it out on our YouTube channel sometime early next year. So January, February, somewhere around there. And then we can't fund the whole thing ourselves. So the idea is that we do this, we do this pilot episode, and then we'll see what kind of interest we can generate.
Eric Topol (19:37):
Well, there will be interest. I am absolutely assured of that. Wow.
Will Flanary (19:42):
Let us know if you know anybody at the Cartoon Network.
Kristin and Will Flanary (19:45):
Yeah, we're open to possibilities. Whatever, Discovery channel. I don't know.
Eric Topol (19:51):
You've gotten to a point now where you're ready for bigger things even because you're the funniest physician couple in medicine today.
Kristin Flanary (20:05):
Well, that's a very low bar, but thank you.
Will Flanary (20:08):
There are some funny ones out there, but yeah, I appreciate that.
Eric Topol (20:11):
Well, I'm a really big comedy fan. Every night I watch the night before, since I'm old now, but of Colbert and Jimmy Kimmel, just to hear the monologues. Trevor Noah, too. And I can appreciate humor. I'll go to see Sebastian Maniscalco or Jim Gaffigan. That's one of the things I was going to ask you about, because when you do these videos, you don't have an audience.
Will Flanary (20:39):
Oh yeah.
Eric Topol (20:40):
You're making it as opposed to when you are doing your live shows, commencement addresses and things like that. What's the difference when you're trying to be humorous, and you have no audience there?
Will Flanary (20:55):
Well, whenever I'm filming a skit, it's just all production. In fact, I feel like it's funny. I think it's funny, but it's really not until I see the response to it, or I show Kristin, or what I have is where I really know if it's going to work. It's great to put the content out there and see the responses, but there's nothing like live interaction. And that's why I keep coming back to performing. And Kristin's been a performer too in her life. And I think we both really enjoy just the personal interaction, the close interaction, the response from people to our story.
Kristin Flanary (21:36):
We do most of our work alone in this room. I do a lot of writing. He does a lot of playing.
Will Flanary (21:44):
Dress up.
Kristin Flanary (21:44):
All the people in his head, and we do that very isolated. And so, it's very lovely to be able to actually put names to faces or just see human bodies instead of just comments on YouTube.
Will Flanary (21:59):
Meet people.
Kristin Flanary (21:59):
It's really nice.
Will Flanary (22:01):
We’ve been doing meet and greets at the live shows and seeing people come up wearing their costumes.
Eric Topol (22:07):
Oh, wow.
Will Flanary (22:11):
Some of them talk about how they tell us their own stories about their own healthcare and talk about how the videos help them get through certain parts of the pandemic or a difficult time in their life. And so, it reinforces that this means something to a lot of people.
Kristin Flanary (22:29):
It's been really fun for me, and probably you too, but to get to see the joy that he has brought so many people. That's really fun to see in person especially.
Eric Topol (22:42):
No question. Now, when you're producing it together, do you ever just start breaking into laughter because it's you know how funny this is? Or is it just you're on kind of a mission to get it done?
Will Flanary (22:54):
Well, the skits I do by myself. And sometimes when I'm writing out the skit, when I'm writing the skit itself, I will laugh at myself sometimes. Not often, but sometimes they're like, oh, I know that's really funny. I just wrote a skit that I'm actually going to be debuting. I'm speaking at the American Academy of PM&R, so the big PM&R conference. I'm writing a skit, it's How to Ace your PM&R residency interview.
Will Flanary (23:28):
I was writing up that skit today and kind of chuckling to myself. So sometimes that happens, but whenever we do our podcast together, we definitely have outtakes.
Kristin Flanary (23:38):
Oh yeah, we've got some.
Will Flanary (23:40):
We crack each other up.
Kristin Flanary (23:41):
We do.
Will Flanary (23:42):
Sometimes we're getting a little punchy toward the end of the day.
Eric Topol (23:47):
And how is the Knock Knock, Hi! podcast going?
Will Flanary (23:51):
It's awesome. Yeah.
Kristin Flanary (23:52):
Yeah. It's a really fun project.
Will Flanary (23:54):
We still enjoy. You can work with your spouse and in close proximity and still be happily married. So it's doable everyone.
Kristin Flanary (24:06):
That's right. And we're in that phase of life that's really busy. We've got kids, we've got a gazillion jobs. House, my parents are around, and so it's like the only time all week that we actually get to sit down and talk to each other. So it's actually kind of like a part of our marriage at this point.
Will Flanary (24:28):
We’re happy to involve the public in our conversations, but we couldn't do it because we have all these things going on, all our hands and all these little places. We can't do it without a team.
Kristin Flanary (24:41):
Yeah, absolutely.
Will Flanary (24:41):
And that's the thing that I've learned, because I've always been a very loner type content creator. I just wanted to do it all myself. It's in my head and I have trouble telling others, describing what's in my head. And Kristin and our producers have helped me to be able to give a little bit of control to others who are really good at what they do. And that's really the only way that we've been able to venture out into all these different things we've talked about.
Eric Topol (25:12):
Well, I think it comes down to, besides your ability to get to people in terms of their laughter receptors, you have this incisive observer capability. And that's one of the things I don't, I can't fathom because when you can understand the nuances of each specialty or of each part of healthcare, and you haven't necessarily interacted with these specialists or at least in recent years, but you nail it every time. I don't know how you do it, really that observational, is that a central quality of a comedian, you think?
Will Flanary (25:52):
There's definitely a big part of that. You got to get the content from somewhere. But for the specialties, it's really first about just getting the personalities down. And that doesn't change over time.
Kristin Flanary (26:08):
Or around the world.
Will Flanary (26:09):
Or around the world. We hear from people from all over the world about, oh, it's the same in Guatemala as it is in the US.
Kristin Flanary (26:18):
Surgeons are the same.
Will Flanary (26:19):
Yeah.
Kristin Flanary (26:20):
Emergency is the same.
Will Flanary (26:21):
Which has been really cool to see. But so, I draw on my experience interacting with all these specialties back in my med school and intern days. You're right, as an ophthalmologist, we don't get out very much.
Eric Topol (26:33):
No.
Will Flanary (26:35):
So very few people have ever seen an ophthalmologist. We do exist. But then beyond that, I do have to include some actual medical things. And so, I actually, I do a lot of research. I find myself learning more about other fields sometimes than I do in my own field. So especially the further out I get from med school, I know less and less.
Eric Topol (27:00):
Yeah, that's what I was thinking. But you're always spot on. It's interesting to get that global perspective from both of you. Now you're still doing surgery and practicing ophthalmology. Have you reduced it because this has just been taking off so much more over the recent years or keeping it the same?
Will and Kristin Flanary (27:21):
Nope, I’m still. Do you know how many years I had to come along on all of this medical training? He is not allowed to give this up.
Will Flanary (27:29):
I know there's something called a sunk cost fallacy, but this is no fallacy. There's enough of a sunk cost. I got to stick with it. No, I still enjoy it. That's the thing. It actually, it informs my comedy, it grounds me. All of the social media stuff is built upon this medical foundation that I have. And if I stopped practicing, I guess I could maybe cut back. But I'm not planning on doing that. If I stop practicing medicine, I feel like it would make my content less.
Kristin Flanary (28:07):
Authentic.
Will Flanary (28:08):
Less authentic, yeah. That's a good way to put it.
Eric Topol (28:09):
Yeah, no, that makes a lot of sense. That's great you can get that balance with all the things you're doing.
Will Flanary (28:17):
And if I stop practicing medicine, they're not going to invite me to any more commencement addresses, Dr. Topol. So I got to draw the line somewhere.
Eric Topol (28:28):
One of the statements you made at some point earlier was, it was easier to go to become a doctor than to try to be a comedian. And yeah, I mean you proven that.
Will Flanary (28:38):
A lot of ways. That's true.
Eric Topol (28:40):
Wow. I am pretty awestruck about the rarefied talent that you bring and what you both have done for medicine today. And the thing is, you're so young, you have so much time ahead to have an impact.
Will Flanary (28:57):
You hear that Kristin, we're young. Look at that.
Kristin Flanary (29:00):
That's getting less and less true.
Will Flanary (29:01):
Kristin, she just turned 40. It's right around the corner for me. So I don't know.
Will Flanary (29:11):
We got some years left.
Eric Topol (29:12):
You're like young puppies. Are you kidding? You're just getting started. But no, I think that what you brought to medicine in terms of comedy, there's no other entity, no person or people like you have done. And just the last thing I want to ask you about is, you have a platform for advocacy. You've been doing that. We talked about co-survivor. We talked about nurturing the human qualities in physicians like creativity and also taking on the insurance companies, which are just monstrous. I'll link a couple of those, but the brain MRI one or the Texaco.
Will Flanary (29:54):
Texaco Mike.
Eric Topol (29:55):
Yeah, that one is amazing. But there is so many. I mean, you've just taken them apart and they deserve every bit of it. Do you have any other targets for advocacy or does that just kind of come up as things go?
Will Flanary (30:08):
It kind of comes up as things go. There's things I keep harping on. The prior authorization reform, which I've helped in a couple of different states. There's a lot of good people around the country doing really good work on prior authorization and reforming that whole process. And I've been able to just play a small part in that in a couple of different ways. And it's been really fun to do that. And so, I do plan on continuing that crusade as it were. There's certain things I'd like to see. I've been learning more about what pharmacists are dealing with as well as a physician. Unfortunately, we are very separate in a lot of ways and just how we come up in medicine. And so, I have had my eyes opened a lot to what community pharmacists are dealing with. For all the terrible things that we have to deal with as physicians in the healthcare system. Pharmacists have just as much, if not more of the things that they're doing that are threatening their livelihoods. And so, I had love to see some more reform on the PBM side of things, pharmacy benefit managers, Caremark, Optum, all of them. They're causing lots of problems.
Eric Topol (31:24):
I couldn't agree with you more. In fact, I'm going to have Mark Cuban on in a few weeks and we're going to get into that. But the pharmacists get abused by these chains.
Will Flanary (31:33):
Oh, it's bad. It's really bad.
Eric Topol (31:35):
Horrible, horrible. I feel, and every time I am in a drugstore working with one of them, I just think what a tough life they have to deal with.
Will Flanary (31:45):
I guess from an advocacy standpoint, the good news is that there's never a shortage of terrible injustices that are being foisted upon the public and physicians and healthcare workers.
Kristin Flanary (31:59):
Yes. The US healthcare system is ripe for advocacy.
Will Flanary (32:01):
Yes. And that's a lesson that I tell people too, and especially the med students coming up, is like, there's work to be done and get in touch with your state societies and there's always work to be done.
Eric Topol (32:18):
Now you've stayed clear of politics. Totally clear, right?
Will Flanary (32:24):
For the most part, yeah. Yeah. It depends on what you consider politics. It depends on what you consider politics.
Eric Topol (32:32):
It being election day, you haven't made any endorsements.
Will Flanary (32:36):
I haven't. And I don't know. I can only handle so much. I've got my things that I really care about. Of course I'm voting, but I want to talk on the things that I feel like I have the expertise to talk about. And I think there's nothing wrong with that. Everybody can't have an opinion on everything, and it means something. So I am happy to discuss the things that I have expertise about, and I'm always on the side of the patient and wanting to make life better for our patients. And that's the side I'm on.
Kristin Flanary (33:25):
I think also he never comes out and explicitly touches on certain topics, but it's not hard to tell where he falls.
Will Flanary (33:34):
If you really want read into it all.
Kristin Flanary (33:38):
It's not like it's a big secret.
Eric Topol (33:40):
I thought that too. I'm glad you mentioned it, Kristin. But it doesn't come out wide open. But yeah, it's inferred for sure.
Eric Topol (33:49):
I think the point being there is that because you have a reach, I think there's no reach that it has 4 million plus people by your posts and no less the tours and keynotes and everything else. So you could go anywhere but sticking to where you're well grounded, it makes a lot of sense. And anyway, I am going to be staying tuned. This is our two-year checkup. I'm hoping you're going to come to San Diego on your next tour.
Kristin Flanary (34:21):
We're working on 2025 plans.
Will Flanary (34:23):
Oh, we got more shows coming up. And we'll hit up other parts of the country too.
Eric Topol (34:28):
I feel like I got to meet you in person, give you a hug or something. I just feel like I'm missing out there. But it's just a joy to have had a chance to work with you on your podcast. And thanks for coming back on one of mine. There's lots of podcasts out there, but having you and joining you is such fun. So thank you.
Will Flanary (34:54):
This has been great. Thank you for having us.
Kristin Flanary (34:55):
Yeah, thank you.
*****************************************
Thank you for reading, listening and and subscribing to Ground Truths.
If you found this fun and informative please share it!
Yes, laughter is the best medicine.
All content on Ground Truths—its newsletters, analyses, and podcasts, are free, open-access.
Paid subscriptions are voluntary. All proceeds from them go to support Scripps Research. Many thanks to those who have contributed—they have greatly helped fund our summer internship programs for the past two years.
Thanks to my producer Jessica Nguyen and to Sinjun Balabanoff for audio and video support at Scripps Research.
Note on Exodus from X/twitter:
Many of you have abandoned the X platform for reasons that I understand. While I intend to continue to post there because of its reach to the biomedical community, I will post anything material here in the Notes section of Ground Truths on a daily basis and cover important topics in the newsletter/analyses.

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
Rachael Bedard: A Geriatric Physician and Champion for Patients Without a Voice
2 nov.· Ground Truths
Above is a brief video snippet from our conversation. Full videos of all Ground Truths podcasts can be seen on YouTube here. The audios are also available on Apple and Spotify.
Transcript with links to audio and external citations
Eric Topol (00:06):
Well, hello it's Eric Topol with Ground Truths, and I'm really delighted to welcome Dr. Rachael Bedard, who is a physician geriatrician in New York City, and is actually much more multidimensional, if you will. She's a writer. We're going to go over some of her recent writings. She's actually quite prolific. She writes in the New Yorker, New York Magazine, New York Times, New York Review of Books. If it has New York in front of it, she's probably writing there. She's a teacher. She works on human rights, civil rights, criminal justice in the prison system. She's just done so much that makes her truly unique. That's why I really wanted a chance to meet her and talk with her today. So welcome, Rachael.
Rachael Bedard (00:52):
Thank you, Dr. Topol. It's an honor to be here.
Eric Topol (00:55):
Well, please call me Eric and it's such a joy to have a chance to get acquainted with you as a person who is into so many different things and doing all of them so well. So maybe we'd start off with, because you're the first geriatrician we've had on this podcast.
Practicing Geriatrics and Internal Medicine
Eric Topol (01:16):
And it’s especially apropos now. I wanted maybe to talk about your practice, how you got into geriatrics, and then we'll talk about the piece you had earlier this summer on aging.
Rachael Bedard (01:32):
Sure. I went into medicine to do social justice work and I was always on a funny interdisciplinary track. I got into the Mount Sinai School of Medicine through what was then called the Humanities and Medicine program, which was an early acceptance program for people who were humanities focused undergrads, but wanted to go into medicine. So I always was doing a mix of politics and activist focused work, humanities and writing, that was always interested in being a doctor. And then I did my residency at the Cambridge Health Alliance, which is a social medicine program in Cambridge, Massachusetts, and my chief residency there.
(02:23):
I loved being an internist, but I especially loved taking care of complex illness and I especially loved taking care of complex illness in situations where the decision making, there was no sort of algorithmic decision-making, where you were doing incredibly sort of complex patient-centered shared decision making around how to come up with treatment plans, what the goals of care were. I liked taking care of patients where the whole family system was sort of part of the care team and part of the patient constellation. I loved running family meetings. I was incredibly lucky when I was senior resident and chief resident. I was very close with Andy Billings, who was one of the founders of palliative care and in the field, but also very much started a program at MGH and he had come to work at Cambridge Hospital in his sort of semi-retirement and we got close and he was a very influential figure for me. So all of those things conspired to make me want to go back to New York to go to the Sinai has an integrated geriatrics and palliative care fellowship where you do both fellowships simultaneously. So I came to do that and just really loved that work and loved that medicine so much. There was a second part to your question.
Eric Topol (03:52):
Is that where you practice geriatrics now?
Rachael Bedard (03:55):
No, now I have ever since finishing fellowship had very unusual practice settings for a geriatrician. So right out of fellowship, I went to work on Rikers Island and then New York City jail system, and I was the first jail based geriatrician in the country, which is a sort of uncomfortable distinction because people don't really like to think about there being a substantial geriatric population in jails. But there is, and I was incredibly lucky when I was finishing fellowship, there was a lot of energy around jail healthcare in New York City and I wrote the guy who was then the CMO and said, do you think you have an aging problem? And he said, I'm not sure, but if you want to come find out, we'll make you a job to come find out. And so, that was an incredible opportunity for someone right out of fellowship.
(04:55):
It meant stepping off the sort of academic track. But I went and worked in jail for six years and took care of older folks and people with serious illness in jail and then left Rikers in 2022. And now I work in a safety net clinic in Brooklyn that takes care of homeless people or people who have serious sort of housing instability. And that is attached to Woodhull Hospital, which is one of the public hospitals in New York City. And there I do a mix of regular internal medicine primary care, but I preferentially see the older folks who come through, which is a really interesting, painful, complicated patient population because I see a fair amount of cognitive impairment in folks who are living in the shelter system. And that's a really hard problem to address.
Frailty, The Aged, and Longevity
Eric Topol (05:54):
Well, there's a theme across your medical efforts. It seems to me that you look after the neglected folks, the prisoners, the old folks, the homeless people. I mean that's kind of you. It's pretty impressive. And there's not enough of people like you in the medical field. Now, no less do you do that, but of course you are a very impressive author, writer, and of many topics I want to get into with you, these are some recent essays you've written. The one that piqued my interest to start to understand who you were and kind of discover this body of work was the one that you wrote related to aging and President Biden. And that was in New York Times. And I do want to put in a quote because as you know very well, there's so much interest in longevity now.
Eric Topol (06:51):
Interrupting the aging process, and this one really stuck with me from that op-ed, “Time marches forward, bodies decline, and the growing expectation that we might all live in perfect health until our 100th birthdays reflects a culture that overprizes longevity to the point of delusion.” So maybe if you could tell us, that was a rich piece, you got into frailty, you related it to the issues that were surrounding President Biden who at that time had not withdrawn from the race. But what were you thinking and what are your thoughts about the ability to change the aging process?
Rachael Bedard (07:36):
I am very interested in, I mean, I'm incredibly interested in the science of it. And so, I guess I think that there are a few things.
(07:49):
One thing is that the framework that, the part that gives me pause the most is this framework that anything less than perfect health is not a life worth living. So if you're going to have a long life, life should not just be long and sort of healthy in relative terms to your age cohort, but healthy that when you're 80 you should feel like you have the health of a 45-year-old is my understanding of the culture of longevity science. And while I understand why that's aspirational and everybody worry about my body's decline, I think it's a really problematic thing to say that sick bodies are bodies that have disability or people who have cognitive difference are somehow leading lesser lives or lives that are not meaningful or not worth living. I think it's a very, very slippery slope. It puts you in a place where it sort of comes up against another trend or another emerging cultural trend, which is really thinking a lot about physician-assisted suicide and end of life choices.
(09:04):
And that in some ways that conversation can also be very focused on this idea that there's just no way that it's worth living if you're sick. And that's just not true, I think, and that's not been true for many, many, many of my patients, some of whom have lived with enormous disability and incredible burden of illness, people who are chronically seriously ill and are still leading lives that for them and for the people who love them are filled with meaning. So that's my concern about the longevity stuff. I'm interested in the science around the longevity stuff for sure. I'm interested in, I think we're living in this really interesting moment where there's so much happening across so many of the chronic disease fields where the things that I think have been leading to body decay over the last several decades for the majority of the population, we're sort of seeing a lot of breakthroughs in multiple fronts all at once. And that's really exciting. I mean, that's really exciting. And so, certainly if it's possible to make it to 100 in wonderful health, that's what I'd wish for all of us. But to hold it up as the standard that we have to achieve, I think is both unrealistic and a little myopic.
Eric Topol (10:28):
Yeah. Well, I certainly agreed with that and I think that that particular essay resonated so well and you really got into frailty and the idea about how it can be potentially prevented or markedly delayed. And I think before we move on to one of those breakthroughs that you were alluding to, any comments about the inevitability of frailty in people who are older, who at some point start to get the dwindles, if you will, what do you have to say about that?
Rachael Bedard (11:11):
Well, from a clinical standpoint, I guess the caveat versus that not everybody becomes frail and dwindles exactly. Some people are in really strong health up until sort of their final years of life or year of life and then something happens, they dwindle quickly and that's how they die. Or some people die of acute events, but the vast majority of us are going to become more frail in our final decades than we are in our middle decades. And that is the normal sort of pattern of wear and tear on the body. And it is an extraordinary framework, I think frailty because the idea of this sort of syndrome of things where it's really not a disease framework, it is a syndrome framework and it's a framework that says many, many small injuries or stressors add up to create a lot of stress and change in a body and trauma for our body. And once you are sort of past a tipping point of an amount of stress, it's very hard to undo those things because you are not sort of addressing one pathologic process. You're addressing, you're trying to mitigate many processes all at once.
(12:31):
When I wrote that piece, it was inspired by the conversation surrounding President Biden's health. And I was particularly struck by, there was a huge amount of clinical speculation about what was going on with him, right? I'm sure you remember there were people, there was all of this talk about whether he had Parkinson's and what his cognitive status was. And it felt to me like there was an opportunity to do some public education around the idea that you need not have one single sort of smoking gun illness to explain decline. What happens to most of us is that we're going to decline in many small ways sort of simultaneously, and it's going to impact function when it tips over a little bit. And that pattern of decline is not going to be steady day over day worsening. It's going to be up and down. And if you slept better the night before, you might have a better day the next day. And if you slept badly, you might have a worse day. And without knowing anything specific about his clinical situation, it felt like a framework that could explain so much of what we were seeing in public. And it was important also, I think to say that nothing was necessarily being hidden from anybody and that this is the kind of thing that, this has accumulated stress over time that then presents suddenly all at once after having been submerged.
Eric Topol (14:01):
Yeah, you reviewed that so well about the wear and tear and everything related to that. And before I move on to the second topic, I want to just circle back to something you alluded to, which is when Peter Attia wrote about this medicine 3.0 and how you would be compressed and you'd have no comorbidities, you'd have no other illnesses and just fall off the cliff. As a geriatrician, do you think that that is even conceivable?
Rachael Bedard (14:35):
No. Do you think it is?
Eric Topol (14:37):
No, but I just wanted to check the reality. I did challenge on an earlier podcast and he came up with his pat answer. But no, there's no evidence of that, that maybe you can delay if there ever was a way to do that. But I think there's this kind of natural phenomena that you just described, and I'll refer people also to that excellent piece that you get into it more.
Rachael Bedard (15:06):
Peter Attia, I mean, he is certainly the sort of standard bearer in my mind of that movement and that science or that framework of thinking about science. And there's stuff in there that's really valuable. The idea of thinking about lifestyle in your middle decades is having meaningful impact on how you will age, what your final years will look like. That seems intuitively true, I think. And so, thinking about his emphasis on exercise, I mean, his emphasis on exercise is particularly intense and not super achievable for the average person, but the idea that you should sort of be thinking about keeping your body strong because it will decline eventually. And so, you want to do that from a higher peak. That makes a lot of sense to me. The idea that where we sort of draw pathologic disease cutoffs is obviously a little bit arbitrary. And so, wanting to think about optimizing pre-disease states and doing prevention, that's obviously, I think pretty appealing and interesting. It's just really in an evidence free zone.
Ozempic for the Indigent
Eric Topol (16:18):
Yeah, that's what I confronted him with, of course, he had a different perspective, but you summed that up really well. Now let's switch to a piece you had in New York magazine. It was entitled, What If Ozempic Is Just a Good Thing? And the reason, of course, this ties into the first thing we're discussing. There's even talk now, the whole GLP-1 family of drugs with the dual triple receptors, pills to come that we're going to be able to interrupt a path towards Alzheimer's and Parkinson's. Obviously you've already seen impact in heart disease, liver disease, kidney disease way before that, diabetes and obesity. So what are your thoughts? Because you wrote a very interesting, you provided a very interesting perspective when you wrote that one.
Rachael Bedard (17:11):
So that piece I wrote because I have this unbelievably privileged, interesting clinical practice. In New York City, there is public health insurance basically available to anybody here, including folks who are undocumented. And the public hospital system has pharmacies that are outpatient pharmacies that have, and New York Medicaid is very generous and they arranged through some kind of brilliant negotiating. I don't quite know how to make Ozempic to make semaglutide available to people who met criteria which meant diabetes plus obesity, but that we could prescribe it even for our very, very poor patients and that they would be able to get it reliably, that we would have it in stock. And I don't know how many other practices in the country are able to reliably provide GLP-1s to marginalized folks like that. I think it feels like a really rare opportunity and a very distinct perspective.
(18:23):
And it has just been the most amazing thing, I think to have this class of drugs come along that, as you say, addresses so many problems all at once with at least in my prescribing experience, a relatively mild tolerable side effect profile. I have not had patients who have become incredibly sick with it. And for folks where making that kind of impact on their chronic illness is so critical to not just their longevity, but their disease status interacts so much with their social burden. And so, it's a very meaningful intervention I think around poverty actually.
(19:17):
I really feel that almost all of the popular press about it has focused very much on use amongst the wealthy and who's getting it off label and how are they getting it and which celebrities are taking it, and what are the implications for eating and diet culture and for people who have eating disorders. And that's a set of questions that's obviously sort of interesting, but it's really interesting in a very rarefied space. There's an unbelievable diabetes epidemic in this country, and the majority of people who have diabetes are not the people who are getting written about over and over again in those pieces. It's the patients that I take care of, and those people are at risk of ending up on dialysis or getting amputations. And so, having a tool this effective is really miraculous feeling to me.
Eric Topol (20:10):
Well, it really gives me some hope because I don't know any program like that one, which is the people who need it the most. It's getting provided for them. And we have been talking about a drug that costs a thousand dollars a month. It may get down to $500 a month, but that's still a huge cost. And of course, there's not much governmental coverage at this point. There might be some more for Medicare, Medicaid, whatever in the future, but it's really the original criteria of diabetes, and it took almost 20 years to get to where we are right now. So what's so refreshing here is to know that there's at least one program that is helping to bridge the inequities and to not make it as was projected, which was, as you say, for celebrities and wealthy people more exclusively, so that's great. And we still don't know about the diverse breadth of these effects, but as you well know, there's trials in Alzheimer's. I spoke to Steve Horvath recently on the podcast and he talked about how it's reset the epigenetic clock, GLP-1.
Rachael Bedard (21:24):
Does he think so?
Eric Topol (21:26):
Whoa. Yeah, there was evidence that was just presented about that. I said, well, if that does correspond to aging, the thing that we spoke about first, that would be very exciting.
Rachael Bedard (21:37):
It’s so wild. I mean, it's so exciting. It's so exciting to me on so many levels. And one of them is it's just exploding my mental model of disease pathogenesis, and it's making me think, oh my goodness, I have zero idea actually how metabolism and the brain and sort of cardiovascular disease, all of those things are obviously, what is happening in the interplay between all of those different systems. It's really so much more complicated and so much more interdependent than I understood it to be. I am really optimistic about the Alzheimer's trial. I am excited for those results, and I think we're going to keep seeing that it prevents different types of tumors.
Eric Topol (22:33):
Yeah, no, and that's been shown at least certainly in obese people, that there’s cancers that gets way reduced, but we never had a potent anti-inflammatory that works at the brain and systemically like this before anyone loses the weight, you already see evidence.
Long Covid and ME/CFS
(22:50):
It is pretty striking. Now, this goes back to the theme that was introduced earlier about looking after people who are neglected, who aren't respected or generally cared for. And I wanted to now get into Long Covid and the piece you wrote in the New Yorker about listening to patients, called “what would it mean for scientists to listen to patients?” And maybe you can talk about myalgic encephalitis/chronic fatigue (ME/CFS), and of course Long Covid because that's the one that is so pervasive right now as to the fact that these people don't get respect from physicians. They don't want to listen to their ailments. There's no blood tests, so there's no way to objectively make a diagnosis supposedly. And they're basically often dismissed, or their suffering is discounted. Maybe you can tell us again what you wrote about earlier this year and any updated thoughts.
Rachael Bedard (24:01):
Have you had my friend Harlan Krumholz on the show to talk about the LISTEN study?
Eric Topol (24:04):
Not yet. I know Harlan very well. Yes.
Eric Topol (24:11):
I know Akiko Iwasaki very well too. They’re very, very close.
Rachael Bedard (24:14):
So, Akiko Iwasaki and Harlan Krumholz at Yale have been running this research effort called the LISTEN study. And I first learned about it sometime in maybe late 2021. And I had been really interested in the emerging discourse around chronic illness in Long Covid in the 2021. So when we were past the most acute phase of the pandemic, and we were seeing this long tail of sequelae in patients, and the conversation had really shifted to one that was about sort of trying to define this new syndrome, trying to understand it, trying to figure out how you could diagnose it, what were we seeing sort of emerge, how are we going to draw boxes around it? And I was so interested in the way that this syndrome was really patient created. It came out of patients identifying their own symptoms and then banning together much, much faster than any kind of institutional science can ever work, getting into message boards together or whatever, and doing their own survey work and then coming up with their own descriptive techniques about what they were experiencing.
(25:44):
And then beyond that, looking into the literature and thinking about the treatments that they wanted to try for themselves. Patients were sort of at the forefront of every step of recognizing, defining, describing this illness presentation and then thinking about what they wanted to be able to do for themselves to address it. And that was really interesting to me. That was incredibly interesting to me. And it was also really interesting because by, I don't know exactly when 2021 or 2022, it was already a really tense landscape where it felt like there were real factions of folks who were in conflict about what was real and what wasn't real, how things ought to be studied, who ought to be studying them, what would count as evidence in this realm. And all of those questions were just really interesting to me. And the LISTEN study was approaching them in this really thoughtful way, which was Harlan and Akiko sort of partnering really closely with patients who enrolled.
(26:57):
And it's a decentralized study and people could enroll from all over the world. There's a portion of patients who do have their blood work evaluated, but you can also just complete surveys and have that data count towards, and those folks would be from anywhere in the world. Harlan did this amazing, amazing work to figure out how to collect blood samples from all over the country that would be drawn at home for people. So they were doing this decentralized study where people from their homes, from within the sort of circumstances of their lives around their chronic illness could participate, which that was really amazing to me. And then they were partnering really thoughtfully with these patients just to figure out what questions they wanted to ask, how they wanted to ask them, and to try to capture a lot of multimodal data all at once.
(27:47):
Survey data, journaling so people could write about their own experience in a freeform journal. They were collecting blood samples, and they were holding these town halls. And the town halls were on a regular basis, Harlan and Akiko, and anybody who was in the study could come on, could log onto a Zoom or whatever, and Harlan and Akiko and their research staff would talk about how things were going, what they were working on, what questions they had, what the roadblocks were, and then they would answer questions from their participants as the study was ongoing. And I didn't think that I had ever heard of something quite like that before. Have you ever heard of anything?
Eric Topol (28:32):
No. I mean, I think this is important to underscore, this was the first condition that was ever patient led, patient named, and basically the whole path was laid by the patient. So yes, and everything you summarize is so well as to the progress that's been made. Certainly, Harlan and Akiko are some of the people that have really helped lead the way to do this properly as opposed to, unfortunately one and a half billion dollars that have been put to the NIH for the RECOVER efforts that haven't yet led to even a significant clinical trial, no less a validated treatment. But I did think it was great that you spotlighted that just because again, it's thematic. And that gets me to the fourth dimension, which is you're the first prison doctor I've ever spoken to. And you also wrote a piece about that called, “the disillusionment of a Rikers Island Doctor” in the New Yorker, I think it was. And I wonder if you could tell us, firstly, now we're four years into Covid, you were for a good part of that at Rikers Island, I guess.
The Rikers Island Prison Doctor During Covid
Rachael Bedard (30:00):
I was, yeah.
Eric Topol (30:00):
Yeah. And what could be a more worrisome spot to be looking after people with Covid in a prison? So maybe you could just give us some insight about all that.
Rachael Bedard (30:17):
Yeah, it was really, I mean, it was the wildest time, certainly in my career probably that I'll ever have. In the end of February and beginning of March of 2020, it became very apparent to my colleagues and I that it was inevitable that this virus that was in Wuhan and in Italy was coming to the US. And jails are, we sort of jokingly described them as the worst cruise ships in the world. They are closed systems where everybody is eating, sleeping, going to the bathroom, everything on top of each other. There's an incredible amount of excess human contact in jails and prisons because people don't have freedom of movement and they don't get to do things for themselves. So every single, somebody brings you your mail, somebody brings you your meals, somebody brings you your medications. If you're going to move from point A to point B, an officer has to walk you there. So for a virus that was going to spread through what we initially thought was droplets and then found out was not just droplets but airborne, it was an unbelievably high-risk setting. It's also a setting where folks tend to be sicker than average for their age, that people bring in a lot of comorbidity to the setting.
(31:55):
And it's not a setting that does well under stress. I mean, jails and prisons are places that are sort of constitutionally violent, and they're not systems that adapt easily to emergency conditions. And the way that they do adapt tends to be through repressive measures, which tends to be violence producing rather than violence quelling. And so, it was just an incredibly scary situation. And in mid-March, Rikers Island, the island itself had the highest Covid prevalence of anywhere in the country because New York City was the epicenter, and Rikers was really the epicenter within New York. It was a wild, wild time. Our first seriously ill patient who ended up getting hospitalized. That was at that time when people were, we really didn't understand very much about what Covid looked like. And there was this guy sitting on the floor and he said, I don't know. I can't really get up.
(32:59):
I don't feel well. And he had an O2 stat of 75 or something. He was just incredibly hypoxic. It's a very scary setting for that kind of thing, right? It's not a hospital, it's not a place where you can't deliver ICU level care in a place like that. So we were also really worried about the fact that we were going to be transferring all of these patients to the city hospitals, which creates a huge amount of extra burden on them because an incarcerated patient is not just the incarcerated patients, the officers who are with that person, and there are special rules around them. They have to be in special rooms and all of these things. So it was just a huge systems crisis and really painful. And we, early on, our system made a bunch of good guesses, and one of our good guesses was that we should just, or one of our good calls that I entirely credit my bosses with is that they understood that we should advocate really hard to get as many people out as we could get out. Because trying to just manage the population internally by moving people around was not going to be effective enough, that we really need to decant the setting.
(34:18):
And I had done all of this work, this compassionate release work, which is work to get people who are sick out of jail so that they can get treatment and potentially die in a free setting. And so, I was sort of involved in trying to architect getting folks who were sort of low enough security risks out of jail for this period of time because we thought that they would be safer, and 1500 people left Rikers in the matter of about six weeks.
Rachael Bedard (34:50):
Which was a wild, wild thing. And it was just a very crazy time.
Eric Topol (34:56):
Yeah. Well, the word compassion and you go together exceptionally well. I think if we learn about you through your writings, that really shines through and what you've devoted your care for people in these different domains. This is just a sampling of your writings, but I think it gives a good cross section. What makes you write about a particular thing? I mean, obviously the Rikers Island, you had personal experience, but why would you pick Ozempic or why would you pick other things? What stimulates you to go after a topic?
Rachael Bedard (35:42):
Sometimes a lot of what I write about relates to my personal practice experience in some way, either to geriatrics or death and dying or to the criminal justice system. I've written about people in death row. I've written about geriatrics and palliative care in sort of a bunch of different ways. I am interested in topics in medicine where things are not yet settled, and it feels very of the moment. I'm interested in what the discourse is around medicine and healthcare. And I am interested in places where I think the discourse, not just that I'm taking a side in that discourse, but where I think the framework of the discourse is a little bit wrong. And I certainly feel that way about the Ozempic discourse. And I felt that way about the discourse around President Biden, that we're having not just a conversation that I have a strong opinion about, but a conversation that I think is a little bit askew from the way that we ought to be thinking about it.
Eric Topol (36:53):
And what I love about each of these is that you bring all that in. You have many different points of view and objective support and they're balanced. They're not just trying to be persuasive about one thing. So, as far as I know, you're extraordinarily unique. I mean, we are all unique, but you are huge standard deviations, Rachael. You cover bases that are, as I mentioned, that are new to me in terms of certainly this podcast just going on for now a couple of years, that is covering a field of both geriatrics and having been on the corrections board and in prison, particularly at the most scary time ever to be working in prison as a physician. And I guess the other thing about you is this drive, this humanitarian theme. I take it you came from Canada.
Rachael Bedard (37:59):
I did.
Eric Topol (37:59):
You migrated to a country that has no universal health.
Rachael Bedard (38:03):
That's right.
Eric Topol (38:03):
Do you ever think about the fact that this is a pretty pathetic situation here?
Rachael Bedard (38:08):
I do. I do think about it all the time.
Eric Topol (38:10):
In our lifetime, we'll probably never see universal healthcare. And then if you just go a few miles up north, you pretty much have that.
Rachael Bedard (38:18):
Yeah, if you've lived in a place that has universal healthcare and you come here, it's really sort of hard to ever get your mind around. And it has been an absolute possessing obsession of my entire experience in the US. I've now been here for over 20 years and still think it is an unbelievably, especially I think if you work with marginalized patients and how much their lack of access compounds the difficulty of their lives and their inability to sort of stabilize and feel well and take care of themselves, it's really frustrating.
Advice for Bringing Humanities to Medicine in a Career
Eric Topol (39:14):
Yeah, yeah. Well, I guess my last question to you, is you have weaved together a career that brings humanities to medicine, that doesn't happen that often. What's your advice to some of the younger folks in healthcare as to how to pull that off? Because you were able to do it and it's not easy.
Rachael Bedard (39:39):
My main advice when people ask me about this, especially to students and to residents who are often the people who are asking is to write when you can or pursue your humanities interests, your critical interests, whatever it is that you're doing. Do it when you can, but trust that your career is long and that you have a lot of time. Because the thing that I would say is I didn't start publishing until I was in fellowship and before that I was busy because I was learning to become a doctor. And I think it's really important that my concern about being a doctor who's a hybrid, which so many of us are now. A doctor or something else is you really do want to be a good doctor. And becoming a good doctor is really hard. And it's okay if the thing that is preoccupying you for the first 10 years of your training is becoming a great clinician. I think that's a really, really important thing to do. And so, for my first 10 years for med school and residency and chief residency and fellowship, I would write privately on the side a fair amount, but not try to publish it, not polish that work, not be thinking in sort of a careerist way about how I was going to become a doctor writer because I was becoming a doctor. And that was really preoccupying.
(41:08):
And then later on, I both sort of had more time and mental space to work on writing. But also, I had the maturity, I think, of being a person who was comfortable in my clinical identity to have real ideas and insights about medicine that felt different and unique to me as opposed to, I barely understand what's going on around me and I'm trying to pull it together. And that's how I would've been if I had done it more, I think when I was younger. Some people are real prodigies and can do it right out the gate, but I wasn't like that.
Eric Topol (41:42):
No, no, I think that's really sound advice because that's kind of the whole foundation for everything else. Is there a book in the works or will there be one someday?
Rachael Bedard (41:53):
There may be one someday. There is not one now. I think about it all the time. And that same advice applies, which is I believe in being a late bloomer and taking your time and figuring out what it is you really want to do.
Eric Topol (42:10):
Yeah. Well, that's great. Have I missed anything? And obviously we only can get to know you in what, 40 minutes to some extent, but have I not touched on something that you want to bring up?
Rachael Bedard (42:23):
No, I don't think so. Thank you for this conversation. It's been lovely.
Eric Topol (42:28):
No, I really enjoyed it. I'll be following your career. It's extraordinary already and you've got decades ahead to make an impact and obviously thinking of all these patients that you look after and have in the past, it’s just extraordinary. So what a joy to talk with you, Rachael, and I hope we'll have a chance to do that again in the times ahead.
Rachael Bedard (42:51):
Me as well. Thank you so much for inviting me.
**********************************************
Thank you for listening, reading or watching!
The Ground Truths newsletters and podcasts are all free, open-access, without ads.
Please share this post/podcast with your friends and network if you found it informative!
Voluntary paid subscriptions all go to support Scripps Research. Many thanks for that—they greatly help fund our education and summer internship programs.
Thanks to my producer Jessica Nguyen and Sinjun Balabanoff for audio and video support at Scripps Research.
Note: you can select preferences to receive emails about newsletters, podcasts, or all I don’t want to bother you with an email for content that you’re not interested in.

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
Episodes manquant?

Cliquez ici pour raffraichir la page manuellement.
Katerina Akassoglou: Blood Clots, Brain Inflammation, and Covid
19 oct.· Ground Truths
Superimposed on an impressive body of work on the blood-brain-barrier and immune system, Prof Akassoglou and her collaborators just published an elegant study in Nature that centered on the direct binding os the SARS-CoV-2 spike protein to fibrin with marked downstream pro-inflammatory effects. The findings and potential treatments have implications beyond Covid, Long Covid to other neurologic diseases.
Full videos of all Ground Truths podcasts can be seen on YouTube here. The audios are also available on Apple and Spotify.
Transcript with links to audio and to relevant papers, graphics
Eric Topol (00:07):
Well, hello this is Eric Topol with Ground Truths, and with me today is Katerina Akassoglou. She is at the Gladstone Institute and she is a remarkable neuroimmunologist who has been doing extraordinary work for three decades to unravel the interactions between the brain, blood vessels and the role of inflammation. So Katerina, there's a lot to discuss, so welcome.
Katerina Akassoglou (00:40):
Thank you. Thank you so much. It's a great pleasure to join.
By Way of Background
Eric Topol (00:43):
It's really interesting going back in your career. First of all, we're thankful that you immigrated here from Greece, and you have become one of the leading scientists in this discipline of important discipline of neuroimmunology, which is not just about Covid that we're going to talk about, but Alzheimer's and neurodegenerative diseases. This is a really big hot area and you're definitely one of the leaders. And what I was impressed is that all these years that you've been working on the integrity of the blood-brain barrier, the importance of fibrinogen and fibrin, and then comes along the Covid story. So maybe what we can do is start with that, which is you've made your mark in understanding this whole interaction between what can get into the brain, through the blood-brain barrier and incite inflammation. So this has been something that you've really taken to the extreme knowledge base. So maybe we can start with your work there before we get into the important seminal Nature paper that you recently published.
Katerina Akassoglou (01:57):
Yes, of course. So since very early on, I was still a graduate student when we made the first discovery and at the time was like mid-90s, so it was really ahead of its time. That dysregulation of cytokine expression in the brain of mice was sufficient to induce the whole cascade of events, triggering neurodegeneration, demyelination in pathological alterations, very reminiscent of multiple sclerosis pathology. And it was really hard to publish that study at the time because it was not yet accepted that this regulation of the immune system modeling the brain can be linked to neurodegeneration. So that was 1995 when we made that discovery, and I became really interested, what are the pathogenic triggers that actually polarized the immune cells in the brain? So with this, of course, this transgenic animal was expressing TNF, it was an artificially made animal that we made, but naturally what were the triggers that would polarize the innate immune cells? So I looked really early on in this mice and what I found was that the very first event was leaks of blood-brain barrier. It was opening of the blood-brain barrier in this mouse before inflammation, before demyelination, before neuronal loss. And this is really what shaped the question that, is it possible that these blood leaks that happened very early in the pathology, could this be the instigators of pathogenic inflammation in the brain?
Eric Topol (03:34):
Yeah. So in a way, you got at this question because of the chicken-and-egg and what happens first, and you got to the temporal saying, which happened first as you said, the leak before you could see evidence of inflammation and being able to study this of course in the experimental model, which you couldn't really do in people. And what I love about the description of your career, which has been quite extraordinary contributions is connecting the dots between the blood, the inflammatory response and the brain. Perhaps no one has done that like you have. And before we get into the recent paper, a lot of people are not aware that a year ago, a group in the UK known as PHOSP-COVID, they published a really important paper in Nature Medicine of over 1,800 people who were hospitalized with Covid and they found that fibrinogen was the best marker for cognitive deficits at 6 and 12 months (Figure below)
(04:40):
So that's just one of many papers, but it's a particularly well done study that already before you got into this work that recently published had emphasized fibrinogen. And by the way, again, having spent a lot of years in clots in the arteries, for me, we have to just get it down to fibrinogen plus thrombin gets you to fibrin. Okay, so fibrin is a major player here when fibrinogen is cleaved. So here we have the basis that you established, which is the fibrinogen leakage into the brain, activating inflammation, activating microglia, which like the macrophages of the brain and inciting the whole process. And before we close, I want to not just talk about Covid, but Alzheimer's too. But now let's get into the study that you did, [Fibrin drives thromboinflammation and neuropathology in COVID-19] which is striking, I mean really striking. And can you kind of take us through, because you not only demonstrated the importance of fibrin in inciting neuroinflammation in this model, but also how you could reverse it or prevent it. So this, and you looked at it in many different ways, this was a systematic approach. Maybe you can take us through how you were able to make such compelling evidence.
The Multimodal Evidence
Katerina Akassoglou (06:09):
Yes, thank you. First of all, thank you for bringing up the human relevance because this was also our inspiration for the work that we did in the Covid study. So as you mentioned in Covid patients, fibrinogen unbiased mass spec analysis was identified as the predictive biomarker for cognitive impairment in Long Covid patients. And this was in addition to also neuropathology data about the abundance of fibrin deposition in the brain. And these were studies that were done by NIH that have found deposition of fibrin in the brain and the reports for the abnormal and puzzling coagulation in Covid that is not setting other infections and also in many cases not always relating with the severity of symptoms. So even mild cases of Covid also had increased coagulation. I was really intrigued by this human, all this evidence in human data, and I thought that maybe the way that we're thinking about this, that it's systemic inflammation that drives the clotting.
(07:24):
Maybe there's another aspect to this. Maybe there is a direct effect of the virus with the coagulation cascade, and in this way maybe this can be an instigator of inflammation. So this was the original idea to be able to reconcile this data from the clinic about why do we have this prevalence of coagulopathy in Covid. And of course, the second question is, could this also be a driver of the disease? And of course, we're in a unique position because we have been studying this pathway now for over 20 years to have all the toolbox, the genetic toolbox, the pharmacologic toolbox to be able to actually really address these questions with genetic loss of function studies, with a blood innate immunity multiomics pipeline that we have set up in the lab. And of course, with preclinical pharmacology in our ABSL3 facility. So we had the infrastructure in place and the source in place to actually really dissect this question with both genetic tools as well as also technology platforms.
Eric Topol (08:29):
And you had in vivo imaging, you're the director of in vivo imaging for Gladstone and UCSF. So you do have the tools to do this.
Katerina Akassoglou (08:38):
Yes. The imaging that you mentioned is really important because this is, we employed that very early in our studies over now 15 years ago. And the reason was sometimes from snapshots of histopathology, you cannot really understand the sequence of events. So by being able to image these processes, both neuronal activity, microglia activation, infiltration of peripheral cells in the brain, this is how we could see the steps that what happens early on and to be able to answer these chicken-and-egg questions that you mentioned. So these were very, they're very important experiments, especially at the beginning because they were hypothesis driving and we were able to ask the right questions to drive our research program.
Eric Topol (09:26):
Now was the binding of the spike protein to one key site in fibrinogen, was that known before? [See outstanding Figure below from Trends in Immunology]
Katerina Akassoglou (09:36):
No, this was not known. So there was evidence that there are abnormal clots in Covid, but it was not known whether the spike protein would directly bind to protein to the coagulation cascade. So one of the key discoveries in our study was to use peptide array mapping and be able to identify not only the binding, but exactly the domains on fibrin that spike binds too. And what we found was two key domains, one the inflammatory domain and the other the plasmin binding site, which is important for fibrin degradation. So this suggested a potential dual deleterious role for this interaction, both by maybe affecting inflammation, but also delaying fibrinolysis, which is the degradation of this toxic protein from the brain. And indeed, we found that this interaction was responsible for all these two aspects, including decreased degradation, more inflammation, but also at the same time increased, increased coagulation. So it was a really pathogenic interaction.
Eric Topol (10:47):
Yeah, actually it's pretty striking. You have these two sites, the plasmin cleavage site of fibrinogen, which as you say, we knew there was a problem with clots. We knew that, but we didn't know exactly the spike protein how exactly it was implicated, particularly with fibrinogen. And then this other site, the CD11b-C18, now that's fancy for surface receptors of macrophages. And basically, this is critical because it's this microglia activation in the brain, and I know you saw it in the lungs as well through this other site that spike protein activated. So you had a twofer here of things that you discovered that the SARS-CoV-2 spike protein was capable of doing. This was a really big revelation. And then you also looked at mice that were genetically manipulated. So maybe you can, because before we get to your antibody monoclonal, the ways that you proved this were, I mean, one thing after another is really systematic. So maybe you can teach us about that.
Thanks for reading Ground Truths! This post is public so feel free to share it.
Establishing Causality
Katerina Akassoglou (12:08):
Yeah, sure. So the first was about chemistry experiment. So this of course, we had to get to the next step to see is there any causality for this pathway. So we employed genetic loss of function studies and we had knockout mice, either fibrinogen knockout mice, this mice have all blood proteins except fibrinogen, and they have a delay in coagulation so they don't clot properly. But we also had a mutant mouse, which is a fibrinogen NK mouse. And this was a mutation only within this inflammatory domain that you mentioned, inflammatory domain that binds to C11b-C18. Other names for this is of course complement receptor 3, Mac-1 (αMβ2). It's the same, many names for this receptor, that as you mentioned, is expressed not only in microglial in the brain, but also peripheral immune cells including macrophages as well as also neutrophils which are CD11b expressing.
(13:12):
So we now have genetic models to be able to look at both complete depletion of fibrinogen, but also a very specific mutation and very selective mutation that only blocks the inflammatory properties without affecting the properties of fibrin in hemostasis. And these mice were made many years ago by a very close collaborator, Jay Degen at the University of Cincinnati. So what we found is that when we block either the inflammatory domain or we completely deplete fibrinogen, there was this profound protection after infection in internasal infection with the virus in lung inflammation. And this was both suppression of oxidative stress and this pathogenic inflammation in the lung, but also decreasing fibrosis, which has been associated with also Long Covid. And the surprise came from the transcriptomic data. So when we did transcriptomic analysis in this mice in the lungs, we found perhaps the expected decrease in the immune signatures in macrophages. This was in line with our previous work in, as you mentioned, Alzheimer's models, multiple sclerosis models. But what also was really surprising is there was that genes that are associated with activation of NK cells were upregulated. And of course this was the first time we had infected these mice, previously we had not done an infection before. So I think that maybe because of this region we had not seen before in our data this immunomodulatory role of fibrin that not only surprises the macrophage response, but also increases these NK cells that are important for viral clearance.
Eric Topol (15:00):
So again, the finding another important unique finding is the natural killer (NK) cells and effect there from the activation of this, as you said, the inflammation site or the CD11b-C18 that we've been talking about. So now another layer of this, a dimension of your Nature paper was that you tested an antibody that you already had developed so-called 5B8. A monoclonal that specifically binds to the domain of the one we're talking about this inflammation domain of fibrinogen. So can you tell us about what that showed?
Katerina Akassoglou (15:45):
Yes, so we tested this antibody in different models of Covid, which were both models with neuroinvasion and models without neuroinvasion. So we used both transgenic mice for hACE2, the human ACE2 infected with Delta, but we also use mouse adapted viruses like Beta that is just in the wild type mice with no transgenic being involved that these are without neuroinvasion. And we wanted to see if the antibody had any potential protective effects. And what we found is that the antibody protected from inflammation in the lung. So the data looked so similar with a genetic mutation of this pathway, protection from inflammation, decreased fibrosis, increased viral clearance, so decreased spike and viral proteins in the lungs. But we also found a protection in the brain. So the brains of this mice, including both the models we used with neuroinvasion and without, they both have had microglia activation in the brain. And we also found neuronal loss in the Delta infected mice and the antibody protected from both neuroinflammation but also improved neuronal survival in the mice. Showing that there can be this despite regardless of which model we used, there was this protective effect suggesting that by blocking fibrin, either the periphery or in the brain, this could be protected for these models.
Eric Topol (17:28):
Yeah, so I mean this is fascinating because until now, until this report of yours and your colleagues at Gladstone, there was knowledge that there would be neuroinflammation from Covid, both in patients from various biomarkers and imaging as well as in experimental model. But what this did was take it to the fibrin story, and I guess that's one of the questions you nailed that how important fibrin is, but that doesn't necessarily rule out other triggers of neuroinflammation, right?
Katerina Akassoglou (18:04):
Oh, absolutely not. So I think that this is one of the mechanisms that can be very important, especially in some patients. But we know that there are additional of course mechanisms of neuroinflammation including auto-antibody responses, as well as also endotheliopathy that are persistent endotheliopathy, this can be interacting also with each other. So I think that it's important for future research that we understand how do these mechanisms feed into each other? Are there a positive feedback loops between autoimmune mechanisms and coagulopathy and endothelial dysfunction with inflammation? But I think most importantly, I think that if we're thinking of this in the context of patients, can we identify patients with mechanism that might be more prevalent in specific cases of Long Covid and tailor our potential future clinical trials towards the needs of Long Covid patients?
Towards Treatment
Eric Topol (19:06):
Absolutely. I did interview some months back on Grounds Truths, Michelle Monje at Stanford, who I'm sure and interact with, and she's also works not so much on the fibrin side, but on neuroinflammation and the likeness between this condition in people and chemo brain because of the inflammation that's seen there. So we've talked about the multiple triggers that could contribute to brain inflammation, which I think most people would say in Long Covid this is one of the most, besides obviously the lack of energy, the profound fatigue and disability, but the cognitive function hit, not just brain fog is often profound. And we've just seen some reports about that, and particularly in hospitalized patients, how bad that can be. So that gets us to a potential treatment. Now, one of the things that's out there dangling, there's many things that people have talked about in terms of why can't we have a treatment for Long Covid?
(20:13):
And now of course this fibrin pathway, if you will, lends itself to many possibilities, whether it's anticoagulants or fibrinolytics like a tPA or things like nattokinase, which is a Japanese food enzyme that you could get at the nutrition centers or whatever. What are your thoughts? Because we don't have any good studies. There are all these little, tiny studies and they don't provide much conclusion, and you have an antibody that could potentially be effective. As I understand it, you set up a company some years ago, Therini Bio and used to be called MedaRed. You're the first woman scientist at Gladstone to develop a spin out company, which is another point of congratulations on that. But could the antibody be tested in patients or what do you think about these other possibilities?
Katerina Akassoglou (21:15):
Yes, yes. These are great questions. So first of all, the different approaches that you mentioned have very different mechanism of action. So degrading fibrin, the degradation products of fibrin also can have deleterious effects. The dimer, for example, can be very pro-inflammatory. So at the same time, blocking coagulation can also have a diverse effects because this can lead to excessive hemorrhage. So the approach that we took was to selectively block the inflammatory properties of fibrin without affecting beneficial effects of the molecule in normal hemostasis. So the challenge when I made the antibody was to be able to dissect these two functions of fibrin. It's our most important clotting factor, but at the same time, a molecule with profound pro-inflammatory capacity. So the observation that these two domains, the clotting domain and inflammatory domain were not overlapping, was really the foundation of this invention was that we could maybe create this antibody to be able to target them in a selective way.
Other Neurologic Conditions
(22:31):
So the antibody I developed is neutralizing blood toxicity by blocking the inflammatory domain of fibrin without adverse coagulation effects. And it's now completing phase one trials. So it has already completed the single ascending dose at 40 milligram per kilogram. It's interim data were announced already for this trial, with no safety signals. So if the antibody completes this year, the phase one trials, then it should be possible to be tested in different patient populations. You mentioned before chemo brain, and I think it's important that we think that blood-brain barrier disruption occurs among many neurological conditions, and it's an early event associated with early disease onset and worse prognosis in multiple sclerosis, Alzheimer's disease, traumatic injuries. So I think that it's by developing a strategy, therapeutic strategy to neutralize blood toxicity, this can have applications in a wide range of neurological conditions with vascular dysfunction.
Eric Topol (23:54):
Yeah, no. In your Nature Immunology 2020 piece [Figure below], you started with the 1883 identification of multiple sclerosis (MS) lesions were “engorged with blood”, the first link between blood leaks and brain inflammation. So this has enormous potential. And what I like about this Katerina is that you've dissected the clot component versus the inflammatory trigger of the fibrinogen and fibrin story. And this is so vital because if you keep throwing these things that just going to work on the clot and not deal with the pro-inflammatory consequences, then you're going to get the wrong impression that clots are not that important. And by the way, you did mention, and I want to come back to that too, endothelial inflammation, which is another feature of Long Covid is another kind of interactive part of this because when the lining of the blood vessel is inflamed, it will attract microthrombi and also be a participant in this whole affair. What do you think about Alzheimer's and the prospects of being able to interfere with Alzheimer's? We have 20 years in someone before this process takes hold and meets clinical manifestations. Would an antibody like this ever be useful along the way?
Katerina Akassoglou (25:29):
Yeah, so well, our antibody was tested first in Alzheimer's, this models when it was originally published, and we performed reversal trials in Alzheimer's models. So we dosed mice when they have established amyloid plaques, microglia activation, neuronal loss, and we could reverse this effect so it could increase cholinergic neurons in mice, reduce inflammation in a very selective way, only the neurotoxic part of inflammation and for genetic depletion of this pathway with akin mice in Alzheimer's disease. Also, improves from cognitive impairment, and we now have a new paper in Cell Press that is showing this effects also with really nice and unbiased machine learning models for behavioral segmentation [Figure below].
So I think that there is the data both from genetic studies and the antibody show projection in Alzheimer's disease. And of course, as you might have read the recent Lancet report from the Lancet committee on dementia that identified the vascular risk factors as the key contributors, especially post sporadic cases of Alzheimer's disease that is over 90% of Alzheimer's disease that is not genetically linked.
(26:58):
So I think that there is a real need in Alzheimer's disease to be able to block this vascular induced pathology. And an antibody like the fibrin neutralizing therapy could be positioned to be protective from the vascular induced immune-mediated neurodegeneration in this disease as well. I mean, ultimately, I think that we need to be thinking the terms of efficacy. So we want to have a drug that is efficacious, but we also want it to be selective. And the selectivity is really important because the immune system has so many protective functions. So if we block phagocytosis, we end up with more debris, decrease of neurorepair, anti-myelination. So by blocking a ligand here and not blocking, not eliminating a cell type or blocking a global pathway in this cell, but biologic a single ligand, I think we have been able to achieve this balance between efficacy, but also safety because we only block this neurotoxic populations and not the entire innate immune response that also has been beneficial for metastatic functions in the brain.
Blocking Neuroinflammation
Eric Topol (28:19):
So you're bringing up another critical concept about targeting the inflammation, this kind of goldilocks story of how much you interfere with the immune response and how much you are able to reduce the adverse pro-inflammatory effects. So that gets me to what if we don't know in any given patient how much fibrin is having a role in their Long Covid. Although we know it has to be a prominent feature because we saw it in, not just a hospitalized patient series that I mentioned we reviewed, but other papers as well. But what about if you just try to take on inflammation like through a GLP-1 drug or cGAS–STING or any of these really strong anti-inflammatory pathways. Do you see a difference in a generalized approach versus a specific approach that is really fibrin centered?
Katerina Akassoglou (29:22):
Yeah, so we have a focus actually on both because we wanted to dissect the downstream intracellular pathways of fibrin, and it's interesting that we can find specific inflammatory mediators that potentially can also be targeted as well, to be able to preserve that specificity, which I think is really important because if we don't preserve the specificity, we'll end up with a lot of adverse effects by eliminating major immune responses. But the point that you raised I think is really important because it's not enough to have an efficacious and selective drug if you don't know the patient population that will benefit from this drug. So I think that in addition to the drug discovery studies, it's important to develop also biomarker programs with both fluid biomarkers, but also imaging biomarkers to be able to identify the patient populations that will benefit from such treatment.
(30:25):
So if for example, a patient population has a fibrin deposition, blocking only downstream might not be enough, and it might be really important to neutralize this fibrin toxicity in the brain of patients. And with our target engagement studies, we show that at least in animal models, the antibody can be there. So I'm very encouraged by also programs that are going on now in the scientific community to develop noninvasive ligands to be able to image fibrin in the brain that are already tested in different patient populations like multiple sclerosis. Because I think we're going to learn so much from the biology as we start interrogating and asking these questions now in different patient populations.
Eric Topol (31:14):
I think that's a vital point you're making because the success of a clinical trial here in a clinical syndrome that is mosaic with lots of different types of pathways. If you can nail down the patients that would have the most to stand to benefit from a particular intervention, that the chance of you not missing the benefit that is matching the marker, what image marker or other markers is so vital. Well, we've talked, I think, about some fascinating discoveries that you and your colleagues have made. I mean, it's really extraordinary, and obviously we need this in Long Covid. But you know what, Katerina, it's almost made me think that you were warming up to this for three decades, that somehow or other you were working on all this stuff and then came Covid. Is that how you see it, that somehow or other you didn't know that all the work you were doing was going to wind up in this space?
Katerina Akassoglou (32:18):
Oh, I never thought I would work in a virology project. This collaboration started over Zoom with Warner Greene. We were both sheltering in place. It was the beginning of the pandemic, and the first reports were coming out about this puzzling coagulopathy. And our labs were hardly operational at the time, as you know, we had to close down our labs for a while. And however, this was a very big problem, and we thought that this is our role as scientists. If we feel that we can contribute and we have the tools to contribute, we felt that it's important that we pivot some part of our research, and even we wouldn't be doing this before, but it was important to pivot a part of our research and collaborate. And I think studies like this, this study would have been impossible without a team of collaborators. As you know, there were over 50 scientists involved at Gladstone, UCSF, UCLA, UCSD, Stanford University. Without collaboration, this study wouldn't be possible. So I'm really grateful to everyone who came together to solve this problem because I think that's what scientists should be doing. We should be solving problems as they arise.
Eric Topol (33:41):
Well, and also, I think a lot of people don't realize that, for example, when the Covid vaccines came along, people think, oh, well, it all got done in 10 months since the sequence of the virus, when in fact it took 30 years at least between all the factors that went into having an mRNA and sequencing virus and nanoparticles. And in many ways, your arc of this work is like that because it took three decades to have all the tools and the basic understanding, the antibody that you had developed for different reasons and this fascinating unraveling of what's going on in the model and undoubtedly in some patients at least as well. So before we wrap up, have I missed anything about this just remarkable work you've done?
Katerina Akassoglou (34:33):
Oh, thank you. I just want to thank you for this discussion and thank you for emphasizing the different areas and the different decisions that this pathway can have implications both for our understanding, our basic understanding of the blood brain immune interface, as well as also potential translation. And I think that the curiosity sometimes of how things work, I never thought it would work on Covid, like you mentioned at the beginning, but I think that basic science and curiosity driven science can sometimes lead to discoveries with translational implications that hopefully might benefit patients one day.
Eric Topol (35:21):
Yeah, well, undoubtedly it will. We're indebted to you, Katerina and all the folks that you have teamed up with, connecting the dots at the neurovascular interface. Phenomenal work and will follow the subsequent with great interest and it will likely not just a story about Long Covid, but other areas as well, so thank you.
*********************************
Thanks for listening, reading or watching!
The Ground Truths newsletters and podcasts are all free, open-access, without ads.
Please share this post/podcast with your friends and network if you found it informative!
Voluntary paid subscriptions all go to support Scripps Research. Many thanks for that—they greatly help fund our summer internship programs.
Thanks to my producer Jessica Nguyen and Sinjun Balabanoff for audio and video support at Scripps Research.
Note: you can select preferences to receive emails about newsletters, podcasts, or all I don’t want to bother you with an email for content that you’re not interested in.

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
Patrick Hsu: A Trailblazer in Digital Biology
13 oct.· Ground Truths
When I think of digital biology, I think of Patrick Hsu—he’s the prototype, a rarified talent in both life and computer science, who recently led the team that discovered bridge RNAs, what may be considered CRISPR 3.0 for genome editing, and is building new generative A.I. models for life science. You might call them LLLMs-large language of life models. He is Co-Founder and a Core Investigator of the Arc Institute and Assistant Professor of Bioengineering and Deb Faculty Fellow at the University of California, Berkeley.
Above is a brief snippet of our conversation. Full videos of all Ground Truths podcasts can be seen on YouTube here. The audios are also available on Apple and Spotify.
Here’s the transcript with links to the audio and external links to relevant papers and things we discussed.
Eric Topol (00:06):
Well hello, it's Eric Topol with Ground Truths and I'm really delighted to have with me today Patrick Hsu. Patrick is a co-founder and core investigator at the Arc Institute and he is also on the faculty at the University of California Berkeley. And he has been lighting things up in the world of genome editing and AI and we have a lot to talk about. So welcome, Patrick.
Patrick Hsu (00:29):
Thanks so much. I'm looking forward to it. Appreciate you having me on, Eric.
The Arc Institute
Eric Topol (00:33):
Well, the first thing I'd like to get into, because you're into so many important things, but one that stands out of course is this Arc Institute with Patrick Collison who I guess if you can tell us a bit about how you two young guys got to meet and developed something that's really quite unique that I think brings together investigators at Stanford, UCSF, and Berkeley. Is that right? So maybe you can give us the skinny about you and Patrick and how all this got going.
Patrick Hsu (01:05):
Yeah, sure. That sounds great. So we started Arc with Patrick C and with Silvana Konermann, a longtime colleague and chemistry faculty at Stanford about three years ago now, though we've been physically operational just over two years and we're an independent research institute working at the interface of biomedical science and machine learning. And we have a few different aspects of our model, but our overall mission is to understand and treat complex human diseases. And we have three pillars to our model. We have this PI driven side of the house where we centrally fund our investigators so that they don't have to write grants and work on their very best ideas. We have a technical staff side of the house more like you'd see in a frontier AI lab or in biotech industry where we have professional teams of R&D scientists working cross-functionally on higher level organizational wide goals that we call our institute initiatives.
(02:05):
One focused on Alzheimer's disease experimentally and one that we call a virtual cell initiative to simulate human biology with AI foundation models. And our third pillar over time is to have things not just end up as academic papers, but really get things out into the real world as products or as medicines that can actually help patients on the translational side. And so, we thought that some really important scientific programs could be unlocked by enabling new organizational models and we are experimenting at the institutional scale with how we can better organize and incentivize and support scientists to reach these long-term capability breakthroughs.
Patrick, Patrick and Silvana
Eric Topol (02:52):
So the two Patrick’s. How did you, one Patrick I guess is a multi-billionaire from Stripe and then there's you who I suspect maybe not quite as wealthy as the other Patrick, how did you guys come together to do this extraordinary thing?
Patrick Hsu (03:08):
Yeah, no, science is certainly expensive. I met Patrick originally through Silvana actually. They actually met, so funny trivia, all three Arc founders did high school science together. Patrick and Silvana originally met in the European version of the European Young Scientist competition in high school. And Silvana and I met during our PhDs in her case at MIT and I was at Harvard, but we met at the Broad Institute sort of also a collaborative Harvard, MIT and Harvard hospitals Institute based in Kendall Square. And so, we sort of in various pairwise combinations known each other for decades and worked together for decades and have all collectively been really excited about science and technology and its potential to accelerate societal progress. Yet we also felt in our own ways that despite a lot of the tremendous progress, the structures in which we do this work, fund it, incentivize it and roll it out into the real world, seems like it's really possible that we'll undershoot that potential. And if you take 15 years ago, we didn't have the modern transformer that launched the current AI revolution, CRISPR technology, single-cell, mRNA technology or broadly addressable LNPs. That’s a tremendous amount of technologies have developed in the next 15 years. We think there's a real unique opportunity for new institutes in the 2020s to take advantage of all of these breakthroughs and the new ones that are coming to continue to accelerate biological progress but do so in a way that's fast and flexible and really focused.
Eric Topol (04:58):
Yeah, I did want to talk with you a bit. First of all before I get to the next related topic, I get a kick out of you saying you've worked or known each other for decades because I think you're only in your early thirties. Is that right?
Patrick Hsu (05:14):
I was lucky to get an early start. I first started doing research at the local university when I was 14 actually, and I was homeschooled actually until college. And so, one of the funny things that you got to do when you're homeschooled is well, you could do whatever you want. And in my case that was work in the lab. And so, I actually worked basically full time as an intern volunteer, cut my teeth in single cell patch clamp, molecular biology, protein biochemistry, two photon and focal imaging and kind of spiraled from there. I loved the lab, I loved doing bench work. It was much more exciting to me than programming computers, which was what I was doing at the time. And I think these sort of two loves have kind of brought me and us to where we are today.
Eric Topol (06:07):
Before you got to Berkeley and Arc, I know you were at Broad Institute, but did you also pick up formal training in computer science and AI or is that something that was just part of the flow?
Patrick Hsu (06:24):
So I grew up coding. I used to work through problems sets before dinner growing up. And so, it's just something that you kind of learn natively just like learning French or Mandarin.
New Models of Funding Life Science
Eric Topol (06:42):
That's what I figured. Okay. Now this model of Arc Institute came along in a kind of similar timeframe as the Arena BioWorks in Boston, where some of the faculty left to go to Arena like my friend Stuart Schreiber and many others. And then of course Priscilla and Mark formed the Chan Zuckerberg Institute and its biohub and its support. So can you contrast for one, these three different models because they’re both very different than of course the traditional NIH pathway, how Arc is similar or different to the others, and obviously the goal here is accelerating things that are going to really make a difference.
Patrick Hsu (07:26):
Yeah, the first thing I would say is zooming out. There have been lots of efforts to experiment with how we do science, the practice of science itself. And in fact, I've recently been reading this book, the Demon Under the Microscope about the history of infectious disease, and it talks about how in the 1910s through the 1930s, these German industrial dye manufacturing companies like Bayer and BASF actually launched what became essentially an early model for industrial scale science, where they were trying to develop Prontosil, Salvarsan and some of these early anti-infectives that targeted streptococcus. And these were some of the major breakthroughs that led to huge medical advances on tackling infectious disease compared to the more academic university bound model. So these trends of industrial versus academic labs and different structures to optimize breakthroughs and applications has been a through current throughout international science for the last century.
(08:38):
And so, the way that we do research today, and that's some of our core tenets at Arc is basically it hasn't always been this way. It doesn't need to necessarily be this way. And so, I think organizational experiments should really matter. And so, there's CZI, Altos, Arena, Calico, a variety of other organizational experiments and similarly we had MRC and Bell Labs and Xerox PARCS, NIBRT, GNF, Google Research, and so on. And so, I think there are lots of different ways that you can organize folks. I think at a high level you can think about ways that you can play with for-profit versus nonprofit structures. Whether you want to be a completely independent organization or if you want to be partnered with universities. If you want to be doing application driven science or really blue sky curiosity driven work. And I think also thinking through internally the types of expertise that you bring together.
(09:42):
You can think of it like a cancer institute maybe as a very vertically integrated model. You have folks working on all kinds of different areas surrounding oncology or immunotherapy and you might call that the Tower of Babel model. The other way that folks have built institutes, you might call the lily pad model where you have coverage of as many areas of biomedical research as possible. Places like the Whitehead or Salk, it will be very broad. You'll have planned epigenetics, folks looking at RNA structural biology, people studying yeast cell cycle, folks doing in vivo melanoma models. It's very broad and I think what we try to do at Arc is think about a model that you might liken more to overlapping Viking shields where there's sort of five core areas that we're deeply investing in, in genetics and genomics, computation, neuroscience, immunology and chemical biology. Now we really think of these as five areas that are maybe the minimal critical mass that you would need to make a dent on something as complicated as complex human diseases. It's certainly not the only thing that you need, but we needed a critical mass of investigators working at least in these areas.
Eric Topol (11:05):
Well, yeah, and they really converge on where the hottest advances are being made these days. Now can you work at Arc Institute without being one of these three universities or is it really that you maintain your faculty and your part of this other entity?
Patrick Hsu (11:24):
So we have a few elements to even just the academic side of the house. We have our core investigators. I'm one of them, where we have dually appointed faculty who retain their latter rank or tenured appointment in their home department, but their labs are physically cited at the Arc headquarters where we built out a lab in Stanford Research Park in Palo Alto. And so, folks move their labs there. They continue to train graduate students based on whatever graduate programs they're formally affiliated with through their university affiliation. And so, we have nearly 40 PhD students across our labs that are training on site every day.
(12:03):
So in addition to our core investigators, we also have what we call our innovation investigators, which is more of a grant program to faculty at our partner universities. They receive unrestricted funding from us to seed a new project or accelerate an existing area in their group and their labs stay at their home campus and they just get that funding to augment their work. The third way is our technical staff model where folks basically just come work at Arc and many of them also are establishing their own research groups focusing on technology R&D areas. And so, we have five of those technology centers working in molecular engineering, multi-omics, complex cellular models, in vivo models, and in machine learning.
Discovery of Bridge RNAs
Eric Topol (12:54):
Yeah, that's a great structure. In fact, just a few months ago, Patrick Collison, the other Patrick came to Stanford HAI where I'm on the board and you've summarized it really well and it's very different than the other models and other entities, companies included that you mentioned. It's really very impressive. Now speaking of impressive on June 26, this past few months ago, which incidentally is coincident with the draft genome in the year 2000, the human sequence. You and your colleagues, perhaps the most impressive jump in terms of an Arc Institute contribution published two papers back-to-back in Nature about bridge RNA: [Bridge RNAs direct programmable recombination of target and donor DNA] and [Structural mechanism of bridge RNA-guided recombination.] And before I get you to describe this breakthrough in genome editing, some would call it genome editing 3.0 or CRISPR 3.0, whatever. But what we have today in the clinic with the approval of CRISPR 1.0 for sickle cell and thalassemia is actually quite crude. I think most people will know it's just a double stranded DNA cleavage with all sorts of issues about repair and it's not very precise. And so, CRISPR 2.0 is supposed to be represented by David Liu's contributions and his efforts at Broad like prime and base editing and then comes yours. So maybe you can tell us about it and how it is has to be viewed as quite an important advance.
Patrick Hsu (14:39):
The first thing I would say before CRISPR, is that we had RNA interference. And so, even before this modern genome editing revolution with programmable CRISPRs, we had this technology that had a lot of the core selling points as well. Any target will now become druggable to us. We simply need to reprogram a guide RNA and we can get genetic access to things that are intracellular. And I think both the discovery of RNA interference by Craig Mello and Andy Fire or the invention or discovery of programmable CRISPR technologies, both depend on the same fundamental biological mechanism. These non-coding guide RNAs that are essentially a short RNA search string that you can easily reprogram to retarget a desired enzyme function, and natively both RNAi and CRISPR are molecular scissors. Their RNA or DNA nucleases that can be reprogrammed to different regions of the genome or the transcriptome to make a cut.
(15:48):
And as bioengineers, we have come up with all kinds of creative ways to leverage the ability to make site specific cuts to do all kinds of incredible things including genome editing or beyond transcriptional up or down regulation, molecular imaging and so on and so forth. And so, the first thing that we started thinking about in our lab was, why would mother nature have stopped only RNAi and CRISPR? There probably are lots of other non-coding RNAs out there that might be able to be programmable and if they did exist, they probably also do more complicated and interesting things than just guide a molecular scissors. So that was sort of the first core kind of intuition that we had. The second intuition that we had on the technology side, I was just wearing my biology hat, I’ll put on my technology hat, is the thing that we call genome editing today hardly involves the genome.
(16:50):
It's really you're making a cut to change an individual base or an individual gene or locus. So really you're doing small scale single locus editing, so you might call it gene level or locus level cuts. And what you really want to be able to do is do things at the genome scale at 100 kb, a megabase at the chromosome scale. And I think that's where I think the field will inevitably go if you follow the technology curves of longer and longer range gene sequencing, longer and longer range gene synthesis, and then longer and longer range gene editing. And so, what would that look like? And we started thinking, could there be essentially recombination technologies that allow you to do cut and paste in a single step. Now, the reason for that is the way that we do gene editing today involves a cut and then a multi-step process of cellular DNA repair that resolves the cut to make the exertion or the error prone deletion or the modification that ends up happening.
(17:59):
And so, it's very complicated and whether that's nucleases or base or prime editing, you're all generally limited to the small-scale single locus changes. However, there are natural mechanisms that have solved this cut and paste problem, right? There are these viruses or bacterial versions of viruses known as phage that have generally been trying to exert their multi kilobase genomes into bacterial hosts and specialize throughout billions of years. So our core thought was, well, if there are these new non-coding RNAs, what kind of functions would we be excited about? Can we look in these mobile genetic elements, these so-called jumping genes for new mechanisms? They're incredibly widespread. Transposons are thought to be some of the most diverse enzyme mechanisms found in nature. And so, we started computationally by asking ourselves a very simple question. If a mobile element inserts itself into foreign DNA and it's able to somehow be programmable, presumably the inside or something encoded in the inside of the element is predictive of some sequence on the outside of the element.
(19:15):
And so, that was the core insight we took, and we thought let's look across the boundaries of many different mobile genetic elements and we zoomed in on a particular sub family of these MGE known as insertion sequence (IS) elements which are the most autonomous minimal transposons. Normally transposons have all kinds of genes that they use to hitchhike around the genomic galaxy and endow the bacterial host with some fitness advantage like some ability to metabolize some copper and some host or some metal. And these IS elements have only the enzymes that they need to jump around. And if you identify the boundaries of these using modern computational methods, this is actually a really non-trivial problem. But if you solve that problem to figure out with nucleotide resolution where the element boundaries end and then you look for the open reading frame of the transposases enzyme inside of this element, you'll find that it's not just that coding sequence.
(20:19):
There are also these non-coding flanks inside of the element boundaries. And when we looked across the non-coding, the entire IS family tree, there are hundreds of these different types of elements. We found that this particular family IS110, had the longest non-coding ends of all IS elements. And we started doing experiments in the lab to try to figure out how these work. And what we found was that these elements are cut and paste elements, so they excise themselves into a circular form and paste themselves back in into a target site linearly. But the circularization of this element brings together two distal ends together, which brings together a -35 and a -10 box that create and reconstitute a canonical bacterial transcriptional promoter. This essentially is like plugging a plug into an electrical socket in the wall and it jacks up transcription. Now you would think this transcription would turn on the transposase enzyme so it can jump around more but it transcribes a non-coding RNA out of this non-coding end.
(21:30):
We're like, holy crap, are these RNAs actually involved in regulating the transposon? Now the boring answer would be, oh, it regulates the expression. It's like an antisense regulate or something. The exciting answer would be, oh, it's a new type of guide RNA and you found an RNA guided integrase. So we started zooming in bound dramatically on this and we undertook a covariation analysis where we were able to show that this cryptic non-coding RNA has a totally novel guide RNA structure, totally distinct from RNAi or CRISPR guide RNAs. And it had a target site that covaried with the target site of the element. And so we're like, oh wow, this could be a programmable transposase. The second thing that we found was even more surprising, there was a second region of complementarity in that same RNA that recognized the donor sequence, which is the circularized element itself. And so, this was the first example of a bispecific guide RNA, and also the first example of RNA guided self-recognition by a mobile genetic element.
Eric Topol (22:39):
It's pretty extraordinary because basically you did a systematic assessment of jumping genes or transposons and you found that they contain things that previously were not at all recognized. And then you have a way to program these to edit, change the genome without having to do any cuts or nicks, right?
Patrick Hsu (23:05):
Yeah. So what we showed in a test tube is when we took this, so-called bridge RNA, which we named because it bridges the target and donor together along with the recombinase enzyme. So the two component system, those are the only two things that you need. They're able to cut and paste DNA and recombine them in a test tube without any DNA repair, meaning that it's independent of cellular DNA repair and it does strand nicking, exchange, junction resolution and religation all in a single mechanism. So that's when we got super excited about its potential applications as bioengineering tool.
Eric Topol (23:46):
Yeah, it's pretty extraordinary. And have you already gone into in vivo assessment?
Patrick Hsu (23:54):
Yes, in our initial set of papers, what we showed is that these are programmable and functional or recombinases in a test tube and in bacterial cells. And by reprogramming the target and donor the right way, you can use these enzymes not just for insertion, but also for flipping and cutting out DNA. And so, we actually have in a single mechanism the ability to do bridge editing, if you will, for universal DNA recombination, insertion, excision or inversion, similar to what folks have been doing for decades with Cre recombinase, but with fully programmable recognition sequences. The work that we're doing now in the lab as you can imagine is to adapt these into robust tools for mammalian genome editing, including of course, human genomes. We're excited about this, we're making good progress. The CRISPR has had thousands of labs over the last 10, 15 years working on it to make these therapeutic level potency and selectivity. We're going to work and follow that same blueprint for getting bridge systems to get to that level of performance, but we're on the path and we're very optimistic for the future.
Exemplar of Digital Biology
Eric Topol (25:13):
Yeah, I think it's quite extraordinary and it's a whole different look to what we've been seeing in the CRISPR era for over the past decade and how that's been advancing and getting more specific and less need for repair and being able to be more versatile. But this takes it to yet another dimension. Now, this brings me to the field that when I think of this term digital biology, I think of you and now our mutual acquaintance, Jensen Huang, who everybody knows now. Back some months ago, he wrote and said at a conference, “Where do I think the next amazing revolution is going to come? And this is going to be flat out one of the biggest ones ever. There’s no question that digital biology is going to be it. For the first time in human history, biology has the opportunity to be engineering, not science.” So can you critique Jensen? Is he right? And tell us how you conceive the field of digital biology.
Patrick Hsu (26:20):
If you look at gene therapy today, the core concepts are actually remarkably simple. They're elegant. Of course, you're missing a broken gene, you need to put it back. And that can be curative. Very simple, powerful concept. However, for complex diseases where you don't have just a single gene that goes wrong, in many cases we actually have no idea what to do. And in fact, when you're trying to put in DNA, that's over more than a gene scale. We kind of very quickly run out of ideas. Is it a CAR and a cytokine, a CAR and a cytokine and another thing? And then we're kind of out of ideas. And so, we started thinking in the lab, how can we actually design genomes where it's not just let's reduce the genome into individual Lego blocks, iGem style with promoters and different genes that we just sort of shuffle the Lego blocks around, but actually use AI to design genome sequences.
(27:29):
So to do that, we thought we would have to first of all, train a model that can learn and decode the foreign language of biology and use that in order to design sequences. And so, we sort of have been training DNA foundation models and virtual cell models at Arc, sort of a major effort of ours where the first thing that we tried was to take a variance of transformer architecture that's used to train ChatGPT from OpenAI, but instead apply this to study the next DNA token, right? Now, the interesting thing about next token prediction in English is that you can actually learn a surprising amount of information by just predicting the next word. You can learn world knowledge is the capital of Azerbaijan, is it Baku or is it London, right? Or if you're walking around in the kitchen, then the next text is, I then left the kitchen or the bathroom, right?
(28:33):
Now you're learning about spatial reasoning, and so you can also learn translation obviously. And so similarly, I think predicting the next token or the next base and DNA can lead you to learn about molecular biochemistry, is the next amino acid residue, hydrophobic or hydrophilic. And it can teach you about the mechanics of some catalytic binding pocket or something. You can learn about a disease mutation. Is the next base, the sick linked base or the wild type base and so on and so forth. And what we found was that at massive scale, DNA foundation models learn about molecular function, not just at the DNA level, but also at the RNA and the protein. And indeed, we could use these to design molecular systems like CRISPR-Cas systems, where you have a protein and the guide RNA. It could also design new DNA transposons, and we could design sequences that look plausibly like real genomes, where we generate a megabase a million bases of continuous genome sequence. And it really looks and feels like it could be a blurry picture of something that you would actually sequence. This has been a wonderful collaboration with Brian Hie, a PI at Stanford and an Arc investigator, and we're really excited about what we've seen in this work because it promises the better performance with even more scale. And so, simply by scaling up these models, by adding in more compute, more training data or more powerful models, they're going to get sharper and sharper.
New A.I. Models in Life Science
Eric Topol (30:25):
Yeah. Well, this whole use of large language models for the language of life, whether it's the genome proteins and on and on, actually RNA and even cells has really taken root. And of course, this is really one of the foundations of that field of digital biology, which brings together generative AI, AI tools and trying to push forward our understanding in biology. And also, obviously what's been emphasized in drug discovery, perhaps it's been emphasized even too much because we still have a lot to learn about biology, but that gets me to these models. Like today, AlphaProteo was announced by DeepMind, as we all know, AlphaFold 1, 2, now 3. They were kind of precursors of being able to predict proteins from amino acid 3D structure. And that kind of took the field by a little bit like ChatGPT for life science, but now it's a new model all the time. So you've been working on various models and Arc Institute, how do you see this unfolding? Are we just going to have every aspect of the language of life being approached in all the different interactions? And this is going to help us get to a much more deep level of understanding.
Patrick Hsu (31:56):
I'll say two things. The first is a lot of models that you just described are what I would call task specific models. A model for de novo design of a binder, a model for protein structure prediction. And there are other models for protein fitness or for RNA structure prediction, et cetera, et cetera. And I think what we're going to move towards are more unifying models where there's different classes of models at different levels of scale. So we will have these atomic level models for looking at generative chemistry or ligand docking. We have models that can unify genomes and their molecules, and then we have models that can unify cells and tissues. And so, for example, if you took an H&E stain of some liver, there are folks building models where you can then predict what the single cell spatial transcriptome will look like of that model. And that's obviously operating at a very different level of abstraction than a de novo protein binder. But in the long run, all of these are going to get, I think unified. I think the reason why this is possible is that biology, unlike physics, actually has this unifying theory of evolution that runs across all of its length scales from atomic, molecular, cellular, organismal to entire ecosystem. And the promise of these models is no short then to make biology a predictive discipline.
Patrick Hsu (33:37):
In physics, the experimentalists win the big prizes for the theorists when they measure gravitational waves or whatever. But in biology, we're very practical people. You do something three times and do a T-test. And I think my prediction is we can actually gauge the success of these LLMs or whatever in biology by how much we respect theory in this field.
The A.I. Scientist
Eric Topol (34:05):
Yeah. Well, that's a really interesting perspective, an important perspective because the proliferation of models, which we're going to get into not just doing the things that you described, but also being able to be “pseudo” scientists, the so-called AI scientist. Maybe you could comment about that concept because that's been the idea that everything from the question that could be asked to the hypothesis and the experiment design and the analysis of data and then the feedback. So what is the role of the scientists, that seems to have been overplayed? And maybe you can put that in context.
Patrick Hsu (34:48):
So yeah, right now there's a lot of excitement that we can use AI agents not just to do software enterprise workflows, but to be a research assistant. And then over time, itself an autonomous research scientist that can read the literature, come up with an idea, maybe run a bunch of robots in the lab or do a bunch of computational analyses and then potentially even analyze data, conclude what is going on and actually write an entire paper. Now, I think the vision of this is compelling in the long term. I think the question is really about timescale. If you break down the scientific method into its constituent parts, like hypothesis generation, doing an experiment, analyzing experiment and iterating, we're clearly going to use AI of some kind at every single step of this cycle. I think different steps will require different levels of maturity. The way that I would liken this is just wet lab automation, folks have dreamed about having pipetting robots that just do their western blots and do their cell culture for them for generations.
(36:01):
But of course, today they don't actually really feel fundamentally different from the same ones that we had in the 90s, let's say. Right? And so, obviously they're getting better, but it seems to me one of the trends I'm very bullish about is the explosion of humanoid robots and robot foundation models that have a world model and a sense of physics and proportionate space loaded onto them. Within five years, we're going to have home robots that can fold your clothes, that can organize your kitchen and do all of this while you're sleeping, so you wake up to a clean home every day.
Eric Topol (36:40):
It’s not going to be just Roomba anymore. There's going to be a lot more, but it isn't just the hardware, it's also the agents playing in software, right?
Patrick Hsu (36:50):
It's the integrated loop of the hardware and the software where the ability to make the same machine generally intelligent will make it adaptable to a broad array of tasks. Now, what I'm excited about is those generally intelligent humanoid robots coming into the lab, where instead of creating a centrifuge or a new type of pipetter that's optimized for your Beckman or Hamilton device, instead you just have robot arms that you snap onto the edge of the bench and then they just work alongside you. And I do think that's coming, although it'll take a lot of hardware and software and computer vision engineering to make that possible.
A Sense of Humor
Eric Topol (37:32):
Yeah, and I think also going back to originating the question, there still is quite a debate about the creativity and the lack of any simulation of AGI, whatever that means anymore. And so, the human in the loop part of this is obviously I think it's still of critical nature. Now, the other thing I learned about you is you have a great sense of humor, which is really important by the way. And recently, which is great that you're active on X or Twitter because that's one way we get to see what you're thinking on a day-to-day basis. But I think you put out a poll which was really quite provocative , and it was about, here's what it said, “do more people in the world *truly* understand transformers or health insurance?” And interestingly, you got 49% for transformers at 51% for health insurance. Can you tell us what you're thinking when you put that poll together? Because obviously a lot of people don't understand either of these.
Patrick Hsu (38:44):
I think the core question is, there are different ways of looking at the world, some of which are very bottom up and some of which are very top down. And one of the very surprising things about transformers is they're taking something that is in principle, an incredibly simple task, which is if you have a string of text, what is the next letter? And somehow at massive, massive scale, you can unlock something that looks an awful lot like reasoning, and you've got these emergent behaviors. Now the bottoms up theory of just the linear algebra that's going on in these models couldn't possibly really help us predict that we have these emerging capabilities. And I think similarly in healthcare, there's a literal set of parts that are operating in some complex way that at massive scale becomes this incredibly confusing and dynamic system for how we can actually incentivize how we make medicines, how we actually take care of people, and how we actually pay for any of this from an economic point of view. And so, I think it was, in some sense if transformers can actually be an explainable by just linear algebra equations, maybe there will be a way to decompose the seemingly incredibly confusing world of healthcare in order to actually build a better way forward.
Computing Power and the GPU Arms Race
Eric Topol (40:12):
Yeah. Well that's great. Now the other thing I wanted to ask you about, we open source and the arms race of GPUs and this whole kind of idea is you touched on the need for coalescing a lot of these tools to exploit the synergy. But we have an issue because many academic labs like here at Scripps Research and so many others, including as I learned even at Stanford, have limited access to GPUs. So computing power of large language models is a problem. And then the models that exist today that can be adopted like Llama or others, and they're somewhat limited. And then we also have a movement towards trying to make things more open source, like for example, recently OpenCRISPR with Profluent Bio that is basically trying to use AI for CRISPR guides. And so, how do you deal with this arms race, computing power, open source, proprietary models that are not easily accessible without a lot of resources?
Patrick Hsu (41:30):
So the first thing I would say is, we are in the academic science sphere really unprepared for the level of resources that are required for doing this type of cutting edge computational work. There are top Stanford computer science professors or computational researchers who have a single GPU in their office, and that's actually what their whole lab runs off of.
(41:58):
The UC Berkeley campus, the grid runs on something like 12 megawatts of power and how are they going to build an on-premises GPU clusters, like a central question that can scale across the entire needs? And these are two of the top computer science universities in the world. And so, I think one of our kind of core beliefs at Arc is, as science both experimentally and computationally has gotten incredibly complex, not just in terms of conceptually, but also just the actual infrastructure and machines and know-how that you need to do things. We actually need to essentially support this. So we have a private GPU cloud that we use to train our models, and we have access to significantly large clusters for large burst kind of train outs as necessary. And I think infrastructurally for running genomics experiments or doing scalable brain organoid screens, right, we're also building out the infrastructure to support that experimentally.
Eric Topol (43:01):
Yeah, no, I think this is one of the advantages of the new model like the Arc Institute because not many centers have that type of plasticity with access to computing power when needed. So that's where a brilliant mind you and the Arc Institute together makes for a formidable recipe for future advances and of course building on the ones you've already accomplished.
The Primacy of Human Talent
Patrick Hsu (43:35):
I would just say, my main skill, if I have one, is to recruit really, really smart people. And so, everything that you're seeing and hearing about is the work of unbelievable colleagues who are curious, passionate, and incredible scientists.
Eric Topol (43:53):
But it also takes the person who can judge those who are in that category set as a role model. And you're certainly doing that. I guess just in closing, I mean, it's just such a delight to get to meet you here and kind of get your thoughts on what is the hottest thing in life science without question, which brings together the fields of AI and what's going on, not just obviously in genome editing, but this digital biology era that we're still in the early phases of, I mean, I think you could say that it's just going to continue to accelerate the exponential curve. We're still kind of on the bottom of that, I would imagine where we're headed. Any other things that you want to bring up that I haven't touched on that will round out this conversation?
Patrick Hsu (44:50):
I mean, I think it's very early days here at Arc.
Patrick Hsu (44:53):
When we founded Arc, we asked ourselves, how do we measure success? We don't have customers or revenue in the way that a typical startup does. And we felt sort of three things. The first was research institutes live and die by their talent. Can we actually hire incredible people when we make offers to people we want to come, do they come? The second was, when those folks do come to Arc, do they feel like they're able to work on important research programs that they couldn't do sort of at their prior university or company? And then longer term, the third thing was, and there's just no shortcut around this, you need to do important work. And I think we've been really excited that there are early signs that we're able to do all three of these things, and we're still, again, just following the same scaling laws that we're seeing in natural language and vision, but for the domain of biology. And so, we're excited about what's ahead and think if there are folks who are interested in learning more about Arc, just shoot me an email or DM.
Eric Topol (46:07):
Yeah, well I would just say, congratulations on what you've already achieved. I know you're going to keep rocking it because you already have in a short time. And for anybody who doesn't know about Arc Institute and your work and your team, I hope this is going to be putting them on notice actually what can be accomplished outside of the usual NIH funded model, which is kind of a risk-free zone where you basically have to have your results nailed down before you send in your proposal frequently, and it doesn't do great things for young people. Really, I think you actually qualify in that demographic where it's hard for them to break in for getting NIH grants and also for this type of work that you're doing. So we'll look for the next bridge beyond bridge RNAs of your just fantastic efforts. So Patrick, thanks so much for joining us today, and we'll be checking back with you and following all the great work that you'll be doing in the times ahead.
Patrick Hsu (47:14):
Thanks so much, Eric. It was such a pleasure to be here today. Appreciate the opportunity.
*******************
Thanks for listening, reading or watching!
The Ground Truths newsletters and podcasts are all free, open-access, without ads.
Please share this post/podcast with your friends and network if you found it informative!
Voluntary paid subscriptions all go to support Scripps Research. Many thanks for that—they greatly help fund our summer internship programs.
Thanks to my producer Jessica Nguyen and Sinjun Balabanoff for audio and video support at Scripps Research.
Note: you can select preferences to receive emails about newsletters, podcasts, or all I don’t want to bother you with an email for content that you’re not interested in.

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
AI Snake Oil—A New Book by 2 Princeton University Computer Scientists
24 sept.· Ground Truths
Arvind Narayanan and Sayash Kapoor are well regarded computer scientists at Princeton University and have just published a book with a provocative title, AI Snake Oil. Here I’ve interviewed Sayash and challenged him on this dismal title, for which he provides solid examples of predictive AI’s failures. Then we get into the promise of generative AI.
Full videos of all Ground Truths podcasts can be seen on YouTube here. The audios are also available on Apple and Spotify.
Transcript with links to audio and external links to key publications
Eric Topol (00:06):
Hello, it's Eric Topol with Ground Truths, and I'm delighted to welcome the co-author of a new book AI SNAKE OIL and it's Sayash Kapoor who has written this book with Arvind Narayanan of Princeton. And so welcome, Sayash. It's wonderful to have you on Ground Truths.
Sayash Kapoor (00:28):
Thank you so much. It's a pleasure to be here.
Eric Topol (00:31):
Well, congratulations on this book. What's interesting is how much you've achieved at such a young age. Here you are named in TIME100 AI’s inaugural edition as one of those eminent contributors to the field. And you're currently a PhD candidate at Princeton, is that right?
Sayash Kapoor (00:54):
That's correct, yes. I work at the Center for Information Technology Policy, which is a joint program between the computer science department and the school of public and international affairs.
Eric Topol (01:05):
So before you started working on your PhD in computer science, you already were doing this stuff, I guess, right?
Sayash Kapoor (01:14):
That's right. So before I started my PhD, I used to work at Facebook as a machine learning engineer.
Eric Topol (01:20):
Yeah, well you're taking it to a more formal level here. Before I get into the book itself, what was the background? I mean you did describe it in the book why you decided to write a book, especially one that was entitled AI Snake Oil: What Artificial Intelligence Can Do, What It Can't, and How to Tell the Difference.
Background to Writing the Book
Sayash Kapoor (01:44):
Yeah, absolutely. So I think for the longest time both Arvind and I had been sort of looking at how AI works and how it doesn't work, what are cases where people are somewhat fooled by the potential for this technology and fail to apply it in meaningful ways in their life. As an engineer at Facebook, I had seen how easy it is to slip up or make mistakes when deploying machine learning and AI tools in the real world. And had also seen that, especially when it comes to research, it's really easy to make mistakes even unknowingly that inflate the accuracy of a machine learning model. So as an example, one of the first research projects I did when I started my PhD was to look at the field of political science in the subfield of civil war prediction. This is a field which tries to predict where the next civil war will happen and in order to better be prepared for civil conflict.
(02:39):
And what we found was that there were a number of papers that claimed almost perfect accuracy at predicting when a civil war will take place. At first this seemed sort of astounding. If AI can really help us predict when a civil war will start like years in advance sometimes, it could be game changing, but when we dug in, it turned out that every single one of these claims where people claim that AI was better than two decades old logistic regression models, every single one of these claims was not reproducible. And so, that sort of set the alarm bells ringing for the both of us and we sort of dug in a little bit deeper and we found that this is pervasive. So this was a pervasive issue across fields that were quickly adopting AI and machine learning. We found, I think over 300 papers and the last time I compiled this list, I think it was over 600 papers that suffer from data leakage. That is when you can sort of train on the sets that you're evaluating your models on. It's sort of like teaching to the test. And so, machine learning model seems like it does much better when you evaluate it on your data compared to how it would really work out in the real world.
Eric Topol (03:48):
Right. You say in the book, “the goal of this book is to identify AI snake oil - and to distinguish it from AI that can work well if used in the right ways.” Now I have to tell you, it's kind of a downer book if you're an AI enthusiast because there's not a whole lot of positive here. We'll get to that in a minute. But you break down the types of AI, which I'm going to challenge a bit into three discrete areas, the predictive AI, which you take a really harsh stance on, say it will never work. Then there's generative AI, obviously the large language models that took the world by storm, although they were incubating for several years when ChatGPT came along and then content moderation AI. So maybe you could tell us about your breakdown to these three different domains of AI.
Three Types of AI: Predictive, Generative, Content Moderation
Sayash Kapoor (04:49):
Absolutely. I think one of our main messages across the book is that when we are talking about AI, often what we are really interested in are deeper questions about society. And so, our breakdown of predictive, generative, and content moderation AI sort of reflects how these tools are being used in the real world today. So for predictive AI, one of the motivations for including this in the book as a separate category was that we found that it often has nothing to do with modern machine learning methods. In some cases it can be as simple as decades old linear regression tools or logistic regression tools. And yet these tools are sold under the package of AI. Advances that are being made in generative AI are sold as if they apply to predictive AI as well. Perhaps as a result, what we are seeing is across dozens of different domains, including insurance, healthcare, education, criminal justice, you name it, companies have been selling predictive AI with the promise that we can use it to replace human decision making.
(05:51):
And I think that last part is where a lot of our issues really come down to because these tools are being sold as far more than they're actually capable of. These tools are being sold as if they can enable better decision making for criminal justice. And at the same time, when people have tried to interrogate these tools, what we found is these tools essentially often work no better than random, especially when it comes to some consequential decisions such as job automation. So basically deciding who gets to be called on the next level of like a job interview or who is rejected, right as soon as they submit the CV. And so, these are very, very consequential decisions and we felt like there is a lot of snake oil in part because people don't distinguish between applications that have worked really well or where we have seen tremendous advances such as generative AI and applications where essentially we've stalled for a number of decades and these tools don't really work as claimed by the developers.
Eric Topol (06:55):
I mean the way you partition that, the snake oil, which is a tough metaphor, and you even show the ad from 1905 of snake oil in the book. You're really getting at predictive AI and how it is using old tools and selling itself as some kind of breakthrough. Before I challenge that, are we going to be able to predict things? By the way, using generative AI, not as you described, but I would like to go through a few examples of how bad this has been and since a lot of our listeners and readers are in the medical world or biomedical world, I'll try to get to those. So one of the first ones you mentioned, which I completely agree, is how prediction of Covid from the chest x-ray and there were thousands of these studies that came throughout the pandemic. Maybe you could comment about that one.
Some Flagrant Examples
Sayash Kapoor (08:04):
Absolutely. Yeah, so this is one of my favorite examples as well. So essentially Michael Roberts and his team at the University of Cambridge a year or so after the pandemic looked back at what had happened. I think at the time there were around 500 studies that they included in the sample. And they looked back to see how many of these would be useful in a clinical setting beyond just the scope of writing a research paper. And they started out by using a simple checklist to see, okay, are these tools well validated? Does the training and the testing data, is it separate? And so on. So they ran through the simple checklist and that excluded all but 60 of these studies from consideration. So apart from 60 studies, none of these other studies even passed a very, very basic criteria for being included in the analysis. Now for these 60, it turns out that if you take a guess about how many were useful, I'm pretty confident most cases would be wrong.
(09:03):
There were exactly zero studies that were useful in a clinically relevant setting. And the reasons for this, I mean in some cases the reasons were as bizarre as training a machine learning model to predict Covid where all of the positive samples of people who had Covid were from adults. But all of the negative samples of people who didn't have Covid were from children. And so, essentially claiming that the resulting classifier can predict who has Covid is bizarre because all the classifier is doing is looking at the checks history and basically predicting which x-ray belongs to a child versus an adult. And so, this is the sort of error in some cases we saw duplicates in the training and test set. So you have the same person that is being used for training the model and that it is also used for evaluating the model. So simply memorizing a given sample of x-rays would be enough to achieve a very high performance. And so, for issues like these, I think all 60 of these studies prove to be not useful in a clinically relevant setting. And I think this is sort of the type of pattern that we've seen over and over again.
Eric Topol (10:14):
Yeah, and I agree with you on that point. I mean that was really a flagrant example and that would fulfill your title of your book, which as I said is a very tough title. But on page 29, and we'll have this in the post. You have a figure, the landscape of AI snake oil, hype, and harm.
And the problem is there is nothing good in this landscape. So on the y-axis you have works, hype, snake oil going up on the y-axis. And on the x-axis, you have benign and harmful. So the only thing you have that works and that's benign is autocomplete. I wouldn't say that works. And then you have works facial recognition for surveillance is harmful. This is a pretty sobering view of AI. Obviously, there's many things that are working that aren't on this landscape. So I just would like to challenge, are you a bit skewed here and only fixating on bad things? Because this diagram is really rough. I mean, there's so much progress in AI and you have in here you mentioned the predicting civil wars, and obviously we have these cheating detection, criminal risk prediction. I mean a lot of problems, video interviews that are deep fakes, but you don't present any good things.
Optimism on Generative AI
Sayash Kapoor (11:51):
So to be clear, I think both Arvind and are somewhat paradoxically optimistic about the future of generative AI. And so, the decision to focus on snake oil was a very intentional one from our end. So in particular, I think at various places in the book we outline why we're optimistic, what types of applications we think we're optimistic about as well. And the reason we don't focus on them is that it basically comes down to the fact that no one wants to read a book that has 300 pages about the virtues of spellcheck or AI for code generation or something like that. But I think I completely agree and acknowledge that there are lots of positive applications that didn't make the cut for the book as well. That was because we wanted people to come to this from a place of skepticism so that they're not fooled by the hype.
(12:43):
Because essentially we see even these positive uses of AI being lost out if people have unrealistic expectations from what an AI tool should do. And so, pointing out snake oil is almost a prerequisite for being able to use AI productively in your work environment. I can give a couple of examples of where or how we've sort of manifested this optimism. One is AI for coding. I think writing code is an application that I do, at least I use AI a lot. I think almost half of the code I write these days is generated, at least the first draft is generated using AI. And yet if I did not know how to program, it would be a completely different question, right? Because for me pointing out that, oh, this syntax looks incorrect or this is not handling the data in the correct way is as simple as looking at a piece of code because I've done this a few times. But if I weren't an expert on programming, it would be completely disastrous because even if the error rate is like 5%, I would have dozens of errors in my code if I'm using AI to generate it.
(13:51):
Another example of how we've been using it in our daily lives is Arvind has two little kids and he's built a number of applications for his kids using AI. So I think he's a big proponent of incorporating AI into children's lives as a force for good rather than having a completely hands-off approach. And I think both of these are just two examples, but I would say a large amount of our work these days occurs with the assistance of AI. So we are very much optimistic. And at the same time, I think one of the biggest hindrances to actually adopting AI in the real world is not understanding its limitations.
Eric Topol (14:31):
Right. Yeah, you say in the book quote, “the two of us are enthusiastic users of generative AI, both in our work and our personal lives.” It just doesn't come through as far as the examples. But before I leave the troubles of predictive AI, I liked to get into a few more examples because that's where your book shines in convincing that we got some trouble here and we need to be completely aware. So one of the most famous, well, there's a couple we're going to get into, but one I'd like to review with you, it's in the book, is the prediction of sepsis in the Epic model. So as you know very well, Epic is the most used IT and health systems electronic health records, and they launched never having published an algorithm that would tell when the patient was hospitalized if they actually had sepsis or risk of sepsis. Maybe you could take us through that, what you do in the book, and it truly was a fiasco.
The Sepsis Debacle
Sayash Kapoor (15:43):
Absolutely. So I think back in 2016/2017, Epic came up with a system that would help healthcare providers predict which patients are most at risk of sepsis. And I think, again, this is a very important problem. I think sepsis is one of the leading causes of death worldwide and even in the US. And so, if we could fix that, I think it would be a game changer. The problem was that there were no external validations of this algorithm for the next four years. So for four years, between 2017 to 2021, the algorithm wasn't used by hundreds of hospitals in the US. And in 2021, a team from University of Michigan did this study in their own hospital to see what the efficacy of the sepsis prediction model is. They found out that Epic had claimed an AUC of between 0.76 and 0.83, and the actual AUC was closer to 0.6, and AUC of 0.5 is making guesses at random.
(16:42):
So this was much, much worse than the company's claims. And I think even after that, it still took a year for sepsis to roll back this algorithm. So at first, Epic's claims were that this model works well and that's why hospitals are adopting it. But then it turned out that Epic was actually incentivizing hospitals to adopt sepsis prediction models. I think they were giving credits of hundreds of thousands of dollars in some cases. If a hospital satisfied a certain set of conditions, one of these conditions was using a sepsis prediction model. And so, we couldn't really take their claims at face value. And finally in October 2022, Epic essentially rolled back this algorithm. So they went from this one size fits all sepsis prediction model to a model that each hospital has to train on its own data, an approach which I think is more likely to work because each hospital's data is different. But it's also more time consuming and expensive for the hospitals because all of a sudden you now need your own data analysts to be able to roll out this model to be able to monitor it.
(17:47):
I think this study also highlights many of the more general issues with predictive AI. These tools are often sold as if they're replacements for an existing system, but then when things go bad, essentially they're replaced with tools that do far less. And companies often go back to the fine print saying that, oh, we should always deploy it with the human in the loop, or oh, it needs to have these extra protections that are not our responsibility, by the way. And I think that gap between what developers claim and how the tool actually works is what is most problematic.
Eric Topol (18:21):
Yeah, no, I mean it's an egregious example, and again, it fulfills like what we discussed with statistics, but even worse because it was marketed and it was incentivized financially and there's no doubt that some patients were completely miscategorized and potentially hurt. The other one, that's a classic example that went south is the Optum UnitedHealth algorithm. Maybe you could take us through that one as well, because that is yet another just horrible case of how people were discriminated against.
The Infamous Optum Algorithm
Sayash Kapoor (18:59):
Absolutely. So Optum, another health tech company created an algorithm to prioritize high risk patients for preemptive care. So I think it was around when Obamacare was being introduced that insurance networks started looking into how they could reduce costs. And one of the main ways they identified to reduce costs is basically preemptively caring for patients who are extremely high risk. So in this case, they decided to keep 3% of the patients in the high risk category and they built a classifier to decide who's the highest risk, because potentially once you have these patients, you can proactively treat them. There might be fewer emergency room visits, there might be fewer hospitalizations and so on. So that's all fine and good. But what happened when they implemented the algorithm was that every machine learning model needs like the target variable, what is being predicted at the end of the day. What they decided to predict was how much patient would pay, how much would they charge, what cost the hospital would incur if they admitted this patient.
(20:07):
And they essentially use that to predict who should be prioritized for healthcare. Now unsurprisingly, it turned out that white patients often pay a lot more or are able to pay a lot more when it comes to hospital visits. Maybe it's because of better insurance or better conditions at work that allow them to take leave and so on. But whatever the mechanism is, what ended up happening with this algorithm was I think black patients with the same level of healthcare prognosis were half as likely or about much less likely compared to white ones of getting enrolled in this high risk program. So they were much less likely to get this proactive care. And this was a fantastic study by Obermeyer, et al. It was published in Science in 2019. Now, what I think is the most disappointing part of this is that Optum did not stop using this algorithm after this study was released. And that was because in some sense the algorithm was working precisely as expected. It was an algorithm that was meant to lower healthcare costs. It wasn't an algorithm that was meant to provide better care for patients who need it most. And so, even after this study was rolled out, I think Optum continued using this algorithm as is. And I think as far as I know, even today this is or some version of this algorithm is still in use across the network of hospitals that Optum serves.
Eric Topol (21:31):
No, it's horrible the fact that it was exposed by Ziad Obermeyer’s paper in Science and that nothing has been done to change it, it's extraordinary. I mean, it's just hard to imagine. Now you do summarize the five reasons predictive AI fails in a nice table, we'll put that up on the post as well. And I think you've kind of reviewed that as these case examples. So now I get to challenge you about predictive AI because I don't know that such a fine line between that and generative AI are large language models. So as you know, the group at DeepMind and now others have done weather forecasting with multimodal large language models and have come up with some of the most accurate weather forecasting we've ever seen. And I've written a piece in Science about medical forecasting. Again, taking all the layers of a person's data and trying to predict if they're high risk for a particular condition, including not just their electronic record, but their genomics, proteomics, their scans and labs and on and on and on exposures, environmental.
Multimodal A.I. in Medicine
(22:44):
So I want to get your sense about that because this is now a coalescence of where you took down predictive AI for good reasons, and then now these much more sophisticated models that are integrating not just large data sets, but truly multimodal. Now, some people think multimodal means only text, audio, speech and video images, but here we're talking about multimodal layers of data as for the weather forecasting model or earthquake prediction or other things. So let's get your views on that because they weren't really presented in the book. I think they're a positive step, but I want to see what you think.
Sayash Kapoor (23:37):
No, absolutely. I think maybe the two questions are sort of slightly separate in my view. So for things like weather forecasting, I think weather forecasting is a problem that's extremely tenable for generative AI or for making predictions about the future. And I think one of the key differences there is that we don't have the problem of feedback loops with humans. We are not making predictions about individual human beings. We are rather making predictions about what happens with geological outcomes. We have good differential equations that we've used to predict them in the past, and those are already pretty good. But I do think deep learning has taken us one step further. So in that sense, I think that's an extremely good example of what doesn't really fit within the context of the chapter because we are thinking about decisions thinking about individual human beings. And you rightly point out that that's not really covered within the chapter.
(24:36):
For the second part about incorporating multimodal data, genomics data, everything about an individual, I think that approach is promising. What I will say though is that so far we haven't seen it used for making individual decisions and especially consequential decisions about human beings because oftentimes what ends up happening is we can make very good predictions. That's not in question at all. But even with these good predictions about what will happen to a person, sometimes intervening on the decision is hard because oftentimes we treat prediction as a problem of correlations, but making decisions is a problem of causal estimation. And that's where those two sort of approaches disentangle a little bit. So one of my examples, favorite examples of this is this model that was used to predict who should be released before screening when someone comes in with symptoms of pneumonia. So let's say a patient comes in with symptoms of pneumonia, should you release them on the day of?
(25:39):
Should you keep them in the hospital or should you transfer them to the ICU? And these ML researchers were basically trying to solve this problem. They found out that the neural network model they developed, this was two decades ago, by the way. The neural network model they developed was extremely accurate at predicting who would basically have a high risk of having complications once they get pneumonia. But it turned out that the model was saying essentially that anyone who comes in who has asthma and who comes in with symptoms of pneumonia is the lowest risk patient. Now, why was this? This was because when in the past training data, when some such patients would come into the hospital, these patients would be transferred directly to the ICU because the healthcare professionals realized that could be a serious condition. And so, it turned out that actually patients who had asthma who came in with symptoms of pneumonia were actually the lowest risk amongst the population because they were taken such good care of.
(26:38):
But now if you use this prediction that a patient comes in with symptoms of pneumonia and they have asthma, and so they're low risk, if you use this to make a decision to send them back home, that could be catastrophic. And I think that's the danger with using predictive models to make decisions about people. Now, again, I think the scope and consequences of decisions also vary. So you could think of using this to surface interesting patterns in the data, especially at a slightly larger statistical level to see how certain subpopulations behave or how certain groups of people are likely to develop symptoms or whatever. But I think when as soon as it comes to making decisions about people, the paradigm of problem solving changes because as long as we are using correlational models, I think it's very hard to say what will happen if we change the conditions, what will happen if the decision making mechanism is very different from one where the data was collected.
Eric Topol (27:37):
Right. No, I mean where we agree on this is that at the individual level, using multimodal AI with all these layers of data that have now recently become available or should be available, that has to be compared ideally in a randomized trial with standard of care today, which doesn't use any of that. And to see whether or not that decision's made, does it change the natural history and is it an advantage, that's yet to be done. And I agree, it's a very promising pathway for the future. Now, I think you have done what is a very comprehensive sweep on the predictive AI failures. You've mentioned here in our discussion, your enthusiasm and in the book about generative AI positive features and hope and excitement perhaps even. You didn't really yet, we haven't discussed much on the content moderation AI that you have discreetly categorized. Maybe you could just give us the skinny on your sense of that.
Content Moderation AI
Sayash Kapoor (28:46):
Absolutely. So content moderation AI is AI that's used to sort of clean up social media feeds. Social media platforms have a number of policies about what's allowed and not allowed on the platforms. Simple things such as spam are obviously not allowed because let's say people start spamming the platform, it becomes useless for everyone. But then there are other things like hate speech or nudity or pornography and things like that, which are also disallowed on most if not all social media platforms today. And I think a lot of the ways in which these policies are enforced today is using AI. So you might have an AI model that runs every single time you upload a photo to Facebook, for instance. And not just one perhaps hundreds of such models to detect if it has nudity or hate speech or any of these other things that might violate the platform's terms of service.
(29:40):
So content moderation AI is AI that's used to make these decisions. And very often in the last few years we've seen that when something gets taken down, for instance, Facebook deletes a post, people often blame the AI for having a poor understanding. Let's say of satire or not understanding what's in the image to basically say that their post was taken down because of bad AI. Now, there have been many claims that content moderation AI will solve social media's problems. In particular, we've heard claims from Mark Zuckerberg who in a senate testimony I think back in 2018, said that AI is going to solve most if not all of their content moderation problems. So our take on content moderation AI is basically this. AI is very, very useful for solving the simple parts of content moderation. What is a simple part? So basically the simple parts of content moderation are, let's say you have a large training data of the same type of policy violation on a platform like Facebook.
(30:44):
If you have large data sets, and if these data sets have a clear line in the sand, for instance, with nudity or pornography, it's very easy to create classifiers that will automate this. On the other hand, the hard part of content moderation is not actually just creating these AI models. The hard part is drawing the line. So when it comes to what is allowed and not allowed on platforms, these platforms are essentially making decisions about speech. And that is a topic that's extremely fraught. It's fraught in the US, it's also fraught globally. And essentially these platforms are trying to solve this really hard problem at scale. So they're trying to come up with rules that apply to every single user of the platform, like over 3 billion users in the case of Facebook. And this inevitably has these trade-offs about what speech is allowed versus disallowed that are hard to say one way or the other.
(31:42):
They're not black and white. And what we think is that AI has no place in this hard part of content moderation, which is essentially human. It's essentially about adjudicating between competing interests. And so, when people claim that AI will solve these many problems of content moderation, I think what they're often missing is that there's this extremely large number of things you need to do to get content moderation right. AI solves one of these dozen or so things, which is detecting and taking down content automatically, but all of the rest of it involves essentially human decisions. And so, this is sort of the brief gist of it. There are also other problems. For example, AI doesn't really work so well for low resource languages. It doesn't really work so well when it comes to nuances and so on that we discussed in the book. But we think some of these challenges are solvable in the medium to long term. But these questions around competing interests of power, I think are beyond the domain of AI even in the medium to long term.
Age 28! and Career Advice
Eric Topol (32:50):
No, I think you nailed that. I think this is an area that you've really aptly characterized and shown the shortcomings of AI and how the human factor is so critically important. So what's extraordinary here is you're just 28 and you are rocking it here with publications all over the place on reproducibility, transparency, evaluating generative AI, AI safety. You have a website on AI snake oil that you're collecting more things, writing more things, and of course you have the experience of having worked in the IT world with Facebook and also I guess also Columbia. So you're kind of off to the races here as one of the really young leaders in the field. And I am struck by that, and maybe you could comment about the inspiration you might provide to other young people. You're the youngest person I've interviewed for Ground Truths, by the way, by a pretty substantial margin, I would say. And this is a field where it attracts so many young people. So maybe you could just talk a bit about your career path and your advice for people. They may be the kids of some of our listeners, but they also may be some of the people listening as well.
Sayash Kapoor (34:16):
Absolutely. First, thank you so much for the kind words. I think a lot of this work is with collaborators without whom of course, I would never be able to do this. I think Arvind is a great co-author and supporter. I think in terms of my career parts, it was sort of like a zigzag, I would say. It wasn't clear to me when I was an undergrad if I wanted to do grad school or go into the industry, and I sort of on a whim went to work at Facebook, and it was because I'd been working on machine learning for a little bit of time, and I just thought, it's worth seeing what the other side has to offer beyond academia. And I think that experience was very, very helpful. One of the things, I talked to a lot of undergrads here at Princeton, and one of the things I've seen people be very concerned about is, what is the grad school they're going to get into right after undergrad?
(35:04):
And I think it's not really a question you need to answer now. I mean, in some cases I would say it's even very helpful to have a few years of industry experience before getting into grad school. That has definitely, at least that has been my experience. Beyond that, I think working in a field like AI, I think it's very easy to be caught up with all of the new things that are happening each day. So I'm not sure if you know, but AI has I think over 500-1,000 new archive papers every single day. And with this rush, I think there's this expectation that you might put on yourself on being successful requires a certain number of publications or a certain threshold of things. And I think more often than not, that is counterproductive. So it has been very helpful for me, for example, to have collaborators who are thinking long term, so this book, for instance, is not something that would be very valued within the CS community, I would say. I think the CS community values peer-reviewed papers a lot more than they do books, and yet we chose to write it because I think the staying power of a book or the longevity of a book is much more than any single paper could do. So the other concrete thing I found very helpful is optimizing for a different metric compared to what the rest of the community seems to be doing, especially when it comes to fast moving fields like AI.
Eric Topol (36:29):
Well, that last piece of advice is important because I think too often people, whether it's computer scientists, life scientists, whoever, they don't realize that their audience is much broader. And that reaching the public with things like a book or op-eds or essays, varied ways that are intended for public consumption, not for, in this case, computer scientists. So that's why I think the book is a nice contribution. I don't like the title because it's so skewed. And also the content is really trying to hammer it at home. I hope you write a sequel book on the positive sides of AI. I did want to ask you, when I read the book, I thought I heard your voice. I thought you had written the book, and Arvind maybe did some editing. You wrote about Arvind this and Arvind that. Did you write the first draft of the book and then he kind of came along?
Sayash Kapoor (37:28):
No, absolutely not. So the way we wrote the book was we basically started writing it in parallel, and I wrote the first draft of half the chapters and he wrote the first draft of the other half, and that was essentially all the way through. So we would sort of write a draft, pass it to the other person, and then keep doing this until we sent it to our publishers.
Eric Topol (37:51):
Okay. So I guess I was thinking of the chapters you wrote where it came through. I'm glad that it was a shared piece of work because that's good, because that’s what co-authoring is all about, right? Well, Sayash, it's really been a joy to meet you and congratulations on this book. I obviously have expressed my objections and my disagreements, but that's okay because this book will feed the skeptics of AI. They'll love this. And I hope that the positive side, which I think is under expressed, will not be lost and that you'll continue to work on this and be a conscience. You may know I've interviewed a few other people in the AI space that are similarly like you, trying to assure its safety, its transparency, the ethical issues. And I think we need folks like you. I mean, this is what helps get it on track, keeping it from getting off the rails or what it shouldn't be doing. So keep up the great work and thanks so much for joining.
Sayash Kapoor (39:09):
Thank you so much. It was a real pleasure.
************************************************
Thanks for listening, reading or watching!
The Ground Truths newsletters and podcasts are all free, open-access, without ads.
Please share this post/podcast with your friends and network if you found it informative!
Voluntary paid subscriptions all go to support Scripps Research. Many thanks for that—they greatly helped fund our summer internship programs for 2023 and 2024.
Thanks to my producer Jessica Nguyen and Sinjun Balabanoff for audio and video support at Scripps Research.
Note: you can select preferences to receive emails about newsletters, podcasts, or all I don’t want to bother you with an email for content that you’re not interested in.

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
Francis Collins: On Truth, Science, Faith and Trust
11 sept.· Ground Truths
Francis Collins is a veritable national treasure. He directed the National Institutes of Health from 2009 to 2021. Prior to that he led the National Human Genetics Research Institute (NHGRI) from 1997-2009, during which the human genome was first sequenced. As a physician-scientist, he has made multiple seminal discoveries on the genetic underpinnings of cystic fibrosis, Huntington’s disease, neurofibromatosis, progeria, and others. This brief summary is barely scratching the surface oh his vast contributions to life science and medicine.
A video clip from our conversation on hepatitis C. Full videos of all Ground Truths podcasts can be seen on YouTube here. The audios are also available on Apple and Spotify.
Transcript with external inks and links to audio
Eric Topol (00:06):
Well, I am really delighted to be able to have our conversation with Francis Collins. This is Eric Topol with Ground Truths and I had the chance to first meet Francis when he was on the faculty at the University of Michigan when I was a junior faculty. And he gave, still today, years later, we're talking about 40 years later, the most dazzling Grand Rounds during his discovery of cystic fibrosis. And Francis, welcome, you inspired me and so many others throughout your career.
Francis Collins (00:40):
Well, Eric, thank you and you've inspired me and a lot of other people as well, so it's nice to have this conversation with you in the Ground Truths format.
Eric Topol (00:49):
Well, thank you. We're at the occasion of an extraordinary book you put together. It's the fifth book, but it stands out quite different from the prior books as far as I can tell. It's called The Road to Wisdom: On Truth, Science, Faith and Trust, these four essential goods that build upon each other. And it's quite a book, Francis, I have to say, because you have these deep insights about these four critical domains and so we'll get into them. But I guess the first thing I thought I'd do is just say, how at some point along the way you said, “the goal of this book is to turn the focus away from hyperpartisan politics and bring it back to the most important sources of wisdom: truth, science, faith and trust, resting upon a foundation of humility, knowledge, morality, and good judgment.” So there's a lot there. Maybe you want to start off with what was in the background when you were putting this together? What were you really aiming at getting across?
Reflections on Covid
Francis Collins (02:06):
I'm glad to, and it's really a pleasure to have a chance to chat with you about this. I guess before Covid came along, I was probably a bit of a naive person when it came to how we make decisions. Yeah, I knew there were kind of wacky things that had gone out there from time to time, but I had a sort of Cartesian attitude that we were mostly rational actors and when presented with evidence that's been well defended and validated that most people will say, okay, I know what to do. Things really ran off the rails in the course of Covid. It was this remarkable paradox where, I don't know what you would say, but I would say the development of the vaccines that were safe and highly effective in 11 months using the mRNA platform was one of the most stunning achievements of science in all of history up until now.
Francis Collins (03:02):
And yet 50 million Americans decided they didn't want any part of it because of information that came to them that suggested this was not safe or there was conspiracies behind it, or maybe the syringes had chips that Bill Gates had put in there or all manner of other things that were being claimed. And good honorable people were distracted by that, lost their trust in other institutions like the CDC, maybe like the government in general like me, because I was out there a lot trying to explain what we knew and what we didn't know about Covid. And as a consequence of that, according to Kaiser Family Foundation, more than 230,000 people died between June of 2021 and April of 2022 because of a decision to reject the opportunity for vaccines that were at that time free and widely available. That is just an incredibly terribly tragic thing to say.
Francis Collins (04:03):
More than four times the number of people who died, Americans who died in the Vietnam War are in graveyards unnecessarily because we lost our anchor to truth, or at least the ability to discern it or we couldn't figure out who to trust while we decided science was maybe not that reliable. And people of faith for reasons that are equally tragic were among those most vulnerable to the misinformation and the least likely therefore, to take advantage of some of these lifesaving opportunities. It just completely stunned me, Eric, that this kind of thing could happen and that what should have been a shared sense of working against the real enemy, which was the SARS-CoV-2 virus became instead a polarized, divisive, vitriolic separation of people into separate camps that were many times driven more by politics than by any other real evidence. It made me begin to despair for where we're headed as a country if we can't figure out how to turn this around.
Francis Collins (05:11):
And I hadn't really considered it until Covid how serious this was and then I couldn't look away. And so, I felt if I have a little bit of credibility after having stepped down after 12 years as the NIH Director and maybe a chance to influence a few people. I just have to try to do something to point out the dangers here and then to offer some suggestions about what individuals can do to try to get us back on track. And that's what this book is all about. And yeah, it's called The Road to Wisdom because that's really how I want to think of all this in terms of truth and science and faith and trust. They all kind of give you the opportunities to acquire wisdom. Wisdom is of course knowledge, but it's not just knowledge, it's also understanding it has a moral character to it. It involves sophisticated judgment about difficult situations where there isn't an obvious answer. We need a lot more of that, it seems we’re at short supply.
Deconvoluting Truth
Eric Topol (06:13):
Well, what I really loved about the book among many things was how you broke things down in just a remarkably thoughtful way. So truth, you have this great diagram like a target with the four different components.
in the middle, necessary truth. And then as you go further out, firmly established facts, then uncertainty and then opinion, and truth is not a dichotomous by any means. And you really got that down and you explained each of these different facets of truth with great examples. And so, this among many other things that you broke down, it wasn't just something that you read somewhere, you really had to think this through and perhaps this experience that we all went through, but especially you. But because you bring so much of the book back to the pandemic at times with each of the four domains, so that and the spider web. The spider web of where your core beliefs
are and then the ones further out on the web and you might be able to work on somebody out further periphery, but it's pretty hard if you're going to get to them in the middle where their main thing is science is untrustworthy or something like that.
Eric Topol (07:36):
So how did you synthesize these because the graphics are quite extraordinary?
Francis Collins (07:44):
Well, I will say the artist for the graphics is a remarkable graphic design student at the University of Michigan who happens to be my granddaughter. So it was nice having that ability to have my scratches turned into something actually looks like artwork. The concepts I got to say, Eric, I was feeling pretty unsure of myself. I never took a course in philosophy. I know there are people who've spent their entire careers going all the way back to Socrates and on up until now about what does truth mean and here's this scientist guy who's trying to say, well, let me tell you what I think about it. I'm glad to hear that you found these circles useful. They have been very useful for me and I hadn't thought about it much until I tried to put it in some sort of framework and a lot of the problems we have right now where somebody says, well, that might be true for you, but it's not true for me, that's fine if you're talking about an opinion, like whether that movie was really good or not.
Francis Collins (08:43):
But it's not fine if it's about an established fact, like the fact that climate change is real and that human activity is the main contributor to the fact that we've warmed up dramatically since 1950. I'm sorry, that's just true. It doesn't care how you feel about it, it's just true. So that zone of established facts is where I think we have to re-anchor ourselves again when something's in that place. I'm sorry, you can't just decide you don't like it, but in our current climate and maybe postmodernism has crept in all kinds of ways we're not aware of, the idea that there is such a thing as objective truth even seems to be questioned in some people's minds. And that is the path towards a terrible future if we can't actually decide that we have, as Jonathan Rauch calls it, a constitution of knowledge that we can depend on, then where are we?
Eric Topol (09:37):
Well, and I never heard of the term old facts until the pandemic began and you really dissect that issue and like you, I never had anticipated there would be, I knew there was an anti-science, anti-vaccine sector out there, but the fact that it would become so strong, organized, supported, funded, and vociferous, it's just looking back just amazing. I do agree with the statement you made earlier as we were talking and in the book, “the development of mRNA vaccines for Covid in record time as one of the greatest medical achievements in human history.” And you mentioned besides the Kaiser Family Foundation, but the Commonwealth Fund, a bipartisan entity saved three million lives in the US, eighteen million hospitalizations. I mean it's pretty extraordinary. So besides Covid, which we may come back to, but you bring in everything, you bring in AI. So for example, you quoted the fellow from Google who lost his job and you have a whole conversation with Blake Lemoine and maybe you can give us obviously, where is AI in the truth and science world? Where do you stand there and what were you thinking when you included his very interesting vignette?
Perspective on A.I.
Francis Collins (11:17):
Well, I guess I was trying to talk about where are we actually at the point of AGI (artificial general intelligence) having been achieved? That is the big question. And here's Blake Lemoine who claimed based on this conversation that I quote in the book between him and the Google AI apparatus called LaMDA. Some pretty interesting comments where LaMDA is talking about having a soul and what its soul looks like and it's a portal to all sorts of other dimensions, and I can sort of see why Blake might've been taken in, but I can also see why a lot of people said, oh, come on, this is of course what an AI operation would say just by scanning the internet and picking out what it should say if it's being asked about a soul. So I was just being a little provocative there. My view of AI, Eric, is that it's applications to science and medicine are phenomenal and we should embrace them and figure out ways to speed them up in every way we can.
Francis Collins (12:17):
I mean here at NIH, we have the BRAIN Initiative that's trying to figure out how your brain works with those 86 billion neurons and all their connections. We're never going to sort that out without having AI tools to help us. It's just too complicated of a problem. And look what AI is doing and things like imaging radiologists are going to be going out of business and the pathologists may not be too far behind because when it comes to image analysis, AI is really good at that, and we should celebrate that. It's going to improve the speed and accuracy of all kinds of medical applications. I think what we have to worry about, and I'm not unique in saying this, is that AI when applied to a lot of things kind of depends on what's known and goes and scrapes through the internet to pull that out. And there's a lot of stuff on the internet that's wrong and a lot of it that's biased and certainly when it comes to things like healthcare, the bias in our healthcare system, health disparities, inadequacies, racial inequities are all in there too, and if we're going to count on AI to fix the system, it's building on a cracked foundation.
Francis Collins (13:18):
So we have to watch out for that kind of outcome. But for the most part, generative AI it’s taking really exciting difficult problems and turning them into solutions, I'm all for it, but let's just be very careful here as we watch how it might be incorporating information that's wrong and we won't realize it and we'll start depending on it more than we should.
Breathtaking Advances
Eric Topol (13:42):
Yeah, no, that's great. And you have some commentary on all the major fronts that we're seeing these days. Another one that is a particularly apropos is way back when you were at Michigan and the years before that when you were warming up to make some seminal gene discoveries and cystic fibrosis being perhaps the first major one. You circle back in the book to CRISPR genome editing and how the success story to talk about some extraordinary science to be able to have a remedy, a cure potentially for cystic fibrosis. So maybe you could just summarize that. I mean that's in your career to see that has to be quite remarkable.
Francis Collins (14:32):
It is breathtaking, Eric. I mean I sort of like to think of three major developments just in the last less than 20 years that I never dreamed would happen in my lifetime. One was the ability to make stem cells from people who are walking around from a skin biopsy or a blood sample that are pluripotent. My whole lab studies diabetes, our main approach is to take induced pluripotent stem cells from people whose phenotypes we know really well and differentiate them into beta cells that make insulin and see how we can figure out how the genetics and other aspects of this determine whether something is going to work properly or not. I mean that's just astounding. The second thing is the ability to do single cell biology.
Francis Collins (15:16):
Which really 15 years ago you just had to have a bunch of cells and studying diabetes, we would take a whole eyelid and grind it up and try to infer what was there, ridiculous. Now we can look at each cell, we even can look at each cell in terms of what's its neighbor, does the beta cell next to an alpha cell behave the same way as a beta cell next to a duct? We can answer those questions, and of course the third thing is CRISPR and gene editing and of course the first version of CRISPR, which is the knockout of a gene was exciting enough, but the ability to go in and edit without doing a double stranded break and actually do a search and replace operation is what I'm truly excited about when it comes to rare genetic diseases including one that we work on progeria, which is this dramatic form of premature aging that is caused almost invariably by a C to T mutation in exon 11 of the LMNA gene and for which we have a viable strategy towards a human clinical trial of in vivo gene editing for kids with this disease in the next two years.
Eric Topol (16:24):
Yeah, it's just the fact that we were looking at potential cures for hundreds and potentially even thousands of diseases where there was never a treatment. I mean that's astounding in itself, no less, the two other examples. The fact that you can in a single cell, you can not only get the sequence of DNA and RNA and methylation and who would've ever thought, and then as you mentioned, taking white cells from someone's blood and making pluripotent stem cells. I mean all these things are happening now at scale and you capture this in the book.
On Humility and Trust
Now the other thing that you do that I think is unique to you, I don't know if it's because of your background in growing up in Staunton, Virginia, a very different type of world, but you have a lot of humility in the book. You go over how you got snickered by Bill Maher, how you had a graduate student who was fabricating images and lots of things, how you might not have communicated about Covid perhaps as well as could. A lot of our colleagues are not able to do that. They don't ever have these sorts of things happening to them. And this humility which comes across especially in the chapter on trust where you break down who do you trust, humility is one of the four blocks as you outlined, competence, integrity, and aligned value
So maybe can you give us a little brief lesson on humility?
Eric Topol (18:06):
Because it's checkered throughout the book and it makes it this personal story that you're willing to tell about yourself, which so few of us are willing to do.
Francis Collins (18:17):
Well, I don't want to sound proud about my humility. That would not be a good thing because I’m not, but thanks for raising it. I do think when we consider one of the reasons we decide to trust somebody, that it does have that humility built into it. Somebody who's willing to say, I don't know. Somebody's willing to say I'm an expert on this issue, but that other issue you just asked me about, I don't know any more than anybody else and you should speak to someone else. We don't do that very well. We tend to plunge right in and try to soak it up. I do feel when it comes to Covid, and I talk about this in the book a bit, that I was one of those trying to communicate to the public about what we think are going to be the ways to deal with this worst pandemic in more than a century.
Francis Collins (19:06):
And I wish Eric, I had said more often what I'm telling you today is the best that the assembled experts can come up with, but the data we have to look at is woefully inadequate. And so, it very well could be that what I'm telling you is wrong, when we get more data, I will come back to you as soon as we have something better and we'll let you know, but don't be surprised if it's different and that will not mean that we are jerking you around or we don't know what we're talking about. It's like this is how science works. You are watching science in real time, even though it's a terrible crisis, it's also an opportunity to see how it works. I didn't say that often enough and neither did a lot of the other folks who were doing the communicating. Of course, the media doesn't like to give you that much time to say those things as you well know, but we could have done a better job of preparing people for uncertainty and maybe there would've been less of a tendency for people to just decide, these jokers don't know what they're talking about.
Francis Collins (20:10):
I'm going to ignore them from now on. And that was part of what contributed to those 230,000 unnecessary deaths, it was just people losing their confidence in the information they were hearing. That's a source of grief from my part.
His Diagnosis And Treatment for Prostate Cancer
Eric Topol (20:24):
Well, it's great and a lesson for all of us. And the other thing that along with that is remarkable transparency about your own health, and there's several things in there, but one that coincides. You mentioned in the book, of course, you wrote an op-ed in the Washington Post back in April 2024 about your diagnosis of prostate cancer. So you touched on it in the book and maybe you could just update us about this because again, you're willing to tell your story and trying to help others by the experiences that you've been through.
Francis Collins (21:00):
Well, I sure didn't want to have that diagnosis happen, but once it did, it certainly felt like an opportunity for some education. We men aren't that good about talking about issues like this, especially when it involves the reproductive system. So going out and being public and saying, yep, I had a five year course of watching to see if something was happening, and then the slow indolent cancer suddenly decided it wasn't slow and indolent anymore. And so, I'm now having my prostate removed and I think I'm a success story, a poster boy for the importance of screening. If I hadn't gone through that process of PSA followed by imaging by MRI followed by targeted biopsies, so you're actually sampling the right place to see if something's going on. I probably would know nothing about it right now, and yet incubating within me would be a Gleason category 9 prostate cancer, which has a very high likelihood if nothing was done to become metastatic.
Francis Collins (22:03):
So I wanted that story to be out there. I wanted men who were squeamish about this whole topic to say, maybe this is something to look into. And I've heard a bunch of follow-ups from individuals, but I don't know how much of it impact it hit. I'm glad to say I'm doing really well. I'm four months out now from the surgery, it is now the case I'm pretty much back to the same level of schedule and energy that I had beforehand, and I'm very happy to say that the post-op value of PSA, which is the best measure to see whether you in fact are now cancer free was zero, which is a really nice number.
Eric Topol (22:45):
Wow. Well, the prostate is the curse of men, and I wish we could all have an automated prostatectomy so we don't have to deal with this. It's just horrible.
Francis Collins (22:58):
It was done by a robot. It wasn't quite automated, I have stab wounds to prove that the robot was actually very actively doing what it needed to do, but they healed quickly.
The Promise of Music As Therapy in Medicine
Eric Topol (23:11):
Right. Well, this gets me to something else that you're well known for throughout your career as a musician, a guitarist, a singer, and recently you hooked up with Renée Fleming, the noted opera singer, and you've been into this music is therapy and maybe you can tell us about that. It wasn't necessarily built up much in the book because it's a little different than the main agenda, but I think it's fascinating because who doesn't like music? I mean, you have to be out there if you don't enjoy music, but can you tell us more about that?
Francis Collins (23:53):
Yeah, I grew up in a family where music was very much what one did after dinner, so I learned to play keyboard and then guitar, and that's always been a source of joy and also a source of comfort sometimes when you were feeling a bit down or going through a painful experience. I think we all know that experience where music can get into your heart and your soul in a way that a lot of other things can't. And the whole field of music therapy is all about that, but it's largely been anecdotal since about World War II when it got started. And music therapists will tell you sometimes you try things that work and sometimes they don't and it's really hard to know ahead of time what's going to succeed. But now we have that BRAIN Initiative, which is pushing us into whole new places as far as the neuroscience of the brain, and it's really clear that music has a special kind of music room in the brain that evolution has put there for an important reason.
Francis Collins (24:47):
If we understood that we could probably make music therapy even more scientifically successful and maybe even get third parties to pay for it. All of this became opportunity for building a lot more visibility because of making friends withRenée Fleming, who I hadn't really known until a famous dinner party in 2015 where we both ended up singing to a trio of Supreme Court justices trying to cheer them up after a bent week. And she has become such an incredible partner in this. She's trained herself pretty significantly in neuroscience, and she's a convener and an articulate spokesperson. So over the course of that, we built a whole program called Sound Health that now has invested an additional $35 million worth NIH research to try to see how we can bring together music therapy, musician performers and neuroscientists to learn from each other, speak each other's language and see what we could learn about this particularly interesting input to the human brain that has such power on us and maybe could be harnessed to do even more good for people with chronic pain or people with PTSD, people with dementia where music seems to bring people back to life who'd otherwise seem to have disappeared into the shadows.
Francis Collins (26:09):
It's phenomenal what is starting to happen here, but we're just scratching the surface.
The Big Miss vs Hepatitis C
Eric Topol (26:14):
Well, I share your enthusiasm for that. I mean, it's something that you could think of that doesn't have a whole lot of side effects, but could have a lot of good. Yeah. Well, now before I get back to the book, I did want to cover one other relatively recent op-ed late last year that you wrote about Hepatitis C. Hepatitis C, one of the most important medical advances in the 21st century that we're squandering. Can you tell us about that? Because I think a lot of people don't realize this is a big deal.
Francis Collins (26:47):
It's a really big deal, and I confess I'm a little obsessed about it. So yes, you may regret bringing it up because I'm really going to want to talk about what the opportunity is here, and I am still the lead for the White House in an initiative to try to find the 4 million Americans who are already infected with this virus and get access to them for treatment. The treatment is fantastic, as you just said, one of the most major achievements of medical research, one pill a day for 12 weeks, 95% cure in the real world, essentially no side effects, and yet the cost is quite high and the people who need it many times do not have great healthcare and maybe also in difficult circumstances because you get hepatitis C from infected blood. And the many ways that happens these days are from shared needles from people who are experimenting with intravenous drugs, but they are family too, and many of them now recovering from that, face the irony of getting over their opioid addiction and then looking down the barrel of a really awful final couple of years dying of liver failure. I watched my brother-in-law die of hepatitis C, and it was just absolutely gruesome and heartbreaking.
Francis Collins (28:04):
So this isn't right. And on top of that, Eric, the cost of all this for all those folks who are going to get into liver failure need a transplant or develop liver cancer, this is the most common cause now of liver cancer it is astronomical in the tens of billions of dollars. So you can make a very compelling case, and this is now in the form of legislation sponsored by Senators Cassidy and Van Hollen that in a five-year program we could find and cure most of those people saving tens of thousands of lives and we would save tens of billions of dollars in just 10 years in terms of healthcare that we will not have to pay for. What's not to love here? There's a lot of things that have to be worked out to make it happen. One thing we've already done is to develop, thanks to NIH and FDA, a point of care viral RNA finger stick test for Hep C. You get an answer in less than an hour.
Francis Collins (29:00):
FDA approved that the end of June. That was a big crash program so you can do test and treat in one visit, which is phenomenally helpful for marginalized populations. The other thing we need to do is to figure out how to pay for this and this subscription model, which was piloted in Louisiana, looks like it ought to work for the whole nation. Basically, you ask the companies Gilead and AbbVie to accept a lump sum, which is more than what they're currently making for Medicaid patients and people who are uninsured and people in the prison system and Native Americans and then make the pills available to those four groups for free. They do fine. The companies come out on this and the cost per patient plummets and it gives you the greatest motivation you can imagine to go and find the next person who's infected because it's not going to cost you another dime for their medicine, it's already paid for. That's the model, and I would say the path we're on right now waiting for the congressional budget office to give the final score, it's looking pretty promising we're going to get this done by the end of this year.
The Pledge
Eric Topol (30:04):
Yeah, that's fantastic. I mean, your work there alone is of monumental importance. Now I want to get back to the book the way you pulled it all together. By the way, if anybody's going to write a book about wisdom, it ought to be you, Francis. You've got a lot of it, but you had to think through how are we going to change because there's a lot of problems as you work through the earlier chapters and then the last chapter you come up with something that was surprising to me and that was a pledge for the Road to Wisdom. A pledge that we could all sign, which is just five paragraphs long and basically get on board about these four critical areas. Can you tell us more about the pledge and how this could be enacted and help the situation?
Francis Collins (31:03):
Well, I hope it can. The initial version of this book, I wrote a long piece about what governments should do and what institutions should do and what universities should do and what K through 12 education should do. And then I thought they're not reading this book and I'm not sure any of those folks are really that motivated to change the status quo. Certainly, politicians are not going to solve our current woes. It seems that politics is mostly performance these days and it's not really about governance. So if there's going to be a chance of recovering from our current malaise, I think it's got to come from the exhausted middle of the country, which is about two thirds of us. We're not out there in the shrill screaming edges of the left and the right we're maybe tempted to just check out because it just seems so discouraging, but we're the solution.
Francis Collins (31:56):
So the last chapter is basically a whole series of things that I think an individual could start to do to turn this around. Beginning with doing a little of their own house cleaning of their worldview to be sure that we are re-anchoring to things like objective truths and to loving your neighbor instead of demonizing your neighbor. But yeah, it does go through a number of those things and then it does suggest as a way of making this not just a nice book to read, but something where you actually decide to make a commitment. Look at this pledge. I've tried the pledge out on various audiences so far and I haven't yet really encountered anybody who said, well, those are ridiculous things to ask of people. They're mostly things that make a lot of sense, but do require a commitment. That you are, for instance, you're not going to pass around information on social media in other ways unless you're sure it's true because an awful lot of what's going on right now is this quick tendency for things that are absolutely wrong and maybe anger inducing or fear inducing to go viral where something that's true almost lands with a thud.
Francis Collins (33:07):
Don't be part of that, that's part of this, but also to make an honest effort to reach out to people who have different views from you. Don't stay in your bubble and try to hear their concerns. Listen, not that you're listening in order to give a snappy response, but listen, so you're really trying to understand. We do far too little of that. So the pledge asks people to think about that, and there is a website now which will be as part of the book up on the Braver Angels website and Braver Angels is a group that has made its mission trying to bring together these divided parties across our country and I'm part of them, and you can then go and sign it there and make a public statement that this is who I am, and it will also give you a whole lot of other resources you could start to explore to get engaged in being part of the solution instead of just shaking your head. I think what we're trying to do is to get people to go beyond the point of saying, this isn't the way it should be to saying, this isn't the way I should be. I'm going to try to change myself as part of fixing our society.
Eric Topol (34:14):
Well, I'm on board for this and I hope it creates a movement. This is as you tell the stories in the book, like the fellow that you wrangled with about the pandemic and how you listened to him and it changed your views and you changed his views and this is the health of different opinions and perspectives and we got to get back there. It used to be that way more at least it wasn't always perfect, and as you said in the book, we all have some entrenched biases. We're never going to get rid of all of them, but your wisdom about the road, the pledge here is I think masterful. So I just want to pass on along and I hope listeners will go to the Brave for Angels website and sign up because if we got millions of people to help you on this, that would say a lot about a commitment to a renewed commitment to the way it should be, not the way it is right now. Well, I've covered a bunch of things, of course, Francis, but did I miss something that you're passionate about or in the book or anything that you want to touch on?
Francis Collins (35:32):
Oh my goodness, yeah. You did cover a lot of ground here, including things that I didn't pay much attention to in the book, but I was glad to talk to you about. No, I think we got a pretty good coverage. The one topic in the book that will maybe appeal particularly to believers is a whole chapter about faith because I am concerned that people of faith have been particularly vulnerable to misinformation and disinformation, and yet they stand on a foundation of principles that ought to be the best antidote to most of the meanness that's going on, and just trying to encourage them to recall that and then build upon the strength that they carry as a result of their faith traditions to try to be part of the solution as well.
Eric Topol (36:12):
I'm so glad you mentioned that. It's an important part of the book, and it is also I think something that you were able to do throughout your long tenure at NIH Director that you were able to connect to people across the aisle. You had senators and the Republicans that were so supportive of your efforts to lead NIH and get the proper funding, and it's a unique thing that you're able to connect with people of such different backgrounds, people of really deep commitment to religion and faith and everything else. And that's one of the other things that we talk about Francis here, and many times I gather is we don't have you at the helm anymore at NIH, and we're worried. We're worried because you're a unique diplomat with all this heavy wisdom and it's pretty hard to simulate your ability to keep the NIH whole and to build on it. Do you worry about it at all?
Francis Collins (37:23):
Well, I was privileged to have those 12 years, but I think it was time to get a new perspective in there, and I appreciate you saying those nice things about my abilities. Monica Bertagnolli is also a person of great skill, and I think on the hill she rapidly acquired a lot of fans by her approach, by some of her background. She's from Wyoming, she's a cancer surgeon. She's got a lot of stories to tell that are really quite inspiring. I think though it's just a very difficult time. She walked in at a point where the partisan attitudes about medical research, which we always hoped would kind of stay out of the conversation and become so prominent, a lot of it politically driven, nasty rhetoric on the heels of Covid, which spills over into lots of other areas of medical research and is truly unfortunate. So she's got a lot to deal with there, but I'm not sure I would be much better than she is in trying to continue stay on message, tell the stories about how medical research is saving lives and alleviating suffering, and we're just getting started, and she does that pretty well.
Francis Collins (38:34):
I just hope the people who need to listen are in a listening mood.
Eric Topol (38:38):
Yeah. Well, that's great to hear your perspective. Well, I can't thank you enough for our conversation and moreover for a friendship that's extended many decades now. We're going to be following not just your progeria research and all the other things that you're up to because juggling a bunch of things still, it isn't like you're slowed down at all. And thanks so much for this book. I think it's a gift. I think it's something that many people will find is a pretty extraordinary, thoughtful and easy read. I mean, it's something that I found that you didn't write it for in technical jargon. You wrote it for the public, you wrote it for non-scientists, non-medical people, and I think hopefully that's what's going to help it get legs in terms of what's needed, which is a sign the darn pledge. Thank you.
Francis Collins (39:42):
Eric, thank you. It has been a privilege being your friend for all these years, and this was a really nice interview and I appreciate that you already had carefully read the book and asked some great questions that were fun to try to answer. So thanks a lot.
*******************************************************
Thanks for listening, reading or watching!
The Ground Truths newsletters and podcasts are all free, open-access, without ads.
Please share this post/podcast with your friends and network if you found it informative!
Voluntary paid subscriptions all go to support Scripps Research. Many thanks for that—they greatly helped fund our summer internship programs for 2023 and 2024.
Thanks to my producer Jessica Nguyen and Sinjun Balabanoff for audio and video support at Scripps Research.
Note: you can select preferences to receive emails about newsletters, podcasts, or all I don’t want to bother you with an email for content that you’re not interested in.

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
Joseph Allen: The Pivotal Importance of Air Quality, Ventilation and Exposures (Such as "Forever Chemicals") For Our Health
5 sept.· Ground Truths
Professor Joseph Allen directs the Healthy Buildings Program at Harvard Chan School of Public Health. His expertise extends far beyond what makes buildings healthy. He has been a leading voice and advocate during the Covid pandemic for air quality and ventilation. He coined the term “Forever Chemicals” and has written extensively on this vital topic, no less other important exposures, which we covered In our wide-ranging conversation. You will see how remarkably articulate and passionate Prof Allen is about these issues, along with his optimism for solutions.
A video snippet of our conversation: buildings as the 1st line of defense vs respiratory pathogens. Full videos of all Ground Truths podcasts can be seen on YouTube here. The audios are also available on Apple and Spotify.
Transcript with External Links and Links to Audio
Eric Topol (00:00:06):
Well, hello. It's Eric Topol from Ground Truths and I am just delighted to have with me, Joseph Allen from the Harvard School of Public Health, where he directs the Healthy Buildings Program that he founded and does a whole lot more that we're going to get into. So welcome, Joe.
Joseph Allen (00:00:24):
Thanks. It's great to be here. I appreciate the invitation.
Joe Allen’s Background As A Detective
Eric Topol (00:00:28):
Well, you have been, as I've learned, rocking it for many years long before the pandemic. There's quite a background about you having been a son of a homicide detective, private eye agency, and then you were going to become an FBI agent. And the quote from that in the article that's the Air Investigator is truly a classic. Yeah, you have in there, “I guarantee I'm the only public health student ever to fail an FBI lie detector polygraph in the morning and start graduate school a few hours later.” That's amazing. That's amazing.
Joseph Allen (00:01:29):
All right. Well, you've done your deep research apparently. That's good. Yeah, my dad was a homicide detective and I was a private investigator. That's no longer my secret. It's out in the world. And I switched careers and it happened to be the day I took the polygraph at the FBI headquarters in Boston, was the same day I started graduate studies in public health.
Sick vs Healthy Buildings (Pre-Covid)
Eric Topol (00:01:53):
Well, you're still a detective and now you're a detective of everything that can hurt us or help us environmentally and my goodness, how grateful we are that you change your career path. I don't know anyone who's had more impact on buildings, on air, and we're going to get into chemicals as well. So if we go back a bit here, you wrote a book before the pandemic, talk about being prescient. It’s called Healthy Buildings: How Indoor Spaces Can Make You Sick - or Keep You Well with John Macomber, your co-author. What was it that gave you the insight to write a book before there was this thing called Covid?
Joseph Allen (00:02:41):
Yeah, well, thanks for making the connection too, my past career to current career. For many years, I thought there wasn't a connection, but I agree. There's actually a lot of similarities and I also am really appreciative. I am lucky I found the field of Public Health, it's clearly where I belong. I feel like I belong here. It's a place to make an impact that I want to make in my career. So yeah, the Healthy Buildings book, we started writing years before the pandemic and was largely motivated by, I think what you and others and other people in my field have known, is that buildings have an outsized impact on our health. Yet it's not something that comes to the forefront when you ask people about what matters for their health. Right, I often start presentations by asking people that, what constitutes healthy living? They'll say, I can't smoke, I have to eat well.
(00:03:30):
I have to exercise. Maybe they'll say, outdoor pollution’s bad for you. Very few people, if any, will say, well, the air I breathe inside my building matters a lot. And over the years I had started my public health career doing forensic investigations of sick buildings. People really can get sick in buildings. It can be anything from headaches and not being able to concentrate all the way to cancer clusters and people dying because of the building. And I've seen this in my career, and it was quite frustrating because I knew, we all knew how to design and operate buildings in a way that can actually keep people healthy. But I was frustrated like many in my field that it wasn't advancing. In other words, the science was there, but the practice wasn't changing. We were still doing things the wrong way around ventilation, materials we put in our building, and I would lecture over and over and give presentations and I decided I want to try something new.
(00:04:22):
I do peer-reviewed science. That's great. I write pieces like you for the public, and I thought we'd try a longer form piece in a book, and it's published by Harvard Press. John Macomber for those who know is a professor at Harvard Business School who's an expert in real estate finance. So he'd been talking about the economic benefits of healthier buildings and some hand waving as he describes around public health. I've been talking about the public health benefits and trying to wave an economic argument. We teamed up to kind of use both of our strengths to, I hope make a compelling case that buildings are good for health and they're also just good business. In other words, try to break down as many barriers as we can to adoption. And then the book was published right as Covid hit.
Indoor Air Quality and Cognition
Eric Topol (00:05:05):
Yeah. I mean, it's amazing. I know that typically you have to have a book almost a year ahead to have it in print. So you were way, way ahead of this virus. Now, I'm going to come back to it later, but there were two things beyond the book that are pretty striking about your work. One is that you did all these studies to show with people wearing sensors to show that when the levels of CO2 were high by sensors that their cognition indoors was suffering. Maybe you could just tell us a little bit about these sensors and why aren't we all wearing sensors so that we don't lose whatever cognitive power that we have?
Joseph Allen (00:05:56):
Well, yeah. First I think we will start having these air quality sensors. As you know, they're starting to become a lot more popular. But yeah, when I first joined the faculty full-time at Harvard, one of the first studies I conducted with my team was to look at how indoor air quality influences cognitive function. And we performed a double-blind study where we took people, office workers and put them in a typical office setting. And unbeknownst to them, we started changing the air they were breathing in really subtle ways during the day, so they didn't know what we were doing. At the end of the day, we administered an hour and a half long cognitive function battery, and like all studies, we control for things like caffeine intake, baseline cognitive performance, all the other factors we want to account for. And after controlling for those factors in a double-blind study, we see that indoor air quality, minor improvements to indoor air quality led to dramatic increases in cognitive function test scores across domains that people recognize as important for everyday life.
(00:06:59):
How do you seek out and utilize information? How do you make strategic decisions? How do you handle yourself during a crisis and importantly recover after that crisis? I don't mean the world's ending crisis. I mean something happens at work that's stressful. How do you handle that and how do you respond? Well, it turns out that amongst all the factors that influence how we respond there, indoor air quality matters a lot. We call that study the COGfx Study for cognitive function. We replicated it across the US, we replicated it across the world with office workers around the world, and again, always showing these links, the subtle impact of indoor air quality on cognitive function performance. Now, that also then starts to be the basis for some of the economic analysis we perform with my colleague at Harvard Business School. We say, well, look, if you perform this much better related to air quality, what would happen if we implemented this at scale in a business?
(00:07:51):
And we estimate that there are just massive economic gains to be had. On a per person basis, we found and published on this, that's about $6,000 to $7,000 per person per year benefit across a company. It could lead to 10% gains to the bottom line performance of the company. And again, I'm a public health professor. My goal is to improve people's health, but we add a lens, mental health, brain health is part of health, and we add the economic lens to say, look, this is good for a worker of productivity and the costs are downright trivial when you compare it against the benefits, even just including the cognitive function benefits, not even including the respiratory health benefit.
Eric Topol (00:08:33):
And I mean, it's so striking that you did these studies in a time before sensors were, and they still are not widely accepted, and it really helped prove, and when we start to fall asleep in a group session indoors, it may not just be because we didn't have enough sleep the night before, right.
Joseph Allen (00:08:56):
It's funny you say that. I talk about that too. It's like, do we actually need the study to tell us to quantify what we've all experienced these bad conference rooms, you get tired, you can't concentrate, you get sleepy while you're driving your car. Yeah, a whole bunch of other factors. Maybe the speaker's boring, but a key factor is clearly indoor air quality and things like good ventilation, the chemical load in the space are all contributing.
Eric Topol (00:09:20):
Yeah. No, it's pretty darn striking. Now we're going to get into the pandemic, and this of course is when your work finally crystallized that you've been working on this for years, and then finally your collaboration with some of the aerosol experts. It was a transdisciplinary synergy that was truly extraordinary. And when you were on 60 Minutes last October, you said, “Think about the public health gains we've made over the past hundred years. We've made improvements to water quality, outdoor air pollution, our food safety, we've made improvements to sanitation: absolute basics of public health. Where has indoor air been in that conversation?” You brought it to us. I mean, you led the Lancet Commission on this. You've done a White House Summit keynote. You had a lot of influence. Why did it take us to finally wake up to this issue that you've been working on for years?
Covid is Airborne, Denial
Joseph Allen (00:10:31):
Yeah. Well, I appreciate that, but I also liked what you started with. I mean, there's been a lot of us pulling on this, and I think one of the magical moments, if you could say that when the pandemic happened was that it forced these collaborations and forced a lot of us in our field to be a bit more vocal. And even that comment about the gains we made in public health, that comes from an article that we co-authored with 40 plus scientists around the world in science, trying to drive home the point that we've ignored one of the key factors that determines our health. We were all frustrated at the beginning of the pandemic. The first piece I wrote was January 2020, talking about healthy buildings as the first line of defense, airborne spread, ventilation, filtration. I could not get it published. I could not get it published.
(00:11:20):
So I moved it to an international paper. I wrote it in the Financial Times in early February, but it wasn't until mid-March that the Times took my piece on this airborne spread buildings ventilation. At the same time, we know people like Linsey Marr, Rich Corsi, many others, Shelly Miller out there publishing, doing the fundamental research, all trying to elevate, and I think we started to find each other and say, hey, someone's trying to hit the medical journals. We're not landing there. I'm trying to hit the Times, and we’re not landing there. We're trying to get the reporters to pay attention. It's not landing there. Let's team up. Let's write these joint pieces. And I think what happened was you saw the benefit of the collective effort and interdisciplinary expertise, right? We could all start to come together, start instead of having these separate voices, a little bit of a unified voice despite important scientific minor disagreements, but start to say, hey, we started elevate each other and said, this is really important. It's the missing component of the messaging in the early days of the pandemic, and to know how to defend yourself.
Eric Topol (00:12:20):
Well, I think a lot of people think the big miss, and I know you agree, was the lack of recognition of aerosol transmission instead of just liquid droplets. But what you brought to this was really your priors on the buildings themselves and the ventilation systems and air quality that was highly, I mean, critical to it isn't just the aerosol, it's obviously how buildings are set up. Now, there's an amazing piece of course that appeared in the summer of 2021 called the Air Investigator, which profiled you, and in it brings up several things that finally are, we're starting to get our act together. I mean, ultimately there was in May 2023 years later, the CDC says, we're going to do something about this. Can you tell us what was this very distinct new path that the CDC was at least saying? And also couple that with whatever action if or not action has been taken.
Joseph Allen (00:13:33):
Yeah. So there really was a monumental shift that took, it was years in development, but we finally won the argument, collectively that airborne spread was the dominant mode of transmission. Okay, we got that. Then the question is, well, what changes? Do we actually get guidance here? And that took a little bit longer. I give Rochelle Walensky a lot of credit when she came into the CDC, we talked with her about this. That's when you start to first see ventilation starts showing up and the guidance, including guidance for schools. So I think that was a big win, but still no one was willing to set an official target or standard around higher ventilation rates. So that's important. Early in the pandemic, some people started to hear a message, yes, ventilation is important. What's the obvious next question, well, how much, what do I need? So in the summer of 2020, actually Shelly Miller and I collaborated on this.
(00:14:23):
We published some guidance on ventilation targets for schools. We said four to six air changes per hour (ACH) and target that. Well, it wasn't until 2023, spring of 2023 that you mentioned that CDC published target ventilation rates, and they went with five air changes per hour, which is right where we were talking about in summer 2020. It's what the Lancet of COVID-19 Commission adopted, but it's momentous in this way. It's the first time in CDCs history they've ever published a ventilation rate target for health. Now, I know this seems slow at the time, and it was, but if we think about some of the permanent gains that will come out of the pandemic. Pandemic changes society and science and policy and practice this, we are never going back. Now buildings will be a first line of defense for respiratory pathogens going forward that can no longer be ignored. And now we have the published target by CDC. That's a big deal because it's not just a recognition, but there's actually something to shoot for out there. It's a target I happen to like, I think there are differences between different scientists, but ultimately we've lifted the floor and said, look, we actually have to raise ventilation rates and we have something to shoot for. The public needed that kind of guidance a lot earlier, of course, but it was a big deal that it happened. It’s just too bad it took until spring 2023.
Eric Topol (00:15:46):
Yeah, I certainly agree that it was momentous, but a year plus later, has there been any change as a result of this major proclamation, if you will?
Joseph Allen (00:15:59):
Well, I actually see a lot of change from a practitioner level, but I want to talk about it in two aspects. I see a lot of schools, universities, major companies that have made this shift. For example, in the 60 Minutes piece, I talk that I advised Amazon and globally they're measuring indoor air quality with real-time sensors in their buildings. I've worked with hundreds of school districts that have made improvements to indoor air quality. I know companies that have shifted their entire approach to how they design and operate their buildings. So it's happening. But what really needs to happen, Eric, if this movement is going to benefit everyone, is that these targets need to be codified. They need to go into building codes. It can't just be, oh, I've heard about this. So I made the decision. I have the resources and the money to make this improvement.
(00:16:44):
To create a healthy building or a healthy school, we need to be sure this gets built into our code. So it just becomes the way it's done. That is not happening. There are some efforts. There are some bills at the national level. Some states are trying to pass bills, and I have to say, this is why I'm optimistic. It feels very slow. I'm as frustrated as anybody. I wanted this done before the pandemic. As soon as the pandemic hit, we saw it. We knew what we needed to get done. It didn't happen. But if we think about the long arc here and the public health gains we're actually, it's remarkable to me that we actually have bills being introduced around indoor air quality that ASHRAE has set a new health focused target for the first time really in their history. CDC, first time. New buildings going up in New York City designed to these public health targets. That's really different. I've been in this field for 20 plus years. I've never seen anything like it. So the pace is still slow, but it really is happening. But it has to reach everybody, and the only way that's going to happen is really this gets into building codes and performance standards.
The Old Efficient Energy Buildings
Eric Topol (00:17:52):
Yeah. Well, I like your optimistic perspective. I do want to go back for a second, back decades ago there was this big impetus to make these energy efficient buildings and to just change the way the buildings were constructed so that there was no leak and it kind of set up this problem or exacerbated, didn't it?
Joseph Allen (00:18:19):
Yeah. I mean, I've written about this a lot. I write in the book our ventilation standards, they've been a colossal mistake. They have cost the public in terms of its health because in the seventies, we started to really tighten up our building envelopes and lower the ventilation rates. The standards were no longer focused on providing people with a healthy indoor space. As I write in the book, they were targeted towards minimally acceptable indoor air quality, bare minimums. By the way that science is unequivocal, is not protective of health, not protective against respiratory pathogens, doesn't promote good cognitive function, not good for allergies. These levels led to more illness in schools, more absences for teachers and students, an absolute disaster from a public health standpoint. We've been in this, what I call the sick building era since then. Buildings that just don't bring in enough clean outdoor air. And now you take this, you have a building stock for 40 years tighter and tighter and tighter bumps up against a novel virus that spread nearly entirely indoors. Is it any wonder we had, the disaster we had with COVID-19, we built these bills. They were designed intentionally with low ventilation and poor filtration.
Optimal Ventilation and Filtration
Eric Topol (00:19:41):
Yeah. Well, it's extraordinary because now we've got to get a reset and it's going to take a while to get this done. We'll talk a bit about cost of doing this or the investment, if you will, but let's just get some terms metrics straight because these are really important. You already mentioned ACH, the number of air changers per hour, where funny thing you recommended between four and six and the CDC came out with five. There's also the minimum efficiency reporting value (MERV). A lot of places, buildings have MERV 8, which is insufficient. We need MERV 13. Can you tell us about that?
Joseph Allen (00:20:23):
Yeah, sure. So I think when we think about how much, you have two ways to capture these respiratory particles, right? Or get rid of them. One is you dilute them out of the building or you capture them on filters. You can inactivate them through UV and otherwise. But let's just stay on the ventilation and filtration side of this. So the air changing per hour is talking about how often the air is change inside. It's an easy metric. There are some strengths to it, there's some weaknesses, but it's intuitive and I'll you some numbers so you can make sense of this. We recommended four to six air changes per hour. Typical home in the US has half an air change per hour. Typical school designed to three air changes per hour, but they operate usually at one and a half. So we tried to raise this up to four, five, or six or even higher. On the filtration side, you mentioned MERV, right? That's just a rating system for filters, and you can think about it this way. Most of the filters that are in a building are cheap MERV 8 filters, I tend to think of them as filters that protect the equipment. A MERV 13 filter may capture 80 or 90% of particles. That's a filter designed to protect people. The difference in price between a MERV 8 and a MERV 13 is a couple of bucks.
(00:21:30):
And a lot of the pushback we got early in the pandemic, some people said, well, look, there's a greater resistance from the better filter. My fan can't handle it. My HVAC system can't handle it. That was nonsense. You have low pressure drop MERV 13 filters. In other words, there really wasn't a barrier. It was a couple extra bucks for a filter that went from a MERV 8 might capture 20 or 30% to a filter, MERV 13 that captures 80 or 90% with very little, if any impact on energy or mechanical system performance. Absolute no-brainer. We should have been doing this for decades because it also protects against outdoor air pollution and other particles we generate indoors. So that was a no-brainer. So you combine both those ventilation filtration, some of these targets are out there in terms of air change per hour. You can combine the metric if we want to get technical to talk about it, but basically you're trying to create an overall amount of clean air. Either you bring in fresh outdoor air or you filter that air. It really is pretty straightforward, but we just didn't have some of these targets set and the standards we're calling for these minimum acceptable levels, which we're not protective of health.
Eric Topol (00:22:37):
So another way to get better air quality are these portable air cleaners, and you actually just wrote about that with your colleagues in the Royal Society of Chemistry, not a journal that I typically read, but this was an important article. Can you give us, these are not very expensive ways to augment air quality. Can you tell us about these PACs ?
Joseph Allen (00:23:06):
These portable air cleaners (PACs), so the same logic applies if people say, well, I can't upgrade my system. That's not a problem for very low cost, you could have, these devices are essentially a fan and a filter, and the amount of clean air you get depends on how strong the fan is and how good the filter is. Really pretty simple stuff here, and you can put one of these in a room if it's sized right. My Harvard team has built tools to help people size this. If you're not quite sure how to do it, we have a technical explainer. Really, if you size it right, you can get that four, five or six air changes per hour, very cheap and very quickly. So this was a tool I thought would be very valuable. Rich Corsi and I wrote about this all through the summer of 2020 to talk about, hey, a stop gap measure.
(00:23:50):
Let's throw out some of these portable air cleaners. You increase the air changes or clean air delivery pretty effectively for very low cost, and they work. And now the paper we just published in my team a couple of days ago starts to advance this more. We used a CFD model, so computational fluid dynamics. Essentially, you can look at the tracers and the airflow patterns in the room, and we learn a couple things that matter. Placement matters, so we like it in the center of the room if you can or as close as possible. And also the airflow matters. So the air cleaners are cleaning the air, but they're also moving the air, and that helps disperse these kind of clouds or plumes when an infected person is breathing or speaking. So you want to have good ventilation, good filtration. Also a lot of air movement in the space to help dilute and move around some of these respiratory particles so that they do get ventilated out or captured in a filter.
Eric Topol (00:24:40):
Yeah. So let me ask you, since we know outdoors are a lot safer. If you could do all these things indoors with filtration, air changing the quality, can you simulate the outdoors to get rid of the risk or markedly reduce the risk of respiratory viruses like SARS-CoV-2 and others?
Joseph Allen (00:25:04):
Yeah, you can't drop it to zero. There's no such thing as zero risk in any of these environments. But yeah, I think some of the estimates we've seen in my own team has produced in the 60-70% reduction range. I mean, if you do this right with really good ventilation filtration, you can drop that risk even further. Now, things like distancing matter, whether or not somebody's wearing a mask, these things are all going to play into it. But you can really dramatically drop the risk by handling just the basics of ventilation and filtration. And one way to think about it is this, distance to the infector still matters, right? So if you and I are speaking closely and I breathe on you, it's going to be hard to interrupt that flow. But you can reduce it through good ventilation filtration. But really what it's doing also is preventing super spreading events.
(00:25:55):
In other words, if I'm in the corner of a room and I'm infectious and you're on the other side, well if that room is sealed up pretty good, poor ventilation, no filtration, the respiratory aerosols are going to build up and your risk is going to increase and we're in there for an hour or two, like you would be in a room or office and you're exposed to infectious aerosol. With good ventilation filtration, those respiratory particles don't have a chance to reach you, or by the time they do, they're much further diluted. Linsey Marr I think was really great early in the pandemic by talking about this in terms of cigarette smoke. So a small room with no ventilation filtration, someone smoking in the corner, yeah, it's going to fill up over time with smoke you're breathing in that secondhand smoke. In a place with great ventilation filtration, that's going to be a lot further reduced, right? You're not going to get the buildup of the smoke and smoke particles are going to operate similarly to respiratory particles. So I think it's intuitive and it's logical. And if you follow public health guidance of harm reduction, risk reduction, if you drop exposure, you drop risk.
(00:26:58):
The goal is to reduce exposure. How do we do that? Well, we can modify the building which is going to play a key role in exposure reduction.
Eric Topol (00:27:06):
Now, to add to this, if I wear a sensor or have a sensor in the room for CO2, does that help to know that you're doing the right thing?
Joseph Allen (00:27:17):
Yeah, absolutely. So people who are not familiar with these air quality sensors. They're small portal air quality sensors. One of the things they commonly measure is carbon dioxide. We're the main source of CO2 inside. It's a really good indicator of ventilation rate and occupancy. And the idea is pretty simple. If the CO2 is low, you don't have a buildup of particles from the respiratory tract, right? And CO2 is a gas, but it's a good indicator of overall ventilation rate. This room I'm in right now at the Harvard School of Public Health has air quality sensors. We have this at Harvard Business School. We have it at the Harvard Health Clinics. Many other places are doing it, Boston Public schools have real-time air quality monitors. Here's the trick with CO2. So first I'll say we have some guidance on this at the Harvard Healthy Buildings page, if people want to go look it up, how to choose an air quality sensor, how to interpret CO2 levels.
Carbon Dioxide Levels
(00:28:04):
But here's a way to think about it. We generally would like to see CO2 levels less than 800 parts per million. Historically, people in my field have said under 1,000 is okay. We like to see that low. If your CO2 is low, the risk is low. If your CO2 is high, it doesn't necessarily mean your risk is high because that's where filtration can come in. So let me say that a little bit better. If CO2 is low, you're diluting enough of the respiratory particles. If it's high, that means your ventilation is low, but you might have excellent filtration happening. Either those MERV 13 filters we talked about or the portable air cleaners. Those filters don't capture CO2. So high CO2 just means you better have a good filter game in place or the risk is going to be high. So if you CO2 is low, you're in good shape. If it's high, you don't quite know. But if you have bad filtration, then the risk is going to be much higher.
Eric Topol (00:29:01):
I like that 800 number because that's a little lower than some of the other thresholds. And why don't we do as good as we can? The other question about is a particulate matter. So we are worried about the less than 5 microns, less than 2.5 microns. Can you tell us about that and is there a way that you can monitor that directly?
Joseph Allen (00:29:25):
Sure. A lot of these same sensors that measure CO2 also measure PM 2.5 which stands for particular matter. 2.5 microns is smaller, one of the key components of outdoor air pollution and EPA just set new standards, right? WHO has a standard for 5 microgram per cubic meter. EPA just lowered our national outdoor limit from 12 to 9 microgram per cubic meter. So that's a really good indicator of how well your filters are working. Here again, in a place like this or where you are, you should see particle levels really under 5 microgram per cubic meter without any major source happening. What's really interesting about those like the room I'm in now, when the wildfire smoke came through the East coast last year, levels were extraordinary outside 100, 200, 300 microgram per cubic meter. But because we have upgraded our filters, so we use MERV 15 here at Harvard, the indoor levels of particles stayed very low.
(00:30:16):
So it shows you how the power of these filters can actually, they do a really good job of capturing particles, whether it be from our lungs or from some other source. So you can measure this, but I'll tell you what's something interesting, if you want to tie it into our discussion about standards. So we think about particles. We have a lot of standards for outdoor air pollution. So there's a national ambient air quality standard 9 microgram per cubic meter. We don't have standards for indoor air quality. The only legally enforceable standard for indoor particles is OSHA's standard, and it's 5,000 microgram per cubic meter 5,000.
(00:30:59):
And it's absurd, right? It's an absurdity. Here we are EPAs, should it be 12, should it be 9, or should it be 8? And for indoors, the legally enforceable limit for OSHA 5,000. So it points to the big problem here. We talked about earlier about the need for these standards to codify some of this. Yes, we have awareness from the public. We have sensors to measure this. We have CDC now saying what we were saying with the Lancet COVID-19 Commission and elsewhere.
This is big movement, but the standards then need to come up behind it and get into code and new standards that are health focused and health based. And we have momentum, but we can't lose it right now because it's the first time in my career I felt like we're on the cusp of really getting this and we are so close. But of course it's always in danger of slipping through our fingers.
Regulatory Oversight for Air
Eric Topol (00:31:45):
Well, does this have anything to do with the fact that in the US there's no regulatory oversight over air as opposed to let's say Japan or other places?
Joseph Allen (00:31:57):
Yeah, I mean, we have regulatory oversight of outdoor air. That's EPA. There's a new bill that was introduced to give EPA more resources to deal with indoor air. EPA has got a great indoor air environments division, but it doesn't have the legally enforceable mandate or statute that we have for outdoor. So they'd give great guidance and have for a long time. I really like that group at EPA, but there's no teeth behind this. So what we have is worker health protections at OSHA to its own admission, says its standards are out of date. So we need an overhaul of how we think about the standards. I like the market driven approach. I think that's being effective, and I think we can do it from voluntary standards that can get adopted into code at the municipal level. I think that's a real path. I see it happening. I see the influence of all this work hitting legislators. So that's where I think the most promising path is for real change.
The Risks of Outdoor Air Pollution
Eric Topol (00:33:03):
Yeah, I think sidestepping, governmental teeth, that probably is going to be a lot quicker. Now, before we get to the cost issue, I do want to mention, as you know very well, the issue of air pollution in Science
a dedicated issue just a few weeks ago, it brought up, of course, that outdoor air pollution we've been talking about indoor is extraordinary risk for cancer, dementia, diabetes, I mean everything. Just everything. And there is an interaction between outdoor pollution and what goes on indoor. Can you explain basically reaffirm your concern about particulate matter outdoors, and then what about this interaction with what goes on indoors?
Joseph Allen (00:33:59):
Yeah, so it's a great point. I mean, outdoor pollution has been one of the most studied environmental pollutants we know. And there's all of these links, new links between Alzheimer's, dementia, Parkinson's disease, anxiety, depression, cardiovascular health, you named it, right? I've been talking about this and very vocal. It's in the book and elsewhere I called the dirty secret of outdoor air pollution. The reality is outdoor air pollution penetrates indoors, and the amount depends on the building structure, the type of filters you have. But let's take an infiltration value of say 50%. So you have a lot of outdoor air pollution, maybe half of that penetrates inside, so it's lower, the concentration is lower, but 90% of the breaths you take are indoor. And if you do the math on it, it's really straightforward. The majority of outdoor air pollution you breathe happens inside.
(00:34:52):
And people, I think when they hear that think, wait, that can't be right. But that's the reality that outdoor pollution comes inside and we're taking so many breaths inside. Your total daily dose of outdoor air pollution is greater from the time you spend inside. I talk about this all the time. You see any article about outdoor air pollution, what's the cover picture? It's someone outside, maybe they're wearing a mask you can't really see. It's smoky hazy. But actually one of the biggest threats is what's happening inside. The nice thing here, again, the solutions are pretty simple and cost-effective. So again, upgrade from MERV 8 to MERV 13, a portable air cleaner. We are just capturing particles on a filter basic step that can really reduce the threat of outdoor air pollution inside. But it's ignored all the time. When the wildfire smoke hit New York City. New York City's orange, I called colleagues who are in the news business.
(00:35:48):
We have to be talking about the indoor threat because the guidance was good, but incomplete. Talk about Mayor Adams in New York City. Go inside, okay, that's good advice. And go to a place that has good filtration or they should have been giving out these low cost air cleaners. So just going inside isn't going to protect your lungs unless you're actually filtering a lot more of that air coming in. So trying to drive home the point here that actually we talk about these in silos. Well, wildfire smoke and particles, Covid and respiratory particles, we're all talking about these different environmental issues that harm our health, but they're all happening through or mediated by the building performance. And if we just get the building performance right, some basics around good ventilation, good filtration, you start to address multiple threats simultaneously. Outdoor air pollution, wildfire smoke, allergens, COVID-19, influenza, RSV, better cognitive function performance, anxiety. You start addressing the root cause or one of the contributors and buildings we can then start to leverage as a true public health tool. We have not taken advantage of the power of buildings to be a true public health tool.
Eric Topol (00:36:59):
Oh, you say it so well, and in fact your Table on page 44 in Healthy Buildings , we’ll link it because it shows quantitatively what you just described about outdoor and indoor cross fertilization if you will. Now before leaving air pollution outdoors, indoors, in order for us to affect this transformation that would markedly improve our health at the public health individual level, we're talking about a big investment. Can you put that in, you did already in some respects, but if we did this right in every school, I think in California, they're trying to mandate that in schools, in the White House, they're mandating federal buildings. This is just a little piece of what's needed. This would cost whatever trillions or hundreds of billions of dollars. What would it take to do this? Because obviously the health benefits would be so striking.
What’s It Gonna Cost?
Joseph Allen (00:38:04):
Well, I think one of the issues, so we can talk about the cost. A lot of the things I'm talking about are intentionally low cost, right? You look at the Lancet of COVID-19 Commission, our report we wrote a report on the first four healthy building strategies every building should pursue. Number one commission your building that's giving your building a tune-up. Well, guess what? That not only improves air quality, it saves energy and therefore saves money. It actually becomes cost neutral. If not provides an ROI after a couple of years. So that's simple. Increase the amount of outdoor air ventilation coming in that has an energy cost, we've written about this. Improved filtration, that's a couple bucks, really a couple bucks, this is small dollars or portable air cleaners, not that expensive. I think one of the big, and Lawrence Berkeley National Lab has written this famous paper people like to cite that shows there's $20 billion of benefits to the US economy if we do this.
(00:38:59):
And I think it points to one of the problems. And what I try to address in my book too, is that very often when we're having this conversation about what's it going to cost, we don't talk about the full cost benefit. In other words, we say, well, it's going to cost X amount. We can't do that. But we don't talk about what are the costs of sick buildings? What are the costs of kids being out of school for an entire year? What are the costs of hormonal disruption to an entire group of women in their reproductive years due to the material choices we make in our buildings? What are the costs to outdoor air pollution and cardiovascular disease, mental health? Because we don't have good filters in our buildings that cost a couple dollars. So in our book, we do this cost benefit analysis in the proforma in our book, we lay out what the costs are to a company. We calculate energy costs. We say these are the CapEx costs, capital costs for fixed costs and the OpEx costs for operating expenditures. That's a classic business analysis. But we factor in the public health benefits, productivity, reduced absenteeism. And you do that, and I don't care how you model it, you are going to get the same answer that the benefits far outweigh the cost by orders of magnitude.
Eric Topol (00:40:16):
Yeah, I want to emphasize orders of magnitude. Not ten hundred, whatever thousand X, right?
Joseph Allen (00:40:23):
What would be the benefit if we said we could reduce influenza transmission indoors in schools and offices by even a small percent because we improve ventilation and filtration? Think of the hospitalization costs, illness costs, out of work costs, out of school costs. The problem is we haven't always done that full analysis. So the conversation gets quickly to well, that's too much. We can't afford that. I always say healthy buildings are not expensive. Sick buildings are expensive. Totally leave human health out of that cost benefit equation. And then it warps this discussion until you bring human health benefits back in.
Forever Chemicals
Eric Topol (00:40:58):
Well, I couldn't agree more with you and I wanted to frame this by giving this crazy numbers that people think it's going to cost to the reality. I mean, if there ever was an investment for good, this is the one that you've outlined so well. Alright, now I want to turn to this other topic that you have been working on for years long before it kind of came to the fore, and that is forever chemicals. Now, forever chemicals, I had no idea that back in 2018 you coined this term. You coined the term, which is now a forever on forever chemicals. And basically, this is a per- and polyfluoroalkyl substances (PFAS), but no one will remember that. They will remember forever chemicals. So can you tell us about this? Because this of course recently, as you know well in May in the New Yorker, there was an expose of 3M, perhaps the chief offender of these. They're everywhere, but especially they were in 3M products and continue to be in 3M products. Obviously they've been linked with all kinds of bad things. What's the story on forever chemicals?
Joseph Allen (00:42:14):
Yeah, they are a class of chemicals that have been used for decades since the forties. And as consumers, we like them, right? They're the things that make your raincoat repel rain. It makes your non-stick pan, your scrambled eggs don't stick to the pan. We put them on carpets for stain resistance, but they came with a real dark side. These per- and polyfluoroalkyl substances, as I say, a name only a chemist could love have been linked with things like testicular cancer, kidney cancer, interference with lipid metabolism, other hormonal disruption. And they are now a global pollutant. And one of the reasons I wrote the piece to brand them as forever chemicals was because I'm in the field of environmental health. We had been talking about these for a long time and I just didn't hear the public aware or didn't capture their attention. And part of it, I think is how we talk about some of these things.
(00:43:14):
I think a lot about this. Per- and polyfluoroalkyl substances, no one's going to, so the forever chemicals is actually a play on their defining feature. So these chemicals, these stain repellent chemicals are characterized by long chains of the carbon fluorine bond. And when we string these together that imparts this and you put them on top of a product that imparts the property of stain resistance, grease resistance, water resistance, but the carbon fluorine bond is the strongest in all of organic chemistry. And these chains of the carbon fluorine bond never fully break down in the environment. And when we talk in my field about persistent organic pollutants, we talk about chemicals that break down on the order of decades. Forever chemicals don't break down. They break down the order of millennia. That's why we're finding them everywhere. We know they're toxic at very low levels. So the idea of talking about forever chemicals, I wanted to talk about their foreverness.
(00:44:13):
This is permanent. What we're creating and the F and the C are the play on the carbon-fluorine bond and I wrote an article trying to raise awareness about this because some companies that have produced these have known about their toxicity for decades, and it's just starting the past couple of years, we're just starting to pay attention to the scale of environmental pollution. Tens of millions of Americans have forever chemicals in their drinking water above the safe limit, tens of millions. I worked as an expert in a big lawsuit for the plaintiffs that were drinking forever chemicals in their water that was dumped into the drinking water supply by a manufacturing company. I met young men with testicular cancer from drinking forever chemicals in their water. These really has escaped the public's consciousness, it wasn't really talked about. Now of course, we know every water body, we use these things in firefighting foams or every airport has water pollution.
(00:45:17):
Most airports do. Firefighters are really concerned about this, high rates of cancer in the firefighter population. So this is a major problem, and the cleanup is not straightforward or easy because they're now a global pollutant. They persist forever. They're hard to remediate and we're stuck with them. So that's the downside, I can talk about the positives. I try to remain an optimist or things we're doing to try to solve this problem, but that's ultimately the story. And my motivation was I just to have people have language to be able to talk about this that didn't require a degree in organic chemistry to understand what they were.
Eric Topol (00:45:52):
Yeah, I mean their pervasiveness is pretty scary. And I am pretty worried about the fact that we still don't know a lot of what they're doing in terms of clinical sequela. I mean, you mentioned a couple types of cancer, but I don't even know if there is a safe threshold.
Joseph Allen (00:46:16):
Eric, I'll tell you one that'll be really interesting for you. A colleague of mine did a famous study on forever chemicals many years ago now and found that kids with higher levels of forever chemicals had reduced vaccine effectiveness related to these chemicals. So your point is, right, a lot of times we're using these industrial chemicals. We know a couple endpoints for their affecting our bodies, but we don't know all of them. And what we know is certainly alarming enough that we know enough to know we shouldn't be using them.
Eric Topol (00:46:51):
And you wrote another masterful op-ed in the Washington Post, 6 forever chemical just 10,000 to go. Maybe you could just review what that was about.
Joseph Allen (00:47:02):
Yeah, I've been talking a lot about this issue I call chemical whack-a-mole. So forever chemical is the perfect example of it. So we finally got people's attention on forever chemicals. EPA just regulated 6 of them. Well, guess what? There are 10,000 if not many more than that. Different variants or what we call chemical cousins. Now that's important for this reason. If you think about how we approach these from a regulatory standpoint, each of the 10,000 plus forever chemicals are treated as different. So by the time EPA regulates 6, that's important. It does free up funding for cleanup and things like this. But already the market had shifted away from those 6. So in other words, in the many thousand products that still use forever chemicals, they're no longer using those 6 because scientists have told people these things are toxic years ago. So they switch one little thing in the chemical, it becomes a new chemical from a regulatory perspective.
(00:47:57):
But to our bodies, it's the same thing. This happens over and over. This has happened with pesticides. It happens with chemicals and nail polish. It happens in chemicals in e-cigarettes. It happens with flame retardant chemicals. I wrote a piece in the Post maybe six years ago talking about chemical whack-a-mole, and this problem that we keep addressing, these one-off, we hit one, it changes just slightly. Chemical cousin pops up, we hit that one. Five years later, scientists say, hey, the next one doesn't look good either. We're doing this for decades. It's really silly. It's ineffective, it's broken, and there are better ways to handle this going forward.
Eric Topol (00:48:31):
And you know what gets me, and it's like in the pharma industry that I've seen the people who run these companies like 3M that was involved in a multi-decade coverup, they're never held accountable. I mean, they know what they're doing and they just play these games that you outlined. They're still using 16,000 products, according to the New Yorker, the employee that exposed them, the whistleblower in the New Yorker article.
Joseph Allen (00:48:58):
That was an amazing article by Sharon Lerner talking to the people who had worked there and she uncovered that they knew the toxicity back in the seventies, and yes, they were still making these products. One of the things that I think has gotten attention of some companies is while the regulations have been behind, the lawsuits are piling up.
Joseph Allen (00:49:21):
The lawsuit I was a part of as an expert for that was about an $800 million settlement in favor of the plaintiffs. A couple months later is another one that was $750 million. So right there, $1.5 billion, there's been several billion dollars. This has caught the attention of companies. This has caught the attention of product manufacturers who are using the forever chemicals, starting to realize they need to reformulate. And so, in a good way now, that's not the way we should be dealing with this, but it has started to get companies to wake up that maybe they had been sleeping on it, that this is a major problem and actually the markets have responded to it.
Eric Topol (00:50:02):
Well, that's good.
Joseph Allen (00:50:03):
Because these are major liabilities on the books.
Eric Topol (00:50:05):
Yeah, I mean, I think what I've seen of course with being the tobacco industry and I was involved with Vioxx of course, is the companies just appeal and appeal and it sounds really good that they've had to pay $800 million, but they never wind up paying anything because they basically just use their muscle and their resources to appeal and put it off forever. So I mean, it's one way to deal with it is a litigation, but it seems like that's not going to be enough to really get this overhauled. I don't know. You may be more sanguine.
Joseph Allen (00:50:44):
No, no, I agree with you. It's the wrong way. I mean, we don't want to, the solution here is not to go after companies after people are sick. We need get in front of this and be proactive. I mentioned it only because I know it has made other companies pay attention how many billion does so-and-so sue for. So that's a good signal that other companies are starting to move away from forever chemicals. But I do want to talk about one of the positive approaches we're doing at Harvard, and we have a lot of other partners in the private sector doing this. We're trying to turn off the spigot of forever chemicals entering the market in the first place. As a faculty advisor to what we call the Harvard Healthier Building Materials Academy, we publish new standards. We no longer buy products that have forever chemicals in them for our spaces.
(00:51:31):
So we buy a chair or carpet. We demand no forever chemicals. What's really neat about this is we also say, we treat them as a whole class. We don't say we don't want PFOA. That's one of the regulated chemicals. We say we don't want any of the 10,000. We are not waiting for the studies to show us they act like the other ones. We've kind of been burned by this for decades. So we're actually telling the suppliers we don't want these chemicals and they're delivering products to us without these chemicals in them. We have 50 projects on our campus built with these new design standards without forever chemicals and other toxic chemicals. We've also done studies that a doctoral student done the study. When we do this, we find lower levels of these chemicals in air and dust, of course. So we're showing that it works.
(00:52:19):
Now, the goal is not to say, hey, we just want to make Harvard a healthier campus and the hell with everybody else. The goal is to show it can be done with no impact to cost, schedule or product performance. We get a healthier environment, products look great, they perform great. We've also now partnered with other big companies in the tech industry in particular to try and grow or influence the market by saying, look how many X amount of purchasing dollars each year? And it's a lot, and we're demanding that our carpets don't have this, that our chairs don't have it, and the supply chain is responding. The goal, of course, is to just make it be the case that we just have healthy materials in the supply chain for everybody. So if you or I, or anybody else goes to buy a chair, it just doesn't have toxic chemicals in it.
Eric Topol (00:53:06):
Right, but these days the public awareness still isn't there, nor are the retailers that are selling whether it's going to buy a rug or a chair or new pots and pans. You can't go in and say, does this have any forever chemicals? They don't even know, right?
Joseph Allen (00:53:24):
Impossible. I study this and it's hard for me when I go out to try and find and make better decisions for myself. This is one of the reasons why we're working, of course, trying to help with the regulatory side, but also trying to change the market. Say, look, you can produce the similar product without these chemicals, save yourself for future lawsuits. Also, there's a market for healthy materials, and we want everybody to be a part of that market and just fundamentally change the supply chain. It's not ideal, but it's what we can do to influence the market. And honestly, we're having a lot of impact. I've been to these manufacturing plants where they have phased out these toxic chemicals.
Eric Topol (00:54:03):
That’s great to hear.
Joseph Allen (00:54:06):
And we see it working on our campus and other companies’ campuses.
Eric Topol (00:54:10):
Well, nobody can ever accuse you of not taking on big projects, okay.
Joseph Allen (00:54:15):
You don't get into public health unless you want to tackle the big ones that are really going to influence.
Micro(nano) Plastics
Eric Topol (00:54:20):
Well, that's true, Joe, but I don't know anybody who's spearheading things like you. So it's phenomenal. Now before we wrap up, there's another major environmental problem which has come to the fore, which are plastics, microplastics, nanoplastics. They're everywhere too, and they're incriminated with all the things that we've been talking about as well. What is your view about that?
Joseph Allen (00:54:48):
Well, I think it's one, well, you see the extent of the pollution. It's a global pollutant. These are petrochemicals. So it's building up, and these are fossil fuel derivatives. So you can link this not just to the direct human health impacts, the ecosystem impacts, but also ecosystem and health impacts through climate change. So we've seen our reliance on plastics grow exponentially over the past several decades, and now we're seeing the price we're paying for that, where we're seeing plastics, but also microplastics kind of everywhere, much like the forever chemicals. Everywhere we look, we find them and we're just starting to scratch a surface on what we know about the environmental impacts. I think there's a lot more that can be done here. Try to be optimistic again, at least if you find a problem, you got to try and point to some kind of solution or at least a pathway towards solutions.
(00:55:41):
But I like some of the stuff from others colleagues at Yale in particular on the principles of green chemistry. I write about them in my book a little bit, but it's this designing for non-permanence or biodegradable materials so that if we're using anything that we're not leaving these permanent and lasting impacts on our ecosystem that then build up and they build up in the environment, then they build up in all of us and in our food systems. So it seems to me that should be part of it. So think about forever chemicals. Should we be using chemicals that never break down in the environment that we know are toxic? How do we do that? As Harvard, one of the motivating things here for forever chemicals too, is how are we ignoring our own science? Everyone's producing this science, but how do we ignore even our own and we feel we have responsibility to the communities next to us and the communities around the world. We're taking action on climate change. How are we not taking action on these chemicals? I put plastics right in there in terms of the environmental pollutants that largely come from our built environment, food products and the products we purchase and use in our homes and in our bodies and in all the materials we use.
Eric Topol (00:56:50):
When you see the plastic show up in our arteries with a three, four-fold increase of heart attacks and strokes, when you see it in our testicles and every other organ in the body, you start to wonder, are we ever going to do something about this plastic crisis? Which is somewhat distinct from the forever chemicals. I mean, this is another dimension of the problem. And tying a lot of this together, you mentioned, we are not going to get into it today, but our climate crisis isn't being addressed fast enough and it's making all these things exacerbating.
Joseph Allen (00:57:27):
Yeah, let me touch on that because I think it is important. It gets to something I said earlier about a lot of these problems we treat as silos, but I think a lot of the problems run through our buildings, and that means buildings are part of the solution set. Buildings consume 40% of global energy.
(00:57:42):
Concrete and steel count for huge percentages of our global CO2 emissions. So if we're going to get climate solved, we're going to have to solve it through our buildings too. So when you start putting this all together, Eric, right, and this is why I talk about buildings as healthy buildings could potentially be one of the greatest public health interventions we have of this century. If we get it right, and I don't mean we get the Covid part, right. We get the forever chemicals part, right. Or the microplastics part, right. If you start getting this all right, good ventilation, better filtration, healthy materials across the board, energy efficient systems, so we're not drawing on the energy demand of our buildings that are contributing to the climate crisis. Buildings that also address climate adaptation and resilience. So they protect us from extreme heat, wildfire smoke, flooding that we know is coming and happening right now.
(00:58:37):
You put that all together and it shows the centrality of buildings on our collective health from our time spent indoors, but also their contribution to environmental health, which is ultimately our collective human health as well. And this is why I'm passionate about healthy buildings as a real good lens to put this all under. If we start getting these right, the decisions we make around our buildings, we can really improve the human condition across all of these dimensions we're talking about. And I actually don't think it's all that hard in all of these. I've seen solutions.
Eric Topol (00:59:12):
I'm with you. I mean, there's innovations that are happening to take the place of concrete, right?
Joseph Allen (00:59:20):
Sure. We have low emission concrete right now that's available. We have energy recovery ventilation available right now. We have real time sensors. We can do demand control ventilation right now. We have better filters right now. We have healthy materials right now.
(00:59:33):
We have this, we have it. And it's not expensive if we quantify the health benefits, the many, many multiple benefits. So it's all within our reach, and it's just about finding these different pathways. Some of its market driven, some of it's regulatory, some of it's at the local level, some of it's about raising awareness, giving people the language to talk about these things. So I do think it's the real beginning of the healthy buildings era. I really, truly believe it. I've never seen change like this in my field. I've been chasing sick buildings for a long time.
Joseph Allen (01:00:11):
And clearly there's pathways to do better.
Eric Topol (01:00:13):
You're a phenom. I mean, really, you not only have all the wisdom, but you articulate it so well. I mean, you’re leading the charge on this, and we're really indebted to you. I'm really grateful for you taking an hour of your busy time to enlighten us on this. I think what you're doing is it's going to keep you busy for your whole career.
Joseph Allen (01:00:44):
Well, the goal here is for me to put myself out of business. We shouldn't have a healthy buildings program. It just should be the way it's done. So I'm looking forward to the time out of business, hopefully have a healthy building future, then I can retire, be happy, and we'll be onto the next big problem.
Eric Topol (01:00:57):
We'll all be following your writings, which are many, and fortunately not just for science publications, but also for the public though, they're so important because the awareness level as I can't emphasize enough, it's just not there yet. And I think this episode is going to help bring that to a higher level. So Joe, thank you so much for everything you're doing.
Joseph Allen (01:01:20):
Well, I appreciate it. Thanks for what you're doing too, and thanks for inviting me on. We can't get the word out unless we start sharing it across our different audiences, so I appreciate it. Thanks so much.
Eric Topol (01:01:28):
You bet.
***********************************************
A Poll
Thanks for listening, reading or watching!
The Ground Truths newsletters and podcasts are all free, open-access, without ads.
Please share this post/podcast with your friends and network if you found it informative!Voluntary paid subscriptions all go to support Scripps Research. Many thanks for that—they greatly helped fund our summer internship programs for 2023 and 2024.
Thanks to my producer Jessica Nguyen and Sinjun Balabanoff for audio and video support at Scripps Research.
Note: you can select preferences to receive emails about newsletters, podcasts, or all I don’t want to bother you with an email for content that you’re not interested in.

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
Steve Horvath: Our Epigenetic Age Clocks
23 août· Ground Truths
Steve Horvath made the seminal discovery of the—Horvath Clock— an epigenetic clock based on DNA methylation, which is now being used extensively in medical research and offered commercially for individuals (←we talk about that!). He was on the faculty at UCLA from 2000-2022 as a Professor of Human Genetics and Biostatistics, and now works on anti-aging research at Altos Labs.
A perspective on the importance of epigenetic clocks this week’s Nature”This insight is crucial for deriving reliable biological markers of ageing in tissues or blood. Such a feat has been accomplished through the ingenious identification of epigenetic clocks in our genome. But these insights are even more important for revealing targets that enable intervention in the ageing process.”
A video snippet on vegetable intake and epigenetic clocks. Full videos of all Ground Truths podcasts can be seen on YouTube here. The audios are also available on Apple and Spotify.
Transcript with links to Audio and External Links
Eric Topol (00:06):
Hello, it's Eric Topol with Ground Truths, and I've got a terrific guest with me today, Steve Horvath. He's a geneticist, a statistician, a mathematician. He's got a lot of background that has led to what is a landmark finding in biomedicine, the Horvath clock. So Steve, welcome.
Steve Horvath (00:30):
Thank you for having me.
Eric Topol (00:33):
Well, it's really fascinating. I followed your work for well over a decade since you introduced the pan-tissue clock in 2013, and it's fascinating to go back a bit on that finding, which initially, I guess was in saliva a couple of years prior, and then you found it everywhere you looked, wherever cells had a nucleus and tissues. And what gave you the sense that these markers of methylation on the DNA would give us some clues about the aging process? How did you even come about to make this discovery?
Serendipity
Steve Horvath (01:17):
It was an accidental discovery because before the methylation clock, I had worked very hard on a gene expression clock, a transcriptomic biomarker. I mean, I was at the height of my energy levels. I worked really on weekends, really eight hour days during the week. But all the weekends I had collected a large set of gene expression data and I dredged the data. And for two years and I couldn't get anywhere, there was nothing I could do. But nowadays, of course, you see various publications where people built transcriptomic clocks. But back in the day when we had these arrays, I just couldn't see a signal. And then at some point I got roped into a study of homosexuality where my collaborator at UCLA wanted to see whether there's an epigenetic correlate of sexual orientation in saliva. And so yeah, being a biostatistician, I said, sure, I analyzed the data and I couldn't find any signal for homosexuality.
(02:48):
But then I just looked for an aging signal in the same, and really within an hour of analyzing the data, I knew that I have to completely drop gene expression. I need to go after methylation. And the signal is so profound, and as you said initially we looked at saliva samples and we thought, isn't it curious? You spit in a cup and you can measure someone's age. And we were of course, hoping that this could become a valuable readout of biologic age, but it took, of course, many years to realize that potential. Nowadays, there's several companies that offer a saliva based methylation clock test. But yeah, many years passed, and it was important to fill in the details and to build the case that methylation clocks are predictive of things we care about time to death or time to various forms of morbidity. So it took many, many years to analyze large cohort studies and to accumulate the evidence that it actually works.
Eric Topol (04:16):
Yeah, I mean, it was pretty amazing back almost a decade ago when I would see, we would take tissue or blood sample and look at your clock and it would say, age of the person is 75 years. And then we look at the actual age of the person who is 75 years to say, wait a minute, how can this be? So I mean, the plausibility of this discovery, if you look back, I mean you say, well, this is just kind of the rust of the pipes, or how do you process that the methylation is such a marker potentially of a person's biologic age? Of course, we're going to get into how it could be a way to intervene to change the aging process. But would it be fair to say that its epigenetic clocks are not the same as biologic aging or how do you put all that together?
Epigenetic Age vs Biologic Age
Steve Horvath (05:21):
Yes, for sure. An epigenetic age estimate is certainly not the same as a biologic age estimate. And the reason why I say it is because biologic age is really determined by so many things and by so many organs. And as I mentioned initially, we had a clock for saliva later for blood and so on. And so, if you only have an epigenetic readout of a certain cell type, it's really too limited to assess the whole organismal state. And arguably you would want to measure also proteomics, readouts and many other data modalities. So I typically avoid the terminology biologic age, because to begin with, we don't have a definition of it. Decades of discussions, nobody really has a precise definition of it.
Second Generation Epigenetic Clocks
Eric Topol (06:35):
Well, from the first generation Horvath clock then became this newer second generation, GrimAge, PhenoAge, the DunedinPACE of aging. How has that helped to advance the field? Because as you touched on, they're measuring different things and what is it meant by kind of a second generation clock?
Steve Horvath (07:03):
Yeah, so a second generation clock truly aims to predict mortality or morbidity risk. As opposed to simply chronologic age or what is known as calendar age. And fortunately, there's no doubt that the second generation clocks can do that. I often finish a talk on GrimAge by telling the audience that I give them a money back guarantee, that it will be predictive of mortality in their cohort study. I'm 100% certain that it works if you analyze a hundred people or so. The question is more whether an individual could benefit from such a test. And there are now many providers of various epigenetic clock tests. These biomarkers have different names, but they're quite pricey. A couple of hundred dollars are needed to get such a measurement. And the question is, is it helpful for the individual should you get such a test? And I would say we are not quite there yet for a variety of reasons. The main reason being we don't have good interventions against accelerated epigenetic age. So because when you think about it, why does a doctor order a test for you? For example, cholesterol levels. Well, because they have a drug against elevated cholesterol levels, the statin. And at the moment, we don't have validated interventions against accelerated epigenetic age. So that's kind of missing.
Eric Topol (09:13):
Yeah, we're going to get to that because obviously a lot of things are in the pipeline there, but are you saying then that these people that are getting these consumer tests, that they're getting a test that really wasn't validated at an individual level, so it predicts their mortality that it may be good at a cohort or population level, but maybe it's not so helpful, accurate, or would you say it is accurate? I mean, GrimAge is a good name because since it says when you're going to die. How do you make the differentiation between the individual level or beyond?
Steve Horvath (09:59):
Yeah, I think it's good to compare to other biomarkers. So take glucose levels, hemoglobin A1C, nobody doubts that these levels predict mortality risk when you study couples a hundred people. But how accurate is such a test for an individual? Clearly there is substantial noise associated with a prediction. Two people could have exactly the same hemoglobin A1C levels, but live very different lifespans. And the same holds for epigenetic clocks. They do predict how long you live. In theory, one could arrive at an estimate of age and death. There's a complicated mathematical formula that allows you to do that, but there would be a substantial error bar associated with it, an order of magnitude plus minus five years. And so, for the individual, such an estimate is not that important because the error bar is substantial. But I want to add that these second generation clocks, they do predict mortality risk. There's no question.
Maximal Lifespan
Eric Topol (11:35):
Well, as you know, the longevity space is now very crowded with all sorts of clubs, and it's like a circus out there. And some of these things are being promoted that really don't have the basis or have a false sense to consumers who want to live forever and be healthy forever. But maybe these markers are not really helping guide them so much. Now, you recently published you and your group a fascinating paper, so getting away from the individual for a second, but now at the species level and in Science Advances, and we'll put this diagram with the podcast, but you looked at 348 mammal species for the maximal lifespan with DNA methylation. And it was amazing to see the display from the desert hamster all the way to the humpback whale with somewhere along the way, the humans. So you could predict maximal lifespan pretty well, right?
Steve Horvath (12:43):
Yes. So I collected this very large dataset over seven years, and one of the reasons was to understand the mystery of maximum lifespan. The bowhead whale can live over 211 years, whereas certain mice only three or four years. And my question was, can methylation teach us something about maximum lifespan? And the answer is a resounding, yes. The methylation profiles very much predict the maximum lifespan of a species. And maybe to use a metaphor to explain the patterns. So one can visualize methylation around the DNA molecule, like a landscape. You want that certain regions exhibit high levels of methylation. These regions must be really shut down and other parts of the DNA as opposed to exhibit very low methylation, for example, a transcriptional start sites. And long lived species have a very hilly landscapes, high hills of methylation and steep valleys of low methylation. Where shorter lived species have flatter landscapes. So that was one of the insights of that study. The other perhaps paradoxical insight was that the locations in our DNA that gain methylation with chronologic age, these regions often differ from regions that determine the maximum lifespan of our species. So that's a bit perhaps paradoxical and counterintuitive, but it just shows that the DNA encodes our species characteristics at different locations from our mortality risk.
The Other Clocks
Eric Topol (15:13):
Right. No, and I mean it's fascinating. I can imagine how it could take seven years to pull all that data together. It's amazing. Now, one of the issues of course, is if you're trying to gauge the biologic age, which we already established is somewhat different than epigenetic age or a clock, there are many different ways to do that. And you mentioned transcriptome clocks, which are not as well perhaps developed. Obviously, none of these others are developed like the Horvath clock and newer generation clocks, but there's immuno aging clocks like iAge, there's proteomic clocks, there's organ clocks with high-throughput proteomics, thousands of proteins. Do you see these as complimentary, like orthogonal where they each add to the story? Or do you really see the methylation as distinct?
Steve Horvath (16:20):
Well, I think ideally you measure all of the above to really get a very granular understanding of different facets of aging. And however, scientists always like to find deep connections between different readouts. For example, it would be wonderful if we could use proteomics instead of methylation, or my group has worked on the opposite. So we can actually estimate protein levels in the plasma based on methylation for about 10% of all plasma proteins, you can estimate their levels based on methylation. So yeah, people who are interested in these deeper programs that ideally link everything, some sort of aging program that underlies these different manifestations of aging, they will want to reduce everything. But until we have a deeper understanding, I think let's air on the side of measuring too much.
Eric Topol (17:45):
Well, what's interesting, as you mentioned, I didn't realize you could basically impute the protein story from the methylation, but one of the issues is if you want to do 11,000 plasma proteins, it could cost a thousand dollars. But if you want to do a bisulfite methylation, you might do that for very inexpensively. So there's a practical part of this too, and the immune characterization is even more expensive and difficult from a practical standpoint. So we go back to that initial work that you did and how you got into an area that is practical, inexpensive compared to some of the alternatives. But as you say, they may have features that are also helpful. Now, this is now the craze, this epigenetic clocks, and I want to mention you probably didn't see it because it's not a journal that you would look at, but just yesterday, July 29th, there were 12 papers published in JAMA Network Open.
Modulating Your Epigenetic Clock
(18:51):
Everything from how loss of loved ones changes your epigenetic clock to PTSD, to vegan diets, to inequities. I mean, just incredible. So it is the rage now. It's taken the biomedical community some years to catch up to where you were. And one of the things of course that we know that from your prior work that is an intervention that helps give a less accelerated epigenetic clock is exercise. And in fact, that was highlighted in our Lancet essay in the first week August issue. But can you comment on that and anything else that we know like plant diets and anything that favorably influence our DNA methylation pattern?
Steve Horvath (19:52):
Yes. So interestingly, vegetable intake really has a strong effect on GrimAge and many other epigenetic clocks. And maybe this is obvious to the listener, everybody knows that vegetable intake is healthy. However, it's very surprising to me as a scientist to contemplate how is it that vegetable intake affects the methylation levels of your blood? How does it affect the hematopoietic stem cells? I just don't understand the mechanism behind it, and however, the effect is very strong. So we studied postmenopausal women in the women's health initiative, and for these women, we had blood measures of carotenoid levels. So this is an objective measure of vegetable intake, and the correlations were substantial. So that's one intervention I'm quite certain about. Other intervention that have a strong effect relate to metabolic syndrome, anything that relates to type 2 diabetes such as obesity, high glucose levels, that part of the biology very much affects our epigenetic clocks. So disturbed metabolism has a strong effect.
Eric Topol (21:37):
Has these findings changed your diet or made you exercise more or anything like that?
Steve Horvath (21:44):
. So I eat a lot of frozen vegetables. My freezer as full frozen vegetables.
Eric Topol (21:56):
That's great. Well, there's a lot of uses today as we touched on in the Lancet piece as we're waiting for more benchmarking and more work on this. But for example, we have a shortage of donor organs, and there are people who might be of calendar age advanced, but their epigenetic clock might put them at a much younger age. Is that ready for use in the transplant world as one application?
Steve Horvath (22:37):
I haven't seen that yet. I've seen several studies that have explored that idea. The idea is rather obvious, but I haven't seen it implemented in practicum.
Eric Topol (22:53):
Another one is that we don't, as you've seen from some of these studies on organ clocks, our organs age at different paces and some people are accelerated heart agers or brain agers. If you had access to tissue to get methylation, would you see the same thing or this is of course of interest because we're trying to understand high risk individuals for age related diseases, whether it's dementia or heart disease or cancer. So is the second generation clocks like PhenoAge just good enough, or would you think that the organ clocks would give you some added insight?
Steve Horvath (23:47):
Yeah, I would say this is literally the frontier of research. Several groups attempt to use blood methylation or saliva or skin or fat adipose as surrogates for various other organs. And I've seen very encouraging results. So I do think this idea makes scientific sense, and which comes back to one of the miracles of methylation that this is even possible because if you had written a grant 10 years ago where you said, I will measure blood methylation to assess cognitive functioning, for example, you wouldn't have received any score, not in no funding, but however, interestingly, blood methylation does relate to cognitive functioning and many other organ functions. And so, the proof of concepts have been established. Blood methylation relates to fatty liver disease, kidney disease, lung disease. It has all been done in epidemiological studies. However, the question is how much could a blood methylation measurement help an individual? Should I measure my blood methylation to learn about my liver? And I would say we are not there yet because arguably there are wonderful plasma biomarkers to assess organ functions. And in certain ways, one needs to provide evidence that a methylation measurement is superior or compliments plasma based biomarker. And that's a hard hurdle to take.
Eric Topol (26:02):
Right. I imagine someday it may become the norm of assessing people's risk, but as you say, we're not there yet because it's a tough bar to meet, for sure. Now, you were a Professor from year 2000 at UCLA in multiple departments in genetics and biostats, and then in more recent times you joined the Cambridge unit of Altos, which is one of the companies that has gotten the most attention for its diverse efforts towards modulating, rejuvenating the aging process. So you and many top scientists around the world were recruited to Altos. I know some here at the San Diego campus. Was this thinking that it could help accelerate the whole idea of modulating aging in a favorably way or where do you see that the biotech world can play a role?
Can We Change the Pace of Aging?
Steve Horvath (27:15):
Yes. I mean, speaking for myself, I was getting tired of writing scientific papers and not affecting clinical care. I felt I needed to help identify or validate rejuvenating interventions because of the great promise, and this is perhaps best done in the setting of a biotech that is focused on translation. And that's why I joined. I'm moving away from biomarker development towards finding interventions that move the needle and ideally rejuvenate multiple organs and cell types at the same time.
Eric Topol (28:09):
Right. Now, there's lots of ideas of how we could do that from senolytics that would get rid of specific senescent cells that are bad actors to epigenetic reprogramming or chemical reprogramming or so many anti-inflammatory, like the recent paper of IL-11 that I'm sure you saw in Nature just a couple of weeks ago and many, many other ways to get there. What are you thinking? Is this going to be possible? Obviously, there's lots of naysayers. Is it going to be possible body wide or only for specific ways? For example, maybe we could bring back the thymus from its involution or we could stop ovarian failure in women so that their loss of advantage is delayed many years. Or do you think we're going to get to body wide anti-aging?
Steve Horvath (29:13):
Yeah, I think of it as divide and conquer. So ultimately I do believe that we can rejuvenate most cell types and tissues. The question is how do you roll out this program? Do you look for this one silver bullet that does it? For example, this idea of interrupted reprogramming based on Yamanaka factor combinations that looks of course very promising and rodent models. But then such silver bullet treatments could be risky for patient keyword malignant transformation, cancer risk, and it could be far safer to focus on one organ system or one tissue. For example, David Sinclair's company Life Biosciences looks at optic nerve regeneration for a reason. It could be safer. And so yeah, I'm very happy that companies explore different strategies. Certain companies focus on one condition, fatty liver disease or NASH. Other companies focus on immune system restoration. But I think many people think of one condition as really a first step to establish safety and efficacy, and then hopefully they could translate it to other body systems and organ systems.
Eric Topol (31:02):
But is it fair to say you're optimistic that we will be able to change the aging pace in people?
Steve Horvath (31:10):
Yes, I think yes. I'm very optimistic and there are several reasons for this optimism. The first is that dramatic results can be achieved in mice and rats. So we and others have published studies that show that you can reduce the epigenetic age by 30% or so and you can extend the lifespan, and you cited this very exciting paper by Stuart Cook on IL-11 inhibition that just came out in Nature. So I keep seeing these kinds of headlines, and then I want to think that one of these will actually work for humans. So the second thing that makes me optimistic is really this combination of artificial intelligence and biomedical research. Then going forward, robotics. So I can see several ways of accelerating biomedical research. So I'm quite optimistic.
The Role of A.I.
Eric Topol (32:24):
Maybe go a little deeper on the AI potential to help here. How does AI come into play?
Steve Horvath (32:33):
So AI can help in so many different ways. The first topic is biomarker development. I of course spent 10 years on a certain statistical model for building biomarkers, which is known as penalized regression. It works well, but AI allows the community to build imaging based biomarkers. So for example, based on MRI images, but also cells growing in a dish, we can say this treatment aged the cells growing in the dish or rejuvenated them. So that's one topic, biomarker discovery. The second is, of course, to design small molecules, keyword, these protein design where it has greatly accelerated drug discovery. And there are several companies working in this space, and again, there's wonderful case studies that look very convincing to me. And the third aspect of AI is another obvious one. AI can read many papers. I mean, you could be a 50-year-old professor who has read papers their entire life, but an AI can really read far better and summarize insights better.
Eric Topol (34:27):
Yeah, the complimentary in terms of the reasoning of that information. So absolutely right now, one of the problems we have here is that aging is not seen as a disease. Of course, we can remember when obesity was not considered a disease and then there was a drug and everything changed. But here we don't have a classification it's a disease. It's considered a natural process that is highly variable in people. But the question is, we can't do studies that are going to wait 20, 30 years to find out if we promoted health span and lifespan. And so, we have to rely on these clocks. So how do you see this playing out? Do you think that we might see a regulatory approval on a surrogate proxy, like an advanced Horvath clock, or do you think that's not going to cut it, that you're going to have to show more to get a anti-aging treatment across the regulatory threshold?
Steve Horvath (35:42):
Yeah, that's a very good question. So I believe that the biomarker community has already assembled enough evidence to offer a battery of tests that could be used as surrogate endpoints of interventional study. And we could discuss the components of this battery. But I would say we already have biomarkers beyond just methylation. One could have the readouts of walking speed or muscle function, many readouts, and they could be aggregated into an index to summarize the biologic age, perhaps, of the individual. So that already exists. At the same time, this field is undergoing explosive growth. You mentioned every day new papers come out in the relatively small field of epigenetic clocks. There's so many papers that it's hard to keep track, but I embrace it. I think it's wonderful because clocks get ever more powerful.
(37:11):
So yeah, I would say there should be different versions. Ideally, a regulatory agency would make an executive decision and say, for the next three years, use the following five biomarkers. Then a few years later, as the science advances, they could come up with an updated version, but even a 90% solution would very much accelerate progress in the whole field of rejuvenating interventions. So I would very much embrace a top down decision on which biomarkers should be made, because the bottom up approach, by the way, simply doesn't work. The minute you put three professors in the room to come up with a decision, which biomarker is best, there will be three different opinions. We need impartial arbiter that makes a decision.
GLP-1 Drugs and Aging
Eric Topol (38:23):
Now, the drug class that's come on the scene, of course it was in incubating for decades for diabetes, but now obesity and so of the obesity related. But now we're seeing the GLP-1 drugs that are showing potential effects in Parkinson's and Alzheimer's and cardiovascular disease, and even in obesity related cancers. And I mean across the board. And you mentioned metabolic derangement as one of the things that accelerate aging. Do you think these class of drugs that has greatly passed our expectations already and it's being tested of course, with even more potent drugs or triple receptors and pills and whatnot, will that be a candidate as one of the anti-aging interventions in the future?
Steve Horvath (39:19):
Yeah, for sure. A couple of months ago, I participated in a conference and one of the speakers showed unpublished results from a study, and they looked good to me. I mean, they registered on epigenetic clocks. This is all unpublished, but it made perfect sense to me because I mentioned the clocks do relate to metabolic health. So I was quite pleased that they registered that intervention.
Eric Topol (39:56):
It's fascinating because we could all be taking GLP-1 drugs someday, not for obesity or not for sleep apnea, but for things that are more far reaching. I didn't know about that unpublished result. That's fascinating.
Steve Horvath (40:15):
Yeah, I have a joke, which is I wish I was chubby because I would be using these drugs, but I'm relatively slender, so I don't have any good reason to take them.
Eric Topol (40:28):
That says a lot. I don't know anybody who knows more about this process than you and is very candid and frank about it. So Steve, this has been terrific to have your insights, the body of work that you should be so proud of that extends over many years and many great years and more contributions to come undoubtedly. So thank you for joining us today, and we will follow this continued evolution of our ability, not just to track the aging process, but also to modulate. So thanks very much.
Steve Horvath (41:06):
Thank you. I really like your podcast Ground Truths, it’s very informative. So thank you for this.
********************************************************
We’re appreciative of your reading and subscribing to Ground Truths. All content of these newsletter/analyses and podcasts are free, with no advertisements.
Thanks to my producer Jessica Nguyen and Sinjun Balabanoff for audio and video support at Scripps Research.
Please share this post, especially if you found it interesting or informative.
If you do elect to be a paid subscriber, all proceeds are donated to Scripps Research and such support has already been extremely helpful for our summer internship programs and other educational activities. Comments and questions from paid subscribers are welcome.

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
Pradeep Natarajan: Preventing Heart Disease
13 août· Ground Truths
Pradeep is a brilliant geneticist and Director of Preventive Cardiology, holds the Paul & Phyllis Fireman Endowed Chair in Vascular Medicine at Mass General Hospital and on faculty at Harvard Medical School and the Broad Institute. His prolific research has been illuminating for the field of improving our approach to reduce the risk of heart disease. That’s especially important because heart disease is the global (and US) #1 killer and is on the increase. We didn’t get into lifestyle factors here since there was so much ground to cover on new tests. drugs, and strategies.
A video snippet of our conversation on ApoB. Full videos of all Ground Truths podcasts can be seen on YouTube here. The audios are also available on Apple and Spotify.
Transcript with links to key publications and audio
Eric Topol (00:06):
Well, welcome to Ground Truths. I'm Eric Topol and with me is Pradeep Natarajan from Harvard. He's Director of Preventative Cardiology at the Mass General Brigham Health System and he has been lighting it up on the field of cardiovascular. We're going to get to lots of different parts of that story and so, Pradeep welcome.
Pradeep Natarajan (00:31):
Thanks Eric, really delighted and honored to be with you and have this discussion.
Eric Topol (00:36):
Well, for years I've been admiring your work and it's just accelerating and so there's so many things to get to. I thought maybe what we'd start off with is you recently wrote a New England Journal piece about two trials, two different drugs that could change the landscape of cardiovascular prevention in the future. I mean, that's one of the themes we're going to get to today is all these different markers and drugs that will change cardiology as we know it now. So maybe you could just give us a skinny on that New England Journal piece.
Two New Lipid Targets With RNA Drugs
Pradeep Natarajan (01:16):
Yeah, yeah, so these two agents, the trials were published at the same time. These phase two clinical trials for plozasiran, which is an siRNA against APOC3 and zodasiran, which is an siRNA against ANGPTL3. The reason why we have medicines against those targets are based on human genetics observations, that individuals with loss of function mutations and either of those genes have reduced lipids. For APOC3, it's reduced triglycerides for ANGPTL3 reduced LDL cholesterol and reduced triglycerides and also individuals that have those loss of function mutations also have lower risk for coronary artery disease. Now that's a very similar parallel to PCSK9. We have successful medicines that treat that target because people have found that carriers of loss of function mutations in PCSK9 lead to lower LDL cholesterol and lower coronary artery disease.
(02:11):
Now that suggests that therapeutic manipulation without significant side effects from the agents themselves for APOC3 and ANGPTL3 would be anticipated to also lower coronary artery disease risk potentially in complementary pathways to PCSK9. The interesting thing with those observations is that they all came from rare loss of function mutations that are enriched in populations of individuals. However, at least for PCSK9, has been demonstrated to have efficacy in large groups of individuals across different communities. So the theme of that piece was really just the need to study diverse populations because those insights are not always predictable about which communities are going to have those loss of function mutations and when you find them, they often have profound insights across much larger groups of individuals.
Eric Topol (03:02):
Well, there's a lot there that we can unpack a bit of it. One of them is the use of small interfering RNAs (siRNA) as drugs. We saw in the field of PCSK9, as you mentioned. First there were monoclonal antibodies directed against this target and then more recently, there’s inclisiran which isn't an RNA play if you will, where you only have to take it twice a year and supposedly it's less expensive and I’m still having trouble in my practice getting patients covered on their insurance even though it's cheaper and much more convenient. But nonetheless, now we're seeing these RNA drugs and maybe you could comment about that part and then also the surprise that perhaps is unexplained is the glucose elevation.
Pradeep Natarajan (03:53):
Yeah, so for medicines and targets that have been discovered through human genetics, those I think are attractive for genetic-based therapies and longer interval dosing for the therapies, which is what siRNAs allow you to do because the individuals that have these perturbations, basically the naturally occurring loss of function mutations, they have these lifelong, so basically have had a one-time therapy and have lived, and so far, at least for these targets, have not had untoward side effects or untoward phenotypic consequences and only reduce lipids and reduce coronary artery disease. And so, instead of taking a pill daily, if we have conviction that that long amount of suppression may be beneficial, then longer interval dosing and not worrying about the pill burden is very attractive specifically for those specific therapeutics. And as you know, people continue to innovate on further prolonging as it relates to PCSK9.
(04:57):
Separately, some folks are also developing pills because many people do feel that there's still a market and comfort for daily pills. Now interestingly for the siRNA for zodasiran at the highest dose, actually for both of them at the highest doses, but particularly for zodasiran, there was an increase in insulin resistance parameters actually as it relates to hyperglycemia and less so as it relates to insulin resistance, that is not predicted based on the human genetics. Individuals with loss of function mutations do not have increased risks in hyperglycemia or type 2 diabetes, so that isolates it related to that specific platform or that specific technology. Now inclisiran, as you'd mentioned, Eric is out there. That's an siRNA against PCSK9 that's made by a different manufacturer. So far, the clinical trials have not shown hyperglycemia or type 2 diabetes as it relates inclisiran, so it may be related to the specific siRNAs that are used for those targets. That does merit further consideration. Now, the doses that the manufacturers do plan to use in the phase three clinical trials are at lower doses where there was not an increase in hyperglycemia, but that does merit further investigation to really understand why that's the case. Is that an expected generalized effect for siRNAs? Is it related to siRNAs for this specific target or is it just related to the platform used for these two agents which are made by the same manufacturer?
Eric Topol (06:27):
Right, and I think the fact that it's a mystery is intriguing at the least, and it may not come up at the doses that are used in the trials, but the fact that it did crop up at high doses is unexpected. Now that is part of a much bigger story is that up until now our armamentarium has been statins and ezetimibe to treat lipids, but it's rapidly expanding Lp(a), which for decades as a cardiologist we had nothing to offer. There may even be drugs to be able to lower people who are at high risk with high Lp(a). Maybe you could discuss that.
What About Lp(a)?
Pradeep Natarajan (07:13):
Yeah, I mean, Eric, as you know, Lp(a) has been described as a cardiovascular disease risk factors for quite so many years and there are assays to detect lipoprotein(a) elevation and have been in widespread clinical practice increasing widespread clinical practice, but we don't yet have approved therapies. However, there is an abundance of literature preclinical data that suggests that it likely is a causal factor, meaning that if you lower lipoprotein(a) when elevated, you would reduce the risk related to lipoprotein(a). And a lot of this comes from similar human genetic studies. The major challenge of just relating a biomarker to an outcome is there are many different reasons why a biomarker might be elevated, and so if you detect a signal that correlates a biomarker, a concentration to a clinical outcome, it could be related to that biomarker, but it could be to the other reasons that the biomarker is elevated and sometimes it relates to the outcome itself.
(08:10):
Now human genetics is very attractive because if you find alleles that strongly relate to that exposure, you can test those alleles themselves with the clinical outcome. Now the allele assignment is established at birth. No other factor is going to change that assignment after conception, and so that provides a robust, strong causal test for that potential exposure in clinical outcome. Now, lipoprotein(a) is unique in that it is highly heritable and so there are lots of different alleles that relate to lipoprotein(a) and so in a well powered analysis can actually test the lipoprotein(a) SNPs with the clinical outcomes and similar to how there is a biomarker association with incident myocardial infarction and incident stroke, the SNPs related to lipoprotein(a) show the same. That is among the evidence that strongly supports that this might be causal. Now, fast forward to many years later, we have at least three phase three randomized clinical trials testing agents that have been shown to be very potent at lowering lipoprotein(a) that in the coming years we will know if that hypothesis is true. Importantly, we will have to understand what are the potential side effects of these medicines. There are antisense oligonucleotides and siRNAs that are primarily in investigation. Again, this is an example where there's a strong genetic observation, and so these genetic based longer interval dosing therapies may be attractive, but side effects will be a key thing as well too. Those things hard to anticipate really can anticipate based on the human genetics for off target effects, for example.
(09:52):
It's clearly a risk signal and hopefully in the near future we're going to have specific therapies.
Eric Topol (09:57):
Yeah, you did a great job of explaining Mendelian randomization and the fact the power of genetics, which we're going to get into deeper shortly, but the other point is that do you expect now that there's these multiple drugs that lower Lp(a) efficiently, would that be enough to get approval or will it have to be trials to demonstrate improved cardiovascular outcomes?
Pradeep Natarajan (10:24):
There is a great regulatory path at FDA for approval just for LDL cholesterol lowering and inclisiran is on the market and the phase three outcomes data has not yet been reported because there is a wide appreciation that LDL cholesterol lowering is a pretty good surrogate for cardiovascular disease risk lowering. The label will be restricted to LDL cholesterol lowering and then if demonstrated to have clinical outcomes, the label could be expanded. For other biomarkers including lipoprotein(a), even though we have strong conviction that it is likely a causal factor there hasn't met the bar yet to get approval just based on lipoprotein(a) lowering, and so we would need to see the outcomes effects and then we would also need to understand side effects. There is a body of literature of side effects for other therapies that have targeted using antisense oligonucleotides. We talked about potential side effects from some siRNA platforms and sometimes those effects could overtake potential benefits, so that really needs to be assessed and there is a literature and other examples.
(11:31):
The other thing I do want to note related to lipoprotein(a) is that the human genetics are modeled based on lifelong perturbations, really hard to understand what the effects are, how great of an effect there might be in different contexts, particularly when introduced in middle age. There's a lot of discussion about how high lipoprotein(a) should be to deliver these therapies because the conventional teaching is that one in five individuals has high lipoprotein(a), and that's basically greater than 75 nanomoles per liter. However, some studies some human genetic studies to say if you want to get an effect that is similar to the LDL cholesterol lowering medicines on the market, you need to start with actually higher lipoprotein(a) because you need larger amounts of lipoprotein(a) lowering. Those are studies and approaches that haven't been well validated. We don't know if that's a valid approach because that's modeling based on this sort of lifelong effect. So I'm very curious to see what the overall effect will be because to get approval, I think you need to demonstrate safety and efficacy, but most importantly, these manufacturers and we as clinicians are trying to find viable therapies in the market that it won't be hard for us to get approval because hopefully the clinical trial will have said this is the context where it works. It works really well and it works really well on top of the existing therapies, so there are multiple hurdles to actually getting it directly to our patients.
How Low Do You Go with LDL Cholesterol?
Eric Topol (13:02):
Yeah, no question about that. I'm glad you've emphasized that. Just as you've emphasized the incredible lessons from the genetics of people that have helped guide this renaissance to better drugs to prevent cardiovascular disease. LDL, which is perhaps the most impressive surrogate in medicine, a lab test that you already touched on, one of the biggest questions is how low do you go? That is Eugene Braunwald, who we all know and love. They're in Boston. The last time I got together with him, he was getting his LDL down to close to zero with various tactics that might be extreme. But before we leave these markers, you're running preventive cardiology at man's greatest hospital. Could you tell us what is your recipe for how aggressive do you go with LDL?
Pradeep Natarajan (14:04):
Yeah, so when I talk to patients where we're newly getting lipid lowering therapies on, especially because many people don't have a readout of abnormal LDL cholesterol when we're prescribing these medicines, it's just giving them a sense of what we think an optimal LDL cholesterol might be. And a lot of this is based on just empirical observations. So one, the average LDL cholesterol in the modern human is about 100 to 110 mg/dL. However, if you look at contemporary hunter gatherers and non-human primates, their average LDL is about 40 to 50 and newborn babies have an LDL cholesterol of about 30. And the reason why people keep making LDL cholesterol lowering medicines because as you stack on therapies, cardiovascular disease events continue to reduce including down to these very low LDL cholesterol values. So the population mean for LDL cholesterol is high and everybody likely has hypercholesterolemia, and that's because over the last 10,000 years how we live our lives is so dramatically different and there has not been substantial evolution over that time to change many of these features related to metabolism.
(15:16):
And so, to achieve those really low LDL cholesterol values in today's society is almost impossible without pharmacotherapies. You could say, okay, maybe everybody should be on pharmacotherapies, and I think if you did that, you probably would reduce a lot of events. You'll also be treating a lot of individuals who likely would not get events. Cardiovascular disease is the leading killer, but there are many things that people suffer from and most of the times it still is not cardiovascular disease. So our practice is still rooted in better identifying the individuals who are at risk for cardiovascular disease. And so, far we target our therapies primarily in those who have already developed cardiovascular disease. Maybe we'll talk about better identifying those at risk, but for those individuals it makes lots of sense to get it as low as possible. And the field has continued to move to lower targets.
(16:07):
One, because we've all recognized, at least based on these empirical observations that lower is better. But now increasingly we have a lot of therapies to actually get there, and my hope is that with more and more options and the market forces that influence that the cost perspective will make sense as we continue to develop more. As an aside, related aside is if you look at the last cholesterol guidelines, this is 2018 in the US this is the first time PCSK9 inhibitors were introduced in the guidelines and all throughout that there was discussions of cost. There are a lot of concerns from the field that PCSK9 inhibitors would bankrupt the system because so many people were on statins. And you look at the prior one that was in 2013 and cost was mentioned once it’s just the cost effectiveness of statins. So I think the field has that overall concern.
(17:01):
However, over time we've gotten comfortable with lower targets, there are more medicines and I think some of this competition hopefully will drive down some of the costs, but also the overall appreciation of the science related to LDL. So long-winded way of saying this is kind of the things that we discussed just to give reassurance that we can go to low LDL cholesterol values and that it's safe and then we think also very effective. Nobody knows what the lower limit is, whether zero is appropriate or not. We know that glucose can get too low. We know that blood pressure can be too low. We don't know yet that limit for LDL cholesterol. I mean increasingly with these trials we'll see it going down really low and then we'll better appreciate and understand, so we'll see 40 is probably the right range.
Eric Topol (17:49):
40, you said? Yeah, okay, I'll buy that. Of course, the other thing that we do know is that if you push to the highest dose statins to get there, you might in some people start to see the hyperglycemia issue, which is still not fully understood and whether that is, I mean it's not desirable, but whether or not it is an issue, I guess it's still out there dangling. Now the other thing that since we're on LDL, we covered Lp(a), PCSK9, the siRNA, is ApoB. Do you measure ApoB in all your patients? Should that be the norm?
Measuring ApoB
Pradeep Natarajan (18:32):
Yeah, so ApoB is another blood test. In the standard lipid panel, you get four things. What's measured is cholesterol and triglycerides, they're the lipids insoluble in blood to get to the different tissues that get packaged in lipoprotein molecules which will have the cholesterol, triglycerides and some other lipids and proteins. And so, they all have different names as you know, right? Low density lipoprotein, high density lipoprotein and some others. But also in the lipid panel you get the HDL cholesterol, the amount of cholesterol in an HDL particle, and then most labs will calculate LDL cholesterol and LDL cholesterol has a nice relationship with cardiovascular disease. You lower it with statins and others. Lower risk for cardiovascular disease, turns out a unifying feature of all of these atherogenic lipoproteins, all these lipoproteins that are measured and unmeasured that relate to cardiovascular disease, including lipoprotein(a), they all have an additional protein called ApoB. And ApoB, at least as it relates to LDL is a pretty good surrogate of the number of LDL particles.
(19:37):
Turns out that that is a bit better at the population level at predicting cardiovascular disease beyond LDL cholesterol itself. And where it can be particularly helpful is that there are some patients out there that have an unexpected ratio between ApoB and LDL. In general, the ratio between LDL cholesterol and ApoB is about 1.1 and most people will have that rough ratio. I verify that that is the expected, and then if that is the expected, then really there is no role to follow ApoB. However, primarily the patients that have features related to insulin resistance have obesity. They may often have adequate looking LDL cholesterols, but their ApoB is higher. They have more circulating LDL particles relative to the total amount of LDL cholesterol, so smaller particles themselves. However, the total number of particles may actually be too high for them.
(20:34):
And so, even if the LDL cholesterol is at target, if the ApoB is higher, then you need to reduce. So usually the times that I just kind of verify that I'm at appropriate target is I check the LDL cholesterol, if that looks good, verify with the ApoB because of this ratio, the ApoB target should be about 10% lower. So if we're aiming for about 40, that's like 36, so relatively similar, and if it's there, I'm good. If it's not and it's higher, then obviously increase the LDL cholesterol lowering medicines because lower the ApoB and then follow the ApoB with the lipids going forward. The European Society of Cardiology has more emphasis on measuring ApoB, that is not as strong in the US guidelines, but there are many folks in the field, preventive cardiologists and others that are advocating for the increasing use of ApoB because I think there are many folks that are not getting to the appropriate targets because we are not measuring ApoB.
Why Aren’t We Measuring and Treating Inflammation?
Eric Topol (21:37):
Yeah, I think you reviewed it so well. The problem here is it could be part of the standard lipid panel, it would make this easy, but what you've done is a prudent way of selecting out people who it becomes more important to measure and moderate subsequently. Now this gets us to the fact that we're lipid centric and we don't pay homage to inflammation. So I wrote a recent Substack on the big miss on inflammation, and here you get into things like the monoclonal antibody to interleukin-6, the trial that CANTOS that showed significant reduction in cardiovascular events and fatal cancers by the way. And then you get into these colchicine trials two pretty good size randomized trials, and here the entry was coronary disease with a high C-reactive protein. Now somehow or other we abandon measuring CRP or other inflammatory markers, and both of us have had patients who have low LDLs but have heart attacks or significant coronary disease. So why don't we embrace inflammation? Why don't we measure it? Why don't we have better markers? Why is this just sitting there where we could do so much better? Even agents that are basically cost pennies like colchicine at low doses, not having to use a proprietary version could be helpful. What are your thoughts about us upgrading our prevention with inflammation markers?
Pradeep Natarajan (23:22):
Yeah, I mean, Eric, there is an urgent need to address these other pathways. I say urgent need because heart disease has the dubious distinction of being the leading killer in the US and then over the last 20 years, the leading killer in the world as it takes over non-communicable diseases. And really since the early 1900s, there has been a focus on developing pharmacotherapies and approaches to address the traditional modifiable cardiovascular disease risk factors. That has done tremendous good, but still the curves are largely flattening out. But in the US and in many parts of the world, the deaths attributable to cardiovascular disease are starting to tick up, and that means there are many additional pathways, many of them that we have well recognized including inflammation. More recently, Lp(a) that are likely important for cardiovascular disease, for inflammation, as you have highlighted, has been validated in randomized controlled trials.
(24:18):
Really the key trial that has been more most specific is one on Canakinumab in the CANTOS trial IL-1β monoclonal antibody secondary prevention, so cardiovascular disease plus high C-reactive protein, about a 15% reduction in cardiovascular disease and also improvement in cancer related outcomes. Major issues, a couple of issues. One was increased risk for severe infections, and the other one is almost pragmatic or practical is that that medicine was on the market at a very high price point for rare autoinflammatory conditions. It still is. And so, to have for a broader indication like cardiovascular disease prevention would not make sense at that price point. And the manufacturer tried to go to the FDA and focus on the group that only had C-reactive protein lowering, but that's obviously like a backwards endpoint. How would you know that before you release the medicine? So that never made it to a broader indication.
(25:14):
However, that stuck a flag in the broader validation of that specific pathway in cardiovascular disease. That pathway has direct relevance to C-reactive protein. C-reactive protein is kind of a readout of that pathway that starts from the NLRP3 inflammasome, which then activates IL-1β and IL-6. C-reactive protein we think is just a non causal readout, but is a reliable test of many of these features and that's debatable. There may be other things like measuring IL-6, for example. So given that there is actually substantial ongoing drug development in that pathway, there are a handful of companies with NLRP3 inflammasome inhibitors, but small molecules that you can take as pills. There is a monoclonal antibody against IL-6 that's in development ziltivekimab that's directed at patients with chronic kidney disease who have lots of cardiovascular disease events despite addressing modifiable risk factors where inflammatory markers are through the roof.
(26:16):
But then you would also highlighted one anti-inflammatory that's out there that's pennies on the dollar, that's colchicine. Colchicine is believed to influence cardiovascular disease by inhibiting NLRP3, I say believed to. It does a lot of things. It is an old medicine, but empirically has been shown in at least two randomized controlled trials patients with coronary artery disease, actually they didn't measure C-reactive protein in the inclusion for these, but in those populations we did reduce major adverse cardiovascular disease events. The one thing that does give me pause with colchicine is that there is this odd signal for increased non-cardiovascular death. Nobody understands if that's real, if that's a fluke. The FDA just approved last year low dose colchicine, colchicine at 0.5 milligrams for secondary prevention given the overwhelming efficacy. Hasn’t yet made it into prevention guidelines, but I think that's one part that does give me a little bit pause. I do really think about it particularly for patients who have had recurrent events. The people who market the medicine and do research do remind us that C-reactive protein was not required in the inclusion, but nobody has done that secondary assessment to see if measuring C-reactive protein would be helpful in identifying the beneficial patients. But I think there still could be more work done on better identifying who would benefit from colchicine because it's an available and cheap medicine. But I'm excited that there is a lot of development in this inflammation area.
Eric Topol (27:48):
Yeah, well, the development sounds great. It's probably some years away. Do you use colchicine in your practice?
Pradeep Natarajan (27:56):
I do. Again, for those folks who have had recurrent events, even though C-reactive protein isn't there, it does make me feel like I'm treating inflammation. If C-reactive protein is elevated and then I use it for those patients, if it's not elevated, it's a much harder sell from my standpoint, from the patient standpoint. At the lower dose for colchicine, people generally are okay as far as side effects. The manufacturer has it at 0.5 milligrams, which is technically not pennies on the dollar. That's not generic. The 0.6 milligrams is generic and they claim that there is less side effects at the 0.5 milligrams. So technically 0.6 milligrams is off label. So it is what it is.
CHIP and Defining High Risk People for CV Disease
Eric Topol (28:40):
It's a lot more practical, that's for sure. Now, before I leave that, I just want to mention when I reviewed the IL-1β trial, you mentioned the CANTOS trial and also the colchicine data. The numbers of absolute increases for infection with the antibody or the cancers with the colchicine are really small. So I mean the benefit was overriding, but I certainly agree with your concern that there's some things we don't understand there that need to be probed more. Now, one of the other themes, well before one other marker that before we get to polygenic risk scores, which is center stage here, defining high risk people. We've talked a lot about the conventional things and some of the newer ways, but you've been one of the leaders of study of clonal hematopoiesis of indeterminate potential known as CHIP. CHIP, not the chips set in your computer, but CHIP. And basically this is stem cell mutations that increase in people as we age and become exceptionally common with different mutations that account in these clones. So maybe you can tell us about CHIP and what I don't understand is that it has tremendous correlation association with cardiovascular outcomes adverse as well as other system outcomes, and we don't measure it and we could measure it. So please take us through what the hell is wrong there.
Pradeep Natarajan (30:14):
Yeah, I mean this is really exciting. I mean I'm a little bit biased, but this is observations that have been made only really over the last decade, but accelerating research. And this has been enabled by advances in genomic technologies. So about 10 years or plus ago, really getting into the early days of population-based next generation sequencing, primarily whole exome sequencing. And most of the DNA that we collect to do these population-based analyses come from the blood, red blood cells are anucleate, so they're coming from white blood cells. And so, at that time, primarily interrogating what is the germline genetic basis for coronary artery disease and early onset myocardial infarction. At the same time, colleagues at the Broad Institute were noticing that there are many additional features that you can get from the blood-based DNA that was being processed by the whole exome data. And there were actually three different groups that converged on that all in Boston that converged on the same observation that many well-established cancer causing mutations.
(31:19):
So mutations that are observed in cancers that have been described to drive the cancers themselves were being observed in these large population-based data sets that we were all generating to understand the relationship between loss of function mutations in cardiovascular disease. That's basically the intention of those data sets for being generated for other things. Strong correlation with age, but it was very common among individuals greater than 70; 10% of them would have these mutations and is very common because blood cancer is extremely, it's still pretty rare in the population. So to say 10% of people had cancer causing driver mutations but didn't have cancer, was much higher than anyone would've otherwise expected. In 2014, there were basically three main papers that described that, and they also observed that there is a greater risk of death. You'd say, okay, this is a precancerous lesion, so they're probably dying of cancer.
(32:17):
But as I said, the absolute incidence rate for blood cancer is really low and there's a relative increase for about tenfold, but pretty small as it relates to what could be related to death. And in one of the studies we did some exploratory analysis that suggested maybe it's actually the most common cause of death and that was cardiovascular disease. And so, a few years later we published a study that really in depth really looked at a bunch of different data sets that were ascertained to really understand the relationship between these mutations, these cancer causing mutations in cardiovascular disease, so observed it in enrichment and older individuals that had these mutations, CHIP mutations, younger individuals who had early onset MI as well too, and then also look prospectively and showed that it related to incident coronary artery disease. Now the major challenge for this kind of analysis as it relates to the germline genetic analysis is prevalence changes over time.
(33:15):
There are many things that could influence the presence of clonal hematopoiesis. Age is a key enriching factor and age is the best predictor for cardiovascular disease. So really important. So then we modeled it in mice. It was actually a parallel effort at Boston University (BU) that was doing the same thing really based on the 2014 studies. And so, at the same time we also observed when you modeled this in mice, you basically perturb introduce loss of function mutations in the bone marrow for these mice to recapitulate these driver mutations and those mice also have a greater burden of atherosclerosis. And Eric, you highlighted inflammation because basically the phenotype of these cells are hyper inflamed cells. Interestingly, C-reactive protein is only modestly elevated. So C-reactive protein is not fully capturing this, but many of the cytokines IL-1β, IL-6, they're all upregulated in mice and in humans when measured as well.
(34:11):
Now there've been a few key studies that have been really exciting about using anti-inflammatories in this pathway to address CHIP associated cardiovascular disease. So one that effort that I said in BU because they saw these cytokines increased, we already know that these cytokines have relationship with atherosclerosis. So they gave an NLRP3 inflammasome inhibitor to the mice and they showed that the mice with or without CHIP had a reduction in atherosclerosis, but there was a substantial delta among the mice that are modeled as having CHIP. Now, the investigators in CANTOS, the manufacturers, they actually went back and they survey where they had DNA in the CANTOS trial. They measured CHIP and particularly TET2 CHIP, which is the one that has the strongest signal for atherosclerosis. As I said, overall about 15% reduction in the primary outcome in CANTOS. Among the individuals who had TET2 CHIP, it was a 64% reduction in event.
(35:08):
I mean you don't see those in atherosclerosis related trials. Now this has the caveat of it being secondary post hoc exploratory, the two levels of evidence. And so, then we took a Mendelian randomization approach. Serendipitously, just so happens there is a coding mutation in the IL-6 receptor, a missense mutation that in 2012 was described that if you had this mutation, about 40% of people have it, you have a 5%, but statistically significant reduction in coronary artery disease. So we very simply said, if the pathway of this NLRP3 inflammasome, which includes IL-6, if you have decreased signaling in that pathway, might you have an even greater benefit from having that mutation if you had CHIP versus those who didn't have CHIP. So we looked in the UK Biobank, those who didn't have CHIP 5% reduction, who had that IL-6 receptor mutation, and then those who did have CHIP, if they had that mutation, it was about a 60% reduction in cardiovascular disease.
(36:12):
Again, three different lines of evidence that really show that this pathway has relevance in the general population, but the people who actually might benefit the most are those with CHIP. And I think as we get more and more data sets, we find that not all of the CHIP mutations are the same as it relates to cardiovascular disease risk. It does hone in on these key subsets like TET2 and JAK2, but this is pretty cool as a preventive cardiologist, new potential modifiable risk factor, but now it's almost like an oncologic paradigm that is being applied to coronary artery disease where we have specific driver mutations and then we're tailoring our therapies to those specific biological drivers for coronary artery disease. Hopefully, I did that justice. There's a lot there.
Why Don’t We Measure CHIP?
Eric Topol (36:57):
Well, actually, it's phenomenal how you've explained that, but I do want to review for our listeners or readers that prior to this point in our conversation, we were talking about germline mutations, the ones you're born with. With CHIP, we're talking about acquired somatic mutations, and these are our blood stem cells. And what is befuddling to me is that with all the data that you and others, you especially have been publishing and how easy it would be to measure this. I mean, we've seen that you can get it from sequencing no less other means. Why we don't measure this? I mean, why are we turning a blind eye to CHIP? I just don't get it. And we keep calling it of indeterminate potential, not indeterminate. It's definite potential.
Pradeep Natarajan (37:51):
Yeah, no, I think these are just overly cautious terms from the scientists. Lots of people have CHIP, a lot of people don't have clinical outcomes. And so, I think from the lens of a practicing hematologists that provide some reassurance on the spectrum for acquired mutation all the way over to leukemia, that is where it comes from. I don't love the acronym as well because every subfield in biomedicine has its own CHIP, so there's obviously lots of confusion there. CH or clinical hematopoiesis is often what I go, but I think continuing to be specific on these mutations. Now the question is why measure? Why aren't we measuring it? So there are some clinical assays out there. Now when patients get evaluated for cytopenias [low cell counts], there are next generation sequencing tests that look for these mutations in the process for evaluation. Now, technically by definition, CHIP means the presence of these driver mutations that have expanded because it's detectable by these assays, not a one-off cell because it can only be detected if it's in a number of cells.
(38:55):
So there has been some expansion, but there are no CBC abnormalities. Now, if there's a CBC abnormality and you see a CHIP mutation that's technically considered CCUS or clonal cytopenia of unknown significance, sometimes what is detected is myelodysplastic syndrome. In those scenarios still there is a cardiovascular disease signal, and so many of our patients who are seen in the cancer center who are being evaluated for these CBC abnormalities will be detected to have these mutations. They will have undergone some risk stratification to see what the malignancy potential is. Still pretty low for many of those individuals. And so, the major driver of health outcomes for this finding may be cardiovascular. So those patients then get referred to our program. Dana-Farber also has a similar program, and then my colleague Peter Libby at the Brigham often sees those patients as well. Now for prospective screening, so far, an insurance basically is who's going to pay for it.
(39:51):
So an insurance provider is not deemed that appropriate yet. You do need the prospective clinical trials because the medicines that we're talking about may have side effects as well too. And what is the yield? What is the diagnostic yield? Will there actually be a large effect estimate? But there has been more and more innovation, at least on the assay and the cost part of the assay because these initial studies, we've been using whole exome sequencing, which is continuing to come down, but is not a widely routine clinical test yet. And also because as you highlighted, these are acquired mutations. A single test is not necessarily one and done. This may be something that does require surveillance for particular high risk individuals. And we've described some risk factors for the prevalence of CHIP. So surveillance may be required, but because there are about 10 genes that are primarily implicated in CHIP, that can substantially decrease the cost of it. The cost for DNA extraction is going down, and so there are research tests that are kind of in the $10 to $20 range right now for CHIP. And if flipped over to the clinical side will also be reasonably low cost. And so, for the paradigm for clinical implementation, that cost part is necessary.
Eric Topol (41:10):
I don't know the $10 or $20 ones. Are there any I could order on patients that I'm worried about?
Pradeep Natarajan (41:17):
Not yet clinical. However, there is a company that makes the reagents for at least the cores that are developing this. They are commercializing that test so that many other cores, research cores can develop it. I think it's in short order that clinical labs will adopt it as well too.
Eric Topol (41:36):
That's great.
Pradeep Natarajan (41:37):
I will keep you apprised.
What About Polygenic Risk Scores?
Eric Topol (41:39):
I think that's really good news because like I said, we're so darn lipid centric and we have to start to respect the body of data, the knowledge that you and others have built about CHIP. Now speaking of another one that drives me nuts is polygenic risk score (PRS) for about a decade, I've been saying we have coronary disease for most people is a polygenic trait. It's not just a familial hypercholesterolemia. And we progressively have gotten better and better of the hundreds of single variants that collectively without a parental history will be and independently predict who is at double, triple or whatever risk of getting heart disease, whereby you could then guide your statins at higher aggressive or pick a statin, use one or even go beyond that as we've been talking about. But we don't use that in practice, which is just incredible because it's can be done cheap.
(42:45):
You can get it through whether it's 23andMe or now many other entities. We have an app, MyGeneRank where we can process that Scripss does for free. And only recently, Mass General was the first to implement that in your patient population, and I'm sure you were a driver of that. What is the reluctance about using this as an orthogonal, if you will, separate way to assess a person's risk for heart disease? And we know validated very solidly about being aggressive about lipid lowering when you know this person's in the highest 5% polygenic risk score. Are we just deadheads in this field or what?
Pradeep Natarajan (43:30):
Yeah, I mean Eric, as you know, lots of inertia in medicine, but this one I think has a potential to make a large impact. Like CHIP mutations, I said news is about 10% in individuals greater than 70. The prospect here is to identify the risk much earlier in life because I think there is a very good argument that we're undertreating high risk individuals early on because we don't know how to identify them. As you highlighted, Dr. Braunwald about LDL cholesterol. The other part of that paradigm is LDL cholesterol lowering and the duration. And as we said, everybody would benefit from really low LDL cholesterol, but again, you might overtreat that if you just give that to everybody. But if you can better identify the folks very early in life, there is a low cost, low risk therapy, at least related to statins that you could have a profound benefit from the ones who have a greater conviction will have future risk for cardiovascular disease.
(44:21):
You highlighted the family history, and the family history has given the field of clues that genetics play a role. But as the genome-wide association studies have gotten larger, the polygenic risk scores have gotten better. We know that family history is imperfect. There are many reasons why a family member who is at risk may or may not have developed cardiovascular disease. A polygenic risk score will give a single number that will estimate the contribution of genetics to cardiovascular disease. And the thing that is really fascinating to me, which is I think some of a clinical implementation challenge is that the alleles for an individual are fixed. The genotyping is very cheap. That continues to be extremely cheap to do this test. But the weights and the interpretation of what the effects should be for each of the SNPs are continually being refined over time.
(45:18):
And so, given the exact same SNPs in the population, the ability to better predict cardiovascular diseases getting better. And so, you have things that get reported in the literature, but literally three years later that gets outdated and those hypotheses need to be reassessed. Today, I'll say we have a great relative to other things, but we have a great polygenic risk score was just reported last year that if you compare it to familial hypercholesterolemia, which has a diagnostic yield of about 1 in 300 individuals, but readily detectable by severe hypercholesterolemia that has about threefold risk for cardiovascular disease. By polygenic risk score, you can find 1 in 5 individuals with that same risk. Obviously you go higher than that, it'll be even higher risk related to that. And that is noble information very early in life. And most people develop risk factors later in life. It is happening earlier, but generally not in the 30s, 40s where there's an opportunity to make a substantial impact on the curve related to cardiovascular disease.
(46:25):
But there is a lot of momentum there. Lots of interest from NIH and others. The major challenge is though the US healthcare system is really not well set up to prevention, as you know, we practice healthcare after patient's developed disease and prevent the complications related to progression. The stakeholder incentives beyond the patient themselves are less well aligned. We've talked a lot here today about payers, but we don't have a single payer healthcare system. And patients at different times of their lives will have different insurers. They'll start early in life with their parents, their first employer, they'll move on to the next job and then ultimately Medicare. There's no entity beyond yourself that really cares about your longevity basically from the beginning and your overall wellness. That tension has been a major challenge in just driving the incentives and the push towards polygenic risk scores. But there are some innovative approaches like MassMutual Life Insurance actually did a pilot on polygenic risk scoring.
(47:33):
They're in the business of better understanding longevity. They get that this is important data. Major challenges, there are federal protections against non-discrimination in the workplace, health insurance, not necessarily life insurance. So I think that there are lots of things that have to be worked out. Everybody recognizes that this is important, but we really have to have all the incentives aligned for this to happen at a system-wide level in the US. So there's actually lots of investment in countries that have more nationalized healthcare systems, lots of development in clinical trials in the UK, for example. So it's possible that we in the US will not be the lead in that kind of evidence generation, but maybe we'll get there.
The GLP-1 Drugs
Eric Topol (48:16):
Yeah, it's frustrating though, Pradeep, because this has been incubating for some time and now we have multi ancestry, polygenic risk scores, particularly for heart disease and we're not using it, and it's not in my view, in the patient's best interest just because of these obstacles that you're mentioning, particularly here in the US. Well, the other thing I want to just get at with you today is the drugs that we were using for diabetes now blossoming for lots of other indications, particularly the glucagon-like peptide 1 (GLP-1) drugs. This has come onto the scene in recent years, not just obviously for obesity, but it's anti-inflammatory effects as we're learning, mediated not just through the brain but also T cells and having extraordinary impact in heart disease for people with obesity and also with those who have heart failure, about half of heart failure for preserved ejection fraction. So recently you and your colleagues recently published a paper with this signal of optic neuropathy. It was almost seven eightfold increase in a population. First, I wanted to get your sense about GLP-1. We're also going to get into the SGLT2 for a moment as well, but how do you use GLP-1? What's your prognosis for this drug class going forward?
Pradeep Natarajan (49:55):
As it relates to the paper, I can't claim credit as one of my former students who is now Mass Eye and Ear resident who participated, but we can talk about that. There's obviously some challenges for mining real world data, but this was related to anecdotes that they were observing at Mass Eye and Ear and then studied and observed an enrichment. In general though, I feel like every week I'm reading a new clinical trial about a new clinical outcome benefit as it relates to GLP-1 receptor agonists. This is kind of one thing that stands out that could be interrogated in these other clinical trials. So I would have that caveat before being cautious about ocular complications. But the data has been overwhelmingly beneficial, I think, because at minimum, obesity and inflammation are relayed to myriad of consequences, and I'm really excited that we have therapies that can address obesity that are safe.
(50:52):
There's a legacy of unsafe medicines for obesity, especially related to cardiovascular disease. So the fact that we have medicines that are safe and effective for lowering weight that also have real strong effects on clinical outcomes is tremendous. We in cardiology are increasingly using a range of diabetes medicines, including GLP-1 receptor agonists and SGLT2 inhibitors. I think that is also the secular changes of what influences cardiovascular disease over time. I talked about over the last 10 years or so with this increase in deaths attributable to cardiovascular disease. If you look at the influences of traditional clinical risk factors today, many of them have decreased in importance because when abnormal, we recognize them, in general we modify them when recognized. And so, many of the things that are unaddressed, especially the features related to insulin resistance, obesity, they start rising in importance. And so, there is a dramatic potential for these kinds of therapies in reducing the residual risks that we see related to cardiovascular disease. So I'm enthusiastic and excited. I think a lot more biology that needs to be understood of how much of this is being influenced specifically through this pathway versus a very effective weight loss medicine. But also interesting to see the insights on how the effect centrally on appetite suppression has profound influences on weight loss as well too. And hopefully that will lead to more innovations in weight management.
The SGLT-2 Drugs
Eric Topol (52:25):
And likewise, perhaps not getting near as much play, but when it came on the cardiovascular scene that an anti-diabetic drug SGLT2 was improving survival, that was big, and we still don't know why. I mean, there's some ideas that it might be a senolytic drug unknowingly, but this has become a big part of practice of cardiology in patients with diabetes or with preserved ejection fraction heart failure. Is that a fair summary for that drug?
Pradeep Natarajan (53:00):
Yeah, I totally agree. I mean, as there has been increased recognition for heart failure preserved ejection fraction, it has been almost disheartening over the last several years that we have not had very specific effective therapies to treat that condition. Now, it is a tremendous boon that we do have medicines interestingly focused on metabolism that are very helpful in that condition for heart failure with preserved ejection fraction. But there is still much more to be understood as far as that condition. I mean, the major challenge with heart failure, as you know, especially with heart failure preserved ejection fraction, it likely is a mix of a wide variety of different etiologies. So in parallel with developing effective therapies that get at some aspect is really understanding what are the individual drivers and then targeting those specific individual drivers. That requires a lot of unbiased discovery work and further profiling to be done. So lot more innovation, but relative to heart failure itself, it is not had widespread recognition as heart failure reduced ejection fraction. So much more to innovate on, for sure.
Eric Topol (54:07):
Right, right. Yeah, I am stunned by the recent progress in cardiovascular medicine. You have been center stage with a lot of it, and we've had a chance to review so much. And speaking of genetics, I wanted to just get a little insight because I recently came across the fact that your mother here at the City of Hope in Southern California is another famous researcher. And is that, I don't know what chromosome that is on regarding parental transmission of leading research. Maybe you can tell me about that.
Pradeep Natarajan (54:41):
Yeah, I mean, I guess it is a heritable trait when a parent has one profession that there is a higher likelihood that the offspring will have something similar. So both of my parents are PhDs, nonphysicians. There is a diabetes department at the City of Hope, so she's the chair of that department. So very active. We do overlap in some circles because she does investigate both vascular complications and renal complications. And then sometimes will ask my advice on some visualization. But she herself has just had a science translational medicine paper, for example, just a couple of months ago. So it's fun to talk about these things. To be honest, because my parents are researchers, I was not totally sure that I would be a researcher and kind of wanted to do something different in medicine. But many of my early observations and just how common cardiovascular disease is around me and in my community and wanting to do something useful is what got me specifically into cardiology.
(55:45):
But obviously there are numerous outstanding, important questions. And as I went through my career, really focused on more basic investigations of atherosclerosis and lipids. What got me excited sort of after my clinical training was the ability to ask many of these questions now in human populations with many new biological data sets, at least first centered on genetics. And the capabilities continue to expand, so now I teach first year Harvard medical students in their genetics curriculum. And when I talk to them just about my career arc, I do remind them they're all doing millions of things and they're exploring lots of things, but when they get to my shoes, the capabilities will be tremendously different. And so, I really advise them to take the different experiences, mainly in an exercise for asking questions, thoughtfully addressing questions, connecting it back to important clinical problems. And then once they start to understand that with a few different approaches, then they'll totally take off with what the opportunities are down the road.
Eric Topol (56:51):
No, it's great. I mean, how lucky somebody could be in the first year of med school with you as their teacher and model. Wow. Pradeep, we've really gone deep on this and it's been fun. I mean, if there's one person I'm going to talk to you about cardiovascular risk factors and the things that we've been into today, you would be the one. So thank you for taking the time and running through a lot of material here today, and all your work with great interest.
Pradeep Natarajan (57:24):
Thanks, Eric. I really appreciate it. It's tremendous honor. I'm a big fan, so I would be glad to talk about any of these things and more anytime.
***************
Thanks for listening, reading or watching!
The Ground Truths newsletters and podcasts are all free, open-access, without ads.
Please share this post/podcast with your friends and network if you found it informative!
Voluntary paid subscriptions all go to support Scripps Research. Many thanks for that—they greatly helped fund our summer internship programs for 2023 and 2024.
Thanks to my producer Jessica Nguyen and Sinjun Balabanoff for audio and video support at Scripps Research.
Note: you can select preferences to receive emails about newsletters, podcasts, or all I don’t want to bother you with an email for content that you’re not interested in.

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
Shane Crotty: A Landmark Study on Upper Airway Mucosal Immunity
31 juil.· Ground Truths
A video snippet of our conversation. Full videos of all Ground Truths podcasts can be seen on YouTube here. The audios are also available on Apple and Spotify.
Shane Crotty: A Landmark Study on Upper Airway Mucosal Immunity
Transcript
This is the first time a Ground Truths podcast is being posted simultaneous with a new publication, this one in Nature, by Professor Shane Crotty and his colleagues at La Jolla Institute for Immunology. Shane is one of the leading immunologists and virologists in the country; he and his group published in 2020 the first detailed analysis for how our immune system responds to SARS-CoV-2. Shane also, among many other notable contributions during COVID, illuminated the role of hybrid immunity vs COVID, the differences between and additivity of vaccination and infection.
Today’s paper in Nature is indeed a landmark contribution doing something that hasn’t been done before—to understand the underpinnings of mucosal immunity of the upper airway. 100 participants had monthly nasal and nasopharyngeal swabs throughout the pandemic. With a median of >100,000 cells per swab recovered, they undertook single-cell sequencing and full characterization of the cells (tissue-resident memory B cells, CD4+ and CD8+ T cells, germinal center follicular helper T cells and B cells, etc.) to determine optimal immune protection of the upper airway, the effect of infections by different variants, breakthrough infections, vaccination, and age.
Here is the transcript of our conversation about the new report with links to the audio:
Eric Topol (00:06):
Hello, it's Eric Topol with Ground Truths, and with me today is Professor Shane Crotty from the La Jolla Institute of Immunology (LJI), not too far away from where I work at Scripps. And Shane has been a go-to immunologist colleague here in the Mesa, and he and his colleagues were the ones that really first published the response to SARS-CoV-2 as far as the immunologic response. And today we're doing something very unique. We're going to go over for the first time in the two year plus history of Ground Truths, going to have a publication with at least simultaneous or near simultaneous podcast. Shane, welcome and congratulations on this really important paper in Nature.
Shane Crotty (00:57):
Thanks, Eric. Thanks for having me. Yeah, somebody asked if I was going to go over to Scripps for the podcast and I was like, yeah, we could.
Eric Topol (01:06):
You could. You could. But no, it's good. And it's nice having the logo of this great institute you work at right in the right corner. And you've done so many contributions with your colleagues at La Jolla Institute. It's really a privilege to have a chance to learn from you and particularly about what we're going to talk about today, which is mucosal immunity to upper airway infections, which is especially germane to COVID. And we're actually in the middle of a significant wave of COVID right now. And I guess it would maybe be fair to say, Shane, that we've never truly understood the underpinnings, the real details of upper airway mucosal immunity. Is that a fair statement?
Shane Crotty (01:53):
Yeah, it is a fair statement.
Eric Topol (01:56):
Okay. So today we're going to crack the case. This paper from you and your colleagues, of course, you're the senior author and first author, Sydney Ramirez did a remarkable study. I mean, just extraordinary. This is why we're doing a special podcast about it. Maybe you could just kind of give us the overview of the design because you were doing things that haven't been done before.
Shane Crotty (02:24):
Sure. And, I would say the genesis even of it goes back to what you were introducing. I mean, during the pandemic, we like a lot of scientists spent a lot of time and energy trying to help understanding immune responses to this virus, and immune memory to this virus, and what was involved in protective immunity. And we're certainly proud of the work that we did. And it was hard work. And after a while we were exhausted and we stopped.
Shane Crotty (02:59):
And then we came back to it after a while and said, well, the virus is still here. And so many people have contributed so much to better understanding the virus and creating vaccines. But there are clearly still things we don't understand. What are those biggest knowledge gaps and where might we be able to contribute? And really to me the biggest one was location, location, location. This is a virus that infects your nose, infects your upper airway—your nose, and throat, and oral cavity. And then obviously if you get severe disease, the severe disease and death are from the lungs. And it's just been a big knowledge gap in terms of understanding what actually occurs in those tissues immunologically and what is associated with protective immunity or what could be associated with protective immunity. And sort of looking forward what might be helpful for mucosal vaccine development from things that we could learn.
Shane Crotty (04:12):
So we started from what we would call the basics, and what does immune memory look like in the upper airways in normal people? And that hasn't been available really even in, and we started this two years ago, even in the biggest atlases published of the human body. There was no upper airway tissue representation at all. And that's because technically it's just tough to access and difficult to reproducibly get at. And so, we recruited people to a group of 20 to 30 people to come to LJI once a month, and just started testing out, published and unpublished sampling techniques to see were there ways where we could reproducibly sample immune cells in the upper airways from people. And once we got things, so the keys for us were you got to have enough cells that you can collect to learn something from. And luckily with modern techniques of flow cytometry and single cell sequencing, you don't need that many cells. And so, we could get a hundred thousand cells on a swab and that's enough to do a lot with. And second, how reproducible was it? So we showed, we had people come in every month for a year and we could reproducibly find the same things in their swab; same cell types in their swabs. And the third thing was that people would come back.
Shane Crotty (06:05):
We found that if you have good nurses doing the techniques, we could find ways that this would be a sampling approach that was tolerable and people would come back for repeat measures, which is really valuable to see what's happening in people over time. So that was what we started from in the study and built from.
Eric Topol (06:27):
And if I am correct, you sampled two places with the swabs, one in the nose and one of the throat. Or, I think one which you have in the paper as the MT for something about the median nasal turbinate and the other adenoid in the back of the throat. Is that right?
Shane Crotty (06:50):
So all the sampling is a swab into your nose. And when we were doing that, we were really excited to see the diversity of immune cells, particularly T cells and B cells, memory T cells and B cells that we isolated. They're like, wow, there's actually a lot of interesting immune memory up in there. And the lab said, oh, by the way, we're seeing T follicular helper cells (TFH). Now that happens to be my favorite cell type.
Eric Topol (07:22):
Why is that, Shane? Of all the cells, why do you say that's your favorite? I know you publish a lot on it.
Shane Crotty (07:31):
Because those are the T cells that are required for basically all neutralizing antibody responses. All high-quality antibody responses depend on—almost all high-quality antibody responses depend on—T cell help. That T cell help comes from T follicular helper cells. Antibody evolution is certainly one of the coolest processes of the immune system. And all of that depends on T follicular helper cells. So the fact that for example, you could get Omicron neutralizing antibodies even after only being vaccinated with ancestral vaccine, that's the immune system making guesses of what variants would look like. And those guesses come about through this antibody evolution that's driven by T follicular helper cells. So, it's really one of the most brilliant things the immune system does, and that's a cell type that's really key, but those processes happen in lymphoid tissue. That's what happens in lymph nodes and spleen. And here we were sampling epithelium, your nasal epithelium, so the cells didn't really belong there.
Shane Crotty (08:37):
And so, that's what turned the study in another direction. And we said, okay, let's figure out why is it that these cells are present in these swabs? And we had a couple of possibilities. One possibility was that the swab was going all the way back to the posterior wall of your nasopharynx, your top of your throat and sampling adenoid tissue. So adenoid tonsils and adenoids are a true lymphoid tissue and they're a mucosal lymphoid tissue. And so, we came up with multiple ways to validate that that's what we were testing. And in fact, it was the Sydney Ramirez, a clinician, and the ENTs involved who said, well, let's just look. And so, they actually did endoscopies with the swab to actually see where the swab went. We've got videos of the swabs going into the adenoid crypt in the back, and then we've got measurements of here are the cells that you find on those swabs.
Shane Crotty (09:58):
And what's cool about it is that, yes, so we did studies with two sets. We then shifted to doing studies with two sets of swabs. One where we essentially went “halfway back” where we were detecting that epithelium of your nasal passages and then one where it was all the way back and detecting the adenoid lymphoid tissue. So here we've got two different sites in your upper airways that are about an inch apart, and we're detecting essentially completely different cells of the immune system at those two places. And we tend to think of the cells present in that epithelial tissue as probably the sentinels, the cells that are sitting there that can potentially immediately react and try and protect you against a viral or bacterial infection. Whereas the lymphoid tissue, the adenoids, is really about generating the immune responses in the first place and priming immune responses. And that's where these germinal centers can occur, which are where the TFH are where you can get antibody evolution. And so, we found in the course of the study that with this non-invasive technique that we can.
Eric Topol (11:14):
By the way, I don't want to be signing up for the one way up there because I mean just a mid-nose enough for me. So wow, I got to give credit to your study participants for coming back every month for a year to have that. Some people call it a brain biopsy.
Video of swab of nasopharyngeal tissue
Shane Crotty (11:33):
Right. So I will tell you, it is a different experience than the COVID nasopharyngeal swab might've gotten through your car window. If you're actually sitting down in a comfortable space and there's a nurse doing it with these particular goals. We really found, we had a hundred people in the study and a total of 300 swabs, and the vast majority of people came back if we asked them to.
Eric Topol (12:06):
That's great.
Shane Crotty (12:07):
And we're certainly very thankful for the volunteers. Obviously they were volunteering in the first place to participate. So I'm a little hesitant about the video because I've told people to not show it to potential volunteers because it definitely doesn't encourage you to volunteer. You're like, wait, that's what's happening? But actually, I've had it done on me.
Video of the swab to the nasopharynx for adenoid (lymphoid tissue) access.
Eric Topol (12:37):
Not that bad.
Shane Crotty (12:39):
It's really pretty compelling. And by doing these repeated samples, we actually now have the capacity to look at ongoing immune responses like after an infection or vaccination in people and see how that results in the immune system changing and what might be the source of the protective immunity that comes up. So we've actually got data in the paper looking at this antibody evolution in real time. So we've got affinity maturation of B cells occurring in just normal healthy adults of mucosal B cells against COVID. And so, that's really helping us learn what's possible, basically to figure out, okay, if you're going to try and make a vaccine, what types of immune cells are even possible to generate in this tissue? And where might you try and generate them? Or if you're trying to study some disease state, what are types of cells that might be problematic?
Eric Topol (13:45):
Yeah, I mean, I think the idea that so many of us have been pushing for a nasal vaccine to induce mucosal immunity because, as you know very well, the current shots are not very good at any durable or substantial protection from upper airway infections of COVID or SARS-CoV-2 and other infections. So I think one of the most important parts of this report is that it lends itself well to helping towards artificially, if you will, make a vaccine to get the protective features that you were able to identify. Maybe you could just [speculate], if you had the ideal nasal airway, what would the cellular profile look like?
Shane Crotty (14:44):
Ah, I see. Yeah, great question. So, first of all, antibodies are great. So most of my career has been dedicated to most licensed vaccines. The correlate of protection is antibodies. Antibodies clearly can be protective, and if you can get them that’s excellent, so certainly I would want, in terms of the non-cellular component, I would want antibodies present, neutralizing antibodies present in it.
Eric Topol (15:26):
Are these IgA or IgG?
Shane Crotty (15:31):
Yeah, in an ideal situation, what would I want? I'd want a mix of both, basically. The IgAs look like they have a little more protective efficacy, but the IgGs, just at a molecular level have a longer half-life, stick around a little. So yeah, I'd want both. And then really the premise for most of what we do is saying, in situations where antibody isn't enough or the antibodies don't stay around long enough, or you've got a variant that now obviates the protective efficacy of that particular antibody, are there other types of protective immunity you can have? And the immune system has other stuff besides antibodies for a reason. Of the lymphocytes in your blood, most of them aren't antibody producing cells. Most of them are other things. And so, well sticking with adjacent to antibodies, those antibodies in the mucosa, I'd want them to be made by cells that were literally right there. So plasma cells living in that site so that you've got basically the highest concentration of antibodies you can get because they're not having to diffuse through the whole body. They're just already at their highest concentration right there. Now antibodies come from B cells, that's what encodes the antibodies.
Shane Crotty (17:03):
And so, the B cells can make neutralizing antibodies if it turns out that you haven't made enough neutralizing antibodies, or if there's a variant that escapes those, maybe there are other B cells that could make, once you get infected, more B cells that could make more antibody rapidly infection, or B cells that recognize this variant that is mismatched to the current antibodies you have. But memory B cells are basically a library of different antibody specificities representing different guesses about what viral variants or structures might look like. And so, I would want memory B cells in that upper airway tissue that could reactivate quickly. There are memory B cells in your blood and we don't know how long it takes. And that's one of the reasons we're hoping we and others build upon this study. But it might take, let's say five days for memory B cells to go from your blood into your upper airway.
Eric Topol (18:06):
Oh, right.
Shane Crotty (18:08):
That's right, you were already quite sick by that point. Instead, if memory B cells are right there, as soon as virus showed up, they got activated. Now maybe after (we’re not sure yet), but maybe after 48 hours those cells are now activated and doing something useful. That would be optimal. So then we can pivot to the T cell side. So there’s a fantastic recognition that T cells being physically present in tissues, tissue resident memory cells, as they’re most often called, can really have fantastic protective capacities. From a lot of mouse model systems where you can see T cells are in the skin or the liver, or whatever [tissue] are already there, they’re more protective than if the cells are in the blood. So if you could also have T cells essentially permanently parked in the epithelium of your nasal passages and in the adenoid, hopefully those could essentially be sentinels for protective immunity, and as soon as you get infected, those T cells would reactivate and start killing off infected cells. ’That’s the mix that I would want to see. And I think there’s at least some reasonable evidence in the context of COVID that people who have T cells in their upper airways maybe manage to control the virus so quickly that it’s a subclinical infection; they never notice when they get infected. And so, building on those types of observations, that’s what I would want.
Eric Topol (19:56):
That sounds good. I like that. I’d like to have that in my nasal airway. Now, just to make sure I’ve got this, what you found, of course, the memory B cells, the T cell memory, CD8+, that is the cell-killing T cells that you mentioned, the resident T cells. One clarification on that, they are not really going to do much until there’s been some cells that have been infected with the virus, right? Then they come alive and kill those cells. So they’re not immediate, but they can work pretty quickly still though, right? If they’re resident T cells?
Shane Crotty (20:45):
Yeah, in theory it might take as little as 12 hours for a virus to infect a cell, and then you get some antigen presentation on that cell that could activate the T cell.
Eric Topol (20:58):
And that’s all happening perhaps within the incubation phase of the virus, right?
Shane Crotty (21:07):
Correct. That’s a tough thing to study, but conceptually that’s the way people tend to sketch it out.
Eric Topol (21:13):
Right. Now the other part of the story is, and you alluded to it earlier, is the lymphoid tissue up there, higher up where there are these germinal centers; is there anything different you want in these germinal centers? Do they contribute to mucosal immunity that you haven’t already mentioned?
Shane Crotty (21:36):
So they really contribute in this forward looking sense or really in the classroom kind of sense. The germinal centers are where you’re basically teaching the B cells in advance of seeing the infection either with your vaccine or with your previous infection, evolving better B cells and better antibodies and hopefully instructing them where to go reside to then be ready for the next infection. If you get really great protection that next time, hopefully then you don’t need to start.
Eric Topol (22:14):
Right. So it’s like the training grounds for this coordinated response, I guess. Now you also noted this, I mean this is a rich paper, which is we’re illuminating something that’s never been done before in human beings. I mean it’s pretty damn important and impressive. But you also found that you had an age relationship. Can you tell us about that?
Shane Crotty (22:39):
Sure. This is one of our favorite parts of the study. I’d say in particular for several of the clinicians who were involved, because the general conversations people have about upper airway lymphoid tissue, like your tonsils and including your adenoids, is that adults don’t really have functional lymphoid tissue in the upper airway that your tonsils atrophy by the time maybe you’re 20 or something. So, immunologically, functionally, what that means is if you have let’s say an intranasal vaccine or you get infected with a new [virus] like SARS-CoV-2, if those would normally be the sites that start your immune response, where does it now happen? And instead what we saw was, we had such a diverse group of people in our studies—we realized we had people from age 18 to 68—and so we could directly ask, in normal healthy individuals across a large age span of adulthood is there functional mucosal lymphoid tissue? And the answer was yes, it was there. But it definitely declines over time, and it's declining on a log scale. Our simplest statement was that 75% of everybody we sampled still had functional tissue, but the younger the people were, the more functional it was, and the more germinal centers actually we saw; again these training grounds.
Eric Topol (24:35):
So this is really important because we know for COVID and obviously for influenza and other respiratory infections that people of advanced age are much more susceptible. And here you are finding something that supports that ,and it's almost like, the thymus, it involutes. After that, what age 20, and our lymphoid tissue [involutes]. We're just set up to fail. Old codgers, like me we're defenseless, I guess, right?
Shane Crotty (25:12):
So what I've liked about that in a positive sense is that it's not that all of these things go to zero. Like for example, naive T cells are definitely less abundant in people over the age of 60 than under, but they're not zero. And the mucosal lymphoid tissue is definitely less abundant in people over the age of 60, but in most people it still wasn't zero. And I always think about these things from a vaccine immunology perspective, and fundamentally the difference between getting vaccinated and infected frequently is that the whole point of the vaccine is you get to generate the immune response on your own time. And so, even if you're starting with five times fewer T cells or five times fewer germinal centers, if you're getting to do all that training ground in advance, you can end up with just as many bispecific T cells as a 20-year-old or just as many memory B cells as a 20-year-old because these things occur on an exponential scale because of the cell divisions. And so, it might take you three extra days, for example, to get to the same level, which again, if you're racing a virus, can be the difference between life and death. But if it's not a race and if you're doing it in the context of a vaccine, it's a much smaller factor. And that's some of what we've been trying to learn.
Eric Topol (26:42):
Now we only have started to scratch the surface of your findings. One of the things that drives me nutty in reading papers, especially from great immunologists like you, is that in each figure there's like 20 different panels. We get to one of the figures, figure three is all the way to panel W. I mean that starts with A. That gives you a little impression of the data. It's rich, another one goes to N or R. I mean we're talking about a lot of data. So I've only started to really deconvolute what you've done here, which is just an amazing study. But what are some other things that we should touch on before wrapping up?
Shane Crotty (27:35):
A lot of the goal in this study was to establish baselines of what is normal in humans in the upper airways. And that's one reason why in this case there actually are a lot of figure panels because we could work out a bunch of individual parts of the immune system that really hadn't been characterized in this way before. And something we really cared about was durability of immune memory. It's often talked about, well, mucosal responses are inherently short-lived. And we're like, well, what does that mean? Does that mean there's just no memory? Is it different kinds of memory? And so, this is the first measurement of memory B cells in this tissue in an antigen specific way. And we were doing it in people who had had recent COVID breakthrough infections. And we saw really the mucosal memory was stable for six months. And so, to me that's quite encouraging that it's not one month and it's gone, at least with an infection, it's at least six months and it looks like it'll project out for substantially longer.
Shane Crotty (28:53):
Amongst those cells, many of them are IgA. IgA is this antibody isotype that's particularly mucosal associated. And only 5% of the memory B cells circulating in blood were IgA. Whereas many of the memory B cells in the local tissue were IgA, which we think is also telling us that there's a lot of immune memory and the immune system in this tissue that we're probably not sampling in the blood. And so, sampling blood's great, right? It's accessible and we can learn a lot from it, but it does look like there is some tissue compartmentalization.
Eric Topol (29:37):
Oh, not a question. And the findings you had of the resident T cell is so indicative of that. And what's really striking, of course Shane, is that as we assess the immune system in people at large, we look at a lymphocyte neutrophil ratio [in the blood], we get almost nothing. And then in the course of the pandemic, you and your colleagues there provided such granular data on B and T cells, CD4 and CD8 T cells, and that you illuminated things that are not done ever clinically. These are research, high tier research labs like yours. The only question I have on before I just wrap up with the nasal vaccine story, interferon wasn't really part of this. As we know SARS-CoV-2 can shut down the interferon response, it's considered a frontline part of the defense. Where does that fit into the mucosal immunity of the upper airway?
Shane Crotty (30:46):
Yeah, it's really important. And that's in this basic divide we do in the immune system, the innate immune system and the adaptive immune system. So everything I was talking about is the B cells, the T cells, and antibodies. That's all the adaptive immune system. That's all virus specific. And then the innate immune system is the generalists, and really sort of the fire alarm, just sensing some danger. And definitely in COVID interferon is very important. I'm quite intrigued to see if using these techniques. I'm curious to see if some of these other aspects of the immune system can compensate somewhat for the fact that this virus. To me, if this virus has one superpower, it's its incredible ability to evade triggering interferon for as long as it does. And that has this massive cascading effect to almost everything about the pandemic essentially. And so, I'm intrigued by whether in people who have immunity are there ways that these other cells of the immune system or even antibodies can do things when a viral infection occurs, that helps trigger the overall immune system to recognize that something's there, even in the absence of type 1 interferons. That's where I think for now it fits in.
Eric Topol (32:14):
Well. I think you've so aptly described, not surprisingly, the superpower of SARS-CoV-2, which I think a lot of people haven't realized that it's so good at shutting down that defense system. Now on the basis of you having really gotten this understanding of the mucosal immunity in the upper airway, does this make you think that the nasal vaccine that we aspire to have is more of a reality? Do you kind of know what the ideal profile might look like to keep people healthy and resist infections? Do you think this is achievable in any durable sense at high level success with a nasal spray vaccine?
Shane Crotty (33:04):
I'm optimistic for several reasons. One is we really saw a lot of different immune memory cell types that were present, that was encouraging and seeing the B cell memory durability for at least six months—pretty flat line for that six months—was encouraging. It looks like the immune system knows how to keep these cells around if it wants to for a significant period of time. We'll have to do more in follow up. But again, it was encouraging. Third, we had some people who were vaccinated only and some people who had breakthrough infections. And really in the vaccinated only, we didn't see T cell memory in the upper airways. And I actually consider that encouraging because it suggests local exposure does give you the memory and exposure in your arm really doesn't. So I think there is something to improve upon. It can be improved upon. And lastly, I get asked all the time, I'm sure you get asked all the time: Why aren't there more intranasal vaccines or inhaled vaccines, more mucosal vaccines in some way?
Shane Crotty (34:25):
And I think there's more than one reason, but I tend to be very practical, and I think one practical reason is there's very little to measure, to guide you in your vaccine development. If you have six ideas or six constructs that you think might work in humans as a nasal vaccine, you basically just have to pick one, try something, and hoping there's not much you can measure it clinical trials for what might be the type of response even. So for example, the FluMist vaccine, it's the only licensed inhaled vaccine, intranasal vaccine. In adults it doesn't have a clear correlate of protection. If you get vaccinated with that, your circulating antibody responses don't increase, but also increases in nasal antibody didn't correlate with protection well. So, what does that mean? That probably means there's other things going on up there that could be indicative of protection but weren't being measured before. So I'm hopeful with these types of approaches. Now, if you're an intranasal vaccine developer, you maybe have 4, 5, 6, 7, 8 ideas or constructs. If you can try those in a few people and make these different measurements and you've got your favorite immune profile that you might, now you have something to, it's more of an engineering problem. It's not a throwing a dart problem. You're like, yeah, this has given me the type of response that I like and I'm going to try and push this into clinical trials. So those are the things that I'm optimistic about moving forward.
Eric Topol (36:04):
Well, I love it because we really need it. And if anybody's optimistic that means a lot; it's yours. What you've done here has been quite extraordinary because you defined for the first time really the underpinnings of the mucosal immune response, the upper airway, you did it by age, you did it by variant, you did it by vaccine and infection. And most importantly, perhaps for longer term is you established what are the desirable features to have, which didn't exist before. It seemed like whatever I read for nasal vaccines, they were measuring some IgA or IgG, and they didn't get down to the memory B cells and the tissue resident T cells, memory cells, and all these other things that you found. You did all this single cell sequencing and flow cytometry. The work is just really fantastic. So Shane, just in closing, I just want to congratulate you.
Eric Topol (37:05):
You made seminal findings along the pandemic. You were the one that really illuminated hybrid immunity, the advantage of if you don't want to have an infection of COVID, but if you did have that and a vaccine, you kind of had some extra synergy, if you will. But here you've done something, you and your team. Unique. Congratulations on that. No surprise that it's in Nature this week. I'm sure a lot of people will share your optimism that we will have something beyond just shots in the future because COVID isn't going away. There's other respiratory pathogens. And finally, somebody did the right study, who knows immunology inside and out. So Shane, thanks very much.
Shane Crotty (37:52):
Thanks Eric. Very much appreciated particularly coming from you.
*****************************************
Thanks for listening, reading or watching!
The Ground Truths newsletters and podcasts are all free, open-access, without ads.
Please share this post/podcast with your friends and network if you found it informative
Voluntary paid subscriptions all go to support Scripps Research. Many thanks for that—they greatly helped fund our summer internship programs for 2023 and 2024.
Thanks to my producer Jessica Nguyen and Sinjun Balabanoff for audio and video support at Scripps Research.
Note: you can select preferences to receive emails about newsletters, podcasts, or all I don’t want to bother you with an email for content that you’re not interested in.

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
Faisal Mahmood: A.I.'s Transformation of Pathology
28 juil.· Ground Truths
Full videos of all Ground Truths podcasts can be seen on YouTube here. The audios are also available on Apple and Spotify.
Thank you for reading Ground Truths. This post is public so feel free to share it.
Transcript with audio and external links
Eric Topol (00:05):
Hello, it's Eric Topol with Ground Truths, and I am really thrilled to have with me Professor Faisal Mahmood, who is lighting it up in the field of pathology with AI. He is on the faculty at Harvard Medical School, also a pathologist at Mass General Brigham and with the Broad Institute, and he has been publishing at a pace that I just can't believe we're going to review that in chronological order. So welcome, Faisal.
Faisal Mahmood (00:37):
Thanks so much for having me, Eric. I do want to mention I'm not a pathologist. My background is in biomedical imaging and computer science. But yeah, I work very closely with pathologists, both at Mass General and at the Brigham.
Eric Topol (00:51):
Okay. Well, you know so much about pathology. I just assume that you were actually, but you are taking computational biology to new levels and you're in the pathology department at Harvard, I take it, right?
Faisal Mahmood (01:08):
Yeah, I'm at the pathology department at Mass General Brigham. So the two hospitals are now integrated, so I'm at the joint department.
Eric Topol (01:19):
Good. Okay. Well, I'm glad to clarify that because as far as I knew you were hardcore pathologist, so you're changing the field in a way that is quite unique, I should say, because a number of years ago, deep learning was starting to get applied to pathology just like it was and radiology and ophthalmology. And we saw some early studies with deep learning whereby you could find so much more on a slide that otherwise would be not even looked at or considered or even that humans wouldn't be able to see. So maybe you could just take us back first to the deep learning phase before these foundation models that you've been building, just to give us a flavor for what was the warmup in this field?
Faisal Mahmood (02:13):
Yeah, so I think around 2016 and 2017, it was very clear to the computer vision community that deep learning was really the state of the art where you could have abstract feature representations that were rich enough to solve some of these fundamental classification problems in conventional vision. And that's around the time when deep learning started to be applied to everything in medicine, including pathology. So we saw some earlier cities in 2016 and 2017, mostly in machine learning conferences, applying this to very basic patch level pathology dataset. So then in 2018 and 2019, there were some studies in major journals including in Nature Medicine, showing that you could take large amounts of pathology data and classify what's known to us and including predicting what's now commonly referred to as non-human identifiable features where you could take a label and this could come from molecular data, other kinds of data like treatment response and so forth, and use that label to classify these images as responders versus non-responders or having a certain kind of mutation or not.
(03:34):
And what that does is that if there is a morphologic signal within the image, it would pick up on that morphologic signal even though humans may not have picked up on it. So it was a very exciting time of developing all of these supervised, supervised foundation models. And then I started working in this area around 2019, and one of the first studies we did was to try to see if we can make this a little bit more data efficient. And that's the CLAM method that we published in 2021. And then we took that method and applied it to the problem of cancers of unknown primary, that was also in 2021.
Eric Topol (04:17):
So just to review, in the phase of deep learning, which was largely we're talking about supervised with ground truth images, there already was a sign that you could pick up things like the driver mutation, the prognosis of the patient from the slide, you could structural variations, the origin of the tumor, things that would never have been conceived as a pathologist. Now with that, I guess the question is, was all this confined to whole slide imaging or could you somehow take an H&E slide conventional slide and be able to do these things without having to have a whole slide image?
Faisal Mahmood (05:05):
So at the time, most of the work was done on slides that were fully digital. So taking a slide and then digitizing the image and creating a whole slide image. But we did show in 2021 that you could put the slide under a microscope and then just capture it with a camera or just with a cell phone coupled to a camera, and then still make those predictions. So these models were quite robust to that kind of domain adaptation. And still I think that even today the slide digitization rate in the US remains at around 4%, and the standard of care is just looking at a glass light under a microscope. So it's very important to see how we can further democratize these models by just using the microscope, because most microscopes that pathologists use do have a camera attached to them. So can we somehow leverage that camera to just use a model that might be trained on a whole slide image, still work with the slide under a microscope?
Eric Topol (06:12):
Well, what you just said is actually a profound point that is only 4% of the slides are being reviewed digitally, and that means that we're still an old pathology era without the enlightenment of machine eyes. I mean these digital eyes that can be trained even without supervised learning as we'll get to see things that we'll never see. And to make, and I know we'll be recalling back in 2022, you and I wrote a Lancet piece about the work that you had done, which is very exciting with cardiac biopsies to detect whether a heart transplant was a rejection. This is a matter of life or death because you have to give more immunosuppression drugs if it's a rejection. But if you do that and it's not a rejection or you miss it, and there's lots of disagreement among pathologists, cardiac pathologists, regarding whether there's a transplant. So you had done some early work back then, and because much of what we're going to talk about, I think relates more to cancer, but it's across the board in pathology. Can you talk about the inner observer variability of pathologists when they look at regular slides?
Faisal Mahmood (07:36):
Yeah. So when I first started working in this field, my kind of thinking was that the slide digitization rate is very low. So how do we get people to embrace and adapt digital pathology and machine learning models that are trained on digital data if the data is not routinely digitized? So one of my kind of line of thinking was that if we focus on problems that are inherently so difficult that there isn't a good solution for them currently, and machine learning provides, or deep learning provides a tangible solution, people will be kind of forced to use these models. So along those lines, we started focusing on the cancers of unknown primary problem and the myocardial biopsy problem. So we know that the Cohen’s kappa or the intra-observer variability that also takes into account agreement by chance is around 0.22. So it's very, very low for endomyocardial biopsies. So that just means that there are a large number of patients who have a diagnosis that other pathologists might not agree with, and the downstream treatment regimen that's given is entirely based on that diagnosis. The same patient being diagnosed by a different cardiac pathologist could be receiving a very different regimen and could have a very, very different outcome.
(09:14):
So the goal for that study is published in Nature of Medicine in 2022, was to see if we could use deep learning to standardize that and have it act as an assistive tool for cardiac pathologists and whether they give more standardized responses when they're given a machine learning based response. So that's what we showed, and it was a pleasure to write that corresponding piece with you in the Lancet.
Eric Topol (09:43):
Yeah, no, I mean I think that was two years ago and so much has happened since then. So now I want to get into this. You've been on a tear every month publishing major papers and leading journals, and I want to just go back to March and we'll talk about April, May, and June. So back in March, you published two foundation models, UNI and CONCH, I believe, both of these and back-to-back papers in Nature Medicine. And so, maybe first if you could explain the foundation model, the principle, how that's different than the deep learning network in terms of transformers and also what these two different, these were mega models that you built, how they contributed to help advance the field.
Faisal Mahmood (10:37):
So a lot of the early work that we did relied on extracting features from a resonant trained on real world images. So by having these features extracted, we didn't need to train these models end to end and allowed us to train a lot of models and investigate a lot of different aspects. But those features that we used were still based on real world images. What foundation models led us do is they leveraged self supervised learning and large amounts of data that would be essentially unlabeled to extract rich feature representations from pathology images that can then be used for a variety of different downstream tasks. So we basically collected as much data as we could from the Brigham and MGH and some public sources while trying to keep it as diverse as possible. So the goal was to include infectious, inflammatory, neoplastic all everything across the pathology department while still being as diverse as possible, including normal tissue, everything.
(11:52):
And the hypothesis there, and that's been just recently confirmed that the hypothesis was that diversity would matter much more than the quantity of data. So if you have lots and lots of screening biopsies and you use all of them to train the foundation model, there isn't enough diversity there that it would begin to learn those fundamental feature representations that you would want it to learn. So we used all of this data and then trained the UNI model and then together with it was an image text model where it starts with UNI and then reinforces the feature representations using images and texts. And that sort of mimics how humans learn about pathology. So a new resident, new trainee learning pathology has a lot of knowledge of the world, but it's perhaps looking at a pathology image for the first time. But besides looking at the image, they're also being reinforced by all these language cues from, whether it's from text or from audio signals. So the hope there was that text would kind of reinforce that and generate better feature representation. So the two studies were made available together. They were published in Nature Medicine back in March, and with that we made both those models public. So at the time we obviously had no idea that they would generate so much interest in this field, downloaded 350,000 times on Hugging Face and used for all kinds of different applications that I would've never thought of. So that's been very exciting to see.
Eric Topol (13:29):
Can you give some examples of some of the things you wouldn't have thought of? Because it seems like you think of everything.
Faisal Mahmood (13:35):
Yeah, people have used it to when there was a challenge for detecting tuberculosis, I think in a very, very different kind of a dataset. It was from the Nightingale Foundation and they have large data sets. So that was very interesting to see. People have used it to create newer data sets that can then be used for training additional foundation models. It's being used to extract rich feature representations from pathology images, corresponding spatial transcriptomic data, trying to predict spatial transcriptomics directly from histology. And there's a number of other options.
Eric Topol (14:27):
Well, yeah, that was March. Before we get to April, you slipped in the spatial omics thing, which is a big deal that is ability to look at tissue, human tissue over time and space. I mean the spatial temporal, it will tell us so much whether an evolution of a cancer process or so many things. Can you just comment because this is one of the major parts of this new era of applying AI to biology?
Faisal Mahmood (15:05):
So I think there are a number of things we can do if we have spatial data spatially resolved omic data with histology images. So the first thing that comes to my mind as a computer scientist would be that can we train a joint foundation model where we would use the spatially resolved transcriptomics to further enforce the pathology signal as a ground truth in a contrastive manner, similar to what we do with text, and can we use that to extract even richer feature representation? So we're doing that. In fact, we made a data set of about a thousand pathology images with corresponding spatial transcriptomic information, both curated from public resources as well as some internal data publicly available so people could investigate that question further. We're entrusted in other aspects of this because there is some indication including a study from James Zou’s group at Stanford showing that we can predict histology, predict the spatial transcriptomic signal directly from histology. So there's early indications that we might also be able to do that in three dimensions. So yeah, it's definitely very interesting. More and more of that data is becoming available and how machine learning can sort of augment that is very exciting.
Eric Topol (16:37):
Yeah, I mean, most of the spatial omics has been a product of single cell sequencing, whether it's single nuclei and different omics, not just DNA, of course, RNA and even methylation, whatnot. So the fact that you could try to impute that from the histologies is pretty striking. Now, that was March and then in April you published to me an extraordinary paper about demographic bias and how generative AI, we're in the generative AI year now since as we discussed with foundation models, here again that gen AI could actually reduce biases and enhance fairness, which of course is so counterintuitive to everything that's been written to date. So maybe you can take us through how we can get a reduction in bias in pathology.
Faisal Mahmood (17:34):
Yeah, so in the study, the study was about, this had been investigated in other fields, but what we try to show is that a model trained on large, diverse, publicly available data. When that's applied internally and we stratify it based on demographic differences, race and so forth, we see these very clear disparities and biases. And we investigated a lot of different solutions that were out there to equalize the distribution of the data to balance the distribution using or sampling and some of these simple techniques. And none of them worked quite well. And then we observed that using foundation models or just having richer feature representations eliminates some of those biases. In parallel, there was another study from Google where they use generative AI to synthesize additional images from those underrepresented groups and then use those images to enhance the training signal. And then they also showed that you could reduce those biases.
(18:49):
So I think the common denominator there is that richer feature representations contribute to reduced biases. So the biases not because there is some inherent signal tied to these subgroups, but the bias is essentially there because the feature representations are not strong enough. Another general observation is that there's some kind of a confounder often there that leads to the bias. And one example would be that patients with socioeconomic disparities might just be diagnosed late and there might not be enough advanced cases in the training dataset. So quite often when you go in and look at what your training distribution looks like and how it varies from your test distribution and what that dataset shift is, you're able to figure out where the bias inherently comes from. But as a general principle, if you use the richest possible feature representation or focus on making your feature representations richer by using better foundation models and so forth, you are able to reduce a lot of the bias.
Eric Topol (19:58):
Yeah, that's really another key point here is about the richer features and the ability counterintuitively to actually reduce bias. And what is important in interrogating data inputs, as you said before, you wind up with a problem with bias. Now, then it comes May since we're just March and April, in May you published TriPath, which is now bringing in the 3D world of pathology. So maybe you can give us a little skinny on that one.
Faisal Mahmood (20:36):
Yeah. So just looking at the spectrum of where pathology is today, I think that we all agree in the community that pathologists often look at extremely sampled tissue. So human tissue is inherently three-dimensional, and by the time it gets to a pathologist, it's been sampled and cut so many times that it often would lack that signal. And there are a number of studies that have shown that if you subsequently cut sections, you get to a different outcome. If you look at multiple slides for a prostate biopsy, you get to a different Gleason score. There are all of these studies that have shown that 3D pathology is important. And with that, there's been a growing effort to build tools, microscopes, imaging tools that can image tissue in 3D. And there are about 10 startups who've built all these different technologies, open-top light-sheet microscopy, microCT and so forth that can image tissue really well in three dimensions, but none of them have had clinical adoption.
(21:39):
And we think that a key reason is that there isn't a good way for a pathologist to examine such a large volume of tissue. If they spend so much time examining this large volume of tissue, they would never be able to get through all the, so the goal here really was to develop a computational tool that would look through the large volume and highlight key regions that a pathologist can then examine. And the secondary goal was that does using three dimensional tissue actually improve patient stratification and does using, essentially using three 3D deep learning, having 3D convolutions extract richer features from the three dimensions that can then be used to separate patients into distinct risk groups. So that's what we did in this particular case. The study relied on a lot of data from Jonathan Liu's group at University of Washington, and also data that we collected at Harvard from tissue that came from the Brigham and Women's Hospital. So it was very exciting to show that what the value of 3D pathology can be and how it can actually translate into the clinic using some of these computational tools.
Eric Topol (22:58):
Do you think ultimately someday that will be the standard that you'll have a 3D assessment of a biopsy sample?
Faisal Mahmood (23:06):
Yeah, I'm really convinced that ultimately 3D would become the standard because the technology to image these tissue is becoming better and better every year, and it's getting closer to a point where the imaging can be fast enough to get to clinical deployment. And then on the computational end, we're increasingly making a lot of progress.
Eric Topol (23:32):
And it seems, again, it's something that human eyes couldn't do because you'd have to look at hundreds of slides to try to get some loose sense of what's going on in a 3D piece of tissue. Whereas here you're again taking advantage, exploiting the digital eyes. Now this culminates to your June big paper PathChat in Nature, and this was a culmination of a lot of work you've been doing. I don't know if you do any sleep or your team, but then you published a really landmark paper. Can you take us through that?
Faisal Mahmood (24:12):
Yeah, so I think that with the foundation models, we could extract very rich feature representation. So to us, the obvious next step was to take those feature representations and link them with language. So a human would start to communicate with a generative AI model where we could ask questions about what's going on in a pathology image, it would be capable of making a diagnosis, it would be capable of writing a report, all of those things. And the reason we thought that this was really possible is because pathology knowledge is a subset of the world's knowledge. And companies like OpenAI are trying to build singular, multimodal, large language models that would harbor the world's information, the world knowledge and pathology is much, much more finite. And if we have the right kind of training data, we should be able to build a multimodal large language model that given any pathology image, it can interpret what's going on in the image, it can make a diagnosis, it can run through grading, prognosis, everything that's currently done, but also be an assistant for research, analyzing lots of images to see if there's anything common across them, cohorts of responders versus non-responders and so forth.
(25:35):
So we started by collecting a lot of instruction data. So we started with the foundation models. We had strong pathology image foundation models, and then we collected a lot of instruction data where we have images, questions, corresponding answers. And we really leveraged a lot of the data that we had here at Brigham and MGH. We're obviously teaching hospitals. We have questions, we have existing teaching training materials and work closely with pathologists at multiple institutions to collect that data. And then finally trained a multimodal large language model where we could give it a whole slide image, start asking questions, what was in the image, and then it started generating all these entrusting morphologic descriptions. But then the challenge of course is that how do you validate this? So then we created validation data sets, validated on what multiple choice questions on free flowing questions where multiple pathologists, we had a panel of seven pathologists look through every response from our model as well as more generic models like the OpenAI, GPT-4 and BiomedCLIP and other models that are publicly available, and then compare how well this pathology specific model does in comparison to some of those other models.
(26:58):
And we found that it was very good at morphologic description.
Eric Topol (27:05):
It's striking though to think now that you have this large language model where you're basically interacting with the slide, and this is rich, but in another way, just to ask you, we talk about multimodal, but what about if you have electronic health record, the person's genome, gut microbiome, the immune status and social demographic factors, and all these layers of data, environmental exposures, and the pathology. Are we going to get to that point eventually?
Faisal Mahmood (27:45):
Yeah, absolutely. So that's what we're trying to do now. So I think that it's obviously one step at a time. There are some data types that we can very easily integrate, and we're trying to integrate those and really have PathChat as being a binder to all of that data. And pathology is a very good binder because pathology is medicine's ground truth, a lot of the fundamental decisions around diagnosis and prognosis and treatment trajectory is all sort of made in pathology. So having everything else bind around the pathology is a very good idea and indication. So for some of these data types that you just mentioned, like electronic medical records and radiology, we could very easily go that next step and build integrative models, both in terms of building the foundation model and then linking with language and getting it to generate responses and so forth. And for other data types, we might need to do some more specific training data types that we don't have enough data to build foundation models and so forth. So we're trying to expand out to other data types and see how pathology can act as a binder.
Eric Topol (28:57):
Well if anybody's going to build it, I'm betting on you and your team there, Faisal. Now what this gets us to is the point that, was it 96% or 95% of pathologists in this country are basically in an old era, we're not eking out so much information from slides that they could, and here you're kind of in another orbit, you're in another world here whereby you're coming up with information. I mean things I never thought really the prognosis of a patient over extended period of time, the sensitivity of drugs to the tumor from the slide, no less the driver mutations to be able to, so you wouldn't even have to necessarily send for mutations of the cancer because you get it from the slide. There's so much there that isn't being used. It's just to me unfathomable. Can you help me understand why the pathology community, now that I know you're not actually a pathologist, but you're actually trying to bring them along, what is the reason for this resistance? Because there's just so much information here.
Faisal Mahmood (30:16):
So there are a number of different reasons. I mean, if you go into details for why digital pathology is not actively happening. Digitizing an entire department is expensive, retaining large amounts of slides is expensive. And then the value proposition in terms of patient care is definitely there. But the financial incentives, reimbursement around AI is not quite there yet. It's slowly getting there, but it's not quite there yet. In the meantime, I think what we can really focus on, and what my group is thinking a lot about is that how can we democratize these models by using what the pathologists already have and they all have a microscope and most of them have a microscope with a camera attached to it. Can we train these models on whole slide images like we have them and adapt them to just a camera coupled to a microscope? And that's what we have done for PathChat2.
(31:23):
I think one of the demos that we showed after the article came out was that you could use PathChat on your computer with the whole slide image, but you can also use it with a microscope just coupled to a camera and you put a glass light underneath. And in an extreme lower source setting, you can also use it with just a cell phone coupled to a microscope. We're also building a lighter weight version of it that wouldn't require internet, so it would just be completely locally deployed. And then it could be active in lower source settings where sometimes sending a consult can take a really, really long time, and quite often it's not very easy for hospitals in lower source settings to track down a patient again once they've actually left because they might've traveled a long distance to get to the clinic and so forth. So the value of having PathChat deployed in a lower source setting where it can run locally without internet is just huge because it can accelerate the diagnosis so much. In particular for very simple things, which it's very, very good at making a diagnosis for those cases.
Eric Topol (32:33):
Oh, sure. And it can help bridge inequities, I mean, all sorts of things that could be an outgrowth of that. But what I still having a problem with from the work that you've done and some of the other people that well that are working assiduously in this field, if I had a biopsy, I want all the information. I don't want to just have the old, I would assume you feel the same way. We're not helping patients by not providing the information that's there just with a little help from AI. If it's going to take years for this transformation to occur, a lot of patients are going to miss out because their pathologists are not coming along.
Faisal Mahmood (33:28):
I think that one way to of course solve this would be to have it congressionally mandated like we had for electronic medical records. And there are other arguments to be made. It's been the case for a number of different hospitals have been sued for losing slides. So if you digitize all your slides and you're not going to lose them, but I think it will take time. So a lot of hospitals are making these large investments, including here at the Brigham and MGH, but it will take time for all the scanners, all the storage solutions, everything to be in place, and then it will also take time for pathologists to adapt. So a lot of pathologists are very excited about the new technology, but there are also a lot of pathologists who feel that their entire career has been diagnosing cases or using a microscope and slide. So it's too big of a transition for them. So I think there'll obviously be some transition period where both would coexist and that's happening at a lot of different institutions.
Eric Topol (34:44):
Yeah, I get what you're saying, Faisal, but when I wrote Deep Medicine and I was studying what was the pathology uptake then of deep learning, it was about 2% and now it's five years later and it's 4% or 5% or whatever. This is a glacial type evolution. This is not keeping up with how the progress that's been made. Now, the other thing I just want to ask you before finishing up, there are some AI pathology companies like PathAI. I think you have a startup model Modella AI, but what can the companies do when there's just so much reluctance to go into the digital era of pathology?
Faisal Mahmood (35:31):
So I think that this has been a big barrier for most pathology startups because around seven to eight years ago when most of these companies started, the hope was that digital pathology would happen much faster than it actually has. So I think one thing that we're doing at Modella is that we understand that the adoption of digital pathology is slow. So everything that we are building, we're trying to enable it to work with the current solutions that exist. So a pathologist can capture images from a pathology slide right in their office with a camera with a microscope and PathChat, for example, works with that. And then the next series of tools that we're developing around generative AI would also be developed in a manner that it would be possible to use just a camera coupled to a microscope. So I think that I do feel that all of these pathology AI companies would have been doing much, much better if everything was digital, because adopting the tools that they developed would very straightforward. Right now, the barrier is that even if you want to deploy an AI driven solution, if your hospital is not entirely digital, it's not possible to do that. So it requires this huge upfront investment.
Eric Topol (37:06):
Yeah, no, it's extraordinary to me. This is such an exciting time and it's just not getting actualized like it could. Now, if somebody who's listening to our conversation has a relative or even a patient or whatever that has a biopsy and would like to get an enlightened interpretation with all the things that could be found that are not being detected, is there a way to send that to a center that is facile with this? Or if that's a no go right now?
Faisal Mahmood (37:51):
So I think at the moment it's not possible. And the reason is that a lot of the generic AI tools are not ready for this. The models are very, very specific for specific purposes. The generalist models are just getting started, but I think that in the years to come, this would be a competitive edge for institutions who do adopt AI. They would definitely have a competitive edge over those who do not. We do from time to time, receive requests from patients who want us to run their slides on the cancers of unknown primary tool that we built. And it depends on whether we are allowed to do so or not, because it has to go through a regular diagnostic first and how much information can we get from the patient? But it's on a case by case basis.
Eric Topol (38:52):
Well, I hope that's going to change soon because you have been, your team there has just been working so hard to eke out all that we can learn from a path slide, and it's extraordinary. And it made me think about what we knew five years ago, which already was exciting, and you've taken that to the fifth power now or whatever. So anyway, just to congratulate you for your efforts, I just hope that it will get translated Faisal. I'm very frustrated to learn how little this is being adopted here in this country, a rich country, which is ignoring the benefits that it could provide for patients.
Faisal Mahmood (39:40):
Yeah. That's our goal over the next five years. So the hope really is to take everything that we have developed so far and then get it in aligned with where the technology currently is, and then eventually deploy it both at our institution and then across the country. So we're working hard to do that.
Eric Topol (40:03):
Well, maybe patients and consumers can get active about this and demand their medical centers to go digital instead of living in an analog glass slide world, right? Yeah, maybe that's the route. Anyway, thank you so much for reviewing at this pace of your publications. It's pretty much unparalleled, not just in pathology AI, but in many parts of life science. So kudos to you, Richard Chen, and your group and so many others that have been working so hard to enlighten us. So thanks. I'll be checking in with you again on whatever the next model that you build, because I know it will be another really important contribution.
Faisal Mahmood (40:49):
Thank you so much, Eric. Thanks.
**************************
Thanks for listening, reading or watching!
The Ground Truths newsletters and podcasts are all free, open-access, without ads.
Please share this post/podcast with your friends and network if you found it informative
Voluntary paid subscriptions all go to support Scripps Research. Many thanks for that—they greatly helped fund our summer internship programs for 2023 and 2024.
Thanks to my producer Jessica Nguyen and Sinjun Balabanoff for audio and video support at Scripps Research.
Side note: My X/twitter account @erictopol was hacked yesterday, 27 July, with no help from the platform to regain access despite many attempts. Please don’t get scammed!

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
Euan Ashley: Exercise may be the single most potent medical intervention ever known
5 juil.· Ground Truths
Recently, a series of papers were published in Nature and Nature journals illuminating the physiologic effects of exercise from an NIH initiative called MoTrPAC. To understand the wealth of new findings, I spoke with Professor Euan Ashley, who, along with Matt Wheeler, heads up the bioinformatics center.
Earlier this week, Stanford announced Euan Ashley will be the new Chair of the Department of Medicine. He has done groundbreaking work in human genomics, including rapid whole genome sequencing for critically ill patients and applying the technology for people with unknown diseases. A few years ago he published The Genome Odyssey book. As you’ll see from our conversation, he has also done extensive work on the science of exercise.
Video snippet from our conversation. Full videos of all Ground Truths podcasts can be seen on YouTube here. The audios are also available on Apple and Spotify.
Transcript with audio and external links
Eric Topol (00:06):
Well, hello, it's Eric Topol with Ground Truths, and I'm really delighted today to welcome my friend, Euan Ashley. He is the Roger and Joelle Burnell Chair of Genomics and Precision Health at Stanford. He's done pioneering work in genomics, but today we're going to talk about something very different, which he also is working in exercise. Exercise the cover of a Nature paper in May regarding this MoTrPAC, which we're going to talk about this big initiative to understand the benefits of exercise. But before I hand it over to Euan, and I just want to mention his description of the paper that he posted to summarize started with, “Exercise may be the single most potent medical intervention ever known.” So Euan welcome.
Euan Ashley (01:01):
Yeah, well, great. It's wonderful to be here, Eric, and so nice to see you.
Eric Topol (01:06):
Yeah. Well, we have a lot to talk about because exercise is a fascinating topic. And I guess maybe we'd start with the MoTrPAC, which is an interesting acronym that you all came up with. Maybe tell us a bit about that with the 800 rats and the 2,400 people and the 17,000 molecules, there’s a lot there.
Euan Ashley (01:24):
Right, right. Yeah. Well, first of all, of course, before you do any scientific study, especially with a large number of people in a consortium, you need a good acronym. So that was where we started with the idea was to focus on the molecular transducers of physical activity. As you pointed out there at the beginning, we really don’t have a more potent medical intervention, especially for prevention of disease. I mean, it’s just such a powerful thing that we have, and yet we don’t really understand how it works. And so, the MoTrPAC Consortium was designed to really work together, bring groups of people across the US together who all have some interest in exercise and some ability to measure molecules and really put together the world's largest study of exercise to try and start answering some of the questions about where the potency of this intervention come from.
Eric Topol (02:20):
So the first crop of papers, and there were several of them that came out all on the same day in Nature publications, was about the rats. The people part is incubating, but can you give us a skinny on, there was a lot there, but maybe you could just summarize what you thought were the main findings.
Key MoTrPAC Findings
Euan Ashley (02:43):
Yeah, of course, of course. And the MoTrPAC Consortium, I'll say first of all, yeah, large group is probably I think 36 principal investigators funded by the Common Fund. And so, it brings together large numbers of people, some of whom who spend most of their time thinking about let’s say animal exercise. Some have spent a lot of time thinking about humans in exercise and many of whom think about measuring technologies. And as you say, these first group of papers were focused on the rat study, but actually the study goes much more broadly than that. But of course, there are some advantages to the animal protocols. We can look at tissue and we'll talk about that in a moment. But the humans, of course, are where we're most interested in the end. And we do have tissues coming from humans blood and adipose tissue and skeletal muscle, but those are obviously the only organs we can really access.
(03:31):
So there's a rat study, which is this one we'll talk about, and that's aerobic exercise and training. There's human studies that include aerobic exercise, strengths studies as well. There's a study in kids, pediatric study and then also a study of people who are very fit because here we're focusing on the change from sedentary to fit. And so that gives us the key exercise signal. So this first crop of papers was really our first look, cross-tissue, cross multi-omics, so multiple different modalities of measurement. And I think, yeah, we were like about nine and a half thousand assays, 19 tissues, 25 different measurement platforms, and then four training points for these rats. So let's talk about the rats for a minute. What do they do? So they normally live at night. They're active at night. In this study, we reverse that so that we can actually do the studies during the day.
(04:25):
So we reverse their at night cycle and they do their treadmill exercise over the course of several weeks. They start with about 20 minutes, and they do more every day. There's a control group of rats that just get placed on the treadmill and then don't do any exercise. And so, this is a controlled study as well. And over the course of time, we work more, it's about eight weeks in total and then two days after each of those bouts of exercise. So it's not an acute study, we measure to see where we are. So we also have this time trajectory of exercise. So what did we find? I mean, I think the first thing I would say, we talked about just how potent exercise is. It's very, very clear from looking at all these tissues that when you exercise regularly, you are just a different person, or in this case a different rat.
(05:15):
Like literally every tissue is changed dramatically and some in quite surprising ways. So I give you a couple of the things that surprised me or that I thought were most interesting. The first thing was this question of how does exercise actually work? Because exercise is a stress. You go out and you pound the pavement or you're on the bike or whatever, and then your body recovers. And so, there's been this idea, it's referred to as hormesis, this idea that some of the benefit of exercise might come from this recurrent stress. So your body learns how to deal with stress. And so given that we were very interested that this heat shock response was so prominent across multiple tissues. So heat shock proteins are molecular chaperones and they take care of protein folding to make sure it's appropriately done and they prevent protein aggregation. And when proteins need degraded because they're damaged, the heat shock system jumps in.
(06:10):
So perhaps not surprising, but pretty interesting that the heat shock proteins were very prominent part of the stress response to exercise. And remember, this is not acute exercise, so these are benefits that are built up over time, so that was one. A surprising one to me, the adrenal gland. So we're used to thinking of adrenaline as an epinephrine, as a stress hormone, but actually we saw dramatic changes in the adrenal gland and we don't necessarily think too much. You think about the exercising muscles, you think about the heart, we think about the lungs, when we think about exercise, you don't necessarily think that you're changing your adrenal gland, but it was one of the most changed tissues. The immune system was a common upregulated system. We saw that. And in fact, some of the tissues in which the immune genes were most changed were somewhat surprising.
(07:02):
So the small intestine, for example, was a place where there was a highest enrichment of immune mediated pathways. And then some tissues changed pretty early, like the small intestine changed after just one or two weeks of training other tissues like the brown adipose tissue. It was more like seven or eight weeks of training before we saw the real changes in there. So just one or two little things that struck out, but I think this really the first molecular map of exercise. So we're looking across the whole system across multiple modalities of measurement across multiple tissues.
Simulating Stress
Eric Topol (07:34):
So as far as understanding the benefits of exercise, does this tell us that it really does simulate stress that it's conditioning the body to deal with stress as reflected by the various points you just summarized?
Euan Ashley (07:51):
Yeah, I think that is exactly right. I mean, part of what we were trying to understand was in what way are you changed after you do exercise regularly? And I think if we think about things that are positive, then the ability to deal with stress at a cellular level, quite literally repair mechanisms seems to be a big part of it. The other aspect that was interesting is that when you're measuring this many analytes, you can also compare that with disease. And so, we understand that exercises is preventive benefit against disease. So in some cases, and this was work highlighted by my colleague Maléne Lindholm in the mitochondrial paper that came along with the main paper and she looked with a team across all mitochondrial changes across all of the tissues of the cell. So these are the workhorses of the individual cells that like the batteries inside the cells of the mitochondria.
(08:54):
And we saw big changes across, it's not surprisingly, but it's the energy source for cells, big changes across many tissues. But interestingly for two specific really important diseases, a liver disease in one case and type 2 diabetes on the other, it was very clear that the training upregulated a network that was exactly the opposite of that of the disease. And so, it really was intervening in a way that was very specifically opposite to the way we know disease mechanisms go. So it does seem like, I mean people talk about an exercise pill. I think this shows that that is just not going to be possible. There may be ways we could mimic some elements of exercise, but there's no pill. This is a multisystem, multi-tissue, multidimensional response to exercise.
Eric Topol (09:44):
Yeah, I think it's really important. That was one of the questions I was going to ask you is whether this would ever be simulated by a drug. And I think you already answered that, and the fact that it's so comprehensively sweeping across every organ and all these different signals, tens thousand plus signals across them, it's really striking. We never really understood the benefits of exercise and not that it's all resolved by any means. Some of the things that were interesting too was the sex specific findings. Maybe you want to comment about that because we don't spend enough time thinking about how sex does have a big effect on physiology.
Sex-Specific Findings
Euan Ashley (10:24):
Yeah, I mean that's a really good point and one that I think was really underlined for us at every corner, every turn of the analysis here. So really no matter which measurement modality, no matter which tissue, no matter which point of training, if we just asked these computer models to sort of separate the data according to the prominent signals without giving it a clue of what to do, the so-called unsupervised models, then sex basically came out every single time. So I think you say you're absolutely right that we so often overlook the difference. For years we've said, oh, it's too expensive to do animal studies in both sexes, so we'll just pick one. And males were picked more often. But there are plenty of studies that were just females, and I mean that clearly is wrong, and we are really, sometimes it appeared like we're almost dealing with two different species.
(11:18):
They were so different. But I think we can also learn from what those differences were. Interestingly, some of them were most profound in adipose tissue, so in fat, and that was the case both at rest, sedentary and amplified by exercise. So we saw big difference between females and males in relation to the kinds of signals that were prominent in the white adipose tissue. So this fat storage tissue, for example, in sedentary females, insulin signaling and the trigger to make fat and store fat was very prominent. But whereas in the males, even before any exercise, the fat signals were more related to metabolism, and we could have wild speculation about in evolutionary terms why that might be. Obviously, males and females have different biological many differences in their biology and obviously thinking about hormone systems and specifically pregnancy of course. And so, we could probably come up with some theories. In reality, all we know now are these observations were found and they're pretty interesting and they show us that we really always need to think separately about both sexes and look at both independently.
Eric Topol (12:39):
Well, and the other thing that you already pointed out, but I just want to underscore, you can't do this stuff in people. You can't just do fat biopsies and whatnot. So I mean, the fact that you can do this multi-omic, multi-organ type assessment is just really an extraordinary opportunity for learning. And while we're on the white fat story just briefly, we would rather have a lot more brown fat, but as we age, and I assume it's the same in rats, they don't have much as they get older brown fat. Does exercise help us get more brown fat or are we just stuck with the white adipose tissue?
Brown vs White Fat
Euan Ashley (13:21):
Yeah, well, it certainly allows us to have less of a white adipose tissue, and I think it's potential that our brown adipose tissue maybe more functional, and for those who are listening who are not familiar, I mean these really are different colors that relate to the actual color of the tissue, but the color is different because the brown adipose tissue contains lots of mitochondria and lipid droplets, and the brown adipose is there to help essentially generate heat. It has a very different function in a way, but even white adipose tissue that we think of as just being about storing energy, people think of fat as a very metabolically neutral or inert tissue, but in reality it's not. It's signaling. It's constantly, it's a tissue that's as alive as any other and not just a storage for excess energy, but exercise definitely appears to alter both in this sexually dimorphic way as we noted already and clearly both in a positive health way where I think the makeup of the brown tissue is different. The white tissue, there is less of it obviously with exercise, which is something that is well known, but not new here for the first time. But still important to have seen that even in the rats.
Eric Topol (14:49):
And there's even, we talked a moment go about drugs, but there are some molecules that are thought to be able to help convert white to brown fat that are understudy and we'll see if they get anywhere that's interesting. But also, you talked about aerobic exercise and with us both being cardiologists, and I know throughout my earlier part of my career, we only talked about aerobic exercise. There was no such thing as strength training, and we even discouraged that or we never talked about it. Now we know how important strength training is and not just strength and resistance training, but balance and posture and all these other things. I assume you can't study that in the rats.
Euan Ashley (15:32):
Well, it's not impossible. This study of course is about endurance, but as you say, and there are some models, I mean I've even seen models in trying to trigger flies to do strength training.
Eric Topol (15:46):
Wow, I didn’t know that.
Intensity of Exercise
Euan Ashley (15:46):
That somewhere, yeah, we'll have something, there are various methods of making animals hang off things, and this was treadmill. So it's a fairly routine and standard I think part of a rat's life to run. So this was not so different. As we mentioned at the beginning in the human study, we do have a strength portion and the endurance portion, which I think is very important because as you say, the benefits of exercise are found really across both of those. And indeed, as you say, flexibility and other often neglected element of physical activity. But yeah, those benefits are there for both aerobic exercise and endurance. And in fact, they are perhaps even higher for higher intensity exercise. Although I think we don't necessarily recommend everybody do higher intensity exercise. I don't think it's necessary to get most of the benefits of exercise, but there is some additional benefit.
(16:42):
One of my favorite facts, I think I first saw it probably on a presentation a few years ago, but I looked up the original and recalculated it. But if you look at this very big study of half a million people and look at their physical activity over the course of years and correlate it with their likelihood of being alive or being dead, then it was clear that one minute of exercise bought you five minutes of extra life. And I just thought that was just a really interesting way of putting it essentially. And actually it's a little more, if you did high intensity exercise, one minute would give you seven or eight minutes of extra life. So I tell this to my patients when they come in and tell me they don't have enough time to exercise. I said, oh, well, one minute of exercise. I'm not very popular when I tell them that, but anyway.
Eric Topol (17:30):
You think it's true. Do you think it's based on good data?
Euan Ashley (17:34):
Well, the data is large, I mean half a million people. I think we've also seen it currently since the early fifties when we were first doing the London bus conductor study that Jerry Morris did that you will know well, where he compared bus conductors on the London to the bus drivers and found a significantly reduced cardiovascular mortality among the conductors because they were on their feet all day up and down stairs and the driver otherwise in the same environment the drivers were sitting. So I think we have a wealth of epidemiologic correlative evidence that exercise leads to a greater length of life, greater longevity, maybe more than for anything else. The causal evidence is less of course, but we do have causal evidence too. There are enough randomized trials and now increasingly some genetic causal evidence that helps us understand that this is really a causal link and that we actually can change our outcome if we do additional exercise.
Mental Health Benefit
Eric Topol (18:32):
Oh, and I don't question at all what you said about the enhancing healthy aging health span and even possibly lifespan. I just wondered about the one to five ratio if we could assert that. I mean that's really interesting and it's a good motivating factor because as you well know by that WHO criteria, one out of four people aren't even close to the modest exercise recommendation. So we got ways to go to get people to spruce up exercise. Now speaking of people, I do want to come back to MoTrPAC and the people plan, but I do want to before that get your sense about a couple of really fascinating studies. So earlier this year there was a study of every exercise study that's been looking at mental health along with SSRIs that name drugs that are used for mental health. And it was a pretty fascinating study. I think I'm just going to pull it up. They looked at everything that this is for depression, walking, jogging, yoga, strength training, SSRIs. And what was fascinating is that dancing, walking, jogging, it made the drugs look like a joke. They didn't seem to work at all. So this was 218 studies with over 14,000 people. And so, I don't know that enough people recognize this fact that this Prozac nation and all this stuff about the SSRIs, but exercise seems to do wonders for people who are depressed, anxious, stressed. What do you think about that?
Euan Ashley (20:26):
Yeah, I mean it's exactly right. I mean I think that it's very clear from the data and as you mentioned, you and I tend to focus first on the cardiovascular benefit, which is very significant, potentially 50% reduction in risk, but there are similar sorts of numbers when you look at mental health and exercise as an intervention for mental health has been very well studied and has these really dramatic benefits. And I think even if we go in the more general population and think about the fact people talk about a runner's high or an exercise high, and many, many of us, myself included, feel that. And a few years ago, I started exercising every morning and now if I don't do that, I really feel like I'm missing something, there’s something in the chemistry of my brain is not quite right. And so, I think that benefit for those who have mental health issues is also very much felt and is real at the brain chemical signaling level and with this few adverse effects as exercise has, I do think we need to think of it earlier and more prominently for almost every disease.
Eric Topol (21:40):
Yeah, you're I think alluding to the opioids that are released with exercise and addiction to exercise, which is what ideally if everybody could be addicted to exercise, that might help a lot of things. As you mentioned in your post that I started with, “its benefits in prevention outstrip any known drugs: 50% reduction in the cardiovascular disease, 50% reduction in risk of many cancers, positive effects on mental health that we just discussed, pulmonary health, GI health, bone health, muscle function. You name it.” So you said it really well there, and that was just one recent report that substantiated the mental health. I want to also mention another report that's fascinating on cancer that is a publication again recently was looking at both mice and people with pancreatic cancer. And what was fascinating about it is the more exercise of the mice and in the people, the more survival that is from pancreatic cancer, which as we both know and all the listeners will know, is that one of the worst cancers of humankind. So the affecting cancer is fascinating. Now can you dial up your immune system response with exercise?
Euan Ashley (23:02):
Yeah, I think you can. And I think we were at some level expecting to see it because it's certainly a known thing, but I think again, this is able, our ability to measure it in this study is just much deeper than we've ever had in any study before. And so, I think when we think about mechanisms that might relate to reduced risk of cancer, as you say, we think first of the immune system and that signal was there in many places. As we mentioned at the very beginning, sometimes to me in some slightly surprising places like the small intestine, we don't think of that necessarily as the seat of immune activation, but I think what we were doing, what we were seeing is those signals really across all the tissues and ultimately the immune system is a distributed system. It senses in multiple places and then obviously has implementation.
(23:53):
Now exactly in what way we've turned up our T or B cells, for example, to be able to attack those cancers or support the therapy that's been given. I don't think we understand that yet. But actually, you bring up another great point, which is part of MoTrPAC was to create this molecular map and analyze it and put the first analysis out there. So that's what we've done, but just as big and maybe even a bigger reason is that to release the data and to make it accessible for everyone and anyone in the world as of the moment this paper came out can go to our data portal at
https://motrpac-data.org/
and download the data and then use that in their own work. They can do their own analysis just of this data, but also what we're hoping is that they'll start to use the data, let's say as control data for a cancer study or for a diabetes study or for others. So we really hope it'll fuel many, many more studies over many years from now.
Eric Topol (24:52):
Yeah, I mean that open science approach to applaud that it's so vital and amplifies what's good to come out of this really important initiative. Now you mentioned the opioids and proteins that are secreted with exercise, exerkines is a term that's used and also I guess these extracellular vesicles (EVs) not electric vehicles. Can you tell us about exerkines and EVs and are they part of the story?
Euan Ashley (25:25):
Yeah, and actually in the human study there's a specific exosome analysis that will be reported there. Yeah, I think that when we think about this multi-system nature of exercise, and one of the fascinating things was to be able to have these omics in multiple tissues and think about how those tissues were signaling to each other. So obviously there are some tissues that are more fundamental to the exercise response. We think of those as the skeletal muscles. They literally the effectors of our ability to exercise. And I think we think of the heart and lungs in particular in the blood system of course, but we were seeing changes everywhere and it's one of the reasons we were seeing changes everywhere is that there are molecules that are essentially secreted into the circulation or locally by these exercising muscles, exerkines that have a number of positive benefits.
(26:21):
And it is possible if there's some mechanism towards mimicking some of what exercise does with a drug, then that's a good place to go look for it. And I think that this will also fuel those thoughts. I think we both, we'd agree that there isn't going to be one pill that will do all the magic of exercise, but I think there are probably things we will learn from the study where we say, well, this was a very positive benefit and it seems to be mediated by this particular molecule, and that's something that could potentially lead towards a more targeted drug. I think we'll definitely get into that and understanding just we're systems people are, again, I think we think in physiology, so when we see the tissues like connecting and communicating with each other, I think that just makes a lot of sense from a systems perspective.
Eric Topol (27:10):
Now getting onto the forthcoming work that's going to come out with the 2,400 people and the different groups that you mentioned, I wonder if it'll include things like biologic aging with DNA methylation, will it have immunomes to characterize the differences in the immune system? What kind of things might we expect? Obviously, you can't get tissue, but for blood samples and things like DNA methylation, can we get some more illumination on what's going on?
Euan Ashley (27:41):
Yeah, I think we can. And of course, ultimately the human is the organism we're most interested in. Interestingly, I'll say interestingly as well, we can get some tissue and huge credit to both the investigators who are doing this and most credit of all to the individuals who agreed to join the study because they actually agreed not just to give blood samples, but actually to give skeletal muscle samples. So a biopsy of the skeletal muscle and a biopsy of the fat pad. So we will actually have two other tissues in the humans, not this obviously vast range that we talked about with the rat study, but we'll have those two other tissues and we'll also then have the rat data, which is the other great thing. So we'll have this foundational insight that we can then bring to the human study with the humans as we mentioned before as well, we'll have not just endurance but strength trained, we'll have it in kids as well, and we'll have these higher intensity exercise.
(28:36):
I think we will be able to connect with this, as you mentioned, longevity literature or the health span literature where we can start to think about DNA methylation. We do have genomes of course, on all of the individuals. It won't be a study powered because it's thousands individuals, these kinds of numbers. It won't be powered to give us genetic predictors. If you think about the studies had to be hundreds of thousands of people and even more now in order to give us, let's say common variant predictive. So we won't be able to do that, but there's lots of connections we'll be able to make by being much closer to the effector systems, which is to say the proteins and the metabolites and those signals we're already seeing are very significant. And so, I do think that there'll be a lot of new signals that we'll see that are specific to humans that will connect into other bodies of work, for example, the longevity, and we'll see those in blood and I hope that we'll be able to connect also the skeletal and adipose tissue data as well.
Eric Topol (29:37):
One of the things that would be wonderful to connect if you can, our mutual friend and your colleague at Stanford, Tony Wyss-Coray has these organ clocks that have been validated now in the UK Biobank, and then you can see what's happening with the wealth of plasma proteins that have been validated across each organ. So without having to do tissue, you might get some real insights about organ clock. So I mean, I'm really looking forward to the people part of this. When do you think the next wave of output's going to come from MoTrPAC?
Euan Ashley (30:11):
Well, I think that another element of the study is that we have ancillary studies, so investigators who said, I want to be able to use MoTrPAC data and use some of the infrastructure, but I'm looking for funding for my parallel study. So some of those ancillary studies will start to come out over time, which I think will be interesting and will be a very good place to see the breadth of activity that has been triggered by this one investment. The human study is coming along. We're actually just now plotting the last two or three years of the consortium. Time has really gone by pretty fast, and we've had to scale back just a little bit on the total numbers of humans, but it should still be, I think probably the largest multi-omics study of humans that there has been. And I think if we were going to plan one of those, then planning it to study around exercise definitely, definitely makes sense. So there is some data that was, of course Covid happened in the middle of this, so that was a major challenge with hitting the original numbers. But there's some data from the humans who were recruited before Covid hit that will be coming out and hopefully in the relatively near future. And then the big study may still be a year or two away to get it finished. But after that, as we say, we hope that the data and the science will continue for I hope decades beyond just the collection of this repository.
Eric Topol (31:41):
That's great. You mentioned Covid and I did want to ask you about the folks with Long Covid who are suffering from fatigue and exercise intolerance and what do you think about this kind of vicious cycle? Because if they could exercise, it could help them get into a better state, but because of not being able to, it's just a negative feedback loop. Any thoughts about that?
Exercise and the Immune System
Euan Ashley (32:13):
I mean, it's such a good point and it's one of course that we talk to many of our patients where they, for whatever reason, sometimes it's because they are struggling with weight or they're struggling with other mobility challenges, and now we have this very large population who are struggling with fatigue. As you mentioned, it's a group that we were somewhat familiar with because of flu and because EBV and other, I mean long syndromes were something we were familiar with. They were just kind of rare, and so there wasn't really much work done on trying to understand them. Now as you've, I think articulated better than anyone, we have this entire population of people because of the scale of Covid who have these symptoms that are recognizable for the first time and including on your podcast, you have had folks on that have discussed it. Some of the insights that have happened from actually applying science, I wish there was an answer that was buried here in MoTrPAC and maybe there is, there will certainly have data from before and after the pandemic and maybe there may be some insights that we can bring to that.
(33:20):
I certainly think we have a lot of insights on the interaction between infection and the immune system. We talked about the potential for the immune system to be ramped up in that potentially being one of the mechanisms through which this might help cancer. There's also the idea of, and we've seen this with the effect of vaccination on Long Covid, which perhaps surprisingly does seem to have a significant benefit for at least a group of people. The assumption there is that we're ramping up the immune system and it's having that extra effect on whether it's actually pools of hidden antigens that are hidden from the immune system or whether it's some other element of the kind of ensemble attack of the immune system that is related to the symptoms. But either way, I think we feel that having a more ramped up immune system is likely to be beneficial, but at a very real human level, the point you made is the hard one. If you're really fatigued and you just feel you can't exercise, then these benefits are just out of reach and you're in this negative feedback cycle and breaking that cycle is hard. I think we try to suggest people do it very gradually because you can get a lot of benefit from just a little exercise and that's something, so that's some way, and then hopefully people can build up slowly over time, but it's a really big challenge.
Eric Topol (34:43):
I hope we can crack the case on that because I know that's something holding these folks back and there's just millions of them out there. Now let's talk about the healthy folks that you see in clinic. What do you advise them about exercise besides the fact that one minute we'll give them five minutes, but do you advise them to have X amount of aerobic and X amount of resistance and in the general person, what would you tell them patients?
Euan Ashley (35:13):
Yeah, yeah, I do. So I suggest habit is everything. So I suggest to people that they exercise every day or take one day of rest because I think there is some benefit with the stress response and having a rest day. So I suggest five or six days a week if possible, trying to get into a habit of doing it. So pick a time that works for you. It could be first thing in the morning, could be last thing at night. The jury's out on when the best time to exercise is. What it's very, very clear is that getting the exercise done is what counts. Accumulating time is also what counts. I mean, if you're not someone who wants to pull on running kit and go out running, that's fine, but accumulating steps, accumulating physical activity and moving is key. So not having people overshoot being too ambitious, but if they're really motivated to do something, then I would say five or six times a week a combination of both aerobic and endurance exercise and strength.
(36:07):
Usually I suggest two to one in favor of aerobic exercise, but it's also possible I think to alternate and do more 50/50. I think the key is that both are featured and then I think a bit neglected because to be honest, our data on it is just not as good, but flexibility is really critical and particularly in the senior population and for a group who sit all day long, I think for those two groups in particular, flexibility is really under-recognized as a major component. Even in my cardiology clinic, I've helped several patients just get over their back pain by teaching them some back stretching exercises. And so, I think that's neglected. So I suggest all three of those and really it's whatever works for the individual. I think the key is to find, it might be working in a group format, it might be going to a gym, it might just be taking regular walks. The key is to get moving and not sit. Get moving and do it regularly and get into the habit.
Individualized Exercise?
Eric Topol (37:09):
Yeah, and actually on that point about potential individualization in the future, I noticed that you and some people that worked in your lab and others, Svexa is a company you started for exercise. Can you tell us about that?
Euan Ashley (37:26):
Yeah, this was a PhD student who was in my lab many years ago and was doing his PhD joint between the Karolinska Institute in Sweden. And of course, the country of Sweden has a long history of exercise physiology, science, and as he came out, we realized that there was the potential for optimization of training for individuals, whether they're recreational athletes or elite athletes in the Olympics. And he was interested in taking this and running with it, which he did. So the company originally Silicon Valley exercise analytics, but shortened now to Svexa builds, builds products to help people basically individualize their training. And we work, say with recreational athletes on an individual basis, we work with a lot of Olympic athletes in multiple countries and the technology building the sort of magic sauce that many of these coaches even up to and including Olympic coaches have into a format that can be spread and amplified to many more people is one of the themes.
(38:29):
And when we think about professional athletes and the company works with a number of well-known brand name teams that are in soccer leagues and in national football league here in the US and really across professional sport, what we're thinking of there is optimizing performance. Of course, all the teams want to win, but reducing injury is the other key part because the management of load, these are professional athletes, they're getting up every day in training and they're trying to optimize their training and their coaches are trying to do that. And it's been a fairly data free zone over the years, but meanwhile, we actually have learned a lot about how to measure individuals and how to measure what training works, and if you think about a team that might be paying 20 million a year for their star player, if that player gets injured, that's a pretty expensive thing. And so, investing a little bit in understanding the training load, helping the coaches understand the data, and then adapting that to each individual in the team so that their chance of injury is lower. That's really a lot of what the company spends its time thinking about.
Eric Topol (39:36):
Now, do you use sensors like lactate and glucose and AI of their body and how do you figure this stuff out?
Euan Ashley (39:45):
Yeah, all of that is possible. It's interesting, some sports have a kind of culture of measurement. For example, lactate measurements, which as your listeners will know, is it requires a small blood sample usually from the finger or from the ear lobe. Some sports like swimming have done that for years. But other sports, it's just not been so much in the culture. So I would say that from the company perspective, we work with whatever data is available and we'll make recommendations if people want to think about wearable devices. Of course, the digital era is around us, and you can get a lot from just a standard watch in terms of heart rate, heart rate variability in terms of accelerometry and movement. You can do a lot with just that, but there's lots more. Many of these teams have GPS signals so they know how far an athlete moves in a given game, how fast they move, how much time they spend at tool speed versus medium speed.
(40:37):
So we can use all of that. And as you say, yes, AI for sure is a large part of what we do and a couple of different ways actually. One is just for the analysis of the data, but another is this idea of scaling expertise. This is something in the AI community. I know you talked about a lot where you could take the expertise of let's say a physician with a very specialized practice or an Olympic coach for a marathon runner and basically make a language model that contains that expertise and then allow many people, thousands of people potentially to benefit from that expertise that we'd otherwise be sort of locked up with next available appointment is 18 months down the road, but if your AI can potentially reflect a lot of what you have, a lot of your expertise, not all of it, we hope, but probably a lot of it, then that expertise could potentially be offered much more broadly. And if it's to help people exercise more and more effectively, it's going to be a lot of good that I think can come from that.
Eric Topol (41:33):
Yeah. No, it's really interesting. I think there's unlimited opportunities there. It's like Moneyball to the 10th power. It's like all this data that's in sports that gets me, I guess to the last question I had for you, and that is the elite athlete or athlete hard. These are people that are working out endurance just to the max, these extremists, and they're prone to heart issues like atrial fibrillation. Why is that? What's going on with these people that they exercise too much? Is it just the lack of moderation, extremism or what's going on?
Euan Ashley (42:10):
Yeah, well, so it's interesting that of course you mentioned atrial fibrillation. I think that really is the only downside of exercise, even fairly extreme exercise that I've ever been, I think that we've ever had really good data for. And I would say that over the years, and I've been one way or another touching the exercise science world for 20 years and more now and certainly have been asked very often, surely these people are doing themselves harm. And the reality is, although every now and again there's a study that shows some harm or they measure troponin, they measure something in the blood and someone says, oh, they must be doing themselves harm. It's been very hard to find it. The reality is atrial fibrillation though really is, especially for those ultra endurance athletes, that's for real. And that is, we don't know that it's associated with a mortality impact necessarily, but it's definitely annoying and it slows down.
Endurance Athletes and Atrial Fibrillation
(43:03):
We have athletes who come in and say they're cycling up a hill and suddenly they drop their power drops and they realize they've gone into atrial fibrillation. I used to play basketball with someone who would go into atrial fibrillation, so I would know when to try and get past him once he went into atrial fibrillation. But that's a real thing, and I think one of your questions was why I think I have a lot of close friends who are ultra endurance runners. They're among some of the most chilled and happiest people I know. I think those benefits of exercise are what they're enjoying, and I think there's a literature on addiction to exercise. So there is a small number of people who get addicted to that feeling and addicted to the chemical matter in their brain and can't stop, and they really do get to the point of doing themselves harm.
(43:53):
Fortunately, I think that's a pretty small number. And overall, although there are many consequences of chronic long-term exercise, almost all of them seem to be positive. The other one that you and I are probably very familiar with is the calcium scans that we see now much more often, it's common for people who've exercised a lot to have more calcium in their hearts. Now they have a lower risk of that. They have lower risk of heart attacks in general, one or two studies muddied the waters just a little. But in general, it's very clear they have very positive health benefits and yet they have more calcium. So they are an exception. We've seen in our sports cardiology clinic here at Stanford, several athletes every month, several will come in with this finding and we are explaining to them, this doesn't mean they have the same risk as someone who hasn't exercised at that level who would have that calcium score. It does seem to be very different, and it may be that there's a stabilization of those plaques in the arteries. I don't think we understand the biology that well, but we understand the epidemiology quite well, which is that their risk really is still low.
Eric Topol (44:59):
Yeah, no, it's interesting that there's still some uncertainties there and MoTrPAC may help guide us or at elucidate some of them. I guess it does bring up one other thing I got to get to with you because we didn't really get to the question of moderate to higher intensity, not to the level of the ultra exercises, but if you just do steps or do you sweat like hell, where do you draw the line? Or is that really part a function of age and ability? When you recommend exercise, because obviously you're rational and there's others out there that are exercising three or four hours a day and they're going to extreme craziness, but just in a reasonable thing, do you think just telling people who are 70 that walking is good enough or do you try to encourage them to push it?
Euan Ashley (45:59):
Yeah, I do encourage people to push it a bit because I think there's clear evidence that higher intensity, some degree of higher intensity exercise really does provide more benefit. But I think my main message first is because for most people, the potential of moderate versus high is in the distance and in the future for most people, we need to get them off the couch and get them on their feet. So my emphasis is that you can go a long way with just a little movement, even a little standing. And then I think if they're really getting into the habit and really doing some exercise then, and if they don't have a prior history of let's say, heart attack or other medical issues that might make high intensity exercise risky, if they don't have those, then I absolutely do get to the point where I recommend some amount of higher intensity exercise, because I think there is some evidence that it has a little extra benefit.
Eric Topol (46:51):
Oh, that's great. Well, this is the most in-depth conversation I've ever had with anybody on exercise, so Euan I really appreciate it. I mean, I knew you from all your work in genomics of course, and we've had some overlap from time to time, but the exercise stuff is fantastic. Did I miss anything?
Euan Ashley (47:09):
No, I don't think so. Just underline again to anyone who's listening if they're interested to play with this data, it's very much out there. It's a tool for the world, and they can go to
https://motrpac-data.org/
and even you can do some analysis without downloading any data either. If you just have a favorite gene or a favorite protein, you can type that in and take a look at some of the tools we have there. But yeah, really appreciate the conversation and very fun to chat about what has been a really, really fun project.
Eric Topol (47:39):
Well, thank you and all the folks at MoTrPAC, all the hard work and of course the funding that got it going to give it that runway of several years. So we'll look forward to more. I hope to convene with you again when some of the other studies come out, and thanks so much.
*****************************************************
Thanks for listening, reading or watching!
The Ground Truths newsletters and podcasts are all free, open-access, without ads.
Please share this post/podcast with your friends and network if you found it informative
Voluntary paid subscriptions all go to support Scripps Research. Many thanks for that—they greatly helped fund our summer internship programs for 2023 and 2024.
Thanks to my producer Jessica Nguyen and Sinjun Balabanoff for audio and video support at Scripps Research.
Note: you can select preferences to receive emails about newsletters, podcasts, or all I don’t want to bother you with an email for content that you’re not interested in.

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
Christopher Labos: Debunking Myths About What We Eat and Drink
26 juin· Ground Truths
A book that reads like a novel; it’s humorous, it’s a love story. Dr. Christopher Labos, an imaginative cardiologist and epidemiologist at McGill University, takes us through multiple longstanding misconceptions about different foods and drinks, and along the way provides outstanding educational value.
Video snippet from our conversation. Full videos of all Ground Truths podcasts can be seen on YouTube here. The audios are also available on Apple and Spotify.
Transcript with external links and links to the audio recording
Eric Topol (00:07):
Hello, it's Eric Topol with Ground Truths, and with me today is a cardiologist, Chris Labos from Montreal, who has written an extraordinary book. I just read it on my Kindle, “Does Coffee Cause Cancer? And 8 More Myths about the Food We Eat. Chris teaches at McGill University. He is a prolific writer at the Montreal Gazette and Canadian broadcast system, CBC, CJAD radio, CTV News. And he also has a podcast on the Body of Evidence and he probably has other stuff, but welcome Chris.
Christopher Labos (00:49):
Hello. Hello. Hello. Thank you for having me. It is a great honor to be on your podcast. I am in awe of the work that you've been doing, I mean, for all your career, but especially during Covid. So it's a big thrill for me to be on the podcast.
Eric Topol (01:03):
Well, for me, I have to say I learned about a person who is not only remarkably imaginative but also humorous. And so, have you ever done standup comedy?
Christopher Labos (01:16):
I have not. Although I was asked to chair the research awards that we did here at McGill one year because I've been doing local media stuff and they said, can you come and be like the MC? And I said, sure. And I said, do you want me to be funny? And they were like, well, if you can. And I went up there and people were laughing and laughing and laughing and then people, like some of my former attendings had come up to me and they're like, Chris, I don't remember you being this funny as a resident. And I was like, well, I guess you come into your own when you start your own career. But I think people were very, it's tough MCing a research awards because you're essentially, it's kind of like a high school graduation where you don't read the names in alphabetical order, right? It's like one name after the other. And I went up there and I tried to throw in a little bit of humor and people seem to like it. So I think that was the first, that was when I started to realize, oh, if you inject a little bit of levity into what you're doing, it tends to resonate a little bit more with people.
Eric Topol (02:13):
Well, no question about that. And what I love about this book is that it wasn't anything like I thought it was going to be.
Eric Topol (02:21):
Amazing. It was a surprise. So basically you took these nine myths, which we'll talk to, hopefully we'll get to several of them, but you didn't just get into that myth. You get into teaching medical statistics, how to read papers, all the myths. I mean, you are the master debunker with entertainment, with funny stuff. It's really great. So this is great, before we get into some of these myths and for you to amplify, but this is a gift of communication, science communication that is you get people to learn about things like p-hacking and you throw in love stories and all kinds of stuff. I mean, I don't know how you can dream this stuff up. I really don't.
Christopher Labos (03:10):
I sort of look back at the inception of this. This book did have sort of a few iterations. And I think the first time I was thinking about it, I mean I wrote it during Covid and so I was really thinking about this type of stuff. It's like how do we educate the public to become better consumers of scientific information? Because there was a lot of nonsense during Covid. So teaching them about confounding, which I think through a lot of people when we started talking about low vitamin D levels and Covid and outcomes and all that. And so, I started like, how do I write this type of book? And I thought, yeah, this should probably be a serious science book. And the first version of it was a very serious science book. And then the idea came and try to make it a conversation. And I think I sort of wrote it.
(04:02):
There's a book that may not be that popular in the US but it was kind of popular here in Canada. It was called The Wealthy Barber. And it was all about personal finance. And the idea of the book was these people would go into a barbershop and the barber would talk to them about how to save money and how to invest in all that. And it was fairly popular and people liked that back and forth. And I said, oh, maybe I could do something like that. And then I wrote the first chapter of the doctor who goes in to talk to the barista and I showed it to a friend of mine. I said, what do you think? Do you think this would work? And her response to me by email was two lines. It was pretty good period. But I kept expecting him to ask her out at the end. And the minute she said that I thought, oh my God, this is a love story. And so, I reshaped everything to make this a love story. And I don't think the publishers were expecting that either because they were like, the first comment from the editor was, most science books don't have a narrative arc to them in character, but this one does. So there you go.
Eric Topol (05:00):
This is a unique book. I hope that people who listen or read the transcript will realize that this is a gift. It's a model of communication and it just is teaching things almost like you don't realize it. You're just learning all this stuff. So let's get into some of these because they're just masterful. I guess I should start ask you, you have nine of them. You could have picked 20 more, but which one is your favorite? Or do you have one?
Christopher Labos (05:31):
I think the one, it's hard to say. I think the first one in the book is the vitamin C one. And I think it's the most interesting one to explain to people, not just because vitamin C to fight the common cold is so pervasive as a product and a thing that people believe. But it also, I think has the greatest opportunity to teach people about what is one of the most important ones, which is subgroup analysis and p-hacking. And it's so easy to bring that back into a comedic level with some of the graphs that I put in there. I think a close second would probably be the coffee one where I was talking about selection bias, because those examples of online dating and then all the jokes that came from it. And it's hard to say how much of it was the subject and how much of it was the character.
(06:21):
Because I'd always heard stories of authors when they say like, oh, the characters will tell me what to say. And I always thought that sounds like bollocks. How could that be possible? You're the author, you write what's on the page. But then the minute I started actually writing it and started envisaging these characters, all of a sudden the characters took on a life of their own and they were dictating how the story ended up. So the coffee one I think is also good too. And I guess it became the title of the book. So I guess that's a good indication that was popular. But when you can really spin it out and make it obvious to people using common examples, I think those are interesting ones. So the vitamin C and the coffee ones, I think were probably the most interesting.
Eric Topol (07:02):
Let's take those first because you've mentioned them and then hopefully we'll get into some others. Now in the vitamin C, you're going on a plane and you hook up with this guy, Jim, on the plane. I know none of this stuff really happened, and you're explaining to him the famous ISIS-2 trial about the Gemini and Libra subgroup. So for those of people who are listening, can you review that? Because that of course is just one of so many things you get into.
Christopher Labos (07:33):
I know it's almost amazing how short a memory we have in medicine, right? And again, this is sort of surprising me. I sort of knew the study and then I went back, and I looked at it and I thought ISIS-2 was in 1988. That's not that long ago. The fact that we didn't give aspirin. So for people who don't know, I mean, we did not give aspirin to people with cardiac disease for a very long time. And it was really from 1988 afterwards. So relatively recently, I mean I realized it's been a couple of decades, but still. So ISIS-2 was really the first trial to show that if you give aspirin to somebodywhen they're having a heart attack, you see a benefit. But what was fascinating in the study was this one subgroup analysis of people in whom it did not work.
(08:19):
And when I give public lectures, I often use this example because it's such a beautiful teaching case, and I go ask people, what do you think it was? And people are like, oh, hemophiliacs, smokers, people who drink alcohol. And then you find out, no, the subgroup in whom aspirin does not work is Geminis and Libras. And everybody sort of laughs and they think it's funny. And it's a beautiful example because a lot of people think it's like, oh, it was a joke or it was sort of silly science. But no, it was actually done purposefully. And the authors put that in there because they wanted to make the point that subgroup analysis are potentially misleading. And I sort of am a little bit in awe of, I mean the power or the intelligence to actually make it a point with the editors like, no, we're going to put this in here essentially as a teaching tool.
(09:09):
And it's amazing to me that we're still using it as a teaching tool decades after the fact. But it was just to show that when you have these tables where you have umpteen subgroup analysis, just by random chance, you will get some spurious results. And though our brain understands that Zodiac signs have nothing to do with the effectiveness of aspirin, you do the same subgroup analysis and diabetics and non-diabetics, and everybody was like, oh yeah, that's plausible. And yeah, it might be, but the computer doesn't know the difference, right. To the computer these are all ones and zeros. So if you don't go into it with a healthy skepticism about the limitations of subgroup analysis, you will eventually get fooled. And the problem with vitamin C research is I think a lot of very smart people have gotten fooled on this because they're like, well, overall the data is negative, but if we slice it up, we can find something that's positive. So maybe there's something here. And the number of people who have fallen in that trap over the years is unfortunately quite high.
Eric Topol (10:10):
No, and it's still happening and it is a famous subgroup story, but I just want to remind everybody that this was in the chapter on vitamin C and it's going into aspirin and subgroups. So each one of these chapters is not confined to the myth. They go into all sorts of other teaching examples in a humorous and fun way through conversations. Here it was with Jim on the plane. Now another one you mentioned, I forgot about this one. In the British Medical Journal, there was a paper, the Miracle of DICE Therapy.
Christopher Labos (10:45):
Miracle of DICE Therapy. Yeah, that's another brilliant one, because again, you couldn't do a study like this today, but basically for people who aren't aware of the paper, I mean, I think it was published in the Christmas issue. So again, just to show you how sometimes even in medical science, the humor is really, really effective. So what researchers did was they went to this neurology conference and they got all the people to participate in this live study, and they gave them dice and said, you're going to roll these dice. And they had white, red, and green dice and said, the exercise is for all of you to roll this dice and then analyze the data and tell us which color dice is off which one has been weighted. Because if you roll a one, two, three, four, or five, the patient has survived their stroke. If they roll a six, the patient died of their stroke.
(11:33):
So you go, you roll these dice dozens of times, generate your data. I mean, what we would do today with a random number generator but they were rolling dice. And they said, you figure out which of these dice is skewed. And so, the people at the conference went, they rolled their dice, they crunched their data, and they said, the red dice are skewed. There's a difference between the red dice and the white and green dice. And then the researchers revealed aha jokes on you. All the dice were the same. And the funniest part about that is that a lot of the people in the room didn't believe them. They refused to believe them that the dice were weighted because, and one of my favorite quotes was when student A refused to believe that his days were really loaded, he rolled one six and then a second and then a third, and he said, the room felt eerily quiet as he rolled a fourth six.
(12:25):
He had never rolled four sixes in a row in his life. And if you're there, I mean, yeah, you're going to be like, how do you doubt the power of your own eyes? You roll four sixes in a row, you think to yourself, gee, this must be the loaded dice. But that thing would happen. You put enough people in a room rolling enough dice, you will eventually get four sixes in a row in the same way that if you put enough monkeys in front of enough typewriters, eventually you're going to get all the works of William Shakespeare. So it's shocking how much our own human biases make us immune to the realization that random things are going to happen. And there was another, I think there was a quote in that paper too, where doctors are very willing to admit that chance affects whether they win a raffle, but they are surprisingly unwilling to admit that chance can affect the results of their medical research. And we don't appreciate it, even though, I mean, the reality is it happens all the time and we don't take the necessary steps to fix it sometimes and to address it, and we keep making the same mistakes over and over again.
Eric Topol (13:32):
Yeah, no, that's a great paper to illustrate. Again, a lot of important teaching points. Now as we get into the coffee, does it cause cancer? It brings up another theme in the book that I noticed. What you do is you pick up on papers or broadcasts that were decades ago that have become inculcated in our minds and our thoughts. And in this case, it was a famous New England Journal paper in 1981 raising the question about does coffee, if you drink too much coffee is that a risk factor for pancreatic cancer? So maybe you could take us through that, and somehow that gets into the NBA, it gets into H. pylori for ulcer. I mean, but maybe you could help get us through this coffee and cancer story.
Christopher Labos (14:23):
Yeah, I mean, well, and it's still happening isn't it, right? In 2018 in California, coffee was declared a carcinogen after that court case. I mean, it was ultimately overturned. So I sort of explained that saga in the chapter as well. And of course, we're going through it now with the decaf coffee, right? There are people trying to petition the FDA to get methylene chloride removed from decaf coffee, even though, I mean, I'm fairly dubious that that's a real significant risk factor in the grand scheme of things. And I was a little bit sort of worried when we were trying to pick a title for the books. I was like, are people going to think this is absurd? Are people going to think this is a pseudoscience book? And I was a little bit worried because people are not going to, they're going to think, oh, this is silly.
(15:03):
Obviously, coffee doesn't cause cancer, and yet we still talk about it. And so, I mean, the 1981 paper just to sort of go way, way back, and this was not a nothing paper. This was in the New England Journal of Medicine with some of heavyweights in the field of epidemiology. And I don’t want to discount what these people did. They have more illustrious careers than I will ever have in the field of epidemiology. But this one paper, they made a mistake. What they did was they went around to the local area hospitals, recruited all the patients with pancreatic cancer, recruited controls from the same hospital, and then gave them questionnaires about what they ate, what they drank, how much they smoked, fairly standard stuff. And so, when they were analyzing the data, they saw some associations with tobacco and alcohol, but they saw this really strong association with cancer where the patients who drank a lot of coffee had a near tripling of their risk of pancreatic cancer.
(16:02):
And so, this made headlines, I mean, this was in all the major US newspapers of the time, interviews people were like, well, maybe we should stop drinking coffee. And they pointed to the Amish and other groups that don't drink coffee and have very low rates of cancer. And what was critical in the critical mistake that they made, which is now taught in intro epidemiology classes we know about it, is that if you pick hospital patients as your control, you have a problem. And it's become so common that actually has a name now it's called Berkson's bias. But the problem with picking hospitalized controls is they are not the same as the general population. And in 1981, why were you going to be admitted to a gastrointestinal ward in a major US hospital? It was probably because you have peptic ulcer disease and you tell this to people now, and of course they have no living memory of this.
(16:53):
They've forgotten that we used to do partial gastrectomies to treat peptic ulcer disease, which is a shocking thing to say out loud. And then it gives you also the opportunity to teach people about H. pylori and everything that happened. And then the discovery and the famous case of the researcher drinking a broth of H. pylori to make himself sick and his wife having to drag him to the hospital throwing up every morning. And really how it changed the field of medicine because now we treat peptic ulcer disease with you eradicate H. pylori with two weeks of antibiotics, and we give people a proton pump inhibitor. But back in the day, the people who were in hospital had peptic ulcer disease and other gastrointestinal complaints because of those gastrointestinal issues. They didn't drink a lot of coffee because it would upset their stomach, because coffee can upset people's stomach a little bit.
(17:48):
And so, it wasn't that the pancreatic cancer patients drank more coffee, it's that the control group drank less, and that's why you saw that discrepancy. Whereas if you did the same study in the general population, which was subsequently done, you see no influence of coffee consumption. And so, it’s a prime example of how selection bias can happen. And it’s a seminal paper because it has become a teaching case, and it’s become, for the most part, so well understood that most people are not going to make the same mistake again. And so, the point of highlighting these things is not to make fun of people, which is an unfortunate trend I've started to see online of people being very, very critical and dismissive of the publish research. Like, no, listen, this is how medicine is supposed to work. It's an evolution. We learn from our mistakes and we move on and we have to keep talking about these stories so that people don't make the mistake because choosing the right control group is important.
(18:44):
And so, that's sort of the message of that chapter because each chapter, you're right, it's about a food, but it's also about an epidemiological concept, be it p-hacking or selection bias or information bias or confounding or reverse causation. So I often joke that if you read this book each chapter, you will become very, very smart at dinner parties. You'll be able to figure out terms that no one's heard of before. They're like, Bob, I know you've heard that red wine is good for you, but are you familiar with the concept of reverse causation? And people are going to be very, very impressed with you and keep inviting you to dinner parties the rest of your life afterwards. So there you go. That's another reason to read the book.
Eric Topol (19:20):
Yeah, really. Well, I do want to get into the red wine story too, because it exemplifies this time instead of that New England Journal, this was a 60 Minutes segment in 1991, and then a paper, I guess I went along with that about how red wine is great to reduce heart disease. It still, here it is, what, 30 some years later, 34 years later. And people still believe this. They still think that red wine is preventing heart disease or reducing it. So can you set the record straight on that one?
Christopher Labos (20:06):
Yeah, listen, if you want to drink red wine, you can. I mean, I have nothing against red wine. I mean, I'm drunk right now. No, I'm not.
Eric Topol (20:15):
By the way, that chapter you were drinking wine with your friend, maybe imaginary friend Alex or Alexi. Anyway, yeah. So it was great to hear you are drinking red wine and you're talking to each other about all the cockamamie stuff about it.
Christopher Labos (20:30):
I mean, yeah, the thing, if you're going to do a story, if you're going to do a book chapter about red wine, I think one of the important things is to have two friends drinking at a conference. I mean, let's be honest, that's what usually happens. And so, throughout the evening, they're sitting there polishing off the wine, and then they go on almost a drunken pub crawl. Not quite, it's not quite that bad, but it was almost fun to sort of introduce that element to it of the story. But the red wine thing is fascinating. I get this a lot. I mean, I'm still practicing. I'm still seeing patients and patients come up. I've had, this is not rare, I have had patients literally come to me in clinic and say things like, doctor, my blood pressure is good. I'm checking it at home. I got my blood tests.
(21:12):
My cholesterol is good. I'm eating healthy, I'm exercising. But I find it really hard to drink two glasses of red wine every day. I just don't like red wine all that much. It’s like, wow. No, please sir. Please, for the love of God, stop. It's still there. And what's fascinating is that if you ever go back and watch the 60 Minutes clip by today's standards, it's very weird. You go back and again, it was a product of its time. They were very, very focused on cheese and fat, which of course now we have a much more nuanced understanding about with regard to cholesterol. I mean, a lot of it's genetically mediated and all that, but you go back, it was partially about the red wine being good for you, but it was also there was this really strange subplot, if you will, where they were saying that milk was bad for you and that we should stop getting kids in the US to drink milk. And they thought that a lot of the cardiovascular risk in the US was attributed to the fact that children drink milk routinely, which again, weird by modern standards. Again, I was aware of the 60 Minutes story, but I'd never seen it and I hadn't seen it at the time. And going back to watch it, you're like, wow, that's odd. That's odd.
(22:26):
Again, this idea that, oh, we should be having kids drink wine at a young age. And it was like, really? Do we really want to start having our kids drink alcohol? I'm not so sure about that. It was weird stuff there. But again, it was all part of this French Paradox, which again was a product of its time in the eighties and nineties, this desire to really understand why was heart disease increasing so much in North America and our real failure to really get a handle on it. And with 30 years of hindsight, I think we're in a much better position now to understand why it was the residual effect of all that smoking. It was the residual effect of our more sedentary lifestyle that was starting to happen post World War II. And I think we've largely got a handle on most of those risk factors now.
(23:13):
But the red wine thing persists because I think people like drinking wine and there are not, what's the word I'm looking for, there is not a significant number of people who still believe this. And we had a change in guidelines up here in Canada where the amount of healthy drinking was really reduced down from 2 drinks a day to 1-2 drinks per week, and it caused a bit of a fury. And there was a local cardiologist here who was going on news and saying is like, I don't believe this, red wine is good for you. And I was a little bit taken a breath like, you're a senior cardiologist at a university hospital. You should not be saying stuff like this. And so, they actually had us on to have a debate, and I think they were expecting us to go at each other.
Eric Topol (23:59):
Oh, wow.
Christopher Labos (24:00):
And I was a little bit diplomatic because I've gotten used to this. I know how to bob and weave and avoid the punches. And then at the end, I think it was either me or the reporter asked him, he's like, so what do you tell your patients? And he was like, well, no, I do tell them to drink less because of the AFib risk and the blood pressure and the blood sugar. So I was like, well, you see, you're telling your patients to drink less alcohol for any number of reasons. And irrespective of the U-shaped associations, which is the main statistical argument of the chapter, there's a lot of other reasons to be wary of alcohol. I mean, I think we've proven pretty conclusively the AFib risk. There was that Australian study where if you get people to abstain, you decrease their AFib burden.
(24:42):
So a lot of sugar in alcohol, I mean the blood pressure and diabetes, there's a lot of reasons to not drink this particular sugary beverage and not to mention sort of the cancer associations too that we've seen as well. So it was an interesting thing to argue with him. But the point of the chapter was really to explain why do we see this U-shaped association? And I'll spoil the chapter for people. The statistical concept is called reverse causation. And that happens because it's not that abstaining from alcohol makes you sick. It's that people who are sick end up abstaining from alcohol. So if you have high blood pressure, diabetes, heart disease, AFib, cancer, you've probably been told don't drink alcohol. And so, if you do just a single cross-sectional study where you ask people, how much do you drink? And they say zero, you're probably identifying a high-risk population because most studies, most, not all, but many studies do not make the distinction between former drinkers and never drinkers. And there's a big difference between somebody who used to drink and then quit and somebody who never drank throughout their whole lives.
Eric Topol (25:47):
Yeah, no, it's great. And I think I just want to come back on that. I think Norway and several other countries are now putting on their alcohol products. This may cause cancer, and the American Cancer Society has put a warning on this. So the cancer story is still out there, but you also make among hundreds of important good points in the book about how these food diaries are notoriously inaccurate. And you already touched on that with the survey thing, but it's hard to get, we don't have randomized trials of people drink a lot or don't drink. You can't drink with adherence to that. So it's out there, and of course, people like to drink their wine, but there's a risk that I think has been consistent through many of these studies that is a bit worrisome. I don't know what you would, if you'd say it's conclusive or you'd say it's kind of unsettled.
Christopher Labos (26:49):
I mean, I think it's as settled as it's going to get because I don't see somebody doing a randomized controlled trial on this. And this is the problem. And there has been this trend recently for people to say, well, if there's no randomized controlled trials, I'm not going to believe it. You're like, okay, look, a fair point. And when you're talking about interventions and therapies, then yes, we should absolutely do randomized controlled trials. And I've made that point vociferously when it comes to vitamin D and a lot of the other stuff. The problem is it's going to be very, very hard to do a randomized controlled trial with alcohol. I mean, that was tried. It fell apart and it fell apart for many reasons, not the least of which was the fact that the alcohol industry seemed to be influencing what outcomes people were going to look at.
(27:34):
So that was problematic. I sort of mentioned it right at the tail end of the chapter as well. So if you're not going to have an NIH funded trial to look at in a randomized fashion, does alcohol effect atherosclerosis or cancer outcomes? You're not going to get it. No private industry is going to do it. You're not going to be able to get it done. So given that we have to live in the real world, and I'm always a firm argument in us basing ourselves in reality and living in the real world, we have to make the best decision we can with the evidence that we have available. And I would say, look, I'm pretty sure alcohol is not good for you. I think it is actually detrimental to your cardiovascular health overall. And I think we can say pretty definitively that any potential benefit that people think exists in terms of myocardial infarction, I think that's all a statistical artifact.
(28:26):
I think if you were to analyze it properly, it would all sort of vanish. And I think it largely does. And there's been some really interesting genetic studies using instrumental variables. So what the Mendelian randomization studies that really do suggest that there really is a linear relationship and that the more you drink, the worse it is. And there's no plateau, there's no floor, there's no J shaped curve. It really does appear to be linear. And I've been, I think, fairly convinced because I think the Mendelian randomization studies are as good as we're going to get on this issue.
Eric Topol (29:01):
No, I think it's an important point. And I think there again, the book will hold on so many of these things, but we keep learning all the time. And for example, going back to coffee, there's many studies now that suggest it will reduce type 2 diabetes, it will improve survival, cardiovascular, the mechanism is unknown. Do you think there's, so not only does coffee not cause cancer, but it actually may make you healthier. Any thoughts about that?
Christopher Labos (29:35):
Well, I can state, again, I'm ruining the book. I can state, I think fairly unequivocally coffee does not cause cancer. I think that is pretty clear. Even protective is harder, I think it's possible that a lot of the benefit that's been seen, because it is very observational, could just be the result of residual confounding. I think that is still possible. And again, we have to learn to live with uncertainty in medical research. And when we talk about Bayesian statistics, which is a subject I love, but probably outside of the topic for today, you have to be able to create a framework for what we're certain about and what we're uncertain about. So if you look at the spectrum of risk, clearly the risk ratio for coffee is not above 1. Is it below 1 or is it really straddling the null value? And I'm a little bit uncertain. I think if there is a benefit, it's probably small. I think a lot of it is residual confounding. The one point that would make though, if we're going to talk about coffee being beneficial, we have to talk about coffee. Not a lot of the stuff they are serving at coffee shops now, which are probably more akin to milkshakes than actual coffee.
Eric Topol (30:52):
Yeah, that’s a really good point. Plus, the other thing is the spike of caffeine at much higher levels than you might have with a standard coffee that is typical, these Grande or super Grande, whatever they are. Now another, since we talked about things that people enjoy like coffee and wine, we have to touch on chocolate. The chapter was fun on chocolate, is it a health food and also about the Nobel Laureates. Can you enlighten us on that one?
Christopher Labos (31:26):
This is another, I mean, again, people are going to think that I hate the New England Journal of Medicine. I don't just, that they provided such great teaching material over the years. And to be fair, the study that we're going to talk about the Nobel Laureate chocolate study, I mean if you read it, it really feels like it was meant to be satire and it probably should have belonged in the BMJ Christmas issue. When you read it and you read the disclosure statement where the author is like, disclosure the author admits to loving chocolate, and you're like, okay, that's a weird thing to write in a serious article. So it was probably meant to be a satire. And when you read some of the interviews that Messerli had given afterwards, it does seem that he was trying to just make a point. But it seems to have taken off a life of its own.
(32:10):
What the study was, and it's again, first time I've ever seen a single author on a New England paper, which probably should have been a warning sign for people because generally New England papers don't have single authors on them. But basically, what he did was he was at a conference as the way the story goes, and he was thinking up this idea. So he went on the internet, went onto Wikipedia, and was basically looking up how many Nobel prizes have been won by various countries, looked up the average chocolate consumption on a variety of other websites and basically plotted out a regression line and showed this really linear association between average chocolate consumption per country and number of Nobel prizes per country with the suggested rules that if you eat chocolate, you'll win a Nobel Prize. Except, and notwithstanding all the jokes that came up later, there was another Nobel laureate, and I'm blanking on his name right now, there is in the book. When he was interviewed, he said something like, I believe this is true. Now, milk chocolate might be fine if you want a Nobel Prize in chemistry or medicine, but if you want a Nobel Prize in physics, it really does have to be dark chocolate.
Christopher Labos (33:20):
He said this to the Associated Press, the Associated Press took the quote and put it on the Newswire, and it got reprinted over and over again. And I think he had to publicly apologize to all the people at his university, which to me seemed ridiculous. He was obviously joking, and people took this study very, very seriously. The explanation for why this study is not true, there's actually a word for this, and it's called the ecological bias. And you have to remember something if you're going to look at chocolate and Nobel prizes and look at it in terms of country as the level of exposure, as the unit of exposure. Countries don't eat chocolate and countries don't win Nobel prizes. People eat chocolate and people eat Nobel prizes. And you can't show that the people eating the chocolate in Switzerland are the ones who are winning the Nobel Prizes.
(34:10):
Right. That's the point you can't show, and this is a humorous example, but we've made this type of mistake before when people were talking about saturated fats causing breast cancer. You can look at countries and show that countries that eat a lot of saturated fat have more breast cancer. But that's also because western countries with other basic differences are the countries where you eat a lot of saturated fats and where women develop higher rates of breast cancer. But that doesn't mean that the women who eat the saturated fats are the ones who get breast cancer. And so, the chocolate one is funny because again, it’s exactly what you said. People like eating chocolate, so they want a reason to believe that it is good for you even when it isn’t. And so, they will latch on to the cardiovascular benefits, which have frankly been disproved in the COSMOS study. They will latch on to the neurological, neurocognitive benefits, which have themselves been disproved. And what's fascinating about the whole story is that you would say, oh, we need a large randomized trial. Well, we had that, it was called the COSMOS study. It got published. I mean, maybe it happened during Covid, people didn't notice, but it got published. It was negative. That should have been the end of the story, and it's not, people still believe it.
Eric Topol (35:23):
Well, there's a lot of confirmation bias there, isn't there? Again, the thread through all the chapters is biases, all the different biases that come in play. And this one, knowing Franz Messerli, he's Swiss, so of course he'd want to, yeah, and he eats a lot of chocolate, by the way. And he also comes into play in the chapter you have on salt. It's really interesting. You have chapters on breakfast. Is it really the most important meal? Were there other chapters that you thought about putting in the book that you didn't wind up there, or if you were to write a second edition that you would add?
Christopher Labos (36:01):
I wanted to do a chapter on fish oils. Actually, there's a tweet that you did that I use in my teaching material, which is two days apart, fish oils are good for you, fish oils are bad for you. Because again, that's one of those things where it's just the cycle of all these studies showing no benefit, and yet there's one study that shows a thing and it just keeps coming back. And so yeah, fish oils would've definitely been one. If there is a sequel to this book, and I'm hoping to make a sequel to it.
Eric Topol (36:30):
You should, you should definitely.
Christopher Labos (36:32):
So fish oils is definitely going to be in there because there were originally going to be ten stories. There's only nine in the book. And because it got to the point where the publisher was like, this book is getting a little long, maybe we've got to wrap it up. Maybe it's time to land the plane. And I was like, okay, fair, fair. So we'll cut it at nine. So we had to drop the fish oil one, but that'll be in the sequel if there is a sequel, I want to do, I have a list. It's just off camera actually. I have a little notepad where I've been jotting down ideas. So like fish oils, artificial sweeteners, I'll throw MSG in there, which is a wild story for anybody who's ever dug into the history of MSG. It is a wild and borderline nonsensical story of why we believe that MSG might be bad for us.
(37:14):
Although, I mean, that was, again, very much a product of the eighties and the nineties. So yeah, there's a lot of stuff out there, but fish oil is definitely one that I want to tackle just because it's so relevant. And I still have patients coming in that are going to pharmacy and buying over the counter fish oil supplements. I have to tell them, it's like, look, the evidence on this is pretty clear. It doesn't help. If anything, maybe it slightly increases risk your AFib risk. There's some stuff there. So yeah, again, you could be easily tempted into thinking this is sort of frivolous and funny, but it actually has an implication for people's daily lives because the people out there walking around the street, they believe these things go stop a hundred random people.
Eric Topol (37:59):
Yeah, no, everything in this book is approaching things that are the dogma still, or at least uncertainty, and you get it straight. I mean, you’re an epidemiologist as well as a cardiologist in your training, but you don't use that in a way that is trying to teach people. You're doing it really subtly. And then the other thing just to bring up is that obviously you're debunking all this stuff, and we live in a time where we got all this misinformation and blurred truths. I mean, that's one of the reasons why I pick Ground Truths for this podcast. But it's diminished or certainly challenged the role of physicians and scientists because things are not reliable. They're not constant. They're changing. You touched on that earlier, but can you address, I mean, one of the things besides communicating in a way that makes it easily understandable and fun, which you do so well, it's also addressing trust. How do we promote trust?
Christopher Labos (39:10):
I think you have to, yeah, that's a really challenging question because I think the old model is not going to work anymore. The model of issuing a guideline statement to be like, this is the truth, people will just ignore it because we have issued new guidelines on alcohol consumption. It didn't change behavior. If you want to get people to drink less, you have to address the underlying reason why they do it, and it's this persistent myth. So I think one of the reasons why pseudoscience succeeds as much as it does is because so much of their communication is about storytelling. You can go at people with these large randomized control trials, and yet they will still latch onto an anecdote, right? Because, oh, my friend Bobby had a bad side effect with the Covid vaccine. That's why I'm not getting vaccinated. And so, storytelling is a really, really powerful tool.
(40:05):
And I think the reason why I thought this type of book format could work is it's a story. Because even if you don't remember the details, I was at a lecture last night and I was speaking to a dermatology friend of mine, which sounds like it's an episode from the book, but it's not. But I was speaking to a dermatology friend of mine, and he had read it. He says, Chris, I read it. I really liked it. He goes, I don't remember a lot of the examples you put up. He is a busy guy. He’s got young kids. He read the book, and I was giving a lecture based on this book and exploring all of these concepts. And he was like, I remember when you started talking about the aspirin. I couldn't remember what the example was, but I remembered your point that it's all about subgroups.
(40:47):
And that's the thing is that even if people don't remember the details, even if people don't remember the New England paper about coughing pancreatic cancer, even if they don't remember the COSMOS study about chocolate, even if they don't remember the Nobel chocolate association, they will remember the take home message, which is that you have to be careful. If somebody is torturing the data, they understand why publication bias is a real problem. So that's the point, is that if you tell a story, it sticks in people's minds. So it's almost very Socratic in a way. If you ever read Plato, he's not writing a philosophical treatise in the same way that other philosophers do. It's a conversation between Socrates and other people, and it's a very one-sided conversation because Socrates is telling everybody why they're wrong. So I tried to sort of nuance that and improve upon that framework, but you take away the general gist of it, and that's what we need to give to people.
(41:48):
We need to tell them, we need to give them the tool so that they can say it's like, oh, well wait a second. You're telling me that broccoli is going to prevent pancreatic cancer? Was this a food questionnaire thing? And you're giving people that little bit of background knowledge that they can ask intelligent questions. And I think that's what we have to do going forward, because we have to introduce that little bit of skepticism into their thought process so that they can question what they see on the internet. Because the reality is a lot of what they see on the internet is going to be wrong because it's clickbait, it's headlines, it's all the issues that we have with our modern communication strategies.
Eric Topol (42:31):
Yeah. Well, I think storytelling and what you just described is so darn important. And so, just to wrap up this book, Does Coffee Cause Cancer?: And 8 More Myths about the Food We Eat is much more than what that title says. I hope you're going to do a sequel. You ought to have a Netflix special.
Christopher Labos (42:54):
Please tell somebody that, I don’t how to get a Netflix special, but use your clout and make it happen, and I'll invite you over for dinner.
Eric Topol (43:01):
Sounds good. We'll have red wine together, and drink a lot of decaffeinated coffee. No, this has been fun. You've definitely had an impact. And I hope everybody takes a chance to get through this book because it's like a novel. A novel, which is somehow you've floated in all this really important stuff in medicine, both content, how to interpret data, how to interpret papers, statistics, somehow invisibly in a novel. You've got it all in there. So congratulations on that. It's a new genre medical book like I've never seen before. And so, we'll be following all your next works, and I'm sure your podcast Body of Evidence must be something along these lines as well. So I'll have to take a look and listen to that too.
Christopher Labos (43:56):
Thank you so much. That is very, very, you have no idea how much it means to me to hear you say something like that, that has warmed the cockles of my heart.
Eric Topol (44:07):
Alright, well Chris, thank you.
***********************
Thanks for listening, reading or watching!
The Ground Truths newsletters and podcasts are all free, open-access, without ads.
Please share this post/podcast with your friends and network if you found it informative
Voluntary paid subscriptions all go to support Scripps Research. Many thanks for that—they greatly helped fund our summer internship programs for 2023 and 2024.
Thanks to my producer Jessica Nguyen and Sinjun Balabanoff for audio and video support at Scripps Research.
Note: you can select preferences to receive emails about newsletters, podcasts, or all I don’t want to bother you with an email for content that you’re not interested in.

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
Charlie Swanton: A Master Class on Cancer
14 juin· Ground Truths
The most enthralling conversation I’ve ever had with anyone on cancer.
It’s with Charlie Swanton who is a senior group leader at the Francis Crick Institute, the Royal Society Napier Professor in Cancer and medical oncologist at University College London, co-director of Cancer Research UK.
Video snippet from our conversation. Full videos of all Ground Truths podcasts can be seen on YouTube here. The audios are also available on Apple and Spotify.
Transcript with audio links and many external links
Eric Topol (00:07):
Well, hello, this is Eric Topol with Ground Truths, and I am really fortunate today to connect us with Charlie Swanton, who is if not the most prolific researcher in the space of oncology and medicine, and he's right up there. Charlie is a physician scientist who is an oncologist at Francis Crick and he heads up the lung cancer area there. So Charlie, welcome.
Charles Swanton (00:40):
Thank you, Eric. Nice to meet you.
Learning from a Failure
Eric Topol (00:43):
Well, it really is a treat because I've been reading your papers and they're diverse. They're not just on cancer. Could be connecting things like air pollution, it could be Covid, it could be AI, all sorts of things. And it's really quite extraordinary. So I thought I'd start out with a really interesting short paper you wrote towards the end of last year to give a sense about you. It was called Turning a failing PhD around. And that's good because it's kind of historical anchoring. Before we get into some of your latest contributions, maybe can you tell us about that story about what you went through with your PhD?
Charles Swanton (01:26):
Yeah, well thank you, Eric. I got into research quite early. I did what you in the US would call the MD PhD program. So in my twenties I started a PhD in a molecular biology lab at what was then called the Imperial Cancer Research Fund, which was the sort of the mecca for DNA tumor viruses, if you like. It was really the place to go if you wanted to study how DNA tumor viruses worked, and many of the components of the cell cycle were discovered there in the 80s and 90s. Of course, Paul Nurse was the director of the institute at the time who discovered cdc2, the archetypal regulator of the cell cycle that led to his Nobel Prize. So it was a very exciting place to work, but my PhD wasn't going terribly well. And sort of 18, 19 months into my PhD, I was summoned for my midterm reports and it was not materializing rapidly enough.
(02:25):
And I sat down with my graduate student supervisors who were very kind, very generous, but basically said, Charlie, this isn't going well, is it? You've got two choices. You can either go back to medical school or change PhD projects. What do you want to do? And I said, well, I can't go back to medical school because I’m now two years behind. So instead I think what I'll do is I'll change PhD projects. And they asked me what I'd like to do. And back then we didn't know how p21, the CDK inhibitor bound to cyclin D, and I said, that's what I want to understand how these proteins interact biochemically. And they said, how are you going to do that? And I said, I'm not too sure, but maybe we'll try yeast two-hybrid screen and a mutagenesis screen. And that didn't work either. And in the end, something remarkable happened.
(03:14):
My PhD boss, Nic Jones, who's a great guy, still is, retired though now, but a phenomenal scientist. He put me in touch with a colleague who actually works next door to me now at the Francis Crick Institute called Neil McDonald, a structural biologist. And they had just solved, well, the community had just solved the structure. Pavletich just solved the structure of cyclin A CDK2. And so, Neil could show me this beautiful image of the crystal structure in 3D of cyclin A, and we could mirror cyclin D onto it and find the surface residue. So I spent the whole of my summer holiday mutating every surface exposed acid on cyclin D to an alanine until I found one that failed to interact with p21, but could still bind the CDK. And that little breakthrough, very little breakthrough led to this discovery that I had where the viral cyclins encoded by Kaposi sarcoma herpes virus, very similar to cyclin D, except in this one region that I had found interactive with a CDK inhibitor protein p21.
(04:17):
And so, I asked my boss, what do you think about the possibility this cyclin could have evolved from cyclin D but now mutated its surface residues in a specific area so that it can't be inhibited by any of the control proteins in the mammalian cell cycle? He said, it's a great idea, Charlie, give it a shot. And it worked. And then six months later, we got a Nature paper. And that for me was like, I cannot tell you how exciting, not the Nature paper so much as the discovery that you were the first person in the world to ever see this beautiful aspect of evolutionary biology at play and how this cyclin had adapted to just drive the cell cycle without being inhibited. For me, just, I mean, it was like a dream come true, and I never experienced anything like it before, and I guess it's sizes the equivalent to me of a class A drug. You get such a buzz out of it and over the years you sort of long for that to happen again. And occasionally it does, and it's just a wonderful profession.
Eric Topol (05:20):
Well, I thought that it was such a great story because here you were about to fail. I mean literally fail, and you really were able to turn it around and it should give hope to everybody working in science out there that they could just be right around the corner from a significant discovery.
Charles Swanton (05:36):
I think what doesn't break you makes you stronger. You just got to plow on if you love it enough, you'll find a way forward eventually, I hope.
Tracing the Evolution of Cancer (TRACERx)
Eric Topol (05:44):
Yeah, no question about that. Now, some of your recent contributions, I mean, it's just amazing to me. I just try to keep up with the literature just keeping up with you.
Charles Swanton (05:58):
Eric, it's sweet of you. The first thing to say is it's not just me. This is a big community of lung cancer researchers we have thanks to Cancer Research UK funded around TRACERx and the lung cancer center. Every one of my papers has three corresponding authors, multiple co-first authors that all contribute in this multidisciplinary team to the sort of series of small incremental discoveries. And it's absolutely not just me. I've got an amazing team of scientists who I work with and learn from, so it's sweet to give me the credit.
Eric Topol (06:30):
I think what you're saying is really important. It is a team, but I think what I see through it all is that you're an inspiration to the team. You pull people together from all over the world on these projects and it's pretty extraordinary, so that's what I would say.
Charles Swanton (06:49):
The lung community, Eric, the lung cancer community is just unbelievably conducive to collaboration and advancing understanding of the disease together. It's just such a privilege to be working in this field. I know that sounds terribly corny, but it is true. I don't think I recall a single email to anybody where I've asked if we can collaborate where they've said, no, everybody wants to help. Everybody wants to work together on this challenge. It's just such an amazing field to be working in.
Eric Topol (07:19):
Yeah. Well I was going to ask you about that. And of course you could have restricted your efforts or focused on different cancers. What made you land in lung cancer? Not that that's only part of what you're working on, but that being the main thing, what drew you to that area?
Charles Swanton (07:39):
So I think the answer to your question is back in 2008 when I was looking for a niche, back then it was lung cancer was just on the brink of becoming an exciting place to work, but back then nobody wanted to work in that field. So there was a chair position in thoracic oncology and precision medicine open at University College London Hospital that had been open, as I understand it for two years. And I don't think anybody had applied. So I applied and because I was the only one, I got it and the rest is history.
(08:16):
And of course that was right at the time when the IPASS draft from Tony Mok was published and was just a bit after when the poster child of EGFR TKIs and EGFR mutant lung cancer had finally proven that if you segregate that population of patients with EGFR activating mutation, they do incredibly well on an EGFR inhibitor. And that was sort of the solid tumor poster child along with Herceptin of precision medicine, I think. And you saw the data at ASCO this week of Lorlatinib in re-arranged lung cancer. Patients are living way beyond five years now, and people are actually talking about this disease being more like CML. I mean, it's extraordinary the progress that's been made in the last two decades in my short career.
Eric Topol (09:02):
Actually, I do want to have you put that in perspective because it's really important what you just mentioned. I was going to ask you about this ASCO study with the AKT subgroup. So the cancer landscape of the lung has changed so much from what used to be a disease of cigarette smoking to now one of, I guess adenocarcinoma, non-small cell carcinoma, not related to cigarettes. We're going to talk about air pollution in a minute. This group that had, as you say, 60 month, five year plus survival versus what the standard therapy was a year plus is so extraordinary. But is that just a small subgroup within small cell lung cancer?
Charles Swanton (09:48):
Yes, it is, unfortunately. It’s just a small subgroup. In our practice, probably less than 1% of all presentations often in never smokers, often in female, never smokers. So it is still in the UK at least a minority subset of adenocarcinomas, but it's still, as you rightly say, a minority of patients that we can make a big difference to with a drug that's pretty well tolerated, crosses the blood-brain barrier and prevents central nervous system relapse and progression. It really is an extraordinary breakthrough, I think. But that said, we're also seeing advances in smoking associated lung cancer with a high mutational burden with checkpoint inhibitor therapy, particularly in the neoadjuvant setting now prior to surgery. That's really, really impressive indeed. And adjuvant checkpoint inhibitor therapies as well as in the metastatic setting are absolutely improving survival times and outcomes now in a way that we couldn't have dreamt of 15 years ago. We've got much more than just platinum-based chemo is basically the bottom line now.
Revving Up Immunotherapy
Eric Topol (10:56):
Right, right. Well that actually gets a natural question about immunotherapy also is one of the moving parts actually just amazing to me how that's really, it's almost like we're just scratching the surface of immunotherapy now with checkpoint inhibitors because the more we get the immune system revved up, the more we're seeing results, whether it's with vaccines or CAR-T, I mean it seems like we're just at the early stages of getting the immune system where it needs to be to tackle the cancer. What's your thought about that?
Charles Swanton (11:32):
I think you're absolutely right. We are, we're at the beginning of a very long journey thanks to Jim Allison and Honjo. We've got CTLA4 and PD-1/PDL-1 axis to target that's made a dramatic difference across multiple solid tumor types including melanoma and lung cancer. But undoubtedly, there are other targets we've seen LAG-3 and melanoma and then we're seeing new ways, as you rightly put it to mobilize the immune system to target cancers. And that can be done through vaccine based approaches where you stimulate the immune system against the patient's specific mutations in their cancer or adoptive T-cell therapies where you take the T-cells out of the tumor, you prime them against the mutations found in the tumor, you expand them and then give them back to the patient. And colleagues in the US, Steve Rosenberg and John Haanen in the Netherlands have done a remarkable job there in the context of melanoma, we're not a million miles away from European approvals and academic initiated manufacturing of T-cells for patients in national health systems like in the Netherlands.
(12:50):
John Haanen's work is remarkable in that regard. And then there are really spectacular ways of altering T-cells to be able to either migrate to the tumor or to target specific tumor antigens. You mentioned CAR-T cell therapies in the context of acute leukemia, really extraordinary developments there. And myeloma and diffuse large B-cell lymphoma as well as even in solid tumors are showing efficacy. And I really am very excited about the future of what we call biological therapies, be it vaccines, an antibody drug conjugates and T-cell therapies. I think cancer is a constantly adapting evolutionary force to be reckoned with what better system to combat it than our evolving immune system. It strikes me as being a future solution to many of these refractory cancers we still find difficult to treat.
Eric Topol (13:48):
Yeah, your point is an interesting parallel how the SARS-CoV-2 virus is constantly mutating and becoming more evasive as is the tumor in a person and the fact that we can try to amp up the immune system with these various means that you just were reviewing. You mentioned the other category that's very hot right now, which is the antibody drug conjugates. Could you explain a bit about how they work and why you think this is an important part of the future for cancer?
Antibody-Drug Conjugates
Charles Swanton (14:26):
That's a great question. So one of the challenges with chemotherapy, as you know, is the normal tissue toxicity. So for instance, neutropenia, hair loss, bowel dysfunction, diarrhea, epithelial damage, essentially as you know, cytotoxics affect rapidly dividing tissues, so bone marrow, epithelial tissues. And because until relatively recently we had no way of targeting chemotherapy patients experienced side effects associated with them. So over the last decade or so, pioneers in this field have brought together this idea of biological therapies linked with chemotherapy through a biological linker. And so one poster chart of that would be the drug T-DXd, which is essentially Herceptin linked to a chemotherapy drug. And this is just the most extraordinary drug that obviously binds the HER2 receptor, but brings the chemotherapy and proximity of the tumor. The idea being the more drug you can get into the tumor and the less you're releasing into normal tissue, the more on tumor cytotoxicity you'll have and the less off tumor on target normal tissue side effects you'll have. And to a large extent, that's being shown to be the case. That doesn't mean they're completely toxicity free, they're not. And one of the side effects associated with these drugs is pneumonitis.
(16:03):
But that said, the efficacy is simply extraordinary. And for example, we're having to rewrite the rule books if you like, I think. I mean I'm not a breast cancer physician, I used to be a long time ago, but back in the past in the early 2000s, there was HER2 positive breast cancer and that's it. Now they're talking about HER2 low, HER2 ultra-low, all of which seem to in their own way be sensitive to T-DXd, albeit to a lower extent than HER2 positive disease. But the point is that there doesn't seem to be HER2 completely zero tumor group in breast cancer. And even the HER2-0 seem to benefit from T-DXd to an extent. And the question is why? And I think what people are thinking now is it's a combination of very low cell service expression of HER2 that's undetectable by conventional methods like immunohistochemistry, but also something exquisitely specific about the way in which HER2 is mobilized on the membrane and taken back into the cell. That seems to be specific to the breast cancer cell but not normal tissue. So in other words, the antibody drug conjugate binds the tumor cell, it's thought the whole receptor's internalized into the endosome, and that's where the toxicity then happens. And it's something to do with the endosomal trafficking with the low level expression and internalization of the receptor. That may well be the reason why these HER2 low tumors are so sensitive to this beautiful technology.
Eric Topol (17:38):
Now I mean it is an amazing technology in all these years where we just were basically indiscriminately trying to kill cells and hoping that the cancer would succumb. And now you're finding whether you want to call it a carry or vector or Trojan horse, whatever you want to call it, but do you see that analogy of the HER2 receptor that's going to be seen across the board in other cancers?
Charles Swanton (18:02):
That's the big question, Eric. I think, and have we just lucked out with T-DXd, will we find other T-DXd like ADCs targeting other proteins? I mean there are a lot of ADCs being developed against a lot of different cell surface proteins, and I think the jury's still out. I'm confident we will, but we have to bear in mind that biology is a fickle friend and there may be something here related to the internalization of the receptor in breast cancer that makes this disease so exquisitely sensitive. So I think we just don't know yet. I'm reasonably confident that we will find other targets that are as profoundly sensitive as HER2 positive breast cancer, but time will tell.
Cancer, A Systemic Disease
Eric Topol (18:49):
Right. Now along these lines, well the recent paper that you had in Cell, called embracing cancer complexity, which we've talking about a bit, in fact it's kind of those two words go together awfully well, but hallmarks of systemic disease, this was a masterful review, as you say with the team that you led. But can you tell us about what's your main perspective about this systemic disease? I mean obviously there's been the cancer is like cardiovascular and cancers like this or that, but here you really brought it together with systemic illness. What can you say about that?
Charles Swanton (19:42):
Well, thanks for the question first of all, Eric. So a lot of this comes from some of my medical experience of treating cancer and thinking to myself over the years, molecular biology has had a major footprint on advances in treating the disease undoubtedly. But there are still aspects of medicine where molecular biology has had very little impact, and often that is in areas of suffering in patients with advanced disease and cancer related to things like cancer cachexia, thrombophilia. What is the reason why patients die blood clots? What is the reason patients die of cancer at all? Even a simple question like that, we don't always know the answer to, on death certificates, we write metastatic disease as a cause of cancer death, but we have patients who die with often limited disease burden and no obvious proximal cause of death sometimes. And that's very perplexing, and we need to understand that process better.
(20:41):
And we need to understand aspects like cancer pain, for example, circadian rhythms affect biological sensitivity of cancer cells to drugs and what have you. Thinking about cancer rather than just sort of a single group of chaotically proliferating cells to a vision of cancer interacting both locally within a microenvironment but more distantly across organs and how organs communicate with the cancer through neuronal networks, for example, I think is going to be the next big challenge by setting the field over the next decade or two. And I think then thinking about more broadly what I mean by embracing complexity, I think some of that relates to the limitations of the model systems we use, trying to understand inter-organ crosstalk, some of the things you cover in your beautiful Twitter reviews. (←Ground Truths link)
I remember recently you highlighted four publications that looked at central nervous system, immune cell crosstalk or central nervous system microbiome crosstalk. It's this sort of long range interaction between organs, between the central nervous system and the immune system and the cancer that I'm hugely interested in because I really think there are vital clues there that will unlock new targets that will enable us to control cancers more effectively if we just understood these complex networks better and had more sophisticated animal model systems to be able to interpret these interactions.
Eric Topol (22:11):
No, it's so important what you're bringing out, the mysteries that still we have to deal with cancer, why patients have all these issues or dying without really knowing what's happened no less, as you say, these new connects that are being discovered at a remarkable pace, as you mentioned, that ground truths. And also, for example, when I spoke with Michelle Monje, she's amazing on the cancer, where hijacking the brain cells and just pretty extraordinary things. Now that gets me to another line of work of yours. I mean there are many, but the issue of evolution of the tumor, and if you could put that in context, a hot area that's helping us elucidate these mechanisms is known as spatial omics or spatial biology. This whole idea of being able to get the spatial temporal progression through single cell sequencing and single cell nuclei, all the single cell omics. So if you could kind of take us through what have we learned with this technique and spatial omics that now has changed or illuminated our understanding of how cancer evolves?
Charles Swanton (23:37):
Yeah, great question. Well, I mean I think it helps us sort of rewind a bit and think about evolution in general. Genetic selection brought about by diverse environments and environmental pressures that force evolution, genetic evolution, and speciation down certain evolutionary roots. And I think one can think about cancers in a similar way. They start from a single cell and we can trace the evolutionary paths of cancers by single cell analysis as well as bulk sequencing of spatially separated tumor regions to be able to reconstruct their subclones. And that's taught us to some extent, what are the early events in tumor evolution? What are the biological mechanisms driving branched evolution? How does genome instability begin in tumors? And we found through TRACERx work, whole genome doubling is a major route through to driving chromosome instability along with mutagenic enzymes like APOBEC that drive both mutations and chromosomal instability.
(24:44):
And then that leads to a sort of adaptive radiation in a sense, not dissimilar to I guess the Cambrian explosion of evolutionary opportunity upon which natural selection can act. And that's when you start to see the hallmarks of immune evasion like loss of HLA, the immune recognition molecules that bind the neoantigens or even loss of the neoantigens altogether or mutation of beta 2 microglobulin that allow the tumor cells to now evolve below the radar, so to speak. But you allude to the sort of spatial technologies that allow us to start to interpret the microenvironments as well. And that then tells us what the evolutionary pressures are upon the tumor. And we're learning from those spatial technologies that these environments are incredibly diverse, actually interestingly seem to be converging on one important aspect I'd like to talk to you a little bit more about, which is the myeloid axis, which is these neutrophils, macrophages, et cetera, that seem to be associated with poor outcome and that will perhaps talk about pollution in a minute.
(25:51):
But I think they're creating a sort of chronic inflammatory response that allows these early nascent tumor cells to start to initiate into frankly tumor invasive cells and start to grow. And so, what we're seeing from these spatial technologies in lung cancer is that T-cells, predatory T-cells, force tumors to lose their HLA molecules and what have you to evade the immune system. But for reasons we don't understand, high neutrophil infiltration seems to be associated with poor outcome, poor metastasis free survival. And actually, those same neutrophils we've recently found actually even tracked to the metastasis sites of metastasis. So it's almost like this sort of symbiosis between the myeloid cells and the tumor cells in their biology and growth and progression of the tumor cells.
Eric Topol (26:46):
Yeah, I mean this white cell story, this seems to be getting legs and is relatively new, was this cracked because of the ability to do this type of work to in the past everything was, oh, it's cancer's heterogeneous and now we're getting pinpoint definition of what's going on.
Charles Swanton (27:04):
I think it's certainly contributed, but it's like everything in science, Eric, when you look back, there's evidence in the literature for pretty much everything we've ever discovered. You just need to put the pieces together. And I mean one example would be the neutrophil lymphocyte ratio in the blood as a hallmark of outcome in cancers and to checkpoint inhibitor blockade, maybe this begins to explain it, high neutrophils, immune suppressive environment, high neutrophils, high macrophages, high immune suppression, less benefit from checkpoint inhibitor therapy, whereas you want lymphocyte. So I think there are biomedical medical insights that help inform the biology we do in the lab that have been known for decades or more. And certainly the myeloid M2 axis in macrophages and what have you was known about way before these spatial technologies really came to fruition, I think.
The Impact of Air Pollution
Eric Topol (28:01):
Yeah. Well you touched on this about air pollution and that's another dimension of the work that you and your team have done. As you well know, there was a recent global burden of disease paper in the Lancet, which has now said that air pollution with particulate matter 2.5 less is the leading cause of the burden of disease in the world now.
Charles Swanton (28:32):
What did you think of that, Eric?
Eric Topol (28:34):
I mean, I was blown away. Totally blown away. And this is an era you've really worked on. So can you put it in perspective?
Charles Swanton (28:42):
Yeah. So we got into this because patients of mine, and many of my colleagues would ask the same question, I've never smoked doctor, I'm healthy. I'm in my mid 50s though they're often female and I've got lung cancer. Why is that doctor? I've had a good diet, I exercise, et cetera. And we didn't really have a very good answer for that, and I don't want to pretend for a minute we solved the whole problem. I think hopefully we've contributed to a little bit of understanding of why this may happen. But that aside, we knew that there were risk factors associated with lung cancer that included air pollution, radon exposure, of course, germline genetics, we mustn't forget very important germline variation. And I think there is evidence that all of them are associated with lung cancer risk in different ways. But we wanted to look at air pollution, particularly because there was an awful lot of evidence, several meta-analysis of over half a million individuals showing very convincingly with highly significant results that increasing PM 2.5 micron particulate levels were associated with increased risk of lung cancer.
(29:59):
To put that into perspective, where you are on the west coast of the US, it's relatively unpolluted. You would be talking about maybe five micrograms per meter cubed of PM2.5 in a place like San Diego or Western California, assuming there aren't any forest fires of course. And we estimate that that would translate to about, we think it's about one extra case of never smoking lung cancer per hundred thousand of the population per year per one microgram per meter cube rise in the pollution levels. So if you go to Beijing for example, on a bad day, the air pollution levels could be upwards of a hundred micrograms per meter cubed because there are so many coal fired power stations in China partly. And there I think the risk is considerably higher. And that's certainly what we've seen in the meta-analyses in our limited and relatively crude epidemiological analyses to be the case.
(30:59):
So I think the association was pretty certain, we were very confident from people's prior publications this was important. But of course, association is not causation. So we took a number of animal models and showed that you could promote lung cancer formation in four different oncogene driven lung cancer models. And then the question is how, does air pollution stimulate mutations, which is what I initially thought it would do or something else. It turns out we don't see a significant increase in exogenous like C to A carcinogenic mutations. So that made us put our thinking caps on. And I said to you earlier, often all these discoveries have been made before. Well, Berenblum in 1947, first postulated that actually tumors are initiated through a two-step process, which we now know involves a sort of pre initiated cell with a mutation in that in itself is not sufficient to cause cancer.
(31:58):
But on top of that you need an inflammatory stimulus. So the question was then, well, okay, is inflammation working here? And we found that there was an interleukin-1 beta axis. And what happens is that the macrophages come into the lung on pollution exposure, engulf phagocytose the air pollutants, and we think what's happening is the air pollutants are puncturing membranes in the lung. That's what we think is happening. And interleukin-1 beta preformed IL-1 beta is being released into the extracellular matrix and then stimulating pre-initiated cells stem cells like the AT2 cells with an activating EGFR mutation to form a tumor. But the EGFR mutation alone is not sufficient to form tumors. It's only when you have the interleukin-1 beta and the activated mutation that a tumor can start.
(32:49):
And we found that if we sequence normal lung tissue in a healthy adult 60-year-old adult, we will find about half of biopsies will have an activating KRAS mutation in normal tissue, and about 15% will have an activating mutation in EGFR in histologically normal tissue with nerve and of cancer. In fact, my friend and colleague who's a co-author on the paper, James DeGregori, who you should speak to in Colorado, fascinating evolutionary cancer biologists estimates that in a healthy 60-year-old, there are a hundred billion cells in your body that harbor an oncogenic mutation. So that tells you that at the cellular level, cancer is an incredibly rare event and almost never happens. I mean, our lifetime risk of cancer is perhaps one in two. You covered that beautiful pancreas paper recently where they estimated that there may be 80 to 100 KRAS mutations in a normal adult pancreas, and yet our lifetime risk of pancreas cancer is one in 70. So this tells you that oncogenic mutations are rarely sufficient to drive cancer, so something else must be happening. And in the context of air pollution associated lung cancer, we think that's inflammation driven by these white cells, these myeloid cells, the macrophages.
Cancer Biomarkers
Eric Topol (34:06):
No, it makes a lot of sense. And this, you mentioned the pancreas paper and also what's going in the lung, and it seems like we have this burden of all you need is a tipping point and air pollution seems to qualify, and you seem to be really in the process of icing the mechanism. And like I would've thought it was just mutagenic and it's not so simple, right? But that gets me to this is such an important aspect of cancer, the fact that we harbor these kind of preconditions. And would you think that cancer takes decades to actually manifest most cancers, or do we really have an opportunity here to be able to track whether it's through blood or other biomarkers? Another area you've worked on a lot whereby let's say you could define people at risk for polygenic risk scores or various cancers or genome sequencing for predisposition genes, whatever, and you could monitor in the future over the course of those high-risk people, whether they were starting to manifest microscopic malignancy. Do you have any thoughts about how long it takes for the average person to actually manifest a typical cancer?
Charles Swanton (35:28):
That's a cracking question, and the answer is we've got some clues in various cancers. Peter Campbell would be a good person to speak to. He estimates that some of the earliest steps in renal cancer can occur in adolescence. We've had patients who gave up smoking 30 or so years ago where we can still see the clonal smoking mutations in the trunk of the tumor's evolutionary tree. So the initial footprints of the cancer are made 30 years before the cancer presents. That driver mutation itself may also be a KRAS mutation in a smoking cigarette context, G12C mutation. And those mutations can precede the diagnosis of the disease by decades. So the earliest steps in cancer evolution can occur, we think can precede diagnoses by a long time. So to your point, your question which is, is there an opportunity to intervene? I'm hugely optimistic about this actually, this idea of molecular cancer prevention.
An Anti-Inflammatory Drug Reduces Fatal Cancer and Lung Cancer
(36:41):
How can we use data coming out of various studies in the pancreas, mesothelioma, lung, et cetera to understand the inflammatory responses? I don't think we can do very much about the mutations. The mutations unfortunately are a natural consequence of aging. You and I just sitting here talking for an hour will have accumulated multiple mutations in our bodies over that period, I'm afraid and there's no escaping it. And right now there's not much we can do to eradicate those mutant clones. So if we take that as almost an intractable problem, measuring them is hard enough, eradicating them is even harder. And then we go back to Berenblum in 1947 who said, you need an inflammatory stimulus. Well, could we do something about the inflammation and dampen down the inflammation? And of course, this is why we got so excited about IL-1 beta because of the CANTOS trial, which you may remember in 2017 from Ridker and colleagues showed that anti IL-1 beta used as a mechanism of preventing cardiovascular events was associated with a really impressive dose dependent reduction in new lung cancer primaries.
(37:49):
Really a beautiful example of cancer prevention in action. And that data weren't just a coincidence. The FDA mandated Novartis to collect the solid tumor data and the P-values are 0.001. I mean it's very highly significant dose dependent reduction in lung cancer incidents associated with anti IL-1 beta. So I think that’s really the first clue in my mind that something can be done about this problem. And actually they had five years of follow-up, Eric. So that’s something about that intervening period where you can treat and then over time see a reduction in new lung cancers forming. So I definitely think there’s a window of opportunity here.
Eric Topol (38:31):
Well, what you’re bringing up is fascinating here because this trial, which was a cardiology trial to try to reduce heart attacks, finds a reduction in cancer, and it’s been lost. It’s been buried. I mean, no one’s using this therapy to prevent cancer between ratcheting up the immune system or decreasing inflammation. We have opportunities that we’re not even attempting. Are there any trials that are trying to do this sort of thing?
Charles Swanton (39:02):
So this is the fundamental problem. Nobody wants to invest in prevention because essentially you are dealing with well individuals. It’s like the vaccine challenge all over again. And the problem is you never know who you are benefiting. There’s no economic model for it. So pharma just won’t touch prevention with a barge pole right now. And that’s the problem. There’s no economic model for it. And yet the community, all my academic colleagues are crying out saying, this has got to be possible. This has got to be possible. So CRUK are putting together a group of like-minded individuals to see if we can do something here and we're gradually making progress, but it is tough.
Eric Topol (39:43):
And it's interesting that you bring that up because for GRAIL, one of the multicenter cancer early detection companies, they raised billions of dollars. And in fact, their largest trial is ongoing in the UK, but they haven't really focused on high-risk people. They just took anybody over age 50 or that sort of thing. But that's the only foray to try to reboot how we or make an early microscopic diagnosis of cancer and track people differently. And there's an opportunity there. You've written quite a bit on you and colleagues of the blood markers being able to find a cancer where well before, in fact, I was going to ask you about that is, do you think there's people that are not just having all these mutations every minute, every hour, but that are starting to have the early seeds of cancer, but because their immune system then subsequently kicks in that they basically kind of quash it for that period of time?
Charles Swanton (40:47):
Yeah, I do think that, I mean, the very fact that we see these sort of footprints in the tumor genome of immune evasion tells you that the immune system's having a very profound predatory effect on evolving tumors. So I do think it's very likely that there are tumors occurring that are suppressed by the immune system. There is a clear signature, a signal of negative selection in tumors where clones have been purified during their evolution by the immune system. So I think there's pretty strong evidence for that now. Obviously, it's very difficult to prove something existed when it doesn't now exist, but there absolutely is evidence for that. I think it raises the interesting question of immune system recognizes mutations and our bodies are replete with mutations as we were just discussing. Why is it that we're not just a sort of epithelial lining of autoimmunity with T-cells and immune cells everywhere? And I think what the clever thing about the immune system is it's evolved to target antigens only when they get above a certain burden. Otherwise, I think our epithelial lining, our skin, our guts, all of our tissues will be just full of T-cells eating away our normal clones.
(42:09):
These have to get to a certain size for antigen to be presented at a certain level for the immune system to recognize it. And it's only then that you get the immune predation occurring.
Forever Chemicals and Microplastics
Eric Topol (42:20):
Yeah, well, I mean this is opportunities galore here. I also wanted to extend the air pollution story a bit. Obviously, we talked about particulate matter and there's ozone and nitric NO2, and there's all sorts of other air pollutants, but then there's also in the air and water these forever chemicals PFAS for abbreviation, and they seem to be incriminated like air pollution. Can you comment about that?
Charles Swanton (42:55):
Well, I can comment only insofar as to say I'm worried about the situation. Indeed, I'm worried about microplastics actually, and you actually cover that story as well in the New England Journal, the association of microplastics with plaque rupture and atheroma. And indeed, just as in parenthesis, I wanted to just quickly say we currently think the same mechanisms that are driving lung cancer are probably responsible for atheroma and possibly even neurodegenerative disease. And essentially it all comes down to the macrophages and the microglia becoming clogged up with these pollutants or environmental particulars and releasing chronic inflammatory mediators that ultimately lead to disease. And IL-1 beta being one of those in atheroma and probably IL-6 and TNF in neurodegenerative disease and what have you. But I think this issue that you rightly bring up of what is in our environment and how does it cause pathology is really something that epidemiologists have spent a lot of time focusing on.
(43:56):
But actually in terms of trying to move from association to causation, we've been, I would argue a little bit slow biologically in trying to understand these issues. And I think that is a concern. I mean, to give you an example, Allan Balmain, who works at UCSF quite close to you, published a paper in 2020 showing that 17 out of 20 environmental carcinogens IARC carcinogens class one carcinogens cause tumors in rodent models without driving mutations. So if you take that to a logical conclusion, in my mind, what worries me is that many of the sort of carcinogen assays are based on driving mutagenesis genome instability. But if many carcinogen aren't driving DNA mutagenesis but are still driving cancer, how are they doing it? And do we actually have the right assays to interpret safety of new chemical matter that's being introduced into our environment, these long-lived particles that we're breathing in plastics, pollutants, you name it, until we have the right biological assays, deeming something to be safe I think is tricky.
Eric Topol (45:11):
Absolutely. And I share your concerns on the nanoplastic microplastic story, as you well know, not only have they been seen in arteries that are inflamed and in blood clots and in various tissues, have they been seen so far or even looked for within tumor tissue?
Charles Swanton (45:33):
Good question. I'm not sure they have. I need to check. What I can tell you is we've been doing some experiments in the lab with fluorescent microplastics, 2.5 micron microplastics given inhaled microplastics. We find them in every mouse organ a week after. And these pollutants even get through into the brain through the olfactory bulb we think.
Charles Swanton (45:57):
Permeate every tissue, Eric.
Eric Topol (45:59):
Yeah, no, this is scary because here we are, we have these potentially ingenious ways to prevent cancer in the future, but we're chasing our tails by not doing anything to deal with our environment.
Charles Swanton (46:11):
I think that's right. I totally agree. Yeah.
Eric Topol (46:15):
So I mean, I can talk to you for the rest of the day, but I do want to end up with a topic that we have mutual interest in, which is AI. And also along with that, when you mentioned about aging, I'd like to get your views on these two, how do you see AI fitting into the future of cancer? And then the more general topic is, can we actually at some point modulate the biologic aging process with or without help with from AI? So those are two very dense questions, but maybe you can take us through them.
Charles Swanton (46:57):
How long have we got?
Eric Topol (46:59):
Just however long you have.
A.I. and Cancer
Charles Swanton (47:02):
AI and cancer. Well, AI and medicine actually in general, whether it's biomedical research or medical care, has just infinite potential. And I'm very, very excited about it. I think what excites me about AI is it's almost the infinite possibilities to work across scale. Some of the challenges we raised in the Cell review that you mentioned, tackling, embracing complexity are perfectly suited for an AI problem. Nonlinear data working, for instance in our fields with CT imaging, MRI imaging, clinical outcome data, blood parameters, genomics, transcriptomes and proteomes and trying to relate this all into something that's understandable that relates to risk of disease or potential identification of a new drug target, for example. There are numerous publications that you and others have covered that allude to the incredible possibilities there that are leading to, for instance, the new identification of drug targets. I mean, Eli Van Allen's published some beautiful work here and in the context of prostate cancer with MDM4 and FGF receptor molecules being intimately related to disease biology.
(48:18):
But then it's not just that, not just drug target identification, it's also going all the way through to the clinic through drug discovery. It's how you get these small molecules to interact with oncogenic proteins and to inhibit them. And there are some really spectacular developments going on in, for instance, time resolved cryo-electron microscopy, where in combination with modeling and quantum computing and what have you, you can start to find pockets emerging in mutant proteins, but not the wild type ones that are druggable. And then you can use sort of synthetic AI driven libraries to find small molecules that will be predicted to bind these transiently emerging pockets. So it's almost like AI is primed to help at every stage in scientific investigation from the bench all the way through to the bedside. And there are examples all the way through there in the literature that you and others have covered in the last few years. So I could not be more excited about that.
Eric Topol (49:29):
I couldn't agree with you more. And I think when we get to multimodal AI at the individual level across all their risks for conditions in their future, I hope someday will fulfill that fantasy of primary prevention. And that is getting me to this point that I touched on because I do think they interact to some degree AI and then will we ever be able to have an impact on aging? Most people conflate this because what we've been talking about throughout the hour has been age-related diseases, that is cancer, for example, and cardiovascular and neurodegenerative, which is different than changing aging per se, body wide aging. Do you think we'll ever changed body wide aging?
Charles Swanton (50:18):
Wow, what a question. Well, if you'd asked me 10 years ago, 15 years ago, do you think we'll ever cure melanoma in my lifetime, I'd have said definitely not. And now look where we are. Half of patients with melanoma, advanced melanoma, even with brain metastasis curd with combination checkpoint therapy. So I never say never in biology anymore. It always comes back to bite you and prove you wrong. So I think it's perfectly possible.
Charles Swanton (50:49):
We have ways to slow down the aging process. I guess the question is what will be the consequences of that?
Eric Topol (50:55):
That's what I was going to ask you, because all these things like epigenetic reprogramming and senolytic drugs, and they seem to at least pose some risk for cancer.
Charles Swanton (51:09):
That's the problem. This is an evolutionary phenomenon. It's a sort of biological response to the onslaught of these malignant cells that are potentially occurring every day in our normal tissue. And so, by tackling one problem, do we create another? And I think that's going to be the big challenge over the next 50 years.
Eric Topol (51:31):
Yeah, and I think your point about the multi-decade challenge, because if you can promote healthy aging without any risk of cancer, that would be great. But if the tradeoff is close, it's not going to be very favorable. That seems to be the main liability of modulation aging through many of the, there's many shots on goal here, of course, as you well know. But they do seem to pose that risk in general.
Charles Swanton (51:58):
I think that's right. I think the other thing is, I still find, I don’t know if you agree with me, but it is an immense conundrum. What is the underlying molecular basis for somatic aging, for aging of normal tissues? And it may be multifactorial, it may not be just one answer to that question. And different tissues may age in different ways. I don't know. It's a fascinating area of biology, but I think it really needs to be studied more because as you say, it underpins all of these diseases we've been talking about today, cardiovascular, neurodegeneration, cancer, you name it. We absolutely have to understand this. And actually, the more I work in cancer, the more I feel like actually what I'm working on is aging.
(52:48):
And this is something that James DeGregori and I have discussed a lot. There's an observation that in medicine around patients with alpha-1 antitrypsin deficiency who are at higher risk of lung cancer, but they're also at high risk of COPD, and we know the associations of chronic obstructive pulmonary disease with lung cancer risk. And one of the theories that James had, and I think this is a beautiful idea, actually, is as our tissues age, and COPD is a reflection of aging, to some extent gone wrong. And as our tissues age, they become less good at controlling the expansion of these premalignant clones, harboring, harboring oncogenic mutations in normal tissue. And as those premalignant clones expand, the substrate for evolution also expands. So there's more likely to be a second and third hit genetically. So it may be by disrupting the extracellular matrices through inflammation that triggers COPD through alpha-1 antitrypsin deficiency or smoking, et cetera, you are less effectively controlling these emergent clones that just expand with age, which I think is a fascinating idea actually.
Eric Topol (54:01):
It really is. Well, I want to tell you, Charlie, this has been the most fascinating, exhilarating discussion I've ever had on cancer. I mean, really, I am indebted to you because not just all the work you've done, but your ability to really express it, articulate it in a way that hopefully everyone can understand who's listening or reading the transcript. So we'll keep following what you're doing because you're doing a lot of stuff. I can't thank you enough for joining me today, and you've given me lots of things to think about. I hope the people that are listening or reading feel the same way. I mean, this has been so mind bending in many respects. We're indebted to you.
Charles Swanton (54:49):
Well, we all love reading your Twitter feeds. Keep them coming. It helps us keep a broader view of medicine and biological research, not just cancer, which is why I love it so much.
******************************************
The Ground Truths newsletters and podcasts are all free, open-access, without ads.
Please share this post/podcast with your friends and network if you found it informative
Voluntary paid subscriptions all go to support Scripps Research. Many thanks for that—they greatly helped fund our summer internship programs for 2023 and 2024.
Thanks to my producer Jessica Nguyen and Sinjun Balabanoff tor audio and video support at Scripps Research.
Note: you can select preferences to receive emails about newsletters, podcasts, or all I don’t want to bother you with an email for content that you’re not interested in.

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
Tom Cech: RNA Takes Center Stage
5 juin· Ground Truths
In this podcast, Thomas Czech, Distinguished Professor at the University of Colorado, Boulder, with a lineage of remarkable contributions on RNA, ribozyme, and telomeres, discuss why RNA is so incredibly versatile.
Video snippet from our conversation. Full videos of all Ground Truths podcasts can be seen on YouTube here. The audios are also available on Apple and Spotify.
Transcript with links to the audio and external links
Eric Topol (00:07):
Well, hello, this is Eric Topol from Ground Truths, and it's really a delight for me to welcome Tom Cech who just wrote a book, the Catalyst, and who is a Nobel laureate for his work in RNA. And is at the University of Colorado Boulder as an extraordinary chemist and welcome Tom.
Tom Cech (00:32):
Eric, I'm really pleased to be here.
The RNA Guy
Eric Topol (00:35):
Well, I just thoroughly enjoyed your book, and I wanted to start out, if I could, with a quote, which gets us right off the story here, and let me just get to it here. You say, “the DNA guy would need to become an RNA guy. Though I didn’t realize it at the time, jumping ship would turn out to be the most momentous decision in my life.” Can you elaborate a bit on that?
Tom Cech (01:09):
As a graduate student at Berkeley, I was studying DNA and chromosomes. I thought that DNA was king and really somewhat belittled the people in the lab next door who were working on RNA, I thought it was real sort of second fiddle material. Of course, when RNA is acting just as a message, which is an important function, a critical function in all life on earth, but still, it's a function that's subservient to DNA. It's just copying the message that's already written in the playbook of DNA. But little did I know that the wonders of RNA were going to excite me and really the whole world in unimaginable ways.
Eric Topol (02:00):
Well, they sure have, and you've lit up the world well before you had your Nobel Prize in 1989 was Sid Altman with ribozyme. And I think one of the things that struck me, which are so compelling in the book as I think people might know, it's divided in two sections. The first is much more on the biology, and the second is much more on the applications and how it's changing the world. We'll get into it particularly in medicine, but the interesting differentiation from DNA, which is the one trick pony, as you said, all it does is store stuff. And then the incredible versatility of RNA as you discovered as a catalyst, that challenging dogma, that proteins are supposed to be the only enzymes. And here you found RNA was one, but also so much more with respect to genome editing and what we're going to get into here. So I thought what we might get into is the fact that you kind of went into the scum of the pond with this organism, which by the way, you make a great case for the importance of basic science towards the end of the book. But can you tell us about how you, and then of course, many others got into the Tetrahymena thermophila, which I don't know that much about that organism.
Tom Cech (03:34):
Yeah, it's related to Tetrahymena is related to paramecium, which is probably more commonly known because it's an even larger single celled animal. And therefore, in an inexpensive grade school microscope, kids can look through and see these ciliated protozoa swimming around on a glass slide. But I first learned about them when I was a postdoc at MIT and I would drive down to Joe Gall's lab at Yale University where Liz Blackburn was a postdoc at the time, and they were all studying Tetrahymena. It has the remarkable feature that it has 10,000 identical copies of a particular gene and for a higher organism, one that has its DNA in the nucleus and does its protein synthesis in the cytoplasm. Typically, each gene's present in two copies, one from mom, one from dad. And if you're a biochemist, which I am having lots of stuff is a real advantage. So 10,000 copies of a particular gene pumping out RNA copies all the time was a huge experimental advantage. And that's what I started working on when I started my own lab at Boulder.
Eric Topol (04:59):
Well, and that's where, I guess the title of the book, the Catalyst ultimately, that grew into your discovery, right?
Tom Cech (05:08):
Well, at one level, yes, but I also think that the catalyst in a more general conversational sense means just facilitating life in this case. So RNA does much more than just serve as a biocatalyst or a message, and we'll get into that with genome editing and with telomerase as well.
The Big Bang and 11 Nobel Prizes on RNA since 2000
Eric Topol (05:32):
Yes, and I should note that as you did early in the book, that there's been an 11 Nobel prize awardees since 2000 for RNA work. And in fact, we just had Venki who I know you know very well as our last podcast. And prior to that, Kati Karikó, Jennifer Doudna who worked in your lab, and the long list of people working RNA in the younger crowd like David Liu and Fyodor Urnov and just so many others, we need to have an RNA series because it's just exploding. And that one makes me take you back for a moment to 2007. And when I was reading the book, it came back to me about the Economist cover. You may recall almost exactly 17 years ago. It was called the Biology’s Big Bang – Unravelling the secrets of RNA. And in that, there was a notable quote from that article. Let me just get to that. And it says, “it is probably no exaggeration to say that biology is now undergoing its neutron moment.”
(06:52):
This is 17 years ago. “For more than half a century the fundamental story of living things has been a tale of the interplay between genes, in the form of DNA, and proteins, which is genes encode and which do the donkey work of keeping living organisms living. The past couple of years, 17 years ago, however, has seen the rise and rise of a third type of molecule, called RNA.” Okay, so that was 2007. It's pretty extraordinary. And now of course we're talking about the century of biology. So can you kind of put these last 17 years in perspective and where we're headed?
Tom Cech (07:34):
Well, Eric, of course, this didn't all happen in one moment. It wasn't just one big bang. And the scientific community has been really entranced with the wonders of RNA since the 1960s when everyone was trying to figure out how messenger RNA stored the genetic code. But the general public has been really kept in the dark about this, I think. And as scientists, were partially to blame for not reaching out and sharing what we have found with them in a way that's more understandable. The DNA, the general public's very comfortable with, it's the stuff of our heredity. We know about genetic diseases, about tracing our ancestry, about solving crimes with DNA evidence. We even say things like it's in my DNA to mean that it's really fundamental to us. But I think that RNA has been sort of kept in the closet, and now with the mRNA vaccines against Covid-19, at least everyone's heard of RNA. And I think that that sort of allowed me to put my foot in the door and say, hey, if you were curious about the mRNA vaccines, I have some more stories for you that you might be really interested in.
RNA vs RNA
Eric Topol (09:02):
Yeah, well, we'll get to that. Maybe we should get to that now because it is so striking the RNA versus RNA chapter in your book, and basically the story of how this RNA virus SARS-CoV-2 led to a pandemic and it was fought largely through the first at scale mRNA nanoparticle vaccine package. Now, that takes us back to some seminal work of being able to find, giving an mRNA to a person without inciting massive amount of inflammation and the substitution of pseudouridine or uridine in order to do that. Does that really get rid of all the inflammation? Because obviously, as you know, there's been some negativism about mRNA vaccines for that and also for the potential of not having as much immune cell long term activation. Maybe you could speak to that.
Tom Cech (10:03):
Sure. So the discovery by Kati Karikó and Drew Weissman of the pseudouridine substitution certainly went a long way towards damping down the immune response, the inflammatory response that one naturally gets with an RNA injection. And the reason for that is that our bodies are tuned to be on the lookout for foreign RNA because so many viruses don't even mess with DNA at all. They just have a genome made of RNA. And so, RNA replicating itself is a danger sign. It means that our immune system should be on the lookout for this. And so, in the case of the vaccination, it's really very useful to dampen this down. A lot of people thought that this might make the mRNA vaccines strange or foreign or sort of a drug rather than a natural substance. But in fact, modified nucleotides, nucleotides being the building blocks of RNA, so these modified building blocks such as pseudoU, are in fact found in natural RNAs more in some than in others. And there are about 200 modified versions of the RNA building blocks found in cells. So it's really not an unusual modification or something that's all that foreign, but it was very useful for the vaccines. Now your other question Eric had to do with the, what was your other question, Eric?
Eric Topol (11:51):
No, when you use mRNA, which is such an extraordinary way to get the spike protein in a controlled way, exposed without the virus to people, and it saved millions of lives throughout the pandemic. But the other question is compared to other vaccine constructs, there's a question of does it give us long term protective immunity, particularly with T cells, both CD8 cytotoxic, maybe also CD4, as I know immunology is not your main area of interest, but that's been a rub that's been put out there, that it isn't just a weaning of immunity from the virus, but also perhaps that the vaccines themselves are not as good for that purpose. Any thoughts on that?
Tom Cech (12:43):
Well, so my main thought on that is that this is a property of the virus more than of the vaccine. And respiratory viruses are notoriously hard to get long-term immunity. I mean, look at the flu virus. We have to have annual flu shots. If this were like measles, which is a very different kind of virus, one flu shot would protect you against at least that strain of flu for the rest of your life. So I think the bad rap here is not the vaccine's fault nearly as much as it's the nature of respiratory viruses.
RNA And Aging
Eric Topol (13:27):
No, that's extremely helpful. Now, let me switch to an area that's really fascinating, and you've worked quite a bit on the telomerase story because this is, as you know, being pursued quite a bit, has thought, not just because telomeres might indicate something about biologic aging, but maybe they could help us get to an anti-aging remedy or whatever you want to call it. I'm not sure if you call it a treatment, but tell us about this important enzyme, the role of the RNA building telomeres. And maybe you could also connect that with what a lot of people might not be familiar with, at least from years ago when they learned about it, the Hayflick limit.
Tom Cech (14:22):
Yes. Well, Liz Blackburn and Carol Greider got the Nobel Prize for the discovery of telomerase along with Jack Szostak who did important initial work on that system. And what it does is, is it uses an RNA as a template to extend the ends of human chromosomes, and this allows the cell to keep dividing without end. It gives the cell immortality. Now, when I say immortality, people get very excited, but I'm talking about immortality at the cellular level, not for the whole organism. And in the absence of a mechanism to build out the ends of our chromosomes, the telomeres being the end of the chromosome are incompletely replicated with each cell division. And so, they shrink over time, and when they get critically short, they signal the cell to stop dividing. This is what is called the Hayflick limit, first discovered by Leonard Hayflick in Philadelphia.
(15:43):
And he, through his careful observations on cells, growing human cells growing in Petri dishes, saw that they could divide about 50 times and then they wouldn't die. They would just enter a state called senescence. They would change shape, they would change their metabolism, but they would importantly quit dividing. And so, we now see this as a useful feature of human biology that this protects us from getting cancer because one of the hallmarks of cancer is immortality of the tumor cells. And so, if you're wishing for your telomeres to be long and your cells to keep dividing, you have to a little bit be careful what you wish for because this is one foot in the door for cancer formation.
Eric Topol (16:45):
Yeah, I mean, the point is that it seems like the body and the cell is smart to put these cells into the senescent state so they can't divide anymore. And one of the points you made in the book that I think is worth noting is that 90% of cancers have the telomerase, how do you say it?
Tom Cech (17:07):
Telomerase.
Eric Topol (17:08):
Yeah, reactivate.
Tom Cech (17:09):
Right.
Eric Topol (17:10):
That's not a good sign.
Tom Cech (17:12):
Right. And there are efforts to try to target telomerase enzyme for therapeutic purposes, although again, it's tricky because we do have stem cells in our bodies, which are the exception to the Hayflick limit rule. They do still have telomerase, they still have to keep dividing, maybe not as rapidly as a cancer cell, but they still keep dividing. And this is critical for the replenishment of certain worn out tissues in our such as skin cells, such as many of our blood cells, which may live only 30 days before they poop out. That's a scientific term for needing to be replenished, right?
Eric Topol (18:07):
Yeah. Well, that gets me to the everybody's, now I got the buzz about anti-aging, and whether it's senolytics to get rid of these senescent cells or whether it's to rejuvenate the stem cells that are exhausted or work on telomeres, all of these seem to connect with a potential or higher risk of cancer. I wonder what your thoughts are as we go forward using these various biologic constructs to be able to influence the whole organism, the whole human body aging process.
Tom Cech (18:47):
Yes. My view, and others may disagree is that aging is not an affliction. It's not a disease. It's not something that we should try to cure, but what we should work on is having a healthy life into our senior years. And perhaps you and I are two examples of people who are at that stage of our life. And what we would really like is to achieve, is to be able to be active and useful to society and to our families for a long period of time. So using the information about telomerase, for example, to help our stem cells stay healthy until we are, until we're ready to cash it in. And for that matter on the other side of the coin, to try to inhibit the telomerase in cancer because cancer, as we all know, is a disease of aging, right? There are young people who get cancer, but if you look at the statistics, it's really heavily weighted towards people who've been around a long time because mutations accumulate and other damage to cells that would normally protect against cancer accumulates. And so, we have to target both the degradation of our stem cells, but also the occurrence of cancer, particularly in the more senior population. And knowing more about RNA is really helpful in that regard.
RNA Drugs
Eric Topol (20:29):
Yeah. Well, one of the things that comes across throughout the book is versatility of RNA. In fact, you only I think, mentioned somewhere around 12 or 14 of these different RNAs that have a million different shapes, and there's so many other names of different types of RNAs. It's really quite extraordinary. But one of the big classes of RNAs has really hit it. In fact, this week there are two new interfering RNAs that are having extraordinary effects reported in the New England Journal on all the lipids, abnormal triglycerides and LDL cholesterol, APOC3. And can you talk to us about this interfering the small interfering RNAs and how they become, you've mentioned in the book over 400 RNAs are in the clinic now.
Tom Cech (21:21):
Yeah, so the 400 of course is beyond just the siRNAs, but these, again, a wonderful story about how fundamental science done just to understand how nature works without any particular expectation of a medical spinoff, often can have the most phenomenal and transformative effects on medicine. And this is one of those examples. It came from a roundworm, which is about the size of an eyelash, which a scientist named Sydney Brenner in England had suggested would be a great experimental organism because the entire animal has only about a thousand cells, and it's transparent so we can look at, see where the cells are, we can watch the worm develop. And what Andy Fire and Craig Mello found in this experimental worm was that double-stranded RNA, you think about DNA is being double-stranded and RNA as being single stranded. But in this case, it was an unusual case where the RNA was forming a double helix, and these little pieces of double helical RNA could turn off the expression of genes in the worm.
(22:54):
And that seemed remarkable and powerful. But as often happens in biology, at least for those of us who believe in evolution, what goes for the worm goes for the human as well. So a number of scientists quickly found that the same process was going on in the human body as a natural way of regulating the expression of our genes, which means how much of a particular gene product is actually going to be made in a particular cell. But not only was it a natural process, but you could introduce chemically synthesized double helical RNAs. There are only 23 base pairs, 23 units of RNA long, so they're pretty easy to chemically synthesize. And that once these are introduced into a human, the machinery that's already there grabs hold of them and can be used to turn off the expression of a disease causing RNA or the gene makes a messenger RNA, and then this double-stranded RNA can suppress its action. So this has become the main company that is known for doing this is Alnylam in Boston, Cambridge. And they have made quite a few successful products based on this technology.
Eric Topol (24:33):
Oh, absolutely. Not just for amyloidosis, but as I mentioned these, they even have a drug that's being tested now, as you know that you could take once or twice a year to manage your blood pressure. Wouldn't that be something instead of a pill every day? And then of course, all these others that are not just from Alnylam, but other companies I wasn't even familiar with for managing lipids, which is taking us well beyond statins and these, so-called PCSK9 monoclonal antibodies, so it's really blossoming. Now, the other group of RNA drugs are antisense drugs, and it seemed like they took forever to warm up, and then finally they hit. And can you distinguish the antisense versus the siRNA therapeutics?
Tom Cech (25:21):
Yes, in a real general sense, there's some similarity as well as some differences, but the antisense, what are called oligonucleotides, whoa, that's a big word, but oligo just means a few, right? And nucleotides is just the building blocks of nucleic acid. So you have a string of a few of these. And again, it's the power of RNA that it is so good at specifically base pairing only with matching sequences. So if you want to match with a G in a target messenger RNA, you put a C in the antisense because G pairs with C, if you want to put an A, if want to match with an A, you put a U in the antisense because A and U form a base pair U is the RNA equivalent of T and DNA, but they have the same coding capacity. So any school kid can write out on a notepad or on their laptop what the sequence would have to be of an antisense RNA to specifically pair with a particular mRNA.
(26:43):
And this has been, there's a company in your neck of the woods in the San Diego area. It started out with the name Isis that turned out to be the wrong Egyptian God to name your company after, so they're now known as Ionis. Hopefully that name will be around for a while. But they've been very successful in modifying these antisense RNAs or nucleic acids so that they are stable in the body long enough so that they can pair with and thereby inhibit the expression of particular target RNAs. So it has both similarities and differences from the siRNAs, but the common denominator is RNA is great stuff.
RNA and Genome Editing
Eric Topol (27:39):
Well, you have taken that to in catalyst, the catalyst, you've proven that without a doubt and you and so many other extraordinary scientists over the years, cumulatively. Now, another way to interfere with genes is editing. And of course, you have a whole chapter devoted to not just well CRISPR, but the whole genome editing field. And by the way, I should note that I forgot because I had read the Codebreaker and we recently spoke Jennifer Doudna and I, that she was in your lab as a postdoc and you made some wonderful comments about her. I don't know if you want to reflect about having Jennifer, did you know that she was going to do some great things in her career?
Tom Cech (28:24):
Oh, there was no question about it, Eric. She had been a star graduate student at Harvard, had published a series of breathtaking papers in magazines such as Science and Nature already as a graduate student. She won a Markey fellowship to come to Colorado. She chose a very ambitious project trying to determine the molecular structures of folded RNA molecules. We only had one example at the time, and that was the transfer RNA, which is involved in protein synthesis. And here she was trying these catalytic RNAs, which we had discovered, which were much larger than tRNA and was making great progress, which she finished off as an assistant professor at Yale. So what the general public may not know was that in scientific, in the scientific realm, she was already highly appreciated and much awarded before she even heard anything about CRISPR.
Eric Topol (29:38):
Right. No, it was a great line you have describing her, “she had an uncanny talent for designing just the right experiment to test any hypothesis, and she possessed more energy and drive than any scientist I'd ever met.” That's pretty powerful. Now getting into CRISPR, the one thing, it's amazing in just a decade to see basically the discovery of this natural system to then be approved by FDA for sickle cell disease and beta thalassemia. However, the way it exists today, it's very primitive. It's not actually fixing the gene that's responsible, it's doing a workaround plan. It's got double strand breaks in the DNA. And obviously there's better ways of editing, which are going to obviously involve RNA epigenetic editing, if you will as well. What is your sense about the future of genome editing?
Tom Cech (30:36):
Yeah, absolutely, Eric. It is primitive right now. These initial therapies are way too expensive as well to make them broadly applicable to the entire, even in a relatively wealthy country like the United States, we need to drive the cost down. We need to get them to work, we need to get the process of introducing them into the CRISPR machinery into the human body to be less tedious and less time consuming. But you've got to start somewhere. And considering that the Charpentier and Doudna Nobel Prize winning discovery was in 2012, which is only a dozen years ago, this is remarkable progress. More typically, it takes 30 years from a basic science discovery to get a medical product with about a 1% chance of it ever happening. And so, this is clearly a robust RNA driven machine. And so, I think the future is bright. We can talk about that some more, but I don't want to leave RNA out of this conversation, Eric. So what's cool about CRISPR is its incredible specificity. Think of the human genome as a million pages of text file on your computer, a million page PDF, and now CRISPR can find one sentence out of that million pages that matches, and that's because it's using RNA, again, the power of RNA to form AU and GC base pairs to locate just one site in our whole DNA, sit down there and direct this Cas9 enzyme to cut the DNA at that site and start the repair process that actually does the gene editing.
Eric Topol (32:41):
Yeah, it's pretty remarkable. And the fact that it can be so precise and it's going to get even more precise over time in terms of the repair efforts that are needed to get it back to an ideal state. Now, the other thing I wanted to get into with you a bit is on the ribosome, because that applies to antibiotics and as you call it, the mothership. And I love this metaphor that you had about the ribosome, and in the book, “the ribosome is your turntable, the mRNA is the vinyl LP record, and the protein is the music you hear when you lower the needle.” Tell us more about the ribosome and the role of antibiotics.
Tom Cech (33:35):
So do you think today's young people will understand that metaphor?
Eric Topol (33:40):
Oh, they probably will. They're making a comeback. These records are making a comeback.
Tom Cech (33:44):
Okay. Yes, so this is a good analogy in that the ribosome is so versatile it's able to play any music that you feed at the right messenger RNA to make the music being the protein. So you can have in the human body, we have tens of thousands of different messenger RNAs. Each one threads through the same ribosome and spills out the production of whatever protein matches that mRNA. And so that's pretty remarkable. And what Harry Noller at UC Santa Cruz and later the crystallographers Venki Ramakrishnan, Tom Steitz, Ada Yonath proved really through their studies was that this is an RNA machine. It was hard to figure that out because the ribosome has three RNAs and it has dozens of proteins as well. So for a long time people thought it must be one of those proteins that was the heart and soul of the record player, so to speak.
RNA and Antibiotics
(34:57):
And it turned out that it was the RNA. And so, when therefore these scientists, including Venki who you just talked to, looked at where these antibiotics docked on the ribosome, they found that they were blocking the key functional parts of the RNA. So it was really, the antibiotics knew what they were doing long before we knew what they were doing. They were talking to and obstructing the action of the ribosomal RNA. Why is this a good thing for us? Because bacterial ribosomes are just enough different from human ribosomes that there are drugs that will dock to the bacterial ribosomal RNA, throw a monkey wrench into the machine, prevent it from working, but the human ribosomes go on pretty much unfazed.
Eric Topol (36:00):
Yeah, no, the backbone of our antibiotics relies on this. So I think people need to understand about the two subunits, the large and the small and this mothership, and you illuminate that so really well in the book. That also brings me to phage bacteria phage, and we haven't seen that really enter the clinic in a significant way, but there seems to be a great opportunity. What's your view about that?
Tom Cech (36:30):
This is an idea that goes way back because since bacteria have their own viruses which do not infect human cells, why not repurpose those into little therapeutic entities that could kill, for example, what would we want to kill? Well, maybe tuberculosis has been very resistant to drugs, right? There are drug resistant strains of TB, yes, of TB, tuberculosis, and especially in immunocompromised individuals, this bug runs rampant. And so, I don't know the status of that. It's been challenging, and this is the way that biomedicine works, is that for every 10 good ideas, and I would say phage therapy for bacterial disease is a good idea. For every 10 such ideas, one of them ends up being practical. And the other nine, maybe somebody else will come along and find a way to make it work, but it hasn't been a big breakthrough yet.
RNA, Aptamers and Proteins
Eric Topol (37:54):
Yeah, no, it's really interesting. And we'll see. It may still be in store. What about aptamers? Tell us a little bit more about those, because they have been getting used a lot in sorting out the important plasma proteins as therapies. What are aptamers and what do you see as the future in that regard?
Tom Cech (38:17):
Right. Well, in fact, aptamers are a big deal in Boulder because Larry Gold in town was one of the discoverers has a company making aptamers to recognize proteins. Jack Szostak now at University of Chicago has played a big role. And also at your own institution, Jerry Joyce, your president is a big aptamer guy. And you can evolution, normally we think about it as happening out in the environment, but it turns out you can also make it work in the laboratory. You can make it work much faster in the laboratory because you can set up test tube experiments where molecules are being challenged to perform a particular task, like for example, binding to a protein to inactivate it. And if you make a large community of RNA molecules randomly, 99.999% of them aren't going to know how to do this. What are the odds? Very low.
(39:30):
But just by luck, there will be an occasional molecule of RNA that folds up into a shape that actually fits into the proteins active sighting throws a monkey wrench into the works. Okay, so now that's one in a billion. How are you going to find that guy? Well, this is where the polymerase chain reaction, the same one we use for the COVID-19 tests for infection comes into play. Because if you can now isolate this needle in a haystack and use PCR to amplify it and make a whole handful of it, now you've got a whole handful of molecules which are much better at binding this protein than the starting molecule. And now you can go through this cycle several times to enrich for these, maybe mutagen it a little bit more to give it a little more diversity. We all know diversity is good, so you put a little more diversity into the population and now you find some guy that's really good at recognizing some disease causing protein. So this is the, so-called aptamer story, and they have been used therapeutically with some success, but diagnostically certainly they are extremely useful. And it's another area where we've had success and the future could hold even more success.
Eric Topol (41:06):
I think what you're bringing up is so important because the ability to screen that tens of thousands of plasma proteins in a person and coming up with as Tony Wyss-Coray did with the organ clocks, and this is using the SomaLogic technology, and so much is going on now to get us not just the polygenic risk scores, but also these proteomic scores to compliment that at our orthogonal, if you will, to understand risk of people for diseases so we can prevent them, which is fulfilling a dream we've never actually achieved so far.
Tom Cech (41:44):
Eric, just for full disclosure, I'm on the scientific advisory board of SomaLogic in Boulder. I should disclose that.
Eric Topol (41:50):
Well, that was smart. They needed to have you, so thank you for mentioning that. Now, before I wrap up, well, another area that is a favorite of mine is citizen science. And you mentioned in the book a project because the million shapes of RNA and how it can fold with all hairpin terms turns and double stranded and whatever you name it, that there was this project eteRNA that was using citizen scientists to characterize and understand folding of RNA. Can you tell us about that?
RNA Folding and Citizen Science
Tom Cech (42:27):
So my friend Rhiju Das, who's a professor at Stanford University, sort of adopted what had been done with protein folding by one of his former mentors, David Baker in Seattle, and had repurposed this for RNA folding. So the idea is to come up with a goal, a target for the community. Can you design an RNA that will fold up to look like a four pointed cross or a five pointed star? And it turned out that, so they made it into a contest and they had tens of thousands of people playing these games and coming up with some remarkable solutions. But then they got a little bit more practical, said, okay, that was fun, but can we have the community design something like a mRNA for the SARS-CoV-2 spike protein to make maybe a more stable vaccine? And quite remarkably, the community of many of whom are just gamers who really don't know much about what RNA does, were able to find some solutions. They weren't enormous breakthroughs, but they got a several fold, several hundred percent increase in stability of the RNA by making it fold more tightly. So I just find it to be a fascinating approach to science. Somebody of my generation would never think of this, but I think for today's generation, it's great when citizens can become involved in research at that level.
Eric Topol (44:19):
Oh, I think it's extraordinary. And of course, there are other projects folded and others that have exemplified this ability for people with no background in science to contribute in a meaningful way, and they really enjoy, it's like solving a puzzle. The last point is kind of the beginning, the origin of life, and you make a pretty strong case, Tom, that it was RNA. You don't say it definitively, but maybe you can say it here.
RNA and the Origin of Life
Tom Cech (44:50):
Well, Eric, the origin of life happening almost 4 billion years ago on our primitive planet is sort of a historical question. I mean, if you really want to know what happened then, well, we don't have any video surveillance of those moments. So scientists hate to ever say never, but it's hard to sort of believe how we would ever know for sure. So what Leslie Orgel at the Salk Institute next to you taught me when I was a starting assistant professor is even though we'll never know for sure, if we can recapitulate in the laboratory plausible events that could have happened, and if they make sense chemically and biologically, then that's pretty satisfying, even if we can never be absolutely sure. That's what a number of scientists have done in this field is to show that RNA is sort of a, that all the chemistry sort of points to RNA as being something that could have been made under prebiotic conditions and could have folded up into a way that could solve the greatest of all chicken and egg problems, which came first, the informational molecule to pass down to the next generation or the active molecule that could copy that information.
(46:32):
So now that we know that RNA has both of those abilities, maybe at the beginning there was just this RNA world RNA copying itself, and then proteins came along later, and then DNA probably much more recently as a useful but a little bit boring of genetic information, right?
Eric Topol (46:59):
Yeah. Well, that goes back to that cover of the Economist 17 years ago, the Big Bang, and you got me convinced that this is a pretty strong story and candidate. Now what a fun chance to discuss all this with you in an extraordinary book, Tom. Did I miss anything that you want to bring up?
Tom Cech (47:21):
Eric, I just wanted to say that I not only appreciate our conversation, but I also appreciate all you are doing to bring science to the non-scientist public. I think people like me who have taught a lot of freshmen in chemistry, general chemistry, sort of think that that’s the level that we need to aim at. But I think that those kids have had science in high school year after year. We need to aim at the parents of those college freshmen who are intelligent, who are intellectually curious, but have not had science courses in a long time. And so, I'm really joining with you in trying to avoid jargon as much as possible. Use simple language, use analogies and metaphors, and try to share the excitement of what we're doing in the laboratory with the populace.
Eric Topol (48:25):
Well, you sure did that it was palpable. And I thought about it when I read the book about how lucky it would be to be a freshman at the University of Boulder and be having you as the professor. My goodness. Well, thank you so much. This has been so much fun, Tom, and I hope everybody's going to get out there and read the Catalyst to get all the things that we didn't even get a chance to dive into. But this has been great and look forward to future interactions with you.
Tom Cech (48:53):
Take care, Eric.
*********************
Thanks for listening or reading this edition of Ground Truths.
Please share this podcast with your friends and network. That tells me you found it informative and makes the effort in doing these worthwhile.
All Ground Truths newsletters and podcast are free. Voluntary paid subscriptions all go to support Scripps Research. Many thanks for that—they greatly helped fund our summer internship programs for 2023 and 2024.
Thanks to my producer Jessica Nguyen and Sinjun Balabanoff for audio and video support at Scripps Research.
Note: you can select preferences to receive emails about newsletters, podcasts, or all I don’t want to bother you with an email for content that you’re not interested in.

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
Venki Ramakrishnan: The New Science of Aging
28 mai· Ground Truths
Professor Venki Ramakrishnan, a Nobel laureate for his work on unraveling the structure of function of the ribosome, has written a new book WHY WE DIE which is outstanding. Among many posts and recognitions for his extraordinary work in molecular biology, Venki has been President of the Royal Society, knighted in 2012, and was made a Member of the Order of Merit in 2022. He is a group leader at the MRC Laboratory of Molecular Biology research institute in Cambridge, UK.
A brief video snippet of our conversation below. Full videos of all Ground Truths podcasts can be seen on YouTube here. The audios are available on Apple and Spotify.
Transcript with links to audio and external links
Eric Topol (00:06):
Hello, this is Eric Topol with Ground Truths, and I have a really special guest today, Professor Venki Ramakrishnan from Cambridge who heads up the MRC Laboratory of Molecular Biology, and I think as you know a Nobel laureate for his seminal work on ribosomes. So thank you, welcome.
Venki Ramakrishnan (00:29):
Thank you. I just want to say that I'm not the head of the lab. I'm simply a staff member here.
Eric Topol (00:38):
Right. No, I don't want to give you more authority than you have, so that was certainly not implied. But today we're here to talk about this amazing book, Why We Die, which is a very provocative title and it mainly gets into the biology of aging, which Venki is especially well suited to be giving us a guided tour and his interpretations and views. And I read this book with fascination, Venki. I have three pages of typed notes from your book.
The Compression of Morbidity
Eric Topol (01:13):
And we could talk obviously for hours, but this is fascinating delving into this hot area, as you know, very hot area of aging. So I thought I'd start off more towards the end of the book where you kind of get philosophical into the ethics. And there this famous concept by James Fries of compression of morbidity that's been circulating for well over two decades. That's really the big question about all this aging effort. So maybe you could give us, do you think there is evidence for compression of morbidity so that you can just extend healthy aging and then you just fall off the cliff?
Venki Ramakrishnan (02:00):
I think that's the goal of most of the sort of what I call the saner end of the aging research community is to improve our health span. That is the number of years we have healthy lives, not so much to extend lifespan, which is how long we live. And the idea is that you take those years that we now spend in poor health or decrepitude and compress them down to just very short time, so you're healthy almost your entire life, and then suddenly go into a rapid decline and die. Now Fries who actually coined that term compression or morbidity compares this to the One-Hoss Shay after poem by Oliver Wendell Holmes from the 19th century, which is about this horse carriage that was designed so perfectly that all its parts wore out equally. And so, a farmer was riding along in this carriage one minute, and the next minute he found himself on the ground surrounded by a heap of dust, which was the entire carriage that had disintegrated.
Venki Ramakrishnan (03:09):
So the question I would ask is, if you are healthy and everything about you is healthy, why would you suddenly go into decline? And it's a fair question. And every advance we've made that has kept us healthier in one respect or another. For example, tackling diabetes or tackling heart disease has also extended our lifespan. So people are not living a bigger fraction of their lives healthily now, even though we're living longer. So the result is we're spending the same or even more number of years with one or more health problems in our old age. And you can see that in the explosion of nursing homes and care homes in almost all western countries. And as you know, they were big factors in Covid deaths. So I'm not sure it can be accomplished. I think that if we push forward with health, we're also going to extend our lifespan.
Venki Ramakrishnan (04:17):
Now the argument against that comes from studies of these, so-called super centenarians and semi super centenarians. These are people who live to be over 105 or 110. And Tom Perls who runs the New England study of centenarians has published findings which show that these supercentenarians live extraordinarily healthy lives for most of their life and undergo rapid decline and then die. So that's almost exactly what we would want. So they have somehow accomplished compression of morbidity. Now, I would say there are two problems with that. One is, I don't know about the data sample size. The number of people who live over 110 is very, very small. The other is they may be benefiting from their own unique genetics. So they may have a particular combination of genetics against a broad genetic background that's unique to each person. So I'm not sure it's a generally translatable thing, and it also may have to do with their particular life history and lifestyle. So I don't know how much of what we learned from these centenarians is going to be applicable to the population as a whole. And otherwise, I don't even know how this would be accomplished. Although some people feel there's a natural limit to our biology, which restricts our lifespan to about 115 or 120 years. Nobody has lived more than 122. And so, as we improve our health, we may come up against that natural limit. And so, you might get a compression of morbidity. I'm skeptical. I think it's an unsolved problem.
Eric Topol (06:14):
I think I'm with you about this, but there's a lot of conflation of the two concepts. One is to suppress age related diseases, and the other is to actually somehow modulate control the biologic aging process. And we lump it all together as you're getting at, which is one of the things I loved about your book is you really give a balanced view. You present the contrarians and the different perspectives, the perspective about people having age limits potentially much greater than 120, even though as you say, we haven't seen anyone live past 122 since 1997, so it's quite a long time. So this, I think, conflation of what we do today as far as things that will reduce heart disease or diabetes, that’s age related diseases, that's very different than controlling the biologic aging process. Now getting into that, one of the things that's particularly alluring right now, my friend here in San Diego, Juan Carlos Belmonte, who went over from Salk, which surprised me to the Altos Labs, as you pointed on in the book.
Venki Ramakrishnan (07:38):
I'm not surprised. I mean, you have a huge salary and all the resources you want to carry out the same kind of research. I wouldn't blame any of these guys.
Rejuvenating Animals With Yamanaka Factors
Eric Topol (07:50):
No, I understand. I understand. It's kind of like the LIV Golf tournament versus the PGA. It's pretty wild. At any rate, he's a good friend of mine, and I visited with him recently, and as you mentioned, he has over a hundred people working on this partial epigenetic reprogramming. And just so reviewing this for the uninitiated is giving the four Yamanaka transcription factors here to the whole animal or the mouse and rejuvenating old mice, essentially at least those with progeria. And then others have, as you point out in the book, done this with just old mice. So one of the things that strikes me about this, and in talking with him recently is it's going to be pretty hard to give these Yamanaka factors to a person, an intravenous infusion. So what are your thoughts about this rejuvenation of a whole person? What do you think?
Venki Ramakrishnan (08:52):
If I hadn't seen some of these papers would've been even more skeptical. But the data from, well, Belmonte's work was done initially on progeria mice. These are mice that age prematurely. And then people thought, well, they may not represent natural aging, and what you're doing is simply helping with some abnormal form of aging. But he and other groups have now done it with normal mice and observed similar effects. Now, I would say reprogramming is one way. It's a very exciting and powerful way to almost try to reverse aging because you're trying to take cells back developmentally. You're taking possibly fully differentiated cells back to stem cells and then helping regenerate tissue, which one of the problems as we age is we start losing stem cells. So we have stem cell depletion, so we can no longer replace our tissues as we do when we're younger. And I think anyone who knows who's had a scrape or been hurt in a fall or something knows this because if I fall and scrape my elbow and get a big bruise and my grandson falls, we repair our tissues at very, very different rates. It takes me days or weeks to recover, and my grandson's fine in two or three days. You can hardly see he had a scrape at all. So I think that's the thing that these guys want to do.
Venki Ramakrishnan (10:48):
And the problem is Yamanaka factors are cancer. Two of them are oncogenic factors, right? If you give Yamanaka factors to cells, you can take them all the way back to what are called pluripotent cells, which are the cells that are capable of forming any tissue in the body. So for example, a fertilized egg or an early embryo cells from the early embryo are pluripotent. They could form anything in the body. Now, if you do that to cells with Yamanaka factors, they often form teratomas, which are these unusual forms of cancer tumors. And so, I think there's a real risk. And so, what these guys say is, well, we'll give these factors transiently, so we'll only take the cells back a little ways and not all the way back to pluripotency. And that way if you start with skin cells, you'll get the progenitor stem cells for skin cells. And the problem with that is when you do it with a population, you're getting a distribution. Some of them will go back just a little, some of them may go back much more. And I don't know how to control all this. So I think it's very exciting research. And of course, if I were one of these guys, I would certainly want to carry on doing that research. But I don't think it's anywhere near ready for primetime in terms of giving it to human beings as a sort of anti-aging therapeutic.
Aging and Cancer Shared Hallmarks
Eric Topol (12:31):
Yeah. Well, I couldn't agree more on that because this is a company that's raised billions of dollars to go into clinical trials. And the question that comes up here, which is a theme in the book and a theme with the aging process to try to artificially, if you will affect it, is this risk of cancer. And as we know, the hallmarks of aging overlap considerably with the hallmarks of cancer. And this is just one example, as you mentioned, where these transcription factors could result in generating cancer. But as you also point out in the book at many places, methylation changes, DNA, repair, and telomeres.
Venki Ramakrishnan (13:21):
And telomeres.
Venki Ramakrishnan (13:24):
All of those are related to cancer as well. And this was first pointed out to me by Titia de Lange, who's a world expert on telomeres at Rockefeller, and she was pointing out to me the intimate connection between cancer and aging and many mechanisms that have evolved to prevent cancer early in life tend to cause aging later in life, including a lot of DNA damage response, which sends cells into senescence and therefore causes aging. Buildup of senescence cells is a problem later in life with aging, but it has a role which is to prevent cancer early in life. And so, I think it's going to be the same problem with stem cell therapy. I think very targeted stem cell therapy, which is involved in replacing certain tissues, the kind of regenerative medicine that stem cells have been trying to address for a very long time, and only now we're beginning to see some of the successes of that. So it's been very slow, even when the goal and target is very specific and well-defined, and there you are using that stem cell to treat a pretty bad disease or some really serious problem. I think with aging, the idea that somebody might take this so they can live an extra 10 years, it's a much higher bar in terms of safety and long-term safety and efficacy. So I don't think that this is going to happen anytime soon, but it’s not to say it'll never happen. There is some serious biology underlying it.
Eric Topol (15:13):
Right. Well, you just touched on this, but of course the other, there's several big areas that are being explored, and one of them is trying to deal with these senescent cells and trying to get rid of them from the body because they can secrete evil humors, if you will. And the problem with that, it seems that these senescent cells are sort of protective. They stop dividing, they're not going to become cancerous, although perhaps they could contribute to that in some way. So like you say, with telomeres and so many things that are trying to be manipulated here, there's this downside risk and it seems like this is what we're going to have to confront this. We have seen Venki with the CAR-T, the T-cell engineering, there's this small risk of engendering cancer while you're trying to deal with the immune system.
Senolytics
Venki Ramakrishnan (16:07):
Yeah, I think with senescent cells, the early in life senescent cells have an important role in biology. They're essentially signaling to the immune system that there's a site that's subject to viral infection or wounds or things like that. So it's a signal to send other kinds of cells there to come and repair the damage. Now, of course, that evolved to help us early in life. And also many senescent cells were a response to DNA damage. And that's again, a way for the body that if your DNA is damaged, you don't want that cell to be able to divide indefinitely because it could become cancerous. And so, you send it into senescence and get it out of harm's way. So early life, we were able to get rid of these senescent cells, we were able to come to the site and then clean up the damage and eventually destroy the senescent cells themselves.
Venki Ramakrishnan (17:08):
But as we get older, the response mechanisms also deteriorate with age. Our immune system deteriorates with age, all the natural signaling mechanisms deteriorate with age. And so, we get this buildup of senescent cells. And there people have asked, well, these senescent cells don't just sit there, they secrete inflammatory compounds, which originally was a feature, not a bug, but then it becomes a problem later in life. And so, people have found that if you target senescent cells in older animals, those animals improve their symptoms of aging improved dramatically or significantly anyway. And so, this has led to this whole field called senolytics, which is being able to specifically target senescent cells. Now there the problem is how would you design compounds that are highly specific for senescent cells and don't damage your other cells and don't have other long-term side effects? So again, I think it's a promising area, but a lot of work needs to be done to establish long-term safety and efficacy.
Eric Topol (18:23):
Right. No, in fact, just today in Nature, there's a feature on killing the zombie cells, and it discusses just what you're pointing out, which is it's not so easy to tag these specifically and target them, even though as you know, there's some early trials and things like diabetic macular edema. And we'll see how that plays out. Now, one of the things that comes up is the young blood story. So in the young blood, whether it's this parabiosis or however you want to get at it, and I guess it even applies to the young microbiome of a gut, but there's this consistent report that there's something special going on there. And of course the reciprocal relationship of giving the old blood to the young mice, whatever, but no one can find the factor, whether it's platelet factor 4, GDF11, or what are your thoughts about this young blood story?
Venki Ramakrishnan (19:25):
I think there's no question that the experiments work because they were reproduced and they were reproduced over quite a long period, and which is that when you connect an old mouse or rat with a young equivalent, then the old mouse or old rat benefits from the young blood from the younger animal. And conversely, the younger animal suffers from the blood from the older animal. And then people were wondering whether this is simply that the young animal has better detoxification and things like that, or whether it's actually the blood. And they gave it just as transfusion without connecting the animals and showed that it really was the blood. And so, this of course then leads to the question, well, what is it about young blood that’s beneficial and what is it about old blood that is bad? But the problem is blood has hundreds of factors. And so, they have to look at which factors are significantly different, and they might be in such small quantities that you might not be able to detect those differences very easily.
Venki Ramakrishnan (20:40):
And then once you've detected differences, then you have to establish their mechanism of action. And first of all, you have to establish that the factor really is beneficial. Then you have to figure out how it works and what its potential side effects could be. And so again, this is a promising area where there's a lot of research, but it has not prevented people from jumping the gun. So in the United States, and I should say a lot of them in your state, California somehow seems to attract all these immortality types. Well, anyway, a lot of companies set up to take blood from young donors, extract the plasma and then give it to rich old recipients for a fee for a healthy fee. And I think the FDA actually shut down one of them on the grounds that they were not following approved procedure. And then they tried to start up under a different name. And then eventually, I don't know what happened, but at one point the CEO said something I thought was very amusing. He said, well, the problem with clinical trials is that they take too long. I'm afraid that's characteristic of some portion of this sort of anti-aging therapeutics community. There's a very mainstream rigorous side to it, but there's also at the other end of the spectrum, kind of the wild west where people just sell whatever they can. And I think this exploits people's fear of getting old and being disabled or things like that and then dying. And I think the fear seems to be stronger in California where people like their lives and don't want to age.
Eric Topol (22:49):
You may be right about that. I like your term in the book immortality merchants, and of course we'll get into a bit, I hope the chapter on the crackpots and prophets that you called it was great. But that quote, by the way, which was precious from, I think it was Ambrosia, the name of the company and the CEO, but there's another quote in the book I want to ask you about. Most scientists working on aging agree that dietary restriction can extend both healthy life and overall life in mice and also lead to reductions in cancer, diabetes, and overall mortality in humans. Is that true? Most scientists think that you can really change these age-related diseases.
Caloric Restriction and Related Pathways
Venki Ramakrishnan (23:38):
I think if you had to pick one area in which there's broad agreement, it is caloric restriction. But I wouldn't say the consensus is complete. And the reason I say that is that most of the comparisons are between animals that can eat as much as they want, called ad libitum diet and mice that are calorically restricted or same with other animals even yeast. You either compared with an extremely rich medium or in a calorically restricted medium. And this is not a great comparison. And people, there's one discrepancy, and that was in monkeys where an NIH study didn't find huge differences, whereas a Wisconsin study found rather dramatic differences between the control group and the calorically restricted group. And so, what was the difference? Well, the difference was that the NIH study, the controlled group didn't have a calorically restricted diet, but still had a pretty reasonable diet.
Venki Ramakrishnan (24:50):
It wasn't given a unhealthy rich diet of all you can eat. And then they tried to somehow reconcile their findings in a later study. But it leads to the question of whether what you can conclude is that a rich all you can eat diet, in other words, gorging on an all you can eat buffet is definitely bad for you. So that's why you could draw that conclusion rather than saying it's actually the caloric restriction. So I think people need to do a little more careful study. There was also a study on mice which took different strains of mice and showed that in some strains, caloric restriction actually shortened lifespan didn't increase lifespan. Now, much of the aging community says, ah, that's just one study. But nobody's actually shown whether there was anything wrong with that study or even tried to reproduce it. So I think that study still stands.
Venki Ramakrishnan (25:51):
So I think it's not completely clear, but the fact is that there's some calorie dependence that's widely been observed across species. So between the control group and the experimental group, whatever you may, however, you may define it as there's been some effective calories intake. And the other interesting thing is that one of the pathways affected by caloric restriction is the so-called TOR pathway and one of the inhibitors of the TOR pathways is rapamycin. And rapamycin in studies has also shown some of these beneficial effects against the symptoms of aging and in lifespan. Although rapamycin has the same issue as with many other remedies, it's an immunosuppressive drug and that means it makes you more prone to infection and wound healing and many other things. I believe one of them was there's a question of whether it affects your libido, but nevertheless, that has not prevented rapamycin clinics from opening up, did I say in California? So I do think that there's often serious science, which leads to sort of promising avenues. But then there are of course people who jump the gun and want to go ahead anyway because they figure by the time trials are done, they'll be dead and they'd rather try act now.
Eric Topol (27:36):
Right. And you make a good, I mean the rapamycin and mTOR pathway, you really developed that quite a bit in the book. It's really quite complex. I mean, this is a pleotropic intervention, whether it's a rapalogs or rapamycin, it's just not so simple at all.
Venki Ramakrishnan (27:53):
Right. It's not at all simple because the TOR pathway has so many consequences. It affects so many different processes in the cell from including my own field of protein synthesis. It's one of the things it does is shut down global protein synthesis, and that's one of the effects of inhibiting TOR. So, and it turns up autophagy, which is this recycling of defective proteins and entirely defective entire organelles. So I think the TOR pathway is like a hub in a very large network. And so, when you start playing with that, you're going to have multiple consequences.
Eric Topol (28:37):
Yeah, no. And another thing that you develop so well is about this garbage disposal waste disposal system, which is remarkably elaborate in the cell, whether it's the proteasome for the proteins and the autophagosome for the autophagy with the lysosomes and the mitochondria mitophagy. Do you want to comment about that? Because this is something I think a lot of people don't appreciate, that waste management in the cell is just, it's a big deal.
Venki Ramakrishnan (29:10):
So we always think of producing things in the cell as being important, making proteins and so on. But the fact is destroying proteins is equally important because sometimes you need proteins for a short time, then they've done their job and you need to get rid of them, or proteins become dysfunctional, they stop working, or even worse, they start clumping together and causing diseases for example you could think of Alzheimer's as a disease, which involves protein tangles. Of course, the relationship between the tangles and the disease is still being worked out, but it's a characteristic of Alzheimer's that you have these protein tangles and the cell has evolved very elaborate mechanisms to constantly turn over defective proteins. Well, for example, it senses when proteins are unfolded and essentially the chain has unraveled and is now sticking to all sorts of things and causing problems. So I think in all of these cases, the cells evolved very elaborate mechanisms to recycle defective products, to have proper turnover of proteins. And in fact, recycling of entire organelles like mitochondria, when they become defective, the whole mitochondria can be recycled. So these systems also break down with aging. And so, as we age, we have more of a tendency to accumulate unfolded proteins or to accumulate defective mitochondria. And it's one of the more serious problems with aging.
Eric Topol (30:59):
Yeah, there's quite a few of them. Unfortunately, quite a few problems. Each of them are being addressed. So there's many different shots on goal here. And as you also aptly point out, they're interconnected. So many of these things are not just standalone strategies. I do want to get your sense about another popular thing, especially here out in California, are the clocks, epigenetic clocks in particular. And these people are paying a few hundred dollars and getting their biologic age, which what is that? And they're also thinking that I can change my future by getting clocks. Some of these companies offer every few months to get a new clock. This is actually remarkable, and I wonder what your thoughts are about it.
Venki Ramakrishnan (31:48):
Well, again, this is an example of some serious biology and then people jumping the gun to use it. So the serious biology comes from the fact that we age at different rates individuals. So anyone who's been to a high school reunion knows this. You'll have classmates who are unrecognizable because they’ve aged so much and others who've hardly changed since you knew them in high school. So of course at my age, that's getting rarer and rarer. But anyway, but you know what I mean. So the thing is that, is there a way that we can ask on an individual level how much has that individual aged? And there are markers that people have identified, some of them are markers on our DNA, which you mentioned in California. Horvath is a very famous scientist who has a clock named after him actually, which has to do with methylation of our DNA and the patterns of methylation affect the pattern of gene expression.
Venki Ramakrishnan (33:01):
And that pattern changes as we age. And they've shown that those patterns are a better predictor of many of the factors of aging. For example, mortality or symptoms of aging. They're a better predictor of that than chronological age. And then of course there are blood markers, for example, levels of various blood enzymes or blood factors, and there are dozens of these factors. So there are many different tests of many different kinds of markers which look at aging. Now the problem is these all work on a population level and they also work on an individual level for time comparison. That is to say, if you want to ask is some intervention working? You could ask, how fast are these markers changing in this person without the intervention and how fast are they changing with the intervention? So for these kind of carefully controlled experiments, they work, but another case is, for example, glycosylation of proteins, especially proteins of your immune system.
Venki Ramakrishnan (34:15):
It turns out that adding sugar groups to your immune system changes with age and causes an immune system to misfire. And that's a symptom of aging. It's called inflammaging. So people have used different markers. Now the problem is these markers are not always consistent with each other because you may be perfectly fine in many respects, but by some particular marker you may be considered old just because they're comparing you to a population average. But how would you say one person said, look, we all lose height as we age, but that doesn't mean if you take a short person, you can consider them old. So it's a difference between an individual versus a population, and it's a difference between what happens to an individual by following that individual over time versus just taking an individual and comparing it to some population average. So that's one problem.
Organ Clocks
Venki Ramakrishnan (35:28):
The other problem is that our aging is not homogeneous. So there's a recent paper from I believe Tony Wyss-Coray group, which talks about the age of different organs in the same person. And it turns out that our organs, and this is not just one paper, there are other papers as well. Our organs don't necessarily age at the same rate. So giving a single person, giving a person a single number saying, this is your biological age, it's not clear what that means. And I would say, alright, even if you do it, what are you going to do about it? What can you do about it knowing your biological, the so-called number of a biological age. So I’m not a big fan. I’m a big fan of using these markers as a tool in research to understand what interventions work because otherwise it would take too long. You’d have to wait 20 years to see some large scale symptoms. And certainly, if you want to look at mortality, you’d have to wait possibly even longer. But if you were to be able to follow track these interventions and see that these markers slowed down with intervention, then you could say, well, your interventions having an effect on something related to aging. So I would say these are very useful research tools, but they’re not meant to be used at $500 a pop in your age.
Venki Ramakrishnan (37:02):
But of course that hasn't stopped lots of companies from doing it.
Eric Topol (37:07):
No, it's just amazing actually. And by the way, we interviewed Tony Wyss-Coray about the organ clock, the paper. I thought it really was quite a great contribution, again, on a research level.
Venki Ramakrishnan (37:19):
He's a very serious scientist. He actually spoke here at the LMB as well. He gave a very nice talk here.
Is Aging A Disease?
Eric Topol (37:26):
He's the real deal. And I think that's going to help us to have that organ specific type of tracking is another edge here to understand the effects. Well, before we wrap up, I want to ask you a question that you asked in the book. Is aging a disease?
Venki Ramakrishnan (37:49):
That's again, a controversial subject. So the WHO, and I believe the FDA decided that aging was not a disease on the grounds that it's inevitable and ubiquitous. It happens to everybody and it's inevitable. So how could something that happens to everybody and inevitable be considered a disease? A disease is an abnormal situation. This is a normal situation, but the anti-aging researchers and especially the anti-aging therapeutics people don't like that because if it's not a disease, how can they run a clinical trial? So they want aging to be considered a disease. And their argument is that if you look at almost every condition of old age, every disease of old age like cancer, diabetes, heart disease, dementia, the biggest risk factor in all of these diseases is age. That's the strongest risk factor. And so, they say, well, actually, you could think of these diseases as secondary diseases, the primary disease being age, and then that results in these other diseases.
Venki Ramakrishnan (39:07):
I am a little skeptical of that idea. I tend to agree with the WHO and the FDA, but I can see both sides of the argument. And as you know, I've laid them both out. My view is that it should be possible to do trials that help with aging regardless of whether you consider aging a disease or not. But that will require the community to agree on what set of markers to use to characterize success. And that's people, for example, Tony Wyss-Coray has his proteome, blood proteome markers, Horvath has his DNA methylation clock. There are a whole bunch of these. And then there are people with glycation or glycosylation of various proteins as markers. These people need to all come together. Maybe we need to organize a nice conference for them in some place like Southern California or Hawaii or somewhere, put them together in a locked room for a week so that they can thrash out a common set of markers and at least agree on what experiments they need to do to even come to that agreement and then use that to evaluate anti-aging therapies. I think that would be a way forward.
Eric Topol (40:35):
Yeah, I think you're bringing up a really valuable point because at the moment, they're kind of competing with one another, whether it's the glycosylated proteins or the transcriptomics or the epigenetics. And we don't know whether these are additive or what they're really measuring.
Venki Ramakrishnan (40:53):
Some of them may be highly correlated, and that's okay, but I think they need to know that. And they also need to come up with some criteria of how do we define age in an individual. It's not one number, just like we have many things that characterize our health. Cholesterol is one, blood pressures another, various other lipids. They're all blood enzymes, liver enzymes. All these things are factors in defining our so-called biological health. So I don't think there's some single number that's going to say this is your age. Just like there isn't one single thing that says you're healthy, you're not healthy.
DNA Repair
Eric Topol (41:38):
Right, that’s well put. Last topic on aging is on about DNA repair, which is an area that you know very well. And one of the quotes in your book, I think is important for people to take in. “Nevertheless, they will make an error once every million or so letters in a genome with a few billion letters. That means several thousand mistakes occur each time a cell divides. So the DNA repair enzyme, as you point out the sentinels of our genome, the better we repair, the better we age.” Can we fix the DNA repair problem?
Venki Ramakrishnan (42:20):
I think maybe, again, I'm not sure what the consequences would be and how much it would take. There's one curious fact, and that is that there was a paradox called Peto’s paradox after the scientist who discovered it, which is why don't big animals get cancer much more frequently than say a mouse? In fact, a mouse gets cancer far more readily than an elephant does, and in reality, the elephant should actually get cancer more because it has many orders of magnitude more cells, and all it takes is for one cell to become cancerous for the animal to get cancer and die. So the chances that one cell would become cancer would be larger if there are many, many more cells. And it turns out that elephants have many copies of DNA repair proteins or DNA damage response proteins, not so much DNA repair, but the response to DNA damage and in particular, a protein called p53. And so, this leads to the question that if you had very good DNA repair or very good DNA damage response, would you then live longer or solve this problem? I'm not entirely sure because it may have other consequences because for example, you don't want to send cells into senescence too easily. So I think these things are all carefully balanced, evolutionarily, depending on what's optimized to optimize fitness for each species.
Venki Ramakrishnan (44:13):
For a mouse, the equation's different than for a large animal because a mouse can get eaten by predators and so on. So there, it doesn't pay for evolution to spend too much select for too much spending of resources in maintenance and repair, for larger animals the equation is different. So I just don't know enough about what the consequences would be.
Eric Topol (44:40):
No, it's really interesting to speculate because as you point out in the book, the elephant has 20 copies of p53, and we have two as humans. And the question is that protection from cancer is very intriguing, especially with the concerns that we've been talking about.
Venki Ramakrishnan (44:57):
And it was also true, I believe they did some analysis of genomics of these whales that live very long, and they found sorts of genes that are probably involved in DNA repair or DNA damage response.
Eric Topol (45:14):
Well, this is a masterful book. Congratulations, Venki. I thoroughly enjoyed it. It's very stimulating. I know a lot of the people that will listen or read the transcript will be grabbed by it.
Crackpots and Prophets
Venki Ramakrishnan (45:28):
I think what I've tried to do is give the general reader a real understanding of the biology of aging so that even a complete non-scientist can get an understanding of the processes, which in turn empowers them to take action to do the sort of things that will actually really help. And also it'll guard them against excessive hype, of which there's a lot in this business. And so, I think that was the goal, and to try and present a balanced view of the field. I'm merely trying to be a realist. I'm not being a pessimist about it, but I also think this excessively optimistic hype is actually bad for the field and bad for science and bad for the public as well.
Eric Topol (46:16):
Well, and you actually were very kind in the chapter you have on crackpots and prophets. You could have been even tougher on some of these guys. You were very relatively diplomatic and gentle, I thought, I don't know if you were holding back.
Venki Ramakrishnan (46:28):
I had two lawyers looked at it, so.
Eric Topol (46:33):
I believe it. And now one thing, apart from what we've been talking about because of your extraordinary contribution on the structural delineation of the ribosome back in the early 2000s and 2009 Nobel Prize. Now, the world of AI now with AlphaFold 3 and all these other large language models, would that have changed your efforts? Would that have accelerated things or is it not really?
Venki Ramakrishnan (47:09):
Well, it would've helped, but you would still need the experimental data to solve something like the ribosome, a large complex like the ribosome. And the other thing that would really change well has changed our world is the advent of cryo-electron microscopy of which Scripps is one of the leading places for it. And that has really changed it so that now nobody would bother to crystallize a ribosome and try to get an X-ray structure out of it. You would just throw it into an EM grid, collect your data and be off to the races. So new ribosome structures are being solved all the time at a fraction, a tiny fraction of the time it took to solve the first one.
Eric Topol (48:02):
Wow, that's fascinating. This has been a real joy for Venki to discuss your book and your work, and thanks so much for what you're doing to enlighten us and keep the balance. And it may not be as popular as the immortality merchants, but it's really important stuff.
Venki Ramakrishnan (48:19):
Yeah, no, I hope actually, I found that many of the public want to read about the biology of aging. They're curious. Humans have been curious ever since we knew about mortality, about why some species live so short lives and other species live such a long time and why we actually have to age and die. So there’s natural curiosity and then it also empowers the public once they understand the basis of aging, to take action, to live healthy lives and do that. It's an empowering book rather than a recipe book.
Venki Ramakrishnan (49:01):
I think a lot of the public actually does appreciate that. And of course, scientists will like the sort of more balanced and tone.
Eric Topol (49:13):
Well, you do it so well. All throughout you have metaphors to help people really understand and the concepts, and I really applaud you for doing this. In fact, a couple of people who we both know, Max and John Brockman, apparently were influential for you to get to do it. So I think it's great that you took it on and all the power to you. So thank you, and I hope that we'll get a chance to visit further as we go forward.
******************
Headshot photo credits by Kate Joyce and Santa Fe Institute
The Ground Truths newsletters and podcasts are all free, open-access, without ads.
Please share this post/podcast with your friends and network if you found it informative
Voluntary paid subscriptions all go to support Scripps Research. Many thanks for that—they greatly helped fund our summer internship programs for 2023 and 2024.
Thanks to my producer Jessica Nguyen and Sinjun Balabanoff tor audio and video support at Scripps Research.
Note: you can select preferences to receive emails about newsletters, podcasts, or all I don’t want to bother you with an email for content that you’re not interested in.
A Poll on Anti-Aging

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
Svetlana Blitshteyn: On the Front Line With Long Covid and POTS
20 mai· Ground Truths
After finishing her training in neurology at Mayo Clinic, Dr. Svetlana Blitshteyn started a Dysautonomia Clinic in 2009. Little did she know what was in store many years later when Covid hit!
Ground Truths podcasts are on Apple and Spotify. The video interviews are on YouTube
Transcript with audio and external links
Eric Topol (00:07):
Well, hello, it's Eric Topol from Ground Truths, and I have with me a really great authority on dysautonomia and POTS. We will get into what that is for those who aren't following this closely. And it's Svetlana Blitshteyn who is a faculty member at University of Buffalo and a neurologist who long before there was such a thing as Covid was already onto one of the most important pathways of the body, the autonomic nervous system and how it can go off track. So welcome, Svetlana.
Svetlana Blitshteyn (00:40):
Thank you so much, Eric for having me. And I want to say it's a great honor for me to be here and just to be on the list with your other guests. It's remarkable and I'm very grateful and congratulations on being on the TIME100 Health list for influential people in 2024. And I am grateful for everything that you've done. As I mentioned earlier, I'm a big fan of your work before the pandemic and of course with Covid I followed your podcast and posts because you became the best science communicator and I'm very happy to see you being a strong advocate and thank you for everything you've done.
Eric Topol (01:27):
Well, that's so kind to you. And I think talking about getting things going before the pandemic, back in 2011, you published a book with Jodi Epstein Rhum called POTS - Together We Stand: Riding the Waves of Dysautonomia. And you probably didn't have an idea that there would be an epidemic of that more than a decade later, I guess, right?
Svetlana Blitshteyn (01:54):
Yeah, absolutely. Of course, SARS-CoV-2 is a new virus and we can technically say that Long Covid and post Covid complications could be viewed as a new entity. But practically speaking, we know that post-infectious syndromes have been happening for many decades. And so, the most common trigger for POTS happened to be infection, whether it was influenza or mononucleosis or Lyme or enterovirus. We knew this was happening. So I think it didn't take long for me and my colleagues to realize that we're going to be seeing a lot of patients with autonomic dysfunction after Covid.
On the Front Line
Eric Topol (02:40):
Well, one of the things that's important for having you on is you're in the front lines taking care of lots of patients with Long Covid and this postural orthostatic tachycardia syndrome (POTS). And I wonder if you could tell us what it's care for these patients because so many of them are incapacitated. As a cardiologist, I see of course some because of the cardiovascular aspects, but you are dealing with this on a day-to-day basis.
Svetlana Blitshteyn (03:14):
Yeah, absolutely. As early as April 2020 when everything was closed, I got a call from a young doctor in New York City saying that he had Covid and he couldn't recover, he couldn't return to the hospital. And his colleagues and cardiology attendants also had the same symptoms and the symptoms were palpitations, orthostatic intolerance, tachycardia, fatigue. Now, how he knew to contact me is that his sister was my patient with POTS before Covid pandemic. So he kind of figured this looked like my sister, let me check this out. And it didn't take long for me to have a lot of patience from the early wave. And then fairly soon, I think within months I was thinking, we have to write this up because this is important. And to some of us it was not news, but I was sure that to many physicians and public health officials, this would be something new.
Svetlana Blitshteyn (04:18):
So because I'm a busy clinician and don't have a lot of time for publications, I had to recruit a graduate student from McMasters and together we had this paper out, which was the first and largest case series on post Covid POTS and other autonomic disorders. And interestingly, even though it came out I think in 2021, by the time it was published, it became the most citable paper for me. And so I think from then on organizations and societies became interested in the work that I do because prior to that, I must say in the kind of a niche specialty was I don't think it was very popular or of interest to me.
How Did You Get Interested in Dysautonomia?
Eric Topol (05:06):
Yeah, so that's why I wanted to just take a step back with you Svetlana, because you had the foresight to be the founder and director of the Dysautonomia Clinic when a lot of people weren't in touch with this as an important entity. What prompted you as a neurologist to really zoom in on dysautonomia when you started this clinic?
Svetlana Blitshteyn (05:28):
Sure. So the reasons are how I ended up in this field is kind of a convoluted road and the reasons are many, but one, I will say that I trained at Mayo Clinic where we received very good training on autonomic disorders and EMG and coming back to returning back to Buffalo, I began working at the large multiple sclerosis clinic because Western New York has a high incidence MS. And so, what they quickly realized in that clinic is that there was a subset of women who did not qualify for the diagnostic criteria of multiple sclerosis, yet they had a lot of the same symptoms and they were certainly very disabled. Now I recognize that these women had autonomic disorders of all sorts and small fiber neuropathy, and I think this population sort of grew and eventually I realized there is no one not only in Buffalo but the entire Western New York who is doing this work.
Svetlana Blitshteyn (06:34):
So I kind of fell into that. But another reason is actually more personal that I haven’t talked about. So years ago I was traveling to Toronto, Canada for a neurology meeting to present my big study on meningioma and hormone replacement therapy using Mayo Clinic database. And so, in that year, the study received top 10 noteworthy studies of the year award from the Society of Neuro-Oncology, and it was profiled in Reuters Health. Now, on the way back from the conference, I had the flu, and when they returned I could no longer walk the same hallways of the hospital where I walked previously. And no matter how hard I try to push my body, we all do this in medicine, we push through, I just couldn’t do it. No amount of wishing or positive thinking. And so, I think that’s how I came to know personally the post-infectious syndromes. And I think it almost became a duality of experiencing this and also practicing it.
Eric Topol (07:52):
No, that’s really striking and it wasn’t so common to hear about this post flu, but certainly it changed in 2020. So how does a person with POTS typically present to you?
Clinical Presentation
Svetlana Blitshteyn (08:08):
So these are very important questions because what I want to stress is though POTS is one of the most common autonomic disorders. Even if you don’t have POTS by the diagnostic criteria, you may still have autonomic dysfunction and significant autonomic symptoms. How do they present? Well, they present like most Long Covid patients, the most common symptoms are orthostatic intolerance, fatigue, exercise intolerance, post exertional malaise, dizziness, tachycardia, brain fog. And these are common themes across the board in Long Covid patients, but also in pre-Covid post-acute infection syndrome patients. And you have to recognize because I think what I tell my colleagues is that oftentimes patients are not going to present to you saying, I have orthostatic intolerance. Many times they will say, I’m very tired. I can no longer go to the gym or when I go to the store, I have to be out of there in 15 minutes because the orthostatic intolerance symptoms come up.
Svetlana Blitshteyn (09:22):
So sometimes the patients themselves don’t recognize that and it’s up to us physicians to ask the right questions to get the information down. History is very important, knowing the pattern. And then of course, as I always say in all of my papers and lectures, you have to do a 10-minute stand test by measuring supine and standing blood pressure and heart rate on every Long Covid patients. And that’s how you spot those that have excessive postural tachycardia or their blood pressure dropping or so forth. So we have the tools. We don’t need fancy autonomic labs. We don’t even need a tilt table test. The diagnostic criteria for POTS is that you need to have either a 10-minute stand test or a tilt table test to get the diagnosis for POTS, orthostatic hypotension or even neurocardiogenic syncope. Now I think it's important to stress that even if a patient doesn't qualify, and let's say many patients with Long Covid will not elevate their heart rate by at least 30 beats per minute, it could be 20, it could be 25. These criteria are of course essential when we do research studies. But I think practically speaking, in patient care where everything is gray and nothing is black or white, especially in autonomic disorders, you really have to make a diagnosis saying, this sounds like autonomic dysfunction. Let me treat the patient for this problem.
Eric Topol (11:07):
Well, you brought up something that’s really important because doctors don’t have much time and they’re inpatient. They don’t wait 10 minutes to do a test to check your blood pressure. They send the patients for a tilt table, which nobody likes to have that test done, and it’s unnecessary added appointment and expense and whatnot. So that’s a good tip right there that you can get the same information just by checking the blood pressure and heart rate on standing for an extended period of time, which 10 minutes is a long time in the clinic of course. Now, what is the mechanism, what do you think is going on with the SARS-CoV-2 virus and its predilection to affect the autonomic nervous system? As you know, so many studies have questioned whether you even actually infect neurons or alternatively, which is more likely this an inflammation of the neural tissue. But what do you think is going on here?
Underpinnings
Svetlana Blitshteyn (12:10):
Right, so I think it’s important to say we don’t have exact pathophysiology of what exactly is going on. I think we can only extrapolate that what’s going on in Long Covid is possibly what’s going on in any post infectious onset dysautonomia. And so there are many hypothesis and there are many suggestions, and we share this disorder with cardiologist and immunologist and rheumatologist. The way I view this is what I described in my paper from a few years ago is that this is likely a central nervous system disorder with multisystemic involvement and it involves the cardiovascular system, immunologic, metabolic, possibly prothrombotic. The pathophysiology of all POTS closely parallels to pathophysiology of Long Covid. Now we don’t know if it’s the same thing and certainly I see that there may be more complications in Long Covid patients in the realm of cardiovascular manifestations in the realm of blood clots and things like that.
Svetlana Blitshteyn (13:21):
So we can’t say it’s the same, but it very closely resembles and I think at the core is going to be inflammation, autoimmunity and immunologic dysfunction. Now there are also other things that are very important and that would be mitochondrial dysfunction, that would be hypercoagulable state, it would be endothelial dysfunction. And I think the silver lining of Long Covid and having so many people invested in research and so many funds is that by uncovering what Long Covid is, we’re now going to be uncovering what POTS and other autonomic disorders are. And I think we also need to mention a couple of other things. One is small fiber neuropathy, small fiber neuropathy and POTS are very much comorbid conditions. And similarly, small fiber neuropathy frequently occurs in patients with Long Covid, so that’s a substrate with the damaged small nerve fibers that they're everywhere in our bodies and also innervate the organs as well.
Svetlana Blitshteyn (14:34):
The second big thing is that needs to be mentioned is hyperactive mast cells. So mast cells, small nerve fibers and capillaries are very much located in proximity. And what I have usually is a slide from an old paper in oral biology that gives you a specimen where you see a capillary vessel, a stain small nerve fiber, and in between them there is a mass cell with tryptase in it stained in black. And so there is a close communication between small nerve fibers between endothelial wall and between mast cells, and that’s what we commonly see as a triad. We see this as a triad in Long Covid patients. We see that as a triad in patients with joint hypermobility syndrome and hypermobile EDS, and you also see this in many of the autoimmune disorders where people develop new allergies and new sensitivities concurrent or preceding the onset of autoimmune disease.
Small Fiber Neuropathy
Eric Topol (15:49):
Yeah, no, it’s fascinating. And I know you’ve worked with this in Ehlers-Danlos syndrome (EDS) as you mentioned, the hypermobility, but just to go back on this, when you want to entertain the involvement of small fiber neuropathy, is that diagnosable? I mean it’s obvious that you can get the tachycardia, the change in position blood pressure, but do you have to do other tests to say there is indeed a small fiber neuropathy or is that a clinical diagnosis?
Svetlana Blitshteyn (16:20):
Absolutely. We have the testing and the testing is skin biopsy. That is simply a punch biopsy that you can do in your clinic and it takes about 15 minutes. You have the free kit that the company of, there are many companies, I don’t want to name specific ones, but there are several companies that do this kind of work. You send the biopsy back to them, they look under the microscope, they stain it. You can also stain it with amyloid stain to rule out amyloidosis, which we do in neurology, and I think that’s quite accessible to many clinicians everywhere. Now we also have another test called QSART (quantitative sudomotor axon reflex test), and that’s a test part of autonomic lab. Mayo Clinic has it, Cleveland Clinic has it, other big labs have it, and it’s hard to get there because the wait time is big.
Svetlana Blitshteyn (17:15):
Patients need to travel. Insurance doesn’t always authorize, so access is a big problem, but more accessible is the skin biopsy. And so, by doing skin biopsy and then correlating with neurologic exam findings, which oftentimes involved reduce pain and temperature sensation in the feet, sometimes in the hands you can conclude that the patient has small fiber neuropathy and that's a very tangible and objective diagnosis. There again, with everything related to diagnostics, some neuropathy is very patchy and the patchy neuropathy is the one that may not be in your feet where you do the skin biopsy. It may be in the torso, it may be in the face, and we don't have biopsy there. So you can totally miss it. The results can come back as normal, but you can have patchy type of small fiber neuropathy and there are also diagnostic tests that might be not sensitive to pick up issues. So I think in everything Long Covid, it highlights the fact that many tests that we use in medicine are outdated perhaps and not targeted towards these patients with Long Covid. Therefore we say, well, we did the workup, everything looks good. MRI looks good, cardiac echo looks great, and yet the patient is very sick with all kinds of Long Covid complications.
Pure Post-Viral POTS?
Eric Topol (18:55):
Right. Now, before we get into the treatments, I want to just segment this a bit. Can you get pure POTS that is no Long Covid just POTS, or as you implied that usually there's some coalescence of symptoms with the usual Long Covid symptoms and POTS added to that?
Svetlana Blitshteyn (19:21):
So the studies have shown for us that about 40% of patients with POTS have post-infectious onset, which means more than a half doesn’t. And so of course you can have POTS from other causes and the most common is puberty, hormonal change, the most common age of onset is about 13, 14 years old and 80% of women of childbearing age and other triggers or pregnancy, hormonal change again, surgery, trauma like concussion, post-concussion, autonomic dysfunction is quite common.
Eric Topol (20:05):
So these are pure POTS without the other symptoms. Is that what you're saying in these examples?
Svetlana Blitshteyn (20:12):
Well, it's a very good question. It depends what you mean by pure POTS, and I have seen especially cardiologists cling to this notion that there is pure POTS and then there is POTS plus. Now I think majority of people don't have pure POTS and by pure POTS I think you mean those who have postural tachycardia and nothing else. And so most patients, I think 80% have a number of symptoms. So in my clinic I almost never see someone who is otherwise well and all they have is postural tachycardia and then they're having a great time. Some patients do exist like that, they tend to be athletic, they can still function in their life, but majority of patients come to us with symptoms like dizziness, like fatigue, like exercise intolerance, decline in functioning. So I think there is this notion that while there is pure POTS, let me just fix the postural tachycardia and the patient will be great and we all want that. Certainly sometimes I get lucky and when I give the patient a beta blocker or ivabradine or a calcium channel blocker, sometimes we use it, certainly they get better, but most patients don't have that because the disability that drives POTS isn't actually postural tachycardia, it's all that other stuff and a lot of it's neurologic, which is why I put this as a central nervous system disorder.
Treatments
Eric Topol (21:58):
Yeah, that's so important. Now you mentioned the treatments. These are drug treatments, largely beta blockers, and can you tell us what's the success rate with the various treatments that you use in your clinic?
Svetlana Blitshteyn (22:13):
So the first thing we'll have to mention is that there are no FDA approved therapies for POTS, just like there are no FDA approved therapies for Long Covid. And so, everything we use is off label. Now, oftentimes people think that because it wasn't evidence-based and there are no big trials. We do have trials, we do have trials for beta blockers and we know they work. We have trials for Midodrine and we know that's working. We also have fludrocortisone, which is a medication that improves sodium and water resorption. So we know that there are certain things we've used for decades that have been working, and I think that's what I was trying to convey in this paper of post Covid autonomic dysfunction assessment and treatment is that when you see these patients, and you can be of any specialty, you can be in primary care, you can be a physiatrist, a cardiologist, there are things to do, there are medications to use.
Svetlana Blitshteyn (23:20):
Oftentimes colleagues would say, well, you diagnose them and then what do you treat them with? And then I can refer them to table six in that paper and say, look at this list. You have a lot of options to try. We have the first line treatment options, which are your beta blockers and Midodrine and Florinef and Mestinon. And then we have the second line therapies you can choose from the stimulants are there Provigil, Nuvigil, Wellbutrin, Droxidopa is FDA approved for neurogenic orthostatic hypotension. Now we don't use it commonly, but it can still be tried in people whose blood pressures are falling on your exam. So we have a number of medications to choose from in addition to non-pharmacologic therapies.
Eric Topol (24:14):
Right now, I'm going to get to the non-pharmacologic in a moment, but the beta blocker, which is kind of the first one to give, it's a little bit paradoxical. It makes people tired, and these people already are, don't have much energy. Is the success rate of beta blocker good enough that that should be the first thing to try?
Svetlana Blitshteyn (24:35):
Absolutely. The first line medication treatment options are beta blockers. Why? Okay, why are they working? They're not only working to reduce heart rate, but they may also decrease sympathetic overactivity, which is the driving mechanism of autonomic dysfunction. And when you reduce that overactivity, even your energy level can improve. Now, the key here is to use a low dose. A lot of the time I see this mistake being done where the doctor is just prescribing 25 milligrams of metoprolol twice a day. Well, this is too high. And so, the key is to use very low doses and to use them and then increase them as needed. We have a bunch of beta blockers to choose from. We have the non-selective propranolol that you can use when someone maybe has a migraine headache or significant anxiety, they penetrate the brain, and we have non-selected beta blockers like atenolol, metoprolol and others that you can use at half a tablet. Sometimes I start my patients at quarter of tablet and then go from there. So low doses will block tachycardia, decrease sympathetic overactivity, and in many cases will allow the patient to remain upright for longer periods of time.
Eric Topol (26:09):
That's really helpful. Now, one of the other things, I believe it's approved in Canada, not in the US, is a vagal neuromodulation device. And I wonder, it seems like it would be nice to avoid drugs if there was a device that worked really well. Is there anything that is in the hopper for that?
Svetlana Blitshteyn (26:32):
Yeah, absolutely. Non-invasive vagus nerve stimulator is in clinical trials for POTS and other autonomic disorders, but we have it FDA for treatment of migraine and cluster headaches, so it's already approved here and it can also be helpful for chronic pain and gastroparesis. So there are studies on mice that show that with the application of noninvasive vagus nerve stimulator, there is reduction of pro-inflammatory cytokines. So here is this very important connection that comes from Kevin Tracey's work that showed inflammatory reflex, and that's a reflex between the vagus nerve and the immune system. So when we talk about sympathetic overactivity, we need to also think about that. That's a mechanism for pro-inflammatory state and possibly prothrombotic state. So anything that decreases sympathetic overactivity and enhancing parasympathetic tone is going to be good for you.
Eric Topol (27:51):
Now, let's go over to, I mean, I'm going to get into this body brain axis in a moment because there's another part of the story here that's becoming more interesting, fascinating, in fact every day. But before I do that, you mentioned the small fiber neuropathy. Is there a specific treatment for that or is that just something that is just an added dimension of the problem without a specific treatment available?
Svetlana Blitshteyn (28:21):
Yeah, we certainly have treatment for small fiber neuropathy. We have symptomatic treatment for neuropathic pain, and these medications are gabapentin, pregabalin, amitriptyline and low dose naltrexone that have been gaining popularity. We used that before the pandemic. We used low dose naltrexone for people with chronic pain related to joint hypermobility. And so, we have symptomatic, we also have patches and creams and all kinds of topical applications for people with neuropathic pain. Then we also have, we try to go for the root cause, right? So the number one cause of small fiber neuropathy in the United States is diabetes. And certainly, you need to control hyperglycemia and in some patients you only need a pre-diabetic state, not even full diabetes to already have peripheral neuropathy. So you want to control blood glucose level first and foremost. Now then we have a big category of autoimmune and immune mediated causes, and that's where it gets very interesting because practical experience from many institutions and many neurologists worldwide have shown that when you give a subset of patients with autoimmune small fiber neuropathy, immunotherapy like IVIG, a lot of patients feel significantly better. And so, I think paralleling our field in dysautonomia and POTS, we are looking forward to immunotherapy being more mainstream rather than exception from the rule because access and insurance coverage is a huge barrier for clinicians and patients, but that may be a very effective treatment options for treatment refractory patients whose symptoms do not improve with symptomatic treatment.
Eric Topol (30:38):
Now, with all these treatments that are on the potential menu to try, and of course sometimes it really is a trial and error to get one that hopefully works for Covid, Long Covid, what is the natural history? Does this persist over years, or can it be completely resolved?
Svetlana Blitshteyn (31:00):
That’s a great question. Everyday Long Covid patients ask me, and I think what we are seeing is that there is a good subset of patients for whom Long Covid is going to be temporary and they will improve and even recover close to normal. Now remember that original case series of patients that I reported in early 2021 based on my 2020 experience in that 20 patient case series, very few recovered, three patients recovered back to normal. Most patients had lingering ongoing chronic symptoms. So of course mine is a kind of a referral bias where I get to see the sickest patients and it looks to be like it’s a problem of chronic illness variety. But I also think there is going to be a subset of patients and then we have to study them. We need to study who got better and who didn’t. And people improve significantly and some even recover close to normal. But I think certain symptoms like maybe fatigue and heat intolerance could persist because those are very heavily rooted in autonomic dysfunction.
Vaccination and POTS
Eric Topol (32:26):
Yeah, well, that’s something that’s sobering and why we need trials and to go after this in much more intensity and priority. Now the other issue here is while with Covid, this is almost always the virus infection, there have been reports of the vaccine inducing POTS and Long Covid, and so what does that tell us?
Svetlana Blitshteyn (32:54):
Well, that’s a big, big topic. Years ago, I was the first one to report a patient with POTS that was developed after HPV vaccine Gardasil. Now, at that time I was a young neurologist. Then the patient came to me saying she was an athlete saying two weeks after Gardasil vaccine, she developed these very disabling symptoms. And I thought it was very interesting and unique and I thought, well, I’ll just publish it. I never knew that this would be the start of a whole different discussion and debate on HPV vaccines. There were multiple reports from numerous countries, Denmark, Mexico, Japan. Japan actually suspended their mass HPV vaccination program. So somehow it became a big deal. Now many people, including my colleagues didn’t agree that POTS can begin POTS, small fiber neuropathy, other adverse neurologic events can begin after vaccination in general. And so, this was a topic that was widely debated and the European medical agencies came back saying, we don't have enough evidence.
Svetlana Blitshteyn (34:20):
Of course, we all want to have a good cancer vaccine. And it was amazing to watch this Covid vaccine issue unfolding where more than one study now have shown that indeed you can develop POTS after Covid vaccines and that the rate of POTS after Covid vaccines is actually slightly higher than before vaccination. So I think it was kind of interesting to see this unfold where I was now invited by Nature Journal to write an editorial on this very topic. So I think it's important to mention that sometimes POTS can begin after vaccination and however, I've always advised my patients to be vaccinated even now. Even now, I have patients who are unvaccinated and I say, I'm worried about you getting a second Covid or third without these vaccines, so please get vaccinated. Vaccines are very important public health measure, but we also have to acknowledge that sometimes people develop POTS, small fiber neuropathy and other complications after Covid vaccines.
Prominence of the Vagus Nerve
Eric Topol (35:44):
Yeah, I think this is important to emphasize here because of all vaccinations can lead to neurologic sequelae. I mean look at Guillain-Barre, which is even more worrisome and that brings in the autoimmune component I think. And of course, the Covid vaccines and boosters have a liability in a small, very small percentage of people to do this. And that can't be discounted because it's a small risk and it's always this kind of risk benefit story when you're getting vaccinated that you are again spotlighting. Now gets us to the biggest thing of all besides the practical pearls you've been coming up with to help everyone in patients and clinicians. In recent weeks, there's been explosion of these intra body circuits. There was a paper from Columbia last week that taught us about the body-brain circuits between the vagus nerve and the caudal Nucleus of the Solitary Tract (cNST) of the brain and how this is basically a master switch for the immune system. And so, the vagus nerve there and then you have this gut to brain story, which is the whole gut microbiome is talking to the brain through the vagus nerve. I mean, everything comes down to the vagus nerve. So you've been working all your career and now everything's coming into this vagus nerve kind of final common pathway that's connecting all sorts of parts of the body that we didn't truly understand before. So could you comment about this because it's pretty striking.
Svetlana Blitshteyn (37:34):
Absolutely. I think this pandemic is highlighting the pitfalls of everything we didn't know but should have in the past. And I think this is one of them. How important is the autonomic nervous system and how important is the vagus nerve that is the longest nerve in the body and carries the parasympathetic outflow. And I think this is a very important point that we have to move forward. We cannot stop at the autonomic knowledge that we've gained thus far. Autonomic neurology and autonomic medicine has always been the field with fellowship, and we have American Autonomic Society as well. But I think now is a great time to move forward and study how the autonomic nervous system communicates with the immunologic system. And again, Kevin Tracey's work was groundbreaking in the sense that he connected the dots and realized that if you stimulate the vagus nerve and the parasympathetic outflow, then you can reduce pro-inflammatory cytokines and that he has shown that you can also improve or significantly such disorders like rheumatoid arthritis and other autoimmune inflammatory conditions.
Svetlana Blitshteyn (39:03):
Now we have the invasive vagus nerve stimulation procedures, and quite honestly, we don't want that to be the mainstream because you don't want to have a neurosurgery as you go to treatment. Of course, you want the non-invasive vagus nerve stimulation being the mainstream therapy. But I think a lot of research needs to happen and it's going to be a very much a multidisciplinary field where we'll have immunology, translational sciences, we'll have neurosurgeons like Kevin Tracey, we'll have rheumatologists, neurologists, cardiologists. We'll have a multidisciplinary collaborative group to further understand what's going on in these autoimmune inflammatory disorders, including those of post-infectious origin.
Eric Topol (40:02):
I certainly agree with all of your points there. I mean, I'm really struck now because the immune system is front and center with so much of what we're seeing with of course Long Covid, but also things like Alzheimer's and Parkinson's and across the board with metabolic diseases. And here we have this connection with your sweet spot of the autonomic nervous system, and we have these pathways that had not been delineated before. I didn't know too much about the cNST of the brain to be such an important connect point for this. And I wonder, so here's another example. Concurrently the glucagon-like peptide 1 (GLP-1) drugs have this pronounced effect on reducing inflammation in the body before the weight loss and in the brain through the gut-brain axis, as we recently discussed with Dan Drucker, have you ever tried a GLP-1 drug or noticed that GLP-1 drugs help people with Long Covid or the POTS problem?
Svetlana Blitshteyn (41:12):
So I have heard anecdotally people with Long Covid using these drugs for other reasons, saying I feel much better. In fact, I recently had a woman who said, I have never been more productive than I am now on this medication. And she used the word productive, which is important because non-productive implies so many things. It's the brain fog, it's the physical fatigue, it's the mental fatigue. So I think we are, first of all, I want to say, I always said that the brain is not separate from the body. And neurologic manifestations of systemic disease is a very big untapped area. And I think it's not going to be surprising for me to see that these drugs can improve many brain parameters and possibly even neuroinflammation. We don't know, but we certainly need to study this.
Eric Topol (42:15):
Yeah, it's interesting because statins had been tried for multiple sclerosis, I think maybe not with very clear cut benefit effects, but here you have a new class of drugs which eventually are going to be in pills and not just one receptor but triple receptor, much more potent than what we're seeing in the clinic today. And you wonder if we're onto an anti-inflammatory for the brain and body that could help in this. I mean, we have a crisis here with Long Covid in POTS without a remedy, without adequate resources that are being dedicated to the clinical trials that are so vital to execute and find treatments. And that's just one candidate of many. I mean, obviously there's so many possible ones on the list. So if you could design studies now based on your extraordinary rich experience with Long Covid and POTS, what would you go after right now? What do you think is the thing that's, would it be to evaluate more of these noninvasive, non-pharmacologic treatments like the vagal nerve stimulation, or are there particular drugs that you find intriguing?
Svetlana Blitshteyn (43:33):
Well, a few years ago we published a case series of patients with severe POTS and nothing helped them, but they improved significantly and some even made close to recovery improvement and were able to return to their careers because they were treated with immunotherapy. So the paper is a subcutaneous immunoglobulin and plasmapheresis and the improvement was remarkable. I say there was one physician there who could not start her residency. She got sick in medical school and could not start her residency due to severe POTS and no amount of beta blockers, Midodrine or Florinef helped her get out the house and out of bed. And therefore, sheer luck, she was able to get subcutaneous immunoglobulin and she improved significantly, finished her residency and is now a practicing physician. So I think when we have these cases, it's important to bring them to scientific community. And I think I'm very excited that hopefully soon we're going to have trials of immunotherapy and immunomodulating treatment options for patients with Long Covid and hopefully POTS in general, I believe in novel, but also repurposed, repurposed treatment.
Svetlana Blitshteyn (45:01):
IVIG has been used for decades, so it's not a new medication. And contrary to popular belief, it's actually quite safe. It is expensive, it's a blood product, but we are very familiar with it in medicine and neurology. So I think we have to look forward to everything. And as I tell my patients, I'm always aggressive with medications when they come to me and their doctor said something like, well, let's see, it's going to go away on its own or keep doing your salt and fluids intake or wear compression sucks. Well, they're already doing it. It's not helping. And now it's a good time to try everything we have. And I would like to have more. I would like to have immunotherapy available. I would like to have immunosuppressants even tried potentially, and maybe we'll be able to try medication for possible viral persistence. Let's see how that works out. We have other inflammatory modalities out there that can potentially give us the tools. You see, I think being that it's a multifactorial disorder, that I don't think it's going to be one thing for everyone. We need to have a toolbox where we're going to choose what's best for your specific case because when we talk about Long Covid, we have to remember there are many different phenotypes under that umbrella.
A Serious Matter
Eric Topol (46:40):
Now, before we wrap up, I mean I guess I wanted to emphasize how there are clinicians out there who discount Long Covid in POTS. They think it's something that is a figment of imagination. Now, on the other hand, you and I especially, you know that people are totally disabled. Certain days they can't even get out of bed, they can't get back to their work, their life. And this can go on and on as we've been discussing. So can you set it straight about, I mean, you are seeing these people every day. What do you have to say to our fellow colleague physicians who tend to minimize and say, this is extremely rare, if it even exists, and that these people have some type of psychiatric problem. And it's really, it's distressing of course, but could you speak to that?
Svetlana Blitshteyn (47:39):
Absolutely. So as I always say, Long Covid is not a psychiatric or psychological disorder, and it's also not a functional neurologic disorder. Now, having said that, as I just mentioned, brain is not separate from the body. And neurologic manifestations of systemic disease are numerous. We just had a paper out on neurologic manifestations of mast cell activation syndrome. So certainly some patients will develop psychiatric manifestations and some patients will develop major depression, anxiety, OCD or functional neurologic disorder. But those are complications of systemic disease, meaning that you cannot diagnose a patient with anxiety and send them off to a psychologist or a psychiatrist without diagnosing POTS and treating it. And in many cases, when you approach an underlying systemic disorder with the right medications, like dysautonomia for example, all of the symptoms including psychological and psychiatric, tend to improve as well. And certainly, there is going to be a small subset of Long Covid patients whose primary problem is psychiatric.
Svetlana Blitshteyn (49:01):
And I think that's totally fine. That is not to say that all Long Covid is psychiatric. Some will have significant psychiatric manifestations. I mean, there are cases of post Covid psychosis and autoimmune encephalitis and all kinds of psychiatric problems that people may develop, but I think we can't really stratify well, this is physiologic and this word functional that I'm not a fan of. This is physiologic as we see it on MRI. But here, because we don't see anything on MRI, it means you are fine and can just exercise your way out of it. So I think with this Long Covid, hopefully we'll get answers as to the pathophysiology, but also most importantly, hopefully we'll get these therapies that millions of people before Covid pandemic were looking for.
Eric Topol (50:02):
Well, I just want to thank you because you were onto this well over 10, 15 years before there was such a thing as Covid, you've dedicated your career to this. These are some of the most challenging patients to try to help and has to be vexing, that you can't get their symptoms resolved no less the underlying problem. And we're indebted to you, Svetlana, because you've really been ahead of the curve here. You were writing a patient book before there were such things as patient activists in Long Covid, as we've seen, which have been so many of the heroes of this whole problem. But thank you for all the work you do. We'll continue to follow. We learned from you about POTS and Long Covid from your work and really appreciate everything you've done. Thank you.
Svetlana Blitshteyn (50:58):
Thank you so much, Eric, for having me. As I said, it's a great honor for me to be here. Remarkable, amazing. And thank you for all this work that you're doing and being an advocate for our field because we always need great champions to help us move forward in these complicated disorders.
********************************
The Ground Truths newsletters and podcasts are all free, open-access, without ads.
Voluntary paid subscriptions all go to support Scripps Research. Many thanks for that—they greatly helped fund our summer internship programs for 2023 and 2024.
Thanks to my producer Jessica Nguyen and Sinjun Balabanoff tor audio and video support at Scripps Research
Note: you can select preferences to receive emails about newsletters, podcasts, or all I don’t want to bother you with an email for content that you’re not interested in.

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
Kate Crawford: A Leading Scholar and Conscience for A.I.
12 mai· Ground Truths
“We haven't invested this much money into an infrastructure like this really until you go back to the pyramids”—Kate Crawford
Transcript with links to audio and external links. Ground Truths podcasts are on Apple and Spotify. The video interviews are on YouTube
Eric Topol (00:06):
Well, hello, this is Eric Topol with Ground Truths, and I'm really delighted today to welcome Kate Crawford, who we're very lucky to have as an Australian here in the United States. And she's multidimensional, as I've learned, not just a scholar of AI, all the dimensions of AI, but also an artist, a musician. We're going to get into all this today, so welcome Kate.
Kate Crawford (00:31):
Thank you so much, Eric. It's a pleasure to be here.
Eric Topol (00:34):
Well, I knew of your work coming out of the University of Southern California (USC) as a professor there and at Microsoft Research, and I'm only now learning about all these other things that you've been up to including being recognized in TIME 2023 as one of 100 most influential people in AI and it's really fascinating to see all the things that you've been doing. But I guess I'd start off with one of your recent publications in Nature. It was a world view, and it was about generative AI is guzzling water and energy. And in that you wrote about how these large AI systems, which are getting larger seemingly every day are needing as much energy as entire nations and the water consumption is rampant. So maybe we can just start off with that. You wrote a really compelling piece expressing concerns, and obviously this is not just the beginning of all the different aspects you've been tackling with AI.
Exponential Growth, Exponential Concerns
Kate Crawford (01:39):
Well, we're in a really interesting moment. What I've done as a researcher in this space for a very long time now is really introduce a material analysis of artificial intelligence. So we are often told that AI is a very immaterial technology. It's algorithms in the cloud, it's objective mathematics, but in actual fact, it comes with an enormous material infrastructure. And this is something that I took five years to research for my last book, Atlas of AI. It meant going to the mines where lithium and cobalt are being extracted. It meant going into the Amazon fulfillment warehouses to see how humans collaborate with robotic and AI systems. And it also meant looking at the large-scale labs where training data is being gathered and then labeled by crowd workers. And for me, this really changed my thinking. It meant that going from being a professor for 15 years focusing on AI from a very traditional perspective where we write papers, we're sitting in our offices behind desks, that I really had to go and do these journeys, these field trips, to understand that full extractive infrastructure that is needed to run AI at a planetary scale.
(02:58):
So I've been keeping a very close eye on what would change with generative AI and what we've seen particularly in the last two years has been an extraordinary expansion of the three core elements that I really write about in Atlas, so the extraction of data of non-renewable resources, and of course hidden labor. So what we've seen, particularly on the resources side, is a gigantic spike both in terms of energy and water and that's often the story that we don't hear. We're not aware that when we're told about the fact that there gigantic hundred billion computers that are now being developed for the next stage of generative AI that has an enormous energy and water footprint. So I've been researching that along with many others who are now increasingly concerned about how we might think about AI more holistically.
Eric Topol (03:52):
Well, let's go back to your book, which is an extraordinary book, the AI Atlas and how you dissected not just the well power of politics and planetary costs, but that has won awards and it was a few years back, and I wonder so much has changed since then. I mean ChatGPT in late 2022 caught everybody off guard who wasn't into this knowing that this has been incubating for a number of years, and as you said, these base models are just extraordinary in every parameter you can think about, particularly the computing resource and consumption. So your concerns were of course registered then, have they gone to exponential growth now?
Kate Crawford (04:45):
I love the way you put that. I think you're right. I think my concerns have grown exponentially with the models. But I was like everybody else, even though I've been doing this for a long time and I had something of a heads up in terms of where we were moving with transformer models, I was also quite taken aback at the extraordinary uptake of ChatGPT back in November 2022 in fact, gosh, it still feels like yesterday it's been such an extraordinary timescale. But looking at that shift to a hundred million users in two months and then the sort of rapid competition that was emerging from the major tech companies that I think really took me by surprise, the degree to which everybody was jumping on the bandwagon, applying some form of large language model to everything and anything suddenly the hammer was being applied to every single nail.
(05:42):
And in all of that sound and fury and excitement, I think there will be some really useful applications of these tools. But I also think there's a risk that we apply it in spaces where it's really not well suited that we are not looking at the societal and political risks that come along with these approaches, particularly next token prediction as a way of generating knowledge. And then finally this bigger set of questions around what is it really costing the planet to build these infrastructures that are really gargantuan? I mean, as a species, we haven't invested this much money into an infrastructure like this really until you go back to the pyramids, you really got to go very far back to say that type of just gargantuan spending in terms of capital, in terms of labor, in terms of all of the things are required to really build these kinds of systems. So for me, that's the moment that we're in right now and perhaps here together in 2024, we can take a breath from that extraordinary 18 month period and hopefully be a little more reflective on what we're building and why and where will it be best used.
Propagation of Biases
Eric Topol (06:57):
Yeah. Well, there's so many aspects of this that I'd like to get into with you. I mean, one of course, you're as a keen observer and activist in this whole space, you've made I think a very clear point about how our culture is mirrored in our AI that is our biases, and people are of course very quick to blame AI per se, but it seems like it's a bigger problem than just that. Maybe you could comment about, obviously biases are a profound concern about propagation of them, and where do you see where the problem is and how it can be attacked?
Kate Crawford (07:43):
Well, it is an enormous problem, and it has been for many years. I was first really interested in this question in the era that was known as the big data era. So we can think about the mid-2000s, and I really started studying large scale uses of data in scientific applications, but also in what you call social scientific settings using things like social media to detect and predict opinion, movement, the way that people were assessing key issues. And time and time again, I saw the same problem, which is that we have this tendency to assume that with scale comes greater accuracy without looking at the skews from the data sources. Where is that data coming from? What are the potential skews there? Is there a population that's overrepresented compared to others? And so, I began very early on looking at those questions. And then when we had very large-scale data sets start to emerge, like ImageNet, which was really perhaps the most influential dataset behind computer vision that was released in 2009, it was used widely, it was freely available.
(09:00):
That version was available for over a decade and no one had really looked inside it. And so, working with Trevor Paglen and others, we analyzed how people were being represented in this data set. And it was really quite extraordinary because initially people are labeled with terms that might seem relatively unsurprising, like this is a picture of a nurse, or this is a picture of a doctor, or this is a picture of a CEO. But then you look to see who is the archetypical CEO, and it's all pictures of white men, or if it's a basketball player, it's all pictures of black men. And then the labeling became more and more extreme, and there are terms like, this is an alcoholic, this is a corrupt politician, this is a kleptomaniac, this is a bad person. And then a whole series of labels that are simply not repeatable on your podcast.
(09:54):
So in finding this, we were absolutely horrified. And again, to know that so many AI models had trained on this as a way of doing visual recognition was so concerning because of course, very few people had even traced who was using this model. So trying to do the reverse engineering of where these really problematic assumptions were being built in hardcoded into how AI models see and interpret the world, that was a giant unknown and remains to this day quite problematic. We did a recent study that just came out a couple of months ago looking at one of the biggest data sets behind generative AI systems that are doing text to image generation. It's called LAION-5B, which stands for 5 billion. It has 5 billion images and text captions drawn from the internet. And you might think, as you said, this will just mirror societal biases, but it's actually far more weird than you might imagine.
(10:55):
It's not a representative sample even of the internet because particularly for these data sets that are now trying to use the ALT tags that are used around images, who uses ALT tags the most on the internet? Well, it's e-commerce sites and it's often stock image sites. So what you'll see and what we discovered in our study was that the vast majority of images and labels are coming from sites like Shopify and Pinterest, these kind of shopping aspirational collection sites. And that is a very specific way of seeing the world, so it's by no means even a perfect mirror. It's a skewed mirror in multiple ways. And that's something that we need to think of particularly when we turn to more targeted models that might be working in say healthcare or in education or even in criminal justice, where we see all sorts of problems emerge.
Exploiting Humans for RLHF
Eric Topol (11:51):
Well, that's really interesting. I wonder to extend that a bit about the human labor side of this. Base models are tweaked, fine-tuned, and one of the ways to do that, of course is getting people to weigh in. And this has been written about quite a bit about how the people that are doing this can be exploited, getting wages that are ridiculously weak. And I wonder if you could comment about that because in the ethics of AI, this seems to be one of the many things that a lot of people don't realize about reinforcement learning.
Kate Crawford (12:39):
Oh, I completely agree. It's quite an extraordinary story. And of course now we have a new category of crowd labor that's called reinforcement learning with human feedback or RLHF. And what was discovered by multiple investigations was that these laborers are in many cases paid less than $2 an hour in very exploitative conditions, looking at results that in many cases are really quite horrifying. They could be accounts of murder, suicide, trauma, this can be visual material, it can be text-based material. And again, the workers in these working for these companies, and again, it's often contract labor, it's not directly within a tech company, it's contracted out. It's very hidden, it's very hard to research and find. But these laborers have been experiencing trauma and are really now in many cases bringing lawsuits, but also trying to unionize and say, these are not acceptable conditions for people to be working under.
(13:44):
So in the case of OpenAI, it was found that it was Kenyan workers who were doing this work for just poverty wages, but it's really across the board. It's so common now that humans are doing the hard work behind the scenes to make these systems appear autonomous. And that's the real trap that we're being told that this is the artificial intelligence. But in actual fact, what Jeff Bezos calls Mechanical Turk is that it's artificial, artificial intelligence otherwise known as human beings. So that is a very significant layer in terms of how these systems work that is often unacknowledged. And clearly these workers in many cases are muzzled from speaking, they're not allowed to talk about what they do, they can't even tell their families. They're certainly prevented from collective action, which is why we've seen this push towards unionization. And finally, of course, they're not sharing in any of the profits that are being generated by these extraordinary new systems that are making a very small number of people, very wealthy indeed.
Eric Topol (14:51):
And do you know if that's improving or is it still just as bad as it has been reported? It's really deeply concerning to see human exploitation, and we all know well about sweatshops and all that, but here's another version, and it's really quite distressing.
Kate Crawford (15:09):
It really is. And in fact, there have been several people now working to create really almost like fair work guidelines. So Oxford has the sort of fair work initiative looking specifically at crowd work. They also have a rating system where they rate all of the major technology companies for how well they're treating their crowd laborers. And I have to say the numbers aren't looking good in the last 12 months, so I would love to see much more improvement there. We are also starting to see legislation be tabled specifically on this topic. In fact, Germany was one of the most recent to start to explore how they would create a strong legislative backing to make sure that there's fair labor conditions. Also, Chile was actually one of the first to legislate in this space, but you can imagine it's very difficult to do because it's a system that is operating under the radar through sort of multiple contracted chains. And even some of the people within tech companies will tell me it's really hard to know if they're working with a company that's doing this in the right way and paying people well. But frankly, I'd like to see far greater scrutiny otherwise, as you say, we're building on this system, which looks like AI sweatshops.
Eric Topol (16:24):
Yeah, no, I think people just have this illusion that these machines are doing everything by themselves, and that couldn't be further from the truth, especially when you're trying to take it to the next level. And there's only so much human content you can scrape from the internet, and obviously it needs additional input to take it to that more refined performance. Now, besides your writing and being much of a conscience for AI, you're also a builder. I mean, I first got to know some of your efforts through when you started the AI Now Institute. Maybe you can tell us a bit about that. Now you're onto the Knowing Machines Project and I don't know how many other projects you're working on, so maybe you can tell us about what it's like not just to be a keen observer, but also one to actually get initiatives going.
Kate Crawford (17:22):
Well, I think it's incredibly important that we start to build interdisciplinary coalitions of researchers, but sometimes even beyond the academic field, which is where I really initially trained in this space, and really thinking about how do we involve journalists, how do we involve filmmakers, how do we involve people who will look at these issues in really different ways and tell these stories more widely? Because clearly this really powerful shift that we're making as a society towards using AI in all sorts of domains is also a public issue. It's a democratic issue and it's an issue where we should all be able to really see into how these systems are working and have a say in how they'll be impacting our lives. So one of the things that I've done is really create research groups that are interdisciplinary, starting at Microsoft Research as one of the co-founders of FATE, a group that stands for fairness, accountability, transparency and ethics, and then the AI Now Institute, which was originally at NYU, and now with Knowing Machines, which is an international group, which I've been really delighted to build, rather than just purely focusing on those in the US because of course these systems are inherently transnational, they will be affecting global populations.
(18:42):
So we really need to think about how do you bring people from very different perspectives with different training to ask this question around how are these systems being built, who is benefiting and who might be harmed, and how can we address those issues now in order to actually prevent some of those harms and prevent the greatest risks that I see that are possible with this enormous turn to artificial intelligence everywhere?
Eric Topol (19:07):
Yeah, and it's interesting how you over the years are a key advisor, whether it's the White House, the UN or the European Parliament. And I'm curious about your experience because I didn't know much about the Paris ENS. Can you tell us about you were Visiting Chair, this is AI and Justice at the École Normale Supérieure (ENS), I don’t know if I pronounce that right. My French is horrible, but this sounds like something really interesting.
Kate Crawford (19:42):
Well, it was really fascinating because this was the first time that ENS, which is really one of the top research institutions in Europe, had turned to this focus of how do we contend with artificial intelligence, not just as a technical question, but as a sort of a profound question of justice of society of ethics. And so, I was invited to be the first visiting chair, but tragically this corresponded with the start of the pandemic in 2020. And so, it ended up being a two-year virtual professorship, which is really a tragedy when you’re thinking about spending time in Paris to be spending it on Zoom. It’s not quite the same thing, but I had the great fortune of using that time to assemble a group of scholars around the world who were looking at these questions from very different disciplines. Some were historians of science, others were sociologists, some were philosophers, some were machine learners.
(20:39):
And really essentially assembled this group to think through some of the leading challenges in terms the potential social impacts and current social impacts of these systems. And so, we just recently published that through the academies of Science and Engineering, and it’s been almost like a template for thinking about here are core domains that need more research. And interestingly, we’re at that moment, I think now where we can say we have to look in a much more granular fashion beyond the hype cycles, beyond the sense of potential, the enormous potential upside that we’re always hearing about to look at, okay, how do these systems actually work now? What kinds of questions can we bring into the research space so that we’re really connecting the ideas that come traditionally from the social sciences and the humanistic disciplines into the world of machine learning and AI design. That’s where I see the enormous upside that we can no longer stay in these very rigorously patrolled silos and to really use that interdisciplinary awareness to build systems differently and hopefully more sustainably as well.
Is Working At Microsoft A Conflict?
Eric Topol (21:55):
Yeah, no, that’s what I especially like about your work is that you’re not a doomsday person or force. You’re always just trying to make it better, but now that's what gets me to this really interesting question because you are a senior principal researcher at Microsoft and Microsoft might not like some of these things that you're advocating, how does that potential conflict work out?
Kate Crawford (22:23):
It's interesting. I mean, people often ask me, am I a technology optimist or a technology pessimist? And I always say I'm a technology realist, and we're looking at these systems being used. I think we are not benefited by discourses of AI doomerism nor by AI boosterism. We have to assess the real politic and the political economies into which these systems flow. So obviously part of the way that I've got to know what I know about how systems are designed and how they work at scale is through being at Microsoft Research where I'm working alongside extraordinary colleagues and all of whom come from, in many cases, professorial backgrounds who are deep experts in their fields. And we have this opportunity to work together and to look at these questions very early on in the kinds of production cycles and enormous shifts in the way that we use technology.
(23:20):
But it is interesting of course that at the moment Microsoft is absolutely at the leading edge of this change, and I've always thought that it's incredibly important for researchers and academics who are in industrial spaces to be able to speak freely, to be able to share what they see and to use that as a way that the industry can, well hopefully keep itself honest, but also share between what it knows and what everybody else knows because there's a giant risk in having those spaces be heavily demarcated and having researchers really be muzzled. I think that's where we see real problems emerge. Of course, one of the great concerns a couple of years ago was when Timnit Gebru and others were fired from Google for speaking openly about the concerns they had about the first-generation large language models. And my hope is that there's been a lesson through that really unfortunate set of decisions made at Google that we need people speaking from the inside about these questions in order to actually make these systems better, as you say, over the medium and long term.
Eric Topol (24:26):
Yeah, no, that brings me to thought of Peter Lee, who I'm sure because he wrote a book about GPT-4 and healthcare and was very candid about its potential, real benefits and the liabilities, and he's a very humble kind of guy. He's not one that has any bravado that I know of, so it speaks well to at least another colleague of yours there at Microsoft and their ability to see all the different sides here, not just what we'll talk about in a minute the arms race both across companies and countries. But before I get to that, there's this other part of you and I wonder if there's really two or three of you that is as a composer of music and art, I looked at your Anatomy of an AI System, I guess, which is on exhibit at the Museum of Modern Art (MoMA) in New York, and that in itself is amazing, but how do you get into all these other parts, are these hobbies or is this part of a main part of your creative work or where does it fit in?
Kate Crawford (25:40):
Eric, didn't I mention the cloning program that I participated in early and that there are many Kate’s and it's fantastic we all work together. Yeah, that explains it. Look, it's interesting. Way back as a teenager, I was fascinated with technology. Of course, it was the early stages of the web at that moment, and I could see clearly that this was, the internet was going to completely change everything from my generation in terms of what we would do in terms of the way that we would experience the world. And as I was also at that time an electronic musician in bands, I was like, this was a really fantastic combination of bringing together creative practice with a set of much larger concerns and interests around at a systems level, how technology and society are co-constituted, how they evolve together and shape each other. And that’s really been the map of how I’ve always worked across my life.
(26:48):
And it’s interesting, I've always collaborated with artists and Vladan Joler who I worked with on anatomy of an AI system. We actually met at a conference on voice enabled AI systems, and it was really looking at the ethics of could it be possible to build an open source, publicly accessible version of say Alexa rather than purely a private model owned by a corporation, and could that be done in a more public open source way? And we asked a different question, we looked at each other and we're like, oh, I haven't met you yet, but I can see that there are some problems here. One of them is it's not just about the data and it's not just about the technical pipelines, it's about where the components come from. It's about the mining structures that needed to make all of these systems. It's about the entire end of life what happens when we throw these devices out from generally between three to four years of use and how they go into these giant e-waste tips.
(27:51):
And we basically started looking at this as an enormous sort of life and death of a single AI system, which for us started out by drawing these things on large pieces of butcher's paper, which just expanded and expanded until we had this enormous systems level analysis of what it takes just to ask Alexa what the weather is today. And in doing that, it taught me a couple of things. One that people really want to understand all of the things that go into making an AI system work. This piece has had a very long life. It's been in over a hundred museums around the world. It's traveled further than I have, but it's also very much about that broader political economy that AI systems aren't neutral, they don't just exist to serve us. They are often sort of fed into corporate structures that are using them to generate profits, and that means that they're used in very particular ways and that there are these externalities in terms of how they produced that linger in our environments that have really quite detrimental impacts on systems of labor and how people are recompensed and a whole range of relationships to how data is seen and used as though it's a natural resource that doesn't actually come from people's lives, that doesn't come with risks attached to it.
(29:13):
So that project was really quite profound for me. So we've continued to do these kinds of, I would call them research art projects, and we just released a new one called Calculating Empires, which looks at a 500 year history of technology and power looking specifically at how empires over time have used new technologies to centralize their power and expand and grow, which of course is part of what we're seeing at the moment in the empires of AI.
Eric Topol (29:43):
And what about the music side?
Kate Crawford (29:45):
Well, I have to say I've been a little bit slack on the music side. Things have been busy in AI Eric, I have to say it's kept me away from the music studio, but I always intend to get back there. Fortunately, I have a kid who's very musical and he's always luring me away from my desk and my research saying, let’s write some music. And so, he'll keep me honest.
Geopolitics and the Arms Races
Eric Topol (30:06):
Well, I think it's striking just because you have this blend of the humanities and you're so deep into trying to understand and improve our approaches in technology. And it seems like a very unusual, I don't know, too many techies that have these different dimensions, so that's impressive. Now let's get back to the arms race. You just were talking about tracing history over hundreds of years and empires, but right now we have a little problem. We have the big tech titans that are going after each other on a daily basis, and of course you know the group very well. And then you have China and the US that are vying to be the dominant force and problems with China accessing NVIDIA chips and Taiwan sitting there in a potentially very dangerous position, not just for Taiwan, but also for the US. And I wonder if you could just give us your sense about the tensions here. They're US based as well of course, because that's some of the major forces in companies, but then they're also globally. So we have a lot of stuff in the background that people don't like to think about, but it's actually happening right now.
Kate Crawford (31:35):
I think it's one of the most important things that we can focus on, in fact. I mean and again, this is why I think a materialist analysis of artificial intelligence is so important because not only does it force you to look at the raw components, where does the energy come from? Where does the water come from? But it means you're looking at where the chipsets come from. And you can see that in many cases there are these infrastructural choke points where we are highly dependent on specific components that sit within geopolitical flashpoints. And Taiwan is really the exemplar of this sort of choke point at the moment. And again, several companies are trying to address this by spinning up new factories to build these components, but this takes a lot of time and an enormous amount of resources yet again. So what we're seeing is I think a very difficult moment in the geopolitics of artificial intelligence.
(32:31):
What we've had certainly for the last decade has been almost a geopolitical duopoly. We've had the US and China not only having enormous power and influence in this space, but also goading each other into producing the most extreme forms of both data extractive and surveillance technologies. And unfortunately, this is just as true in the United States that I commonly hear this in rooms in DC where you'll hear advisors say, well, having any type of guardrails or ethical considerations for our AI systems is a problem if it means that China's going to do it anyway. And that creates this race to the bottom dynamic of do as much of whatever you can do regardless of the ethical and in some cases legal problems that will create. And I think that's been the dynamic that we've seen for some time. And of course the last 18 months to two years, we've seen that really extraordinary AI war happening internally in the United States where again, this race dynamic I think does create unfortunately this tendency to just go as fast as possible without thinking about potential downsides.
(33:53):
And I think we're seeing the legacy of that right now. And of course, a lot of the conversations from people designing these systems are now starting to say, look, being first is great, but we don’t want to be in a situation as we saw recently with Google’s Gemini where you have to pull an entire model off the shelves and you have to say, this is not ready. We actually have to remove it and start again. So this is the result I think of that high pressure, high speed dynamic that we’ve been seeing both inside the US but between the US and China. And of course, what that does to the rest of the world is create this kind of client states where we've got the EU trying to say, alright, well we'll export a regulatory model if we're not going to be treated as an equivalent player here. And then of course, so many other countries who are just seen as spaces to extract low paid labor or the mineralogical layer. So that is the big problem that I see is that that dynamic has only intensified in recent years.
A.I. and Medicine
Eric Topol (34:54):
Yeah, I know it's really another level of concern and it seems like it could be pretty volatile if for example, if the US China relations takes another dive and the tensions there go to levels that haven't been seen so far. I guess the other thing, there's so much that is I think controversial, unsettled in this space and so much excitement. I mean, just yesterday for example, was the first AI randomized trial to show that you could save lives. When I wrote that up, it was about the four other studies that showed how it wasn't working. Different studies of course, but there's so much excitement at the same time, there's deep concerns. You've been a master at articulating these deep concerns. What have we missed in our discussion today, I mean we've covered a lot of ground, but what do you see are other things that should be mentioned?
Kate Crawford (36:04):
Well, one of the things that I've loved in terms of following your work, Eric, is that you very carefully walk that line between allowing the excitement when we see really wonderful studies come out that say, look, there's great potential here, but also articulating concerns where you see them. So I think I'd love to hear, I mean take this opportunity to ask you a question and say what's exciting you about the way that this particularly new generation AI is being used in the medical context and what are the biggest concerns you have there?
Eric Topol (36:35):
Yeah, and it's interesting because the biggest advance so far in research and medicine was the study yesterday using deep learning without any transformer large language model effort. And that's where that multiplicative of opportunity or potential is still very iffy, it's wobbly. I mean, it needs much more refinement than where we are right now. It's exciting because it is multimodal and it brings in the ability to bring all the layers of a human being to understand our uniqueness and then do much better in terms of, I got a piece coming out soon in Science about medical forecasting and how we could really get to prevention of conditions that people are at high risk. I mean like for example today the US preventive task force said that all women age 40 should have mammograms, 40.
Kate Crawford (37:30):
I saw that.
Eric Topol (37:30):
Yeah, and this is just crazy Looney Tunes because here we have the potential to know pretty precisely who are those 12%, only 12% of women who would ever get breast cancer in their lifetime, and why should we put the other 88% through all this no less the fact that there are some women even younger than age 40 that have significantly high risk that are not picked up. But I do think eventually when we get these large language models to actualize their potential, we'll do really great forecasting and we'll be able to not just prevent or forestall cancer, Alzheimer’s and so many things. It's quite exciting, but it's the earliest, we're not even at first base yet, but I think I can see our way to get there eventually. And it's interesting because the discussion I had previously with Geoffrey Hinton, and I wonder if you think this as well, that he sees the health medical space as the only really safe space. He thinks most everything else has got more concerns about the downsides is the sweet spot as he called it. But I know that's not particularly an area that you are into, but I wonder if you share that the excitement about your health could be improved in the future with AI.
Kate Crawford (38:52):
Well, I think it's a space of enormous potential, but again, enormous risk for the same reasons that we discussed earlier, which is we have to look at the training data and where it's coming from. Do we have truly representative sources of data? And this of course has been a consistent problem certainly for the last hundred years and longer. When we look at who are the medical patients whose data is being collected, are we seeing skews? And that has created all sorts of problems, particularly in the last 50 years in terms of misdiagnosing women, people of color, missing and not taking seriously the health complaints of people who are already seen as marginalized populations, thus then further skewing the data that is then used to train AI models. So this is something that we have to take very seriously, and I had the great fortune of being invited by Francis Collins to work with the NIH on their AI advisory board.
(39:50):
They produced a board to look just at these questions around how can this moment in AI be harnessed in such a way that we can think about the data layer, think about the quality of data and how we train models. And it was a really fascinating sort of year long discussion because in the room we had people who were just technologists who just wanted as much data as possible and just give us all that data and then we'll do something, but we'll figure it out later. Then there were people who had been part of the Human Genome Project and had worked with Francis on questions around the legal and ethical and social questions, which he had really centered in that project very early on. And they said, no, we have to learn these lessons. We have to learn that data comes from somewhere. It's not divorced of context, and we have to think about who's being represented there and also who's not being represented there because that will then be intensified in any model that we train on that data.
Humans and Automation Bias
(40:48):
And then also thinking about what would happen in terms of if those models are only held by a few companies who can profit from them and not more publicly and widely shared. These were the sorts of conversations that I think at the absolute forefront in terms of how we're going to navigate this moment. But if we get that right, if we center those questions, then I think we have far greater potential here than we might imagine. But I'm also really cognizant of the fact that even if you have a perfect AI model, you are always going to have imperfect people applying it. And I'm sure you saw that same study that came out in JAMA back in December last year, which was looking at how AI bias, even slightly biased models can worsen human medical diagnosis. I don’t know if you saw this study, but I thought it was really extraordinary.
(41:38):
It was sort of 450 doctors and physician's assistants and they were really being shown a handful of cases of patients with acute respiratory failure and they really needed come up with some sort of diagnosis and they were getting suggestions from an AI model. One model was trained very carefully with highly accurate data, and the other was a fairly shoddy, shall we say, AI model with quite biased data. And what was interesting is that the clinicians when they were working with very well-trained AI model, we're actually producing a better diagnosis across the board in terms of the cases they were looking at. I think their accuracy went up by almost 4.5 percentage points, but when they were working with the less accurate model, their capacity actually dropped well below their usual diagnostic baseline, something like almost 12 percentage points below their usual diagnostic quality. And so, this really makes me think of the kind of core problem that's been really studied for 40 years by social scientists, which is called automation bias, which is when even an expert, a technical system which is giving a recommendation, our tendency is to believe it and to discard our own knowledge, our own predictions, our own sense.
(42:58):
And it's been tested with fighter pilots, it's been tested with doctors, it's been tested with judges, and it's the same phenomenon across the board. So one of the things that we're going to need to do collectively, but particularly in the space of medicine and healthcare, is retaining that skepticism, retaining that ability to ask questions of where did this recommendation come from with this AI system and should I trust it? What was it trained on? Where did the data come from? What might those gaps be? Because we're going to need that skepticism if we're going to get through particularly this, as you say, this sort of early stage one period where in many cases these models just haven't had a lot of testing yet and people are going to tend to believe them out of the box.
The Large Language Model Copyright Issue
Eric Topol (43:45):
No, it's so true. And one of the key points is that almost every study that's been published in large language models in medicine are contrived. They're using patient actors or they're using case studies, but they're not in the real world. And that's where you have to really learn, as you know, that's a much more complex and messy world than the in silico world of course. Now, before wrapping up, one of the things that's controversial we didn't yet hit is the fact that in order for these base models to get trained, they basically ingest all human content. So they've ingested everything you've ever written, your books, your articles, my books, my articles, and you have the likes of the New York Times suing OpenAI, and soon it's going to run out of human content and just use synthetic content, I guess. But what's your sense about this? Do you feel that that's trespassing or is this another example of exploiting content and people, or is this really what has to be done in order to really make all this work?
Kate Crawford (44:59):
Well, isn't it a fascinating moment to see this mass grabbing of data, everything that is possibly extractable. I actually just recently published an article in Grey Room with the legal scholar, Jason Schultz, looking at how this is producing a crisis in copyright law because in many ways, copyright law just cannot contend with generative AI in particular because all of the ways in which copyright law and intellectual property more broadly has been understood, has been premised around human ideas of providing an incentive and thus a limited time monopoly based on really inspiring people to create more things. Well, this doesn't apply to algorithms, they don't respond to incentives in this way. The fact that, again, it's a longstanding tradition in copyright that we do not give copyright to non-human authors. So you might remember that there was a very famous monkey selfie case where a monkey had actually stepped on a camera and it had triggered a photograph of the monkey, and could this actually be a copyright image that could be given to the monkey?
(46:12):
Absolutely not, is what the court's decided. And the same has now happened, of course, for all generative AI systems. So right now, everything that you produce be that in GPT or in Midjourney or in Stable Diffusion, you name it, that does not have copyright protections. So we're in the biggest experiment of production after copyright in world history, and I don't think it's going to last very long. To be clear, I think we're going to start to see some real shifts, I think really in the next 6 to 12 months. But it has been this moment of seeing this gigantic gap in what our legal structures can do that they just haven't been able to contend with this moment. The same thing is true, I think, of ingestion, of this capturing of human content without consent. Clearly, many artists, many writers, many publishing houses like the New York Times are very concerned about this, but the difficulty that they're presented with is this idea of fair use, that you can collect large amounts of data if you are doing something with that, which is sufficiently transformative.
(47:17):
I'm really interested in the question of whether or not this does constitute sufficiently transformative uses. Certainly if you looked at the way that large language models a year ago, you could really prompt them into sharing their training data, spitting out entire New York Times articles or entire book chapters. That is no longer the case. All of the major companies building these systems have really safeguarded against that now but nonetheless, you have this question of should we be moving towards a system that is based on licensing, where we're really asking people if we can use their data and paying them a license fee? You can see how that could absolutely work and would address a lot of these concerns, but ultimately it will rely on this question of fair use. And I think with the current legal structures that we have in the current case law, that is unlikely to be seen as something that's actionable.
(48:10):
But I expect what we'll look at is what really happened in the early 20th century around the player piano, which was that I'm sure you remember this extraordinary technology of the player piano. That was one of the first systems that automated the playing of music and you'd have a piano that had a wax cylinder that almost like code had imprinted on a song or a piece of music, and it could be played in the public square or in a bar or in a saloon without having to pay a single artist and artists were terrified. They were furious, they were public hearings, there were sort of congressional hearings and even a Supreme Court case that decided that this was not a copyright infringement. This was a sufficiently transformative use of a piece of music that it could stand. And in the end, it was actually Congress that acted.
(49:01):
And we from that got the 1908 Copyright Act and from that we got this idea of royalties. And that has become the basis of the music industry itself for a very long time. And now we're facing another moment where I think we have a legislative challenge. How would you actually create a different paradigm for AI that would recognize a new licensing system that would reward artists, writers, musicians, all of the people whose work has been ingested into training data for AI so that they are recognized and in some ways, recompensed by this massive at scale extraction?
Eric Topol (49:48):
Wow, this has been an exhilarating conversation, Kate. I've learned so much from you over the years, but especially even just our chance to talk today. You articulate these problems so well, and I know you're working on solutions to almost everything, and you're so young, you could probably make a difference in the decades ahead. This is great, so I want to thank you not just for the chance to visit today, but all the work that you've been doing, you and your colleagues to make AI better, make it fulfill the great promise that it has. It is so extraordinary, and hopefully it'll deliver on some of the things that we have big unmet needs, so thanks to you. This has really been fun.
Kate Crawford (50:35):
This has been wonderful. And likewise, Eric, your work has just been a fantastic influence and I've been delighted to get to know you over the years and let's see what happens. It's going to be a wild ride from now to who knows when.
Eric Topol (50:48):
No question, but you'll keep us straight, I know that. Thank you so much.
Kate Crawford (50:52):
Thanks so much, Eric.
*******************************
Your support of subscribing to Ground Truths, and sharing it with your network of friends and colleagues, is much appreciated.
The Ground Truths newsletters and podcasts are all free, open-access, without ads.
Voluntary paid subscriptions all go to support Scripps Research. Many thanks for that—they greatly helped fund our summer internship programs for 2023 and 2024.
Thanks to my producer Jessica Nguyen and Sinjun Balabanoff tor audio and video support at Scripps Research
Note: you can select preferences to receive emails about newsletters, podcasts, or all I don’t want to bother you with an email for content that you’re not interested in.
Comments for this post are welcome from all subscribers.

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
Akiko Iwasaki: The Immunology of Covid and the Future
4 mai· Ground Truths
If there’s one person you’d want to talk to about immunology, the immune system and Covid, holes in our knowledge base about the complex immune system, and where the field is headed, it would be Professor Iwasaki. And add to that the topic of Women in Science. Here’s our wide-ranging conversation.
A snippet of the video, Full length Ground Truths videos are posted here and you can subscribe.
Ground Truths is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.
Transcript with many external link and links to the audio, recorded 30 April 2024
Eric Topol (00:06):
Hello, it's Eric Topol and I'm really thrilled to have my friend Akiko Iwasaki from Yale, and before I start talking with Akiko, I just want to mention there aren't too many silver linings of the pandemic, but one for me was getting to know Professor Iwasaki. She is my go-to immunologist. I've learned so much from her over the last four years and she's amazing. She just, as you may know, she was just recently named one of the most influential people in the world by TIME100. [and also recognized this week in TIME 100 Health]. And besides that, she's been elected to the National Academy of Medicine, National Academy of Sciences. She's the president of the American Association of Immunologists and she's a Howard Hughes principal investigator. So Akiko, it's wonderful to have you to join into an extended discussion of things that we have of mutual interest.
Akiko Iwasaki (01:04):
Thank you so much, Eric, for having me. I equally appreciate all of what you do, and I follow your blog and tweets and everything. So thank you Eric.
Eric Topol (01:14):
Well, you are a phenom. I mean just, that's all I can say because I think it was so appropriate that TIME recognize your contributions, not just over the pandemic, but of course throughout your career, a brilliant career in immunology. I thought we'd start out with our topic of great interest on Long Covid. You've done seminal work here and this is an evolving topic obviously. I wonder what your latest thoughts are on the pathogenesis and where things are headed.
Long Covid
Akiko Iwasaki (01:55):
Yeah, so as I have been saying throughout the pandemic, I think that Long Covid is not one disease. It's a collection of multiple diseases and that are sort of ending up in similar sets of symptoms. Obviously, there are over 200 symptoms and not everyone has the same set of symptoms, but what we are going for is trying to understand the disease drivers, so persistent viral infection is one of them. There are overwhelming evidence for that theory now, all the way from autopsy and biopsy studies to looking at peripheral blood RNA signatures as well as circulating spike protein and nucleocapsid proteins that are detected in people with Long Covid. Now whether that persistent virus or remnants of virus is driving the disease itself is unclear still. And that's why trials like the one that we are engaging with Harlan Krumholz on Paxlovid should tell us what percentage of the people are suffering from that type of driver and whether antivirals like Paxlovid might be able to mitigate those. If I may, I'd like to talk about three other hypotheses.
Eric Topol (03:15):
Yeah, I'd love for you to do that.
Akiko Iwasaki (03:18):
Okay, great. So the second hypothesis that we've been working on is autoimmune disease. And so, this is clearly happening in a subset of people, again, it's a heterogeneous disease, but we can actually not only look at reactogenicity of antibodies from people with Long Covid where we can transfer IgG from patients with Long Covid into an animal, a healthy animal, and really measure outcomes of a pathogenesis. So that's a functional evidence that antibodies in some people with Long Covid is really actually causing some of the damages that are occurring in vivo. And the third hypothesis is the reactivation of herpes viruses. So many of us adults have multiple latent herpes virus family members that are just dormant and are not really causing any pathologies. But in people with Long Covid, we're seeing elevated reactivation of viruses like Epstein-Barr virus (EBV) or Varicella-zoster virus (VZV) and that may again be just a signature of Long Covid, but it may also be driving some of the symptoms that people are suffering from.
(04:32):
So that's again, we see the signature over and over, not just our group, but multiple other groups, Michael Peluso's group, Jim Heath, and many others. So that's also an emerging evidence from multiple groups showing that. And finally, we think that inflammation that occurs during the acute phase can sort of chronically change some tissue tone. For instance, in the brain with Michelle Monje’s team, we developed a sort of localized mild Covid model of infection and showed that changes in microglia can be seen seven weeks post infection even though the virus is completely gone. So that means that inflammation that's established as a result of this initial infection can have prolonged sequence and sequela within the person and that may also be driving disease. And Eric, the reason we need to understand these diseases separately is because not only for diagnostic purposes, but for therapeutic purposes because to target a persistent virus is very different approach from targeting autoantibodies, for example.
Eric Topol (05:49):
Well, that's great. There's a lot to unpack there as you laid out four distinct paths that could result in the clinical syndrome and sequelae. I think you know I had the chance to have a really fun conversation with Michelle about their joint work that you've done, and she reminded me how she made a cold call to you to start as a collaboration, which I thought was fantastic. Look what that yielded. But yeah, this is fascinating because as I think you're getting at is that it may not be the same pathogenesis in any given individual so that all these, and even others might be operative. I guess maybe I first delve into the antibody story as you're well aware, we see after people get Covid a higher rate of autoimmune diseases crop up, which is really interesting because it seems to rev up self-directed immune response. And this I think many people haven't really noted yet, although obviously you're well aware of this, it's across all the different autoimmune diseases, connective tissue disease, not just one in particular. And it's, as you say, the idea that you could take the blood from a person suffering from Long Covid and give it to an experimental animal model and be able to recapitulate some of the abnormalities, it's really pretty striking. So the question I guess is if you were to do plasmapheresis and try to basically expunge these autoantibodies, wouldn't you expect people to have some symptomatic benefit pretty rapidly or is it just that the process is already far from the initiating step?
Akiko Iwasaki (07:54):
That's a great question. Plasmapheresis may be able to transiently improve the person if they're suffering from these autoantibody mediated diseases. People have reported, for example, IVIG treatment has dramatically improved their symptoms, but not in everybody. So it's really critical to understand who's suffering from this particular driver and appropriately treat those people. And there are many other very effective therapies in autoimmune disease field that can be repurposed for treating these patients as well.
Eric Topol (08:34):
The only clinical trial that has clicked so far, interestingly, came out of Hong Kong with different types of ways to manipulate the gut microbiome, which again, you know better than me is a major modulator of our immune system response. What are your thoughts about taking advantage of that way to somehow modulate this untoward immune response in people with this condition?
Akiko Iwasaki (09:07):
Yeah, so that is an exciting sort of development, and I don't mean to discount the importance of microbiome at all. It's just the drivers that are mentioning are something that can be directly linked to disease, but certainly dysbiosis and translocation of metabolites and microbiome itself could trigger Long Covid as well. So it's something that we're definitely keeping our eyes on. And as you say, Eric, the immune system is in intimate contact with the gut microbiome and also the gut is intimate contact with the brain. So there's a lot of connections that we really need to be paying attention to. So yeah, absolutely. This is a very exciting development.
Eric Topol (09:57):
And it is intriguing of course, the reactivation of viruses. I mean, we’ve learned in recent years how important EBV is in multiple sclerosis (MS). The question I have for you on that pathway, is this just an epiphenomena or do you actually think that could be a driving force in some people?
Akiko Iwasaki (10:19):
Yeah, so that's really hard to untangle in people. I mean, David Putrino and my team we're planning a clinical trial using Truvada. Truvada obviously is an HIV drug, but it has reported antiviral activity to Epstein-Barr virus (EBV) and others. So potentially we can try to interrogate that in people, but we're also developing mouse models that can sort of recapitulate EBV like viral reactivation and to see whether there's any sort of causal link between the reactivation and disease process.
Eric Topol (10:57):
Right now, recently there's been a bunch of anecdotes of people who get the glucagon-like peptide one (GLP-1) drugs which have a potent anti-inflammatory, both systemic and in the brain. I'd love to test these drugs, but of course these companies that make them or have other interests outside of Long Covid, do you think there's potential for a drug like that?
Akiko Iwasaki (11:23):
Yeah, so those drugs seem to have a lot of miraculous effects on every disease. So obviously it has to be used carefully because many people with Long Covid have issues with liver functions and other existing conditions that may or may not be conducive to taking those types of GLP-1 agonists. But in subset of people, maybe this can be tried, especially due to the anti-inflammatory properties, it may benefit again, a subset of people. I don't expect a single drug to cure everyone. That would be pretty amazing, but unlikely.
Eric Topol (12:09):
Absolutely. And it's unfortunate we are not further along in this whole story of clinical trials, testing treatments and applauding your efforts with my friend Harlan there to get into the testing which we had hoped RECOVER was going to do with their more than billion dollars or allocation, which didn't get us too far in that. Now before we leave Long Covid, which we could speak about for hours, I mean it's so darn important because so many people are really out there disabled or suffering on a daily basis or periodically they get better and then get worse again. There's been this whole idea that, oh, it's going away and that reinfections don't pose a threat. Maybe you could straighten that story out because I think there seems to be some miscues about the risk of Long Covid even as we go along with the continued circulating virus.
Akiko Iwasaki (13:11):
Right, so when you look at the epidemiological evidence of Long Covid, clearly in the beginning when we had no vaccines, no antivirals, no real good measure against Covid, the incident of developing Long Covid per infection was higher than a current date where we do have vaccines and Omicron may have changed its property significantly. So if you compare, let's say the Delta period versus Omicron period, there seems to be a reduced risk per infection of Long Covid. However, Omicron is super infectious. It's infected millions of people, and if you look at the total number of people suffering from Long Covid, we're not seeing a huge decline there at all because of the transmissibility of Omicron. So I think it's too early for us to say, okay, the rates are declining, we don't need to worry about it. Not at all, I think we still have to be vigilant.
(14:14):
We need to be up to date on vaccines and boosters because those seem to reduce the risk for Long Covid and whether Paxlovid can reduce the rate of Long Covid at the acute phase for the high risk individual, it seems to be yes, but for people who are not at high risk may or may not be very effective. So again, we just need to be very cautious. It's difficult obviously, to be completely avoiding virus at this time point, but I think masking and anything you can do, vaccination boosters is going to be helpful. And a reinfection does carry risk for developing Long Covid. So that prior infection is not going to prevent Long Covid altogether, even though the risk may be slightly reduced in the first infection. So when you think about these risks, again we need to be cognizant that reinfection and some people have multiple infections and then eventually get Long Covid, so we're just not safe from Long Covid yet.
Nasal Vaccines and Mucosal Immunity
Eric Topol (15:24):
Right. No, I think that's the problem is that people have not acknowledged that there's an ongoing risk and that we should continue to keep our guard up. I want to applaud you and your colleagues. You recently put out [Yale School of Public Health] this multi-panel about Covid, which we'll post with this podcast that gave a lot of the facts straight and simple diagrams, and I think this is what you need is this is kind of like all your threads on Twitter. . They're always such great educational ways to get across important information. So now let's go onto a second topic of great mutual interest where you've also been a leader and that's in the mucosal nasal vaccine story. I had the privilege of writing with you a nice article in Science Immunology back in 2022 about Operation Nasal Vaccine, and unfortunately we don't have a nasal vaccine. We need a nasal vaccine against Covid. Where do we stand with this now?
Akiko Iwasaki (16:31):
Yeah, so you're right. I mean nasal vaccines, I don't really know what the barrier is because I think the preclinical models all support the effectiveness against transmission and infection and obviously disease. And there is a White House initiative to support rapid development of next generation vaccine, which includes mucosal vaccine, so perhaps that's sort of pushing some of these vaccine candidates forward. You're probably more familiar than me about those kinds of events that are happening. But yeah, it's unfortunate that we don't have an approved mucosal booster vaccine yet, and our research has shown that as simple as a spray of recombinant spike protein without any adjuvants are able to restimulate immune response and then establish mucosal immunity in the nasal cavity, which goes a long way in preventing infection as well as transmission. So yeah, I mean I'm equally frustrated that things like that don't exist yet.
The Neomycin and Neosporin Surprise
Eric Topol (17:52):
Well, I mean the work that you and many other groups around the world have published on this is so compelling and this is the main thing that we don't have now, which is a way to prevent infection. And I think most of us would be very happy to have a spray that every three or four months and gave us much higher levels of protection than we're ever going to get from shots. And your whole concept of prime and spike, I mean this is something that we could have had years ago if there was a priority, and unfortunately there never has been. Now, the other day you came with a surprise in a paper on Neomycin as an alternate or Neosporin ointment. Can you tell us about that? Because that one wasn't expected. This was to use an antibiotic in a way to reduce Covid and other respiratory virus.
Akiko Iwasaki (18:50):
Right. So yeah, that's a little known fact. I mean, of course widespread use of antibiotics has caused some significant issues with resistance and so on. However, when you look at the literature of different types of antibiotics, we have reported in 2018 that certain types of antibiotics known as aminoglycoside, which includes Neosporin or neomycin, has this sort of unintended antiviral property by triggering Toll-like receptor 3 in specialized cell types known as conventional dendritic cell type 1. And we published that for a genital herpes model that we were working on at the time. But because it's acting on the host, the Toll-like receptor 3 on the host cell to induce interferon and interferon stimulated genes to prevent the replication of the virus, we knew that it could be pan-viral. It doesn't really matter what the virus is. So we basically leverage that discovery that was made by a postdoc Smita Gopinath when she was in the lab to see if we can use that in the nasal cavity.
(20:07):
And that's what Tianyang Mao, a former graduate student did, in fact. And yeah, little spray of neomycin in the nose of the mice reduce this infection as well as disease and can even be used to treat shortly after the infection disease progress and using hamster models we also showed that hamsters that are pretreated with neomycin when they were caged with infected hamsters, the transmission rate was much reduced. And we also did with Dr. Charles Dela Cruz, a small clinical trial, randomized though into placebo and Neosporin arms of healthy volunteers. We asked them to put in a pea size amount of Neosporin on a cotton swab into the nose, and they were doing that twice a day for seven days. We measured the RNA from the nose of these people and indeed see that more than half the participants in the Neosporin group had elevated interferon stimulated genes, whereas the control group, which were given Vaseline had no response. So this sort of shows the promise of using something as generic and cheap as Neosporin to trigger antiviral state in the nose. Now it does require a much larger trial making sure that the safety profiles there and effectiveness against viral infection, but it's just a beginning of a story that could develop into something useful.
New Frontiers in Immunology and Tx Cells
Eric Topol (21:51):
Yeah, I thought it was fascinating, and it does bring up, which I think has also been underdeveloped, is our approaches for interferon a frontline defense where augmenting that, just getting that exploiting the nasal mucosa, the entry site, whether it be through that means or of course through even more potent a nasal vaccine, it's like a missing, it's a hole in our whole defense of against this virus that's led to millions of people not just dying, but of course also sick and also with Long Covid around the world. So I hope that we'll see some progress, but I thought that was a really fascinating hint of something to come that could be very helpful in the meantime while we're waiting for specific nasal vaccines. Now added to all these things recently, like last week you published a paper in Cell with your husband who's in the same department, I think at Yale. Is that right? Can you tell us about that and this paper about the whole new perspectives in immunology?
Akiko Iwasaki (23:05):
Yeah, so my husband Ruslan Medzhitov is a very famous immunologist who's in the same department, and we've written four or five review and opinion pieces together over the years. This new one is in Cell and it's really exploring new perspectives in immunology. We were asked by the editors to celebrate the 50th anniversary of the Cell journal with a perspective on the immune system. And the immune response is just a beautiful system that is triggered in response to specific pathogens and can really provide long-term or even sometimes lifelong immunity and resistance against pathogens and it really saves our lives. Much has been learned throughout the last 20, 30 years about the innate and adaptive immune system and how they're linked. In this new perspective, we are trying to raise some issues that the current paradigm cannot explain properly, some of the mysteries that are still remaining in the immune system.
(24:22):
And we try to come up with new concepts about even the role of the immune system in general. For instance, is the immune system only good for fighting pathogens or can it be repurposed for conducting normal physiology in the host? And we came up with a new subset of T-cells known as, or we call it Tx cells, which basically is an interoceptive type of T-cells that monitor homeostasis in different tissues and are helping with the normal process of biology as opposed to fighting viruses or bacteria or fungi. But these cells, when they are not appropriately regulated, they are also the source of autoimmune diseases because they are by design reactive against auto antigens. And so, this is a whole new framework to think about, a different arm of the immune function, which is really looking inside of our body and not really fighting against pathogens, but we believe these cells exist, and we know that the counterpart of Tx cells, which is the T regulatory cells, are indeed well known for its physiological functions. So we're hoping that this new perspective will trigger a new set of approaches in the field to try to understand this interceptive property of T-cells.
Eric Topol (25:59):
Yeah, well, I thought it was fascinating, of course, and I wanted to get into that more because I think what we're learning is this immune system not only obviously is for cancer whole. We're only starting to get warmed up with immunotherapy where checkpoint inhibitors were just the beginning and now obviously with vaccines and all these different ways that we can take the CAR-T cells, engineered T-cells, take the immune system to fight cancer and potentially to even use it as a way to prevent cancer. If you have these, whether it's Tx or Tregs or whatever T-cells can do this. But even bigger than that is the idea that it's tied in with the aging process. So as you know, again, much more than I do, our senescent immune cells are not good for us. And the whole idea is that we could build immune resilience if we could somehow figure out these mysteries that you're getting at, whereby we get vulnerable just as we were with Covid. And as we get older, we get vulnerable to not just infections, but everything going wrong, whether it's the walls of our arteries or whether it's the cancer or the immunity that's going on in our brain for Alzheimer's and neurodegenerative diseases. How can we fix the immune system so that we age more healthily
The Immune System and Healthy Aging
Akiko Iwasaki (27:37):
Oh yeah. A lot of billionaires are also interested in that question and are pouring money into this question. It's interesting, but when you think about the sort of evolutionary perspective, we humans are only living so long. In the very recent decades, our life expectancy used to be much shorter and all we had to survive was to reproduce and generate the next progeny. But nowadays, because of this amazing wealth and health interventions and food and everything else, we're just living so much longer than even our grandparents. The immune system didn't evolve to deal with such one to begin with. So we were doing fine living up to 30 years of age or whatever. But now that we're living up to a hundred years, the immune system isn't really designed to keep up with this kind of stressors. But I think you're getting at a very important kind of more engineering questions of how do we manipulate the immune system or rejuvenate it so that we can remain healthy into the later decades? And it is well known that the immune system itself ages and that our ability to produce new lymphocytes, for example, decline over time and thymus that is important for T-cell development shrinks over time. And so anatomically it's impossible to help stop that process. However, is there a way of, for example, transferring some factors or engineering the immune cells to remain healthy and even like hematopoiesis itself can be manipulated to perhaps rejuvenate the whole immune system in their recent papers showing that. So this is a new frontier.
Eric Topol (29:50):
Do you think that some point in the future, we'll ex vivo inject Yamanaka factors into these cell lines and instead of this idea that you know get young plasma to old folks, and I mean since we don't know what's in there and it doesn't specifically have an effect on immune cells, who knows how it's working, but do you foresee that that might be a potential avenue going forward or even an in vivo delivery of this?
Akiko Iwasaki (30:22):
Yeah, it's not impossible, right? There are really rapidly evolving technologies and gene therapies that are becoming online. So it's not impossible to think about engineering in situ as you're suggesting, but we also have to be certain that we are living longer, but also healthy. So we do have to not only just deal with the aging immune system, but preventing neurodegenerative diseases and so on. And the immune system may have a role to play there as well. So there's a lot of, I mean, I can't think of a non-genetically mediated disease that doesn't involve the immune system.
Eric Topol (31:03):
Sure. No, I mean, it's just, when I think about this, people keep talking about the digital era of digital biology, but I actually think of it more as digital immunobiology, which is driving this because it's center stage and in more and more over time. And the idea that I'm concerned about is that we could rejuvenate the relevant immune cells or the whole immune response, but then it's such a delicate balance that we could actually wind up with untoward, whether it's autoimmune or overly stimulated immune system. It's not such a simple matter, as I'm sure you would agree. Now, this gets me to a broader thing which you've done, which is a profound contribution in life science and medicine, which is being an advocate for women in science. And I wonder if you could speak to that because you have been such a phenomenal force propelling the importance of women in science and not just doing that passively, but also standing up for women, which is being an activist is how you get things to change. So can you tell us about your thoughts there?
An Activist for Women in Science
Akiko Iwasaki (32:22):
Yeah, so I grew up in Japan, and part of the reason I left Japan at the age of 16 was that I felt very stifled because of the societal norm and expectation of what a woman should be. And I felt like I didn't have the opportunity to develop my skills as a scientist remaining in Japan. And maybe things have changed over the years, but at the time when I was growing up, that's how I felt. And so, I was very cognizant of biases in society. And so, in the US and in Canada where I also trained, there's a lot less barrier to success, and we are able to do pretty much anything we want, which is wonderful, and that's why I think I'm here. But at the same time, the inequity still exists, even in pay gaps and things like that that are easy to fix but are still kind of insidious and it's there.
(33:32):
And Yale School of Medicine has done a great job partly because of the efforts of women who spoke up and who actually started to collect evidence for pay gap. And now there's very little pay gap because there's active sort of involvement of the dean and everyone else to ensure equity in the medical school. But it's just a small segment of the society. We really need to expand this to other schools and making sure that women are getting paid equally as men in the same ranks. And also, I see still some sexual harassment or more just toxic environment for people in general in academia. Some PIs get away with a lot of behavior that's not conducive to a healthy environment, so I have written about that as well and how we can have antidotes for such toxic environments. And it really does require the whole village to act on it. It's not just one person speaking up. And there should be measures placed to make sure that those people who does have this tendency of abusive behavior that they can get training and just being aware of these situations and corrective behavior. So I think there's still a lot of work left in academia, but things have obviously improved dramatically over the last few decades, and we are in a very, very good place, but we just have to keep working to achieve true equity.
Why Don’t We Have Immunome Check-Ups?
Eric Topol (35:25):
Well applauding your efforts for that, and I'm still in touch with that. We got a ways to go, and I hope that we'll see steady and even more accelerated and improvement to get to parity, which is what it should be. And I really think you've been a model for doing this. It isn't like you aren't busy with everything else, so to fit that in is wonderful. In closing up, one of the things that I wonder about is our ability to assess back to the immune system for a moment isn't what it should be. That is we do a CBC and we have how many lymphocytes, how many this, why don't we have an immunome, why doesn’t everybody serially have an immune system checkup? Because that would tell us if we’re starting to go haywire and then maybe hunt for reactivated viruses or what’s going on. Do you foresee that we could ever get to a practical immunome as we go forward? Because it seems like it’s a big missing link right now.
Akiko Iwasaki (36:33):
Yeah, I think that’s a great idea. I mean, I’ll be the first one to sign up for the immunome.
Eric Topol (36:40):
But I’m depending on you to make it happen.
Akiko Iwasaki (36:44):
Well, interestingly, Eric, there are lots of amazing technologies that are developed even during the pandemic, which is monitoring everything from antibody reactivity to reactivated viruses to the cytokines to every cell marker you can imagine. So the technologies out there, it’s just I think a matter of having the right set of panels that are relatively affordable because some of these things are thousands of dollars per sample to analyze, and then of course clinical validation, something that’s CLIA approved, and then we can start to, I guess the insurance company needs to also cover this, right? So we need to demonstrate the benefit to health in the long run to be able to afford this kind of immunome analysis. But I think that very wealthy people can already get this done.
Eric Topol (37:43):
Yeah, well, we want to make it so it's a health equity story, not of course, only for the crazy ones that are out there that are taking 112 supplements a day and whatnot. But it's intriguing because I think we might be able to get ahead of things if we had such an easy means. And as you said during the pandemic, for example, my friends here in La Jolla at La Jolla Immunology did all kinds of T-cell studies that were really insightful and of course done with you and others around the country and elsewhere to give us insights that you didn't get just from neutralizing antibodies. But it isn't something that you can get done easily. Now, I think this immunome hopefully will get us to another level in the future. One of the most striking things I've seen in our space clinically before wrapping up is to take the CD19 CAR T therapies to deplete the B cells of people with lupus, systemic sclerosis and other conditions, and completely stop their autoimmune condition. And when the B cells come back, they're not fighting themselves. They're not self-directed anymore. Would you have predicted this? This seems really striking and it may be a clue to the kind of mastering approaches to autoimmune diseases in the future.
Akiko Iwasaki (39:19):
Yeah, absolutely. So for multiple sclerosis, for example, where B cells weren't thought to be a key player by doing anti-CD20 depletion, there's this remarkable clinical effects. So I think we can only find the answer experimentally in people when they do these clinical trials and show this remarkable effects. That's when we say, aha, we don't really understand immunology. You know what I mean? That's when we have to be humble about what we think we understand. We really don't know until we try it. So that's a really good lesson learned. And these may be also applicable to people with autoimmune phenotype in Long Covid, right? We may be able to benefit from similar kinds of depletion therapy. So I think we have a lot to learn still.
Eric Topol (40:14):
Yeah, that's why, again, going back to the paper you just had in Cell about the mysteries and about some new ideas and challenging the dogma is so important. I still consider the immune system most complex one in the body by far, and I'm depending on you Akiko to unravel it, not to put any weight on your shoulders. Anyway, this has been so much fun. You are such a gem and always learning from you, and I can't thank you enough for all the work. And the fact is that you've got decades ahead of you to keep building on this. You've already done enough for many people, many scientists in your career, and I know you'll keep going. So we're all going to be following you with great interest in learning from you on a frequent basis. And I hope we'll build on some of the things we've talked about like a Long Covid treatment, treatments that are effective nasal vaccines, maybe even some dab of Neosporin, and keep on the momentum we’ve had with the understanding of the immune system, and finally, someday achieving the true parity of gender and science. And so, thank you for all that you do.
Akiko Iwasaki (41:35):
Thank you so much, Eric.
************************
Credits
Headshot photo credits by Robert Lisak, Yale School of Medicine
My producer for Ground Truths is Jessica Nguyen, Scripps Research and our technical support for audio/video is by SInjun Balabanoff at Scripps Research.
I hope you found the spot informative. Please share it
The Ground Truths newsletters and podcasts are all free, open-access, without ads.
Voluntary paid subscriptions all go to support Scripps Research. Many thanks for that—they greatly helped fund our summer internship programs for 2023 and 2024.
Note: you can select preferences to receive emails about newsletters, podcasts, or all I don’t want to bother you with an email for content that you’re not interested in.
Comments are welcome from all subscribers.

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
Aviv Regev: The Revolution in Digital Biology
28 avril· Ground Truths
“Where do I think the next amazing revolution is going to come? … There’s no question that digital biology is going to be it. For the very first time in our history, in human history, biology has the opportunity to be engineering, not science.” —Jensen Huang, NVIDIA CEO
Aviv Regev is one of the leading life scientists of our time. In this conversation, we cover the ongoing revolution in digital biology that has been enabled by new deep knowledge on cells, proteins and genes, and the use of generative A.I .
Transcript with audio and external links
Eric Topol (00:05):
Hello, it's Eric Topol with Ground Truths and with me today I've really got the pleasure of welcoming Aviv Regev, who is the Executive Vice President of Research and Early Development at Genentech, having been 14 years a leader at the Broad Institute and who I view as one of the leading life scientists in the world. So Aviv, thanks so much for joining.
Aviv Regev (00:33):
Thank you for having me and for the very kind introduction.
The Human Cell Atlas
Eric Topol (00:36):
Well, it is no question in my view that is the truth and I wanted to have a chance to visit a few of the principal areas that you have been nurturing over many years. First of all, the Human Cell Atlas (HCA), the 37 trillion cells in our body approximately a little affected by size and gender and whatnot, but you founded the human cell atlas and maybe you can give us a little background on what you were thinking forward thinking of course when you and your colleagues initiated that big, big project.
Aviv Regev (01:18):
Thanks. Co-founded together with my very good friend and colleague, Sarah Teichmann, who was at the Sanger and just moved to Cambridge. I think our community at the time, which was still small at the time, really had the vision that has been playing out in the last several years, which is a huge gratification that if we had a systematic map of the cells of the body, we would be able both to understand biology better as well as to provide insight that would be meaningful in trying to diagnose and to treat disease. The basic idea behind that was that cells are the basic unit of life. They're often the first level at which you understand disease as well as in which you understand health and that in the human body, given the very large number of individual cells, 37.2 trillion give or take, and there are many different characteristics.
(02:16):
Even though biologists have been spending decades and centuries trying to characterize cells, they still had a haphazard view of them and that the advancing technology at the time – it was mostly single cell genomics, it was the beginnings also of spatial genomics – suggested that now there would be a systematic way, like a shared way of doing it across all cells in the human body rather than in ways that were niche and bespoke and as a result didn't unify together. I will also say, and if you go back to our old white paper, you will see some of it that we had this feeling because many of us were computational scientists by training, including both myself and Sarah Teichmann, that having a map like this, an atlas as we call it, a data set of this magnitude and scale, would really allow us to build a model to understand cells. Today, we call them foundational models or foundation models. We knew that machine learning is hungry for these kinds of data and that once you give it to machine learning, you get amazing things in return. We didn't know exactly what those things would be, and that has been playing out in front of our eyes as well in the last couple of years.
Spatial Omics
Eric Topol (03:30):
Well, that gets us to the topic you touched on the second area I wanted to get into, which is extraordinary, which is the spatial omics, which is related to the ability to the single cell sequencing of cells and nuclei and not just RNA and DNA and methylation and chromatin. I mean, this is incredible that you can track the evolution of cancer, that the old word that we would say is a tumor is heterogeneous, is obsolete because you can map every cell. I mean, this is just changing insights about so much of disease health mechanisms, so this is one of the hottest areas of all of life science. It's an outgrowth of knowing about cells. How do you summarize this whole era of spatial omics?
Aviv Regev (04:26):
Yeah, so there's a beautiful sentence in the search for lost time from Marcel Proust that I'm going to mess up in paraphrasing, but it is roughly that going on new journeys is not about actually going somewhere physically but looking with new eyes and I butchered the quote completely.[See below for actual quote.] I think that is actually what single cells and then spatial genomics or spatial omics more broadly has given us. It's the ability to look at the same phenomenon that we looked at all along, be it cancer or animal development or homeostasis in the lung or the way our brain works, but having new eyes in looking and because these new eyes are not just seeing more of something we've seen before, but actually seeing things that we couldn't realize were there before. It starts with finding cells we didn't know existed, but it's also the processes that these cells undergo, the mechanisms that actually control that, the causal mechanisms that control that, and especially in the case of spatial genomics, the ways in which cells come together.
(05:43):
And so we often like to think about the cell because it's the unit of life, but in a multicellular organism we just as much have to think about tissues and after that organs and systems and so on. In a tissue, you have this amazing orchestration of the interactions between different kinds of cells, and this happens in space and in time and as we're able to look at this in biology often structure is tightly associated to function. So the structure of the protein to the function of the protein in the same way, the way in which things are structured in tissue, which cells are next to each other, what molecules are they expressing, how are they physically interacting, really tells us how they conduct the business of the tissue. When the tissue functions well, it is this multicellular circuit that performs this amazing thing known as homeostasis.
(06:36):
Everything changes and yet the tissue stays the same and functions, and in disease, of course, when these connections break, they're not done in the right way you end up with pathology, which is of course something that even historically we have always looked at in the level of the tissue. So now we can see it in a much better way, and as we see it in a better way, we resolve better things. Yes, we can understand better the mechanisms that underlie the resistance to therapeutics. We can follow a temporal process like cancer as it unfortunately evolves. We can understand how autoimmune disease plays out with many cells that are actually bent out of shape in their interactions. We can also follow magnificent things like how we start from a single cell, the fertilized egg, and we become 37.2 trillion cell marvel. These are all things that this ability to look in a different way allows us to do.
Eric Topol (07:34):
It's just extraordinary. I wrote at Ground Truths about this. I gave all the examples at that time, and now there's about 50 more in the cardiovascular arena, knowing with single cell of the pineal gland that the explanation of why people with heart failure have sleep disturbances. I mean that's just one of the things of so many now these new insights it's really just so remarkable. Now we get to the current revolution, and I wanted to read to you a quote that I have.
Digital Biology
Aviv Regev (08:16):
I should have prepared mine. I did it off the top of my head.
Eric Topol (08:20):
It's actually from Jensen Huang at NVIDIA about the digital biology [at top of the transcript] and how it changes the world and how you're changing the world with AI and lab in the loop and all these things going on in three years that you've been at Genentech. So maybe you can tell us about this revolution of AI and how you're embracing it to have AI get into positive feedbacks as to what experiment to do next from all the data that is generated.
Aviv Regev (08:55):
Yeah, so Jensen and NVIDIA are actually great partners for us in Genentech, so it's fun to contemplate any quote that comes from there. I'll actually say this has been in the making since the early 2010s. 2012 I like to reflect on because I think it was a remarkable year for what we're seeing right now in biology, specifically in biology and medicine. In 2012, we had the beginnings of really robust protocols for single cell genomics, the first generation of those, we had CRISPR happen as a method to actually edit cells, so we had the ability to manipulate systems at a much better way than we had before, and deep learning happened in the same year as well. Wasn't that a nice year? But sometimes people only realize the magnitude of the year that happened years later. I think the deep learning impact people realized first, then the single cells, and then the CRISPR, then the single cells.
(09:49):
So in order maybe a little bit, but now we're really living through what that promise can deliver for us. It's still the early days of that, of the delivery, but we are really seeing it. The thing to realize there is that for many, many of the problems that we try to solve in biomedicine, the problem is bigger than we would ever be able to perform experiments or collect data. Even if we had the genomes of all the people in the world, all billions and billions of them, that's just a smidge compared to all of the ways in which their common variants could combine in the next person. Even if we can perturb and perturb and perturb, we cannot do all of the combinations of perturbations even in one cell type, let alone the many different cell types that are out there. So even if we searched for all the small molecules that are out there, there are 10 to the 60 that have drug-like properties, we can't assess all of them, even computationally, we can't assess numbers like that.
(10:52):
And so we have to somehow find a way around problems that are as big as that and this is where the lab in the loop idea comes in and why AI is so material. AI is great, taking worlds, universes like that, that appear extremely big, nominally, like in basic numbers, but in fact have a lot of structure and constraint in them so you can reduce them and in this reduced latent space, they actually become doable. You can search them, you can compute on them, you can do all sorts of things on them, and you can predict things that you wouldn't actually do in the real world. Biology is exceptionally good, exceptionally good at lab sciences, where you actually have the ability to manipulate, and in biology in particular, you can manipulate at the causes because you have genetics. So when you put these two worlds together, you can actually go after these problems that appear too big that are so important to understanding the causes of disease or devising the next drug.
(11:51):
You can iterate. So you start, say, with an experimental system or with all the data that you have already, I don't know from an initiative like the human cell atlas, and from this you generate your original model of how you think the world works. This you do with machine learning applied to previous data. Based on this model, you can make predictions, those predictions suggest the next set of experiments and you can ask the model to make the most optimized set of predictions for what you're trying to learn. Instead of just stopping there, that's a critical point. You go back and you actually do an experiment and you set up your experiments to be scaled like that to be big rather than small. Sometimes it means you actually have to compromise on the quality of any individual part of the experiment, but you more than make up for that with quantity.
The A.I. Lab-in-the-Loop
(12:38):
So now you generate the next data from which you can tell both how well did your algorithm actually predict? Maybe the model didn’t predict so well, but you know that because you have lab results and you have more data in order to repeat the loop, train the model again, fit it again, make the new next set of predictions and iterate like this until you're satisfied. Not that you've tried all options, because that's not achievable, but that you can predict all the interesting options. That is really the basis of the idea and it applies whether you're solving a general basic question in biology or you're interested in understanding the mechanism of the disease or you're trying to develop a therapeutic like a small molecule or a large molecule or a cell therapy. In all of these contexts, you can apply this virtual loop, but to apply it, you have to change how you do things. You need algorithms that solve problems that are a little different than the ones they solved before and you need lab experiments that are conducted differently than they were conducted before and that's actually what we're trying to do.
Eric Topol (13:39):
Now I did find the quote, I just want to read it so we have it, “biology has the opportunity to be engineering, not science. When something becomes engineering, not science, it becomes exponentially improving. It can compound on the benefits of previous years.” Which is kind of a nice summary of what you just described. Now as we go forward, you mentioned the deep learning origin back at the same time of CRISPR and so many things happening and this convergence continues transformer models obviously one that's very well known, AlphaFold, AlphaFold2, but you work especially in antibodies and if I remember correctly from one of your presentations, there's 20 to the 32nd power of antibody sequences, something like that, so it's right up there with the 10 to the 60th number of small molecules. How do transformer models enhance your work, your discovery efforts?
Aviv Regev (14:46):
And not just in antibodies, I'll give you three brief examples. So absolutely in antibodies it's an example where you have a very large space and you can treat it as a language and transformers are one component of it. There's other related and unrelated models that you would use. For example, diffusion based models are very useful. They're the kind that people are used to when you do things, you use DALL-E or Midjourney and so on makes these weird pictures, think about that picture and not as a picture and now you're thinking about a three-dimensional object which is actually an antibody, a molecule. You also mentioned AlphaFold and AlphaFold 2, which are great advances with some components related to transformers and some otherwise, but those were done as general purpose machines for proteins and antibodies are actually not general purpose proteins. They're antibodies and therapeutic antibodies are even further constrained.
(15:37):
Antibodies also really thrive, especially for therapeutics and also in our body, they need diversity and many of these first models that were done for protein structure really focused on using conservation as an evolutionary signal comparison across species in order to learn the model that predicts the structure, but with antibodies you have these regions of course that don't repeat ever. They're special, they're diverse, and so you need to do a lot of things in the process in order to make the model fit in the best possible way. And then again, this loop really comes in. You have data from many, many historical antibodies. You use that to train the model. You use that model in order to make particular predictions for antibodies that you either want to generate de novo or that you want to optimize for particular properties. You make those actually in the lab and in this way gradually your models become better and better at this task with antibodies.
(16:36):
I do want to say this is not just about antibodies. So for example, we develop cancer vaccines. These are personalized vaccines and there is a component in making a personalized cancer vaccine, which is choosing which antigens you would actually encode into the vaccine and transformers play a crucial role in actually making this prediction today of what are good neoantigens that will get presented to the immune system. You sometimes want to generate a regulatory sequence because you want to generate a better AAV-like molecule or to engineer something in a cell therapy, so you want to put a cis-regulatory sequence that controls gene expression. Actually personally for me, this was the first project where I used a transformer, which we started years ago. It was published a couple of years ago where we learned a general model that can predict in a particular system. Literally you throw a sequence at that model now and it will predict how much expression it would drive. So these models are very powerful. They are not the be all and end all of all problems that we have, but they are fantastically useful, especially for molecular therapeutics.
Good Trouble: Hallucinations
Eric Topol (17:48):
Well, one of the that has been an outgrowth of this is to actually take advantage of the hallucinations or confabulation of molecules. For example, the work of David Baker, who I'm sure you know well at University of Washington, the protein design institute. We are seeing now molecules, antibodies, proteins that don't exist in nature from actually, and all the things that are dubbed bad in GPT-4 and ChatGPT may actually help in the discovery in life science and biomedicine. Can you comment about that?
Aviv Regev (18:29):
Yeah, I think much more broadly about hallucinations and what you want to think about is something that's like constrained hallucination is how we're creative, right? Often people talk about hallucinations and they shudder at it. It sounds to them insane because if you think about your, say a large language model as a search tool and it starts inventing papers that don't exist. You might be like, I don't like that, but in reality, if it invents something meaningful that doesn't exist, I love that. So that constrained hallucination, I'm just using that colloquially, is a great property if it's constrained and harnessed in the right way. That's creativity, and creativity is very material for what we do. So yes, absolutely in what we call the de novo domain making new things that don't exist. This generative process is the heart of drug discovery. We make molecules that didn't exist before.
(19:22):
They have to be imagined out of something. They can't just be a thing that was there already and that's true for many different kinds of therapeutic molecules and for other purposes as well, but of course they still have to function in an effective way in the real world. So that's where you want them to be constrained in some way and that's what you want out of the model. I also want to say one of the areas that personally, and I think for the field as a whole, I find the most exciting and still underused is the capacity of these models to hallucinate for us or help us with the creative endeavors of identifying the causes of processes, which is very different than the generative process of making molecules. Thinking about the web of interactions that exist inside a cell and between cells that drives disease processes that is very hard for us to reason through and to collect all the bits of information and to fill in blanks, those fillings of the blanks, that's our creativity, that's what generates the next hypothesis for us. I'm very excited about that process and about that prospect, and I think that's where the hallucination of models might end up proving to be particularly impressive.
A.I. Accelerated Drug Discovery
Eric Topol (20:35):
Yeah. Now obviously the field of using AI to accelerate drug discovery is extremely hot, just as we were talking about with spatial omics. Do you think that is warranted? I mean you've made a big bet on that you and your folks there at Genentech of course, and so many others, and it's a very crowded space with so many big pharma partnering with AI. What do you see about this acceleration? Is it really going to reap? Is it going to bear fruit? Are we going to see, we've already seen some drugs of course, that are outgrowths, like Baricitinib in the pandemic and others, but what are your expectations? I know you're not one to get into any hyperbole, so I'm really curious as to what you think is the future path.
Aviv Regev (21:33):
So definitely my hypothesis is that this will be highly, highly impactful. I think it has the potential to be as impactful as molecular biology has been for drug discovery in the 1970s and 1980s. We still live that impact. We now take it for granted. But, of course that's a hypothesis. I also believe that this is a long game and it's a deep investment, meaning decorating what you currently do with some additions from right and left is not going to be enough. This lab in the loop requires deep work working at the heart of how you do science, not as an add-on or in addition to or yet another variant on what has become a pretty established approach to how things are done. That is where I think the main distinction would be and that requires both the length of the investment, the effort to invest in, and also the willingness to really go all out, all in and all out.
(22:36):
And that takes time. The real risk is the hype. It's actually the enthusiasm now compared to say 2020 is risky for us because people get very enthusiastic and then it doesn't pay off immediately. No, these iterations of a lab in the loop, they take time and they take effort and they take a lot of changes and at first, algorithms often fail before they succeed. You have to iterate them and so that is actually one of the biggest risks that people would be like, but I tried it. It didn't work. This was just some over-hyped thing. I'm walking away and doing it the old way. So that's where we actually have to keep at it, but also keep our expectations not low in magnitude. I think that it would actually deliver, but understanding that it's actually a long investment and that unless you do it deeply, it's not going to deliver the goods.
Eric Topol (23:32):
I think this point warrants emphasis because the success already we've seen has not been in necessarily discovery and in preliminary validation of new molecules, but rather data mining repurposing, which is a much easier route to go quicker, but also there's so many nodes on past whereby AI can make a difference even in clinical trials, in synthetic efforts to project how a clinical trial will turn out and being able to do toxic screens without preclinical animal work. There's just so many aspects of this that are AI suited to rev it up, but the one that you're working on, of course is the kind of main agenda and I think you framed it so carefully that we have to be patient here, that it has a chance to be so transformative. Now, you touched on the parallels to things like DALL-E and Midjourney and large language models. A lot of our listeners will be thinking only of ChatGPT or GPT-4 or others. This is what you work on, the language of life. This is not text of having a conversation with a chatbot. Do you think that as we go forward, that we have to rename these models because they're known today as language models? Or do you think that, hey, you know what, this is another language. This is a language that life science and biomedicine works with. How do you frame it all?
Large Non-Human Language Models
Aviv Regev (25:18):
First of all, they absolutely can remain large language models because these are languages, and that's not even a new insight. People have treated biological sequences, for example, in the past too, using language models. The language models were just not as great as the ones that we have right now and the data that were available to train models in the past were not as amazing as what we have right now. So often these are really the shifts. We also actually should pay respect to human language. Human language encodes a tremendous amount of our current scientific knowledge and even language models of human language are tremendously important for this scientific endeavor that I've just described. On top of them come language models of non-human language such as the language of DNA or the language of protein sequences, which are also tremendously important as well as many other generative models, representation learning, and other approaches for machine learning that are material for handling the different kinds of data and questions that we have.
(26:25):
It is not a single thing. What large language models and especially ChatGPT, this is an enormous favor for which I am very grateful, is that I think it actually convinced people of the power. That conviction is extremely important when you're solving a difficult problem. If you feel that there's a way to get there, you're going to behave differently than if you're like, nothing will ever come out of it. When people experience ChatGPT actually in their daily lives in basic things, doing things that felt to them so human, this feeling overrides all the intellectual part of things. It's better than the thinking and then they're like, in that case, this could actually play out in my other things as well. That, I think, was actually materially important and was a substantial moment and we could really feel it. I could feel it in my interactions with people before and after how their thinking shifted. Even though we were on this journey from before.
Aviv Regev (27:30):
We were. It felt different.
Eric Topol (27:32):
Right, the awareness of hundreds of millions of people suddenly in end of November 2022 and then you were of course going to Genentech years before that, a couple few years before that, and you already knew this was on the move and you were redesigning the research at Genentech.
Aviv Regev (27:55):
Yes, we changed things well before, but it definitely helps in how people embrace and engage feels different because they've seen something like that demonstrated in front of them in a way that felt very personal, that wasn't about work. It's also about work, but it's about everything. That was very material actually and I am very grateful for that as well as for the tool itself and the many other things that this allows us to do but we have, as you said, we have been by then well on our way, and it was actually a fun moment for that reason as well.
Eric Topol (28:32):
So one of the things I'm curious about is we don't think about the humans enough, and we're talking about the models and the automation, but you have undoubtedly a large team of computer scientists and life scientists. How do you get them to interact? They're of course, in many respects, in different orbits, and the more they interact, the more synergy will come out of that. What is your recipe for fostering their crosstalk?
Aviv Regev (29:09):
Yeah, this is a fantastic question. I think the future is in figuring out the human question always above all and usually when I draw it, like on the slide, you can draw the loop, but we always put the people in the center of that loop. It's very material to us and I will highlight a few points. One crucial thing that we've done is that we made sure that we have enough critical mass across the board, and it played out in different ways. For example, we built a new computational organization, gRED Computational Sciences, from what was before many different parts rather than one consolidated whole. Of course within that we also built a very strong AI machine learning team, which we didn't have as much before, so some of it was new people that we didn't have before, but some of it was also putting it with its own identity.
(29:56):
So it is just as much, not more, but also not less just as much of a pillar, just as much of a driver as our biology is, as our chemistry and molecule making is, as our clinical work is. This equal footing is essential and extremely important. The second important point is you really have to think about how you do your project. For example, when we acquired Prescient, at the time they were three people, tiny, tiny company became our machine learning for drug discovery. It's not tiny anymore, but when we acquired them, we also invested in our antibody engineering so that we could do antibody engineering in a lab in the loop, which is not how we did it before, which meant we invested in our experiments in a different way. We built a department for cell and tissue genomics so we can conduct biology experiments also in a different way.
(30:46):
So we changed our experiments, not just our computation. The third point that I think is really material, I often say that when I'm getting asked, everyone should feel very comfortable talking with an accent. We don't expect our computational scientists to start behaving like they were actually biology trained in a typical way all along, or chemists trained in a typical way all along and by the same token, we don't actually expect our biologists to just embrace wholeheartedly and relinquish completely one way of thinking for another way of thinking, not at all. To the contrary, we actually think all these accents, that's a huge strength because the computer scientist thinks about biology or about chemistry or about medical work differently than a medical doctor or a chemist or a biologist would because a biologist thinks about a model differently and sometimes that is the moment of brilliance that defines the problem and the model in the most impactful way.
(31:48):
We want all of that and that requires both this equal footing and this willingness to think beyond your domain, not just hand over things, but actually also be there in this other area where you're not the expert but you're weird. Talking with an accent can actually be super beneficial. Plus it's a lot of fun. We're all scientists, we all love learning new things. So that's some of the features of how we try to build that world and you kind of do it in the same way. You iterate, you try it out, you see how it works, and you change things. It's not all fixed and set in stone because no one actually wrote a recipe, or at least I didn't find that cookbook yet. You kind of invent it as you go on.
Eric Topol (32:28):
That's terrific. Well, there's so much excitement in this convergence of life science and the digital biology we've been talking about, have I missed anything? We covered human cell atlas, the spatial omics, the lab in the loop. Is there anything that I didn't touch on that you find important?
Aviv Regev (32:49):
There's something we didn't mention and is the reason I come to work every day and everyone I work with here, and I actually think also the people of the human cell atlas, we didn't really talk about the patients.
(33:00):
There's so much, I think you and I share this perspective, there's so much trepidation around some of these new methods and we understand why and also we all saw that technology sometimes can play out in ways that are really with unintended consequences, but there's also so much hope for patients. This is what drives people to do this work every day, this really difficult work that tends not to work out much more frequently than it works out now that we're trying to move that needle in a substantial way. It's the patients, and that gives this human side to all of it. I think it's really important to remember. It also makes us very responsible. We look at things very responsibly when we do this work, but it also gives us this feeling in our hearts that is really unbeatable, that you're doing it for something good.
Eric Topol (33:52):
I think that emphasis couldn't be more appropriate. One of the things I think about all the time is that because we're moving into this, if you will, hyper accelerated phase of discovery over the years ahead with this just unparallel convergence of tools to work with, that somebody could be cured of a condition, somebody could have an autoimmune disease that we will be able to promote tolerogenicity and they wouldn't have the autoimmune disease and if they could just sit tight and wait a few years before this comes, as opposed to just missing out because it takes time to get this all to gel. So I'm glad you brought that up, Aviv, because I do think that's what it's all about and that's why we're cheering for your work and so many others to get it done, get across the goal line because there's these 10,000 diseases out there and there's so many unmet needs across them where we don't have treatments that are very effective or have all sorts of horrible side effects. We don't have cures, and we've got all the things now, as we've mentioned here in this conversation, whether it's genome editing and ability to process massive scale data in a way that never could be conceived some years ago. Let's hope that we help the patients, and go ahead.
Aviv Regev (35:25):
I found the Proust quote, if you want it recorded correctly.
Eric Topol (35:29):
Yeah, good.
Aviv Regev (35:30):
It's much longer than what I did. It says, “the only true voyage, the only bath in the Fountain of Youth would be not to visit strange lands but to possess other eyes, to see the universe through the eyes of another, of a hundred others, to see the hundred universes that each of them sees, that each of them is; and this we do, with great artists; with artists like these we do fly from star to star.”—Marcel Proust
Eric Topol (35:57):
I love that and what a wonderful way to close our conversation today. Aviv, I look forward to more conversations with you. You are an unbelievable gem. Thanks so much for joining today.
Aviv Regev (36:10):
Thank you so much.
*************************************
Thanks for listening or reading to this Ground Truths Podcast.
Please share if you found it of interest
The Ground Truths newsletters and podcasts are all free, open-access, without ads.
Voluntary paid subscriptions all go to support Scripps Research. Many thanks for that—they greatly helped fund our summer internship programs for 2023 and 2024.
Note: you can select preferences to receive emails about newsletters, podcasts, or all I don’t want to bother you with an email for content that you’re not interested in.
Comments are welcome from all subscribers.

Get full access to Ground Truths at erictopol.substack.com/subscribe
- Écouter Écoute encore Continuer Écoutez...
- Écoutez plus tard Écoutez plus tard
Montre plus

Episodes

The Glaucomfleckens: The Best in Medical Comedy

Rachael Bedard: A Geriatric Physician and Champion for Patients Without a Voice

Katerina Akassoglou: Blood Clots, Brain Inflammation, and Covid

Patrick Hsu: A Trailblazer in Digital Biology

AI Snake Oil—A New Book by 2 Princeton University Computer Scientists

Francis Collins: On Truth, Science, Faith and Trust

Joseph Allen: The Pivotal Importance of Air Quality, Ventilation and Exposures (Such as "Forever Chemicals") For Our Health

Steve Horvath: Our Epigenetic Age Clocks

Pradeep Natarajan: Preventing Heart Disease

Shane Crotty: A Landmark Study on Upper Airway Mucosal Immunity

Faisal Mahmood: A.I.'s Transformation of Pathology

Euan Ashley: Exercise may be the single most potent medical intervention ever known

Christopher Labos: Debunking Myths About What We Eat and Drink

Charlie Swanton: A Master Class on Cancer

Tom Cech: RNA Takes Center Stage

Venki Ramakrishnan: The New Science of Aging

Svetlana Blitshteyn: On the Front Line With Long Covid and POTS

Kate Crawford: A Leading Scholar and Conscience for A.I.

Akiko Iwasaki: The Immunology of Covid and the Future

Aviv Regev: The Revolution in Digital Biology