Plain English with Derek Thompson - The Most Exciting Month of Medical Breakthroughs in Years

Episode Date: June 16, 2026

For years, scientists worried that medical progress was slowing down. Drug development became more expensive than ever with more complex clinical trials, and even then, many new treatments offered only modest gains. But over the past month, a series of breakthroughs has raised hopes that medicine may be entering a new era. Researchers unveiled a massively promising new therapy for pancreatic cancer, a gene-editing treatment that could dramatically reduce the risk of heart disease, and an experimental obesity drug that not only produces unprecedented weight loss but also improves a huge range of related conditions. Cancer and heart disease are America’s two biggest killers, but if these treatments fulfill their promise, they could transform public health and extend millions of lives. Today’s guest is Matthew Herper, senior writer at STAT News. We discuss this remarkable month in medicine, why so many advances are arriving at once, and what they could mean for the future of human health. Subscribe to our YouTube channel here:https://www.youtube.com/@PlainEnglishwithDerekThompson If you have questions, observations, or ideas for future episodes, email us at [e-mail]. Host: Derek ThompsonGuest: Matthew HerperProducer: Devon BaroldiAdditional Production Support: Ben Glicksman All Matches Streaming Live. Watch 3 Days Free. Offers are subject to change. See fox.com for complete terms and conditions. Learn more about your ad choices. Visit podcastchoices.com/adchoices

Today, a miracle month in medicine. There's a famous quote often attributed to Vladimir Lenin, that there are decades when nothing happens and there are weeks when decades happen. For the last few years, I've spoken to so many scientists who told me they were deeply, deeply frustrated that America was moving too slowly on solving the most important diseases,

especially cancer and Alzheimer's. The cost of developing drugs, and in particular the cost of clinical trials, soared in the last few decades, while many of the most profitable drugs to come out of those clinical trials only helped a small number of people. But this past month has proved Lenin Wright, which is something I do not say very often on this show. This was a month when it felt like decades of medical progress happened at warp speed. In prepping just for this episode, I was literally exhausting my web browser's ability to hold tabs,

just trying to keep track of all the unbelievably good news. at this year's ASCO conference for the American Society of Clinical Oncology, scientists revealed perhaps the best news in the history of pancreatic cancer therapy, with a miraculous drug that had cancer scientists standing and cheering like they never have before. In May, a small gene editing study dramatically lowered cholesterol by disabling a gene, PCSK-9, that's associated with heart disease. The therapy raises the possibility that we might be on the verge of a virus, a one and done shot to end some kinds of heart disease,

the number one killer in America. Surely, you might say, that's sufficient reason for exuberance. But I might have saved the best part for last. Those are the results from Eli Lilly's phase three trials for its new GLP1 drug, Retutide. Actually, maybe you should call this a GLP3 drug because it targets not one hormone, GLP1, but three,

GLP1, GIP, and Glucagon. Lilly's new drug was found to double the weight loss effects of OZempic, along with huge positive effects on reducing sleep apnea, inflammation, systolic blood pressure, knee pain, type of diabetes, triglycerides, and bad cholesterol. Other clinical trial data suggests it reduces visceral fat around the liver by up to 80%. Now, we'll have plenty of time for informed skepticism on this show in just a moment, but let me offer some unhinged optimism right now. Heart disease and cancer are by far the two biggest killers in America.

Medicines that could slash the mortality rates of both heart disease and cancer would completely transform the mortality picture of the United States. And we might, might be on the verge of getting them. Today's guest is Matthew Herper, a senior writer at Stat News where he covers medicine. We talk about this miracle month in medicine, why we're suddenly singing this bonanza of progress, and what it would mean for our lifespans and our health if all of this went right,

and if we were indeed on the verge of a golden age of cures. I'm Derek Thompson. This is plain English. Matthew Herper, welcome with the show. Thanks for having me. It's really fun. So I am so excited to talk about this miracle. month in medicine with you. And I think we should start with Reda Trutide, the new GLP1 drug that the

medical world is buzzing about. I want to round up some of the wins that I saw from phase three clinical trial data. Weight loss, almost double OZempic, 28% versus 15%, huge effects on sleep apnea, inflammation, systolic blood pressure, knee pain, type 2 diabetes, triglycerides, bad cholesterol. There's phase two clinical data suggesting that it reduced visceral fat around the liver by up to 80%. Number one, do you buy the hype here? And number two, what in this constellation of benefits really stands out to you? Do I buy the hype?

It depends which hype. I think GLP1 drugs are changing the world. I think, right of true tide, or as we used to call it, triple G, which I really love that name, is the most potent of them that we've seen. I think there will be more potent ones coming along. I think exactly who does what is a bit of a question, but like there are a lot of these. There are also the oral versions now. Big picture, they are changing.

They're the first really effective obesity drugs. Obesity is bad for you in a whole bunch of ways, and it seems to affect those, and it seems to have other salutatory benefits. We haven't seen drugs that kind of have this kind of broad impact since the cholesterol-lowering drugs called statins, and these are drugs people actually want to be on because they want to lose weight. They are absolutely phenomenal. The data look great. There are some side effects. They're not perfect for everybody. No medicine is, but this is a powerful class of medicines, and this is a powerful member of that class, and this is what you see when there is a company like

Eli Lilly that is at the head of the pack, and they are making sure to do every single study that they can do to show you how great this medicine is, and they're all coming up heads. Do you think five to ten years from now, we're going to think of this category of drugs primarily as a weight loss category? Or do you think there are so many different positive benefits, positive side effects, it seems,

from this drug category, that maybe a decade from now, there'll be people taking this drug that have no issue with type 2 diabetes, no issue with obesity, but they look at the effect on cholesterol, they look at the effect on fatty liver disease, they look at the effect on, I don't know, triglycerides, knee pain, sleep apnea.

And you have a lot of people taking weaker versions, lower dosage, oral versions of these drugs, just because they want the side effect profile, even if they don't want the number one thing that these drugs are advertised for, which is weight loss. Well, I think absolutely, but I want to put a caveat on that. And the caveat's mostly that a lot of those effects are probably, mediated by weight loss.

Weight gain is pretty bad for us. We're not supposed to be running this heavy. We're supposed to be, I don't love evolutionary arguments for anything, but we're supposed to be walking around a lot and be pretty skinny, right? Like, that's how we're evolved. There seem to be even more benefits beyond that. There may be benefits for addictions. There may be benefits in psychiatry.

Those are very new. I don't want to be encouraging people to take the medicines for that. But absolutely, we could end. up in a place where a lot of people are taking these medicines for a lot of things. And one of the big questions is going to be, do you want a lower dose? Do you want a weaker drug? Like, how do you, you know, I mean, it can be hard to drink water on these medicines sometimes, you know, like it's, they're not always benign.

A lot of people, you know, I, David Kessler was written about trying a lot of them, you know, but a lot of people seem to go, former FDA commissioner. A lot of people do seem to go... From my guest to this show. And at a recent set event. But people go on and off them, and there's not great medical data on that, but that may end up being the paradigm. I don't think we know the paradigm. But, I mean, gosh, we've got a class of drugs that are improving all sorts of measures

that have hard outcome data for some of the early metrics, which are people not following. Hard outcome data. I mean, like, they reduce heart attacks and strokes. Like these are drugs that may be making people live longer. So, like, absolutely these could be medicines that a lot of people end up on, especially as they get easier to take. And if the small molecules end up being, meaning the pills end up being as great as the peptides that people are injecting. They're really great drugs. On top of everything else, it might, might cure cancer.

Yet another thing that happened in June is that we got word of a retrospective analysis of more than 100,000 women. between the ages of 45 and 80 who were taking GLP1 drugs for years, they were found to be 30% less likely to develop breast cancer than the population that was not on GLP1 medications. And one of that U-PN studies authors, Elizabeth McDonald, a professor at UPenn, offered the following context, quote, while our study was observational and does not definitely confirm an association

between GLP1 medications and reduced breast cancer incidence, it does add to the growing body of evidence. And quote, Matt, how strong is a signal here that on top of all the other things it's doing, gLP ones might be protective against some forms of cancer? Well, something we know particularly about breast cancer is that obesity increases risk. So the mechanism is actually, you keep saying it's doing all these things. There's a good debate on whether it's doing all these things or whether it's mostly doing one thing, which is reversing a thing that's very bad for us.

There are people who will talk about body positivity, but generally it seems getting weight off is generally healthy for people so long as you're not doing it in an unhealthy way. So that may be the mechanism. I'd be cautious here. I lived through the statin wars. I mentioned before, the cholesterol-lowering drugs. These are medicines that like tens of millions of people are taking. They're among the most prescribed drugs in the country. I think Lipitor, which is off-patent, may actually be the most prescribed.

drug in the country. Those are amazing drugs. They reduce cholesterol, and there were all these studies saying they did all these other things, which they may be kind of do, but it's really hard to tell with observational studies. One of the things that people outside of drug development have a lot of trouble getting their heads around is that you really do not know whether a medicine is effective. until you randomly assigned to people get that medicine or something else. We can't control for all the other things that happen.

You have to do what's called a randomized control trial. And these are risks from observational data, and there could be other things going on. Would I be surprised if gLP ones reduce the risk of breast cancer? No, absolutely not. I would not be surprised by that at all. But I wouldn't take this study to the bank either. one point I want to make sure that we may because we've talked to a couple different people, including Eli Lilly's CEO Dave Ricks on this show.

And while I think you are right that overwhelmingly the thing that GLP1 drugs do is reduce obesity, it seems like there are some study suggesting that even among populations receiving GLP1 drugs who are not losing weight still see some benefits in categories like reduced visceral fat around the liver or reduced inflammation markers. So I just want to add my own two cents here, which is that, well, I think I get where you were headed, which is that, Derek, you're talking about this constellation of side effects that some people might think are happening in parallel to the weight loss. It's possible that a lot of them are merely downstream of weight loss. I get that point.

There is a separate point, though, that these drugs seem to be pressing a few different buttons in the body, maybe one in the weight loss category and another in the inflammation category that we should think about as we think about what these mysterious drugs are doing. But I'll pass the baton back to you so you can say what you meant. The addiction that I mentioned is a totally different, and there do seem to be brain-mediated effects. I would, these were for the first, I started covering GOP-1s in the early 2000s.

These trucks been around a long time. The first one derived as I will never tire of mentioning from Heelam Monster Spit. Of course. These are originally diabetes drugs, affecting blood sugar, and you know what, that's bad for you too. So they do a bunch of things that are good for you. I think the point I was making is that they may not be a whole bunch of different things.

It may be affecting one kind of metabolic path that seems to be pretty good for people. Like, there are few people that have problems, and the biggest ones seem to be related to gastric slowing or nausea and vomiting. But a lot of people are taking these drugs and seem to be a lot of satisfied customers. Before we move on to the next category of Miracle News in Medicine, I wonder in a world where you have the one single agonist, semaglutide, you've got the double agonist, terseptide, you've got the triple agonist, the triple G, red at true tide. This is going to keep happening. It's going to keep going on. New drugs are going to be more and more powerful and new versions of those drugs are going to come online that offer people new ways of getting into this drug category.

where is this all headed, do you think? Well, it is headed toward a lot of people taking these drugs, whether this becomes more than tens of millions, which is about where drug markets tend to cap out in the U.S., how much of the population are we talking? I think that's a hard question. You know, out of 300 million people, could there be a drug that 100 million people take?

We don't have manufacturing experience for that. But more importantly, we don't have pricing and access. And I think that's the bigger question. You know, I mentioned statins before. These are cheap drugs. There is the problem that people are not nearly as eager to be on them because they don't make you skinny. But here you have a similar set of benefits, drugs that are really beneficial. It's hard to keep people on them.

People don't stay on them. So I still think the idea that this is just going to change everything. I think the problems we have to solve socially around that are how do you make sure people can afford them? I mean, we have a health care system in this country that is a health care system in this country that is, a barrier for people to take medicine. We also have high drug prices compared to the rest of the world. A lot of that's going to have to work out to get to that world. And so more has to change than you'd probably immediately think. But you're going to have huge impacts on society way before that. Moving to cancer, there was a huge, huge pancreatic cancer breakthrough at ASCO.

the oncology conference. And I'd really love you to set up just how significant this breakthrough was. Why has pancreatic cancer in your mind been considered undruggable? And what does it have to do with this protein family called RAS? There has been a revolution in cancer over the past 25 years. And it's because we figured out, you know, biology genes make proteins which do things. And we figured out the genes and the proteins that are key to certain cancers.

And this has led to a lot of amazing medicines that have had big impacts, and there's been this one gene and protein called K-RAS that's been out there. The proteins called RAS that seemed like a great, what drug developers call a target. It's a protein that if you get a drug to block it,

the gene can drive cells to divide uncontrollably. It could be useful in a bunch of cancers, including pancreatic cancer. But it was considered undruggable because I've heard it compared to like when I was in a kid, we used to play water polo with a greased watermelon. It's impossible to hold. This is a smooth molecule. You can't get the proteins to hook.

And there were a couple of discoveries, one in 2013 by Kavana Choucada biologist at UCSF. Some more work by people working with a research named Greg Verdeen who founded a company called WarpDrived Bio, which then merged with Revolution Medicine, which is developing this drug, where they figured out how to drug this undrugable target. People have been hoping they could drug Rast. They were hoping it would work. And what we see with these data is that it really, really worked. an amazing drug that slows down cancer, extends life. It's not a cure. It's not perfect.

But it's one of the best results we've seen in pancreatic cancer. And there was already early data of combining another drug with it and getting more of what doctors call a response, which just means you shrank tumors. It's the first way to measure if a drug is working. So there's a lot of hope. And that's why people at ASCO were so excited that this was such a breakthrough. And, you know, I haven't seen cancer doctors this excited in a while. It's been years, you know.

So the drug's name is Diraxon Rassib, and the company that made it is Revolution Medicines. I would just like to slow down here just a little bit. And have you explained to me how it solves the greased watermelon question? Because, like, this is a really interesting, fascinating medical mystery. You've got this incredibly significant, nefarious protein. You've got to find some way to stop it. The only way to stop it is to grip onto it. It's ungrippable.

It's a pool cue. It's a greased watermelon. So how did these medical geniuses solve the greased watermelon problem? So a colleague of mine, Angus Chen, award-winning cancer reporter at Statt, has actually told this story, and I'm just going to read it to you.

A scientist at Harvard University named Greg Verdine had been trying a different approach, Like Choucate, Verdine had figured the best way to get at RAS mutant proteins would be to hit it in its on state. But rather than pouring over the protein to find a microscopic handhold, Ferdine began to wonder if there were any instances where evolution had found a way to bind a flat target within a cell. We would go looking for an answer in nature, for Dine said. The inspiration came from compounds known as molecular glues. A classic example is rapamycin, which was originally discovered in soil bacteria.

Rapamycin is a small molecule drug that hits a flat target, the protein mTOR, which only stands for molecular target of rapamycin, by the way. But it doesn't do it directly. Instead, it first binds to a different protein called FKBP12, a ubiquitous enzyme that helps fold proteins in the cell. When the two molecules combine, they form a new surface that can lock into the mTOR protein. The three compounds create a tric complex with the small molecule rapypomase. and sandwich between the two proteins. Nature figured out if I bind the small molecule, not on its own, but in a complex, that will give the extra oomph that's needed to contact the target, for Dean said.

This is nature doing its thing, man. If you could go out and willy-nilly change the surface of this molecule, you could reprogram it to target RAS. You could pick a new target and dial in the molecule. That's the guy whose company was bought by Revolution Medicines. That's basically how it works. Wow. So is it like, it's like a bear hug? It's like we created a drug that's like giving the evil protein a little bit of a bear hug and strangling it so it can't go do its nefarious business. Yeah, we found another protein that nature had figured it had to target a thing that was hard to target and we just stole it and made it target something else. It's all hacking. It's all hacking. It's all hacking. So you got, there's obviously so much excitement about this because pancreatic cancer is so deadly and it's been so expensively researched with nothing. coming close to a breakthrough like this.

You know, now a lot of people, as this is the media's one, getting really excited about the idea, this is the first step toward a cure for pancreatic cancer. How far are we from a drug like this and a cure for pancreatic cancer? This isn't the first step. The first step happened maybe 30 years ago. You should view this as,

There was a drug that was on the cover of all the magazines 25 years ago that did this in another disease called Chronic Myelaglis leukemia called Levec. There are, we are learning to develop these targeted medicines. They are mostly not cures. They extend lives a lot. There was a trial of another one, a drug called lorlatinib, developed by Pfizer and non-small cell lung. cancer that's caused by a particular mutation called ALC that also showed the longest survival without cancer returning called progression-free survival. The longest time before your cancer comes back that we've seen with a targeted drug. I think it was seven years. These drugs can have really

dramatic effects. This is a revolution that's been ongoing. This is a revolution that has had other standing ovations. This is a big win. in that revolution. But you understand it better if you see it as part of that series of battles and they come along every couple years we get to cover one of these drugs and it has a huge immediate impact on a disease

and then you move it earlier and it has an even bigger impact. You combine it with other drugs. It has a bigger impact. This is how the war on cancer is being fought. There's a lot of people that I talk to in the frontier AI labs where you say, you know,

do you think this technology, artificial intelligence is dangerous? And they say, yes, we think it's very dangerous. And then you say, do you think it's going to destroy a lot of jobs? And they say, yes, it's going to destroy tens of millions of jobs potentially. And you go, well, why are you building this? And they always say, like almost all of them say, well, look, artificial intelligence is going to cure all disease. It's going to cure cancer.

What is the state of artificial intelligence curing cancer right now? I've been talking to a lot of people about this on the drug side. And I would say the jury is still out. But I want to tell you there's a range of things that artificial intelligence is doing for trying to cure cancer. One of the biggest and clearest. I actually had an interview with Marquese Levine, who is the CEO of a company called Zara, which has raised a ton of money to develop, use AI to develop drugs. And his big targets are kind of cases like Dirax-on-Rassib, where, like K-RAS, where you have a target you can't drug and you think AI is. is going to help you make the antibody or the molecule that allows you to drug the undrugable.

That's a really good use case for making some progress pretty quickly. There are also things that sound smaller, but they're a big deal, like enrolling people in clinical trials fast or identifying people to be in clinical trials. Conducting clinical trials is one of the biggest roadblocks here. But the really big win is that the AI can understand biology better than us and predict what drives. drug will work. And the problem is it's a little bit like you're you're asking Claude something, you're getting an answer, and you find out if it's right after you the clinical trial, which takes eight years if you're lucky. So there's a real chance of you think you have the AI that solves the biology that fixes things. And you don't do any better than drug developers do now,

which is for years we've done about, about out of every 20 things that enter human beings in clinical trials, one or two make it to the market. And this is a financially brutal thing to the extent that tech people have failed at this before. They have thought they were succeeding before. And whereas the semiconductor industry talks about Moore's law, the drug industry has a term called E-Room's law, which is Moore's law backward because the costs go up that fast.

So I think we have to be careful of irrational exuberance while realizing it could work. How about on clinical trial efficiency? You know, it's one thing that... Oh, it's huge. What I hear about all the time is just to set you up. No, absolutely. But it will allow us to experiment faster.

And it's possible that accelerating the speed of trial and error will help us cure cancer in our lifetimes rather than the lifetimes of our grandchildren, right? Like speed is not a cure, but speed compresses the distance

between where we are now and a cure. So could you just talk a little bit about how you've heard folks in AI talk about clinical trial efficiency because this sounds nerdy and esoteric maybe to a lot of people,

but if we could have clinical trial readouts faster than eight years and for less than what $7 billion, I mean that would be enormous for medicine not just in cancer but you know throughout all medicine Right well the three to

the $3 billion plus number comes from that includes how many things fail so your biggest impact is stop picking things that are going to fail but yes I wrote a piece a few years ago where the thesis was this is biology century

which is what we're talking about and we're not ready for it. And the main way I said we weren't ready for it wasn't all the other stuff people worried about. It's we're too slow at clinical trials. We don't know a medical question. You go into the doctor's office. They don't know what to give you.

You should be enrolled in a trial. I mean, we could answer so many things so fast if we did that. If we did the kind of AB testing that tech companies do all the time. Yes, with informed consent. Yes, with people knowing. And tech has been a potential big enabular for that.

But, it's not it's not the only one and AI is definitely speeding things up using AI to identify patients instead of having people do it by hand using AI to track also just technologies that track data in real time

all these things are great but you know China's enrolling clinical trials a lot faster than the US and running them faster and some of that's lack of ethical barriers that might exist but some It's also having everybody in the same EMR and identifying them in a culture that says if you are offered a trial, you go into it. That's one of the big advantage that Chinese companies have over U.S. companies in biotech and is something that is freaking biotech investors out. They're also freaked out about IP.

But yes, absolutely. Any AI technology that can help us run clinical trials faster is a godsend. more than that, I would say that we should remember, there's often the temptation to think the AI can understand the data. We don't need the randomized trial. The AI can look at big populations and the AI will figure out why, what the confounders are, right? I think we have to be very careful about that. I think we haven't found anything that's like a randomized controlled trial in medicine,

and we shouldn't forget that. The AI should help us run them. not replace them. One thing that Lilly, CEO, Dave Rick said on the show that I thought was really useful is that one thing that makes large language models so effective at answering certain kinds of questions is when there's a deep corpus of information in the question that you're asking. So if I have a question about, hey, can you explain the Habsburg Empire? Can you explain Adam Smith's wealth of nations? Well, there's been trillions of words written about these subjects. So it's very easy for Claude or OpenAI to essentially synthesize some answer based on that enormous corpus of text.

There is no really high quality corpus of information about the entire molecular grammar of our bodies. This is in many ways still a mystery for us. There is no perfect internet of the human body. And so he said that's something that he's trying to build. And I think Nvidia is trying to build it with a lot of other companies that can. How do we create, essentially, the training set that would allow AI to be as masterful at protein folding, as masterful as it is now at, say, you know, figuring out protein folding problems, that masterful at every other question about knock-on effects of introducing some small or large molecule in the body,

that that's a huge, huge impediment to AI being a really fantastic tool for drug developers. Do you generally agree with that notion? I generally agree with the approach in their other companies, Genentech, has, a whole idea called the lab and the loop, right, where everything's looping back and it's very AI forward. The thing you have to be careful about, and I know Lily in particular is, Lily is doing everything right now. They have this huge influx of cash, and they're trying all sorts of things, and they're buying a lot of things, and they're doing really out there stuff, and they're doing, you know, basic drug development. I think you want to be careful about not losing the discovery

techniques that have worked so far. And I don't know how fast this happens. I wouldn't be shocked if suddenly some drug company has an internal engine that lets it really predict what's going to work and what's not. I also wouldn't be shocked to see somebody build one of those and find out that it's predicting the wrong things and they make all the wrong bets, right? Like, I think this is more complicated than it sounds. And I do kind of think the, you know, you ask Grock about vaccines experience can be interesting. Like the training set matters and you will hit points where with large language models where suddenly it doesn't understand something. And how do you deal with that when it's something you don't understand? Like molecular biology, something you can't test

another way. I think that's the big problem for AI in drug discovery is how do you know when you've made the model right? How do you validate it? I think that's a really hard problem because we're making drugs and putting them into people after all the animal testing, after all the petri dishes, and they don't work almost all the time. So how do I know if the AI is smarter than me? Before we move on to our third category

that I really want to talk to you about a final question in the realm of cancer. You know, you've got Diracson Rassib, you've got other breakthroughs in the realm of checkpoint inhibitors. We talked a little bit about the frontier of artificial intelligence. intelligence. I mean, how optimistic are you that this is a really special moment in the history of the war on cancer? How optimistic should we be that we're living through a kind of golden age

in drug development against cancer? I mean, my pushback on that is actually how much has happened over the past few years. I think there's a problem with medicines that people don't realize, you know, I think of my father having his gallbladder out. Um, you know, it was a brutal surgery. You got a big scar. It's done laparoscopically now. People don't really think about that every time someone has a gallbladder out. People don't really think about the extent to which heart attacks were deadly and frequent 20 years ago, 40 years ago. We have made huge progress at treating a lot of things, including types of cancer. I do think we are making accelerating process. I think you do better viewing it as something that's building and where we're doing a lot of and recognizing a

lot of incredible things than thinking that this is the moment. My experience covering this is that you often think this is the moment and it slows down. If you told me five years ago, or maybe 10 years ago, if you told me around when we were all really excited about cancer immunotherapy that we would not find another target like PD1, which the checkpoint inhibitors hit. We all thought there was going to be a flood of new cancer medicines that worked on the immune system. There's been a trickle. And it's kind of not the main approach people are pursuing. These things come in amazingly fast fits and starts. And suddenly an amazing thing happens, but it's taken 15 years to get to this point.

And sometimes it leads to the next amazing thing and sometimes you have to wait. So now we're moving from cancer, which is the number two killer in this country, to heart disease, which is the number one killer in this country. In May, Verve, a company, had a small preliminary study where they found that an experimental gene editing treatment

dramatically lowered cholesterol levels, perhaps permanently, after just one infusion. The New York Times reported this as potentially if it worked out as a kind of one-and-done shot for preventing vast swaths of heart disease. Matthew, how big a deal is this? Well, depends what part you're asking. I think the idea of this gene therapy is potentially revolutionary and is a big deal. I've thought that since Sikar Katherison, who founded this company, who's a cardiovascular, geneticists, I know them pretty well, had the idea.

And it's a radical solution to the problem that we know that bad cholesterol, low-density lipoprotein causes heart attacks. We know that drugs that target PCSK-9, which is the target of this gene editing, reduce that risk. We know that statins, which also hit LDL, reduce that risk. we know that people don't take them so sakes argument was well they're probably people

who'd be willing to edit their genomes and then they have lifetime low risk and the scientific argument there makes sense I'd say this is a very early result we saw this is the second gene therapy

gene editing treatment that sakes group has developed that verve which is actually owned by lily now to do this the first one had a safety issue there was a hiccup I don't foresee a hiccup with this one, but I've been covering drug development for 25 years

and I've learned not to be too confident in a result this early. That said, this is one of the most amazing drug targets in medicine. And there are people who are, we learned about PCS canine, which is a gene that makes a protein. Because there are people who have double knockouts of PCS canine, which means they don't have a functional. copy of the gene, they're fine, their LDL's low, they don't have heart disease. And we know from giving statins in so many gigantic trials that drugs that lower LDL prevent heart attacks, strokes, and deaths. So the idea is not as crazy as it sounds. I do think it's a bit of a Rorschach test because I think some people hear this and like, yeah, I want to edit my genome so I

won't have heart disease. And some people hear this and are like, I'm never editing my genome. Why would I edit my genome when I can take a drug? And I think that's a big cultural fight will eventually get to have, even though we're not there yet. Right. I mean, one advantage of editing your genome is that there's no issue with adherence, right? It doesn't matter if you forget to take your statin every day. It doesn't matter if you forget to take your statin for a year, a decade. You've already edited the genome to do the thing that the statins would be trying to do anyway. You said this is one of your favorite targets, PCSK-9. Can you just say a little bit about what PCSK-9 is, why it's such a popular target and, me, like, how this therapy works? So this is, it is a really cool target. I remember

learning about it in a Chinese restaurant in Manhattan. So it's, but PCSK-9 stands for the pro-protein-Kin-Type 9. Yeah, it rolls right up the time. Which, rolls right off the tongue, it's an enzyme, like most drug targets. The type 9th enzyme. But it's been a really great drug target because there was some research, particularly done by a researcher named Helen Hobbs, where she looked at big populations of people, a lot of them were African-American, and found people who had defective copies of this gene, meaning the gene didn't work. And what happens when you have a defective copy of PCSK9 is that you have lower LDL or bad cholesterol. And you can see in these databases of people where they looked at their

genes and were looking at their health outcomes, that they had lower risk of heart disease. And more than that, there were people who had double knockouts of PCSK9. I remember being told about one who was an aerobics instructor, right? Like, this is, this is like drug developers love this. They love human proof. They call them human genetic knockouts of people who, where a gene doesn't work right and they're better off.

Yeah. Someone described it to me like discovering X-Men within the human population, right? It's like discovering some mutant who, that have defective genes, which you think would be bad, except this defective disabled gene, in fact, makes you,

practically invincible to heart disease, which is like an extraordinary, you know, comic book hero mutant power. So I have this PCSK-19. It's really boring to be standing next to Wolverine and your power is you don't have heart attacks. Longer life, though.

But you outlive him, yeah. I mean, maybe his healing factor does that. But, um, the, uh, but yes, and there was, there was this, there's this aerobics instructor and she just doesn't have, She has really low cholesterol. I forget how low it was. So companies went and developed drugs based on this. They made antibodies which knock out this enzyme. And it turned out to be a way to lower cholesterol. And the two main drugs were originally from Regeneron, really interesting biotech company that focuses on genetics and Amgen. And they ward it out in the market and they were unlucky enough to come after another medical revolution. which was new drugs for Hep C, which had broken the bank for a lot of people. And the kind of insurance system rebelled and was like, well, if everybody gets these, they'll be really expensive.

So we're going to show we can tamp down on them. And it was so extreme that the drugs never really took off. And Amgen kind of stayed in the game and eventually got a blockbuster, but they're disappointing sellers. But Merck has a pill that works on it. Astrosenica is working on one, two. but it's just this amazing case of the genetics are so clear. You want a drug that knocks out this enzyme. You don't need this enzyme.

It's actually good for you if you knock it out in our, at least in our modern world where we all end up with really high LDL. The other problem for these drugs is that the statins, which I mentioned, which are through a different genetic mechanism that also lowers cholesterol, are cheap generic drugs and are taken by a large portion of the country. So, but it makes total sense as a gene to edit if you want to edit genes. The question is whether, really, whether people who are going to be editing their genes, whether we're really going to start in heart disease, where people can be a lot of risk,

or whether we're going to start in something that sounds more severe or that is more severe. I do want to say, though, there is a benefit that you didn't mention. It's not just that you don't forget. it's that we start treating high cholesterol late. Biologically. And we're probably never going to do a drug trial that tries starting to lower your cholesterol early, and there are some safety reasons not to start too early, particularly for women of childbearing age, at least in theory.

And I'm not actually sure there is a safety. But people, you tend to start statins later. Although some people do start them early if they're at high risk, people with that age. But we're careful about that. But it's really your lifetime exposure to LDL that lipidologists think adds up over time. You get this plaque in your arteries. It becomes inflamed. It's likely to burst.

It bursts. It forms a clot. It blocks the artery that causes a heart attack or a stroke if it's in the wrong place. It's kind of a random process. And so there might be a benefit someday. Like you're talking everybody being on GLP ones. I could certainly imagine a world where people edit their genes really early so that they never really have to worry about heart disease. I think that's a very, very, very long way away.

I think culturally we're not ready for that. I think medically we're not ready for that. I think you'd want a lot more evidence that editing your genes in your liver doesn't do something else that we haven't thought of. Right. than a small study in the New England Journal of Medicine with really just a few patients. I want to put some of these breadcrumbs together and see if we can, you know, make bread, so to speak.

We've talked about GLP1s. We've talked about the breakthroughs in cancer. We've talked about this gene therapy for heart disease. When I put all of this side by side, and I think about more and more people taking GLP ones, reduced obesity, reduced visceral fat around organs, maybe even some knock-on effects for certain types of cancer like breast cancer,

among middle-aged women. You also, with the cancer news might see, you know, new breakthroughs and drugging previously undruggable cancer proteins, even maybe using AI to find new protein targets and hopefully at some point reduce both the cost and time of clinical trials so that we have answers faster and cheaper. And then you add to that, this frontier of gene therapy that we're seeing in this, as you said, small study in the New England Journal of Medicine showing that it can have the ability to knock out PCSK9 and maybe significantly reduce heart disease for many people.

I mean, when you put all of this together, it's a pretty exciting picture. And I wonder if you think I'm a little bit over my skis thinking about the possibility of dramatically, dramatically, reducing the mortality of cancer and heart disease in our lifetime. Not at all, but I'd give you this opposite thought experiment. What if instead we lived in walkable places where we all moved around a lot and we didn't sit at desks and we weren't living in a food environment that was dramatically unhealthy in ways we don't fully understand? Would that have a bigger or smaller effect? I don't really know the answer. I do also want to add the caution that the big story of medicine,

is that we've made breakthroughs and we can't always get them to people. People don't always get the medicines they need. Too many people die of cancer because it's caught too late. Too many people don't get the drug that would help them. It's a problem when we talk about these gene-targeted drugs. There are a lot of people who have these mutations

who don't find out because their tumors aren't sequenced. So it's... I always say inventing a new drug is harder than putting a person on the mood. This is harder than rocket science. But actually, Dan Skavronsky-Lilley has talked about how hard the problem of how you get medicines to people are. And what other kinds of systemic fixes do we need when we're developing all these great medicines to make sure they get to people? And I think we have we have trouble even talking about those problems. They often devolve into, well, the drugs have to be expensive.

Look, there does need to be a big return on investment so long as drug development is this risky. That also doesn't mean that every high drug price is excused by that. There are certainly cases where industry is priced too high or behaved in ways that were not great. But I think solving those problems is every bit as hard, if not harder, than inventing the breakthrough drugs. Let's pause on just one of these problems because we're not going to solve

every problem of medicine in the next five to ten minutes. But I am very interested in the problem of price, the fact that Americans in particular pay much, much higher prices for new branded drugs before they go generic. And the answer that we keep hearing

from the pharmaceutical companies is, well, R&D is incredibly expensive. clinical trials are expensive. If we can't make that money back from our drugs, then not only does the company maybe die, but then nobody has money to put into R&D. And I'm interested in solutions that exist at the federal level. Like I'm interested, for example,

in the possibility of golden tickets or prizes where the federal government is essentially saying, you know, X company, whatever. Lily, you expect to make, I don't know, $10 billion from this drug. Maybe we'll say if you make a drug in this capital, category, we'll buy it for a guaranteed $5 to $7 billion. This is what we did for the COVID vaccines. We essentially told every pharmaceutical company, if you make a drug and it's good enough and it

passes phase through clinical trials, we are going to buy it from you, even if it's the 27th COVID vaccine that comes out. We'll still guarantee your billions of dollars in terms of a payout. That's one solution. Another solution also experimented with during Operation Warp Speed are something called advanced market commitments, where you essentially say, you know, it's similar to the prize. If you build this thing, then we promise to buy a certain amount of it. And that gets some money into scale and gets you over the problem of having to depend on consumers and insurance companies to pay the pharmaceutical companies rather than the government. Do you like those ideas of the government essentially paying upfront? Or what are other ideas that you have for

being down prices in the short term before the drugs go generic? I have another. I have another idea I like better, but I want to answer those first. I like advanced market commitments better than prizes, and I can explain why really simply. If you look at the net present value of something like a GLP1, you could be looking at something on the order that's bigger than the entire annual budget of the NIH. Your prizes are just too small. Which is for, for, wait, $40 billion? Yeah, I'm thinking like it's a $50 billion. I'm thinking 5x sales, right? Like you got a $10 billion drug, 5x sales, $50 billion is probably cheap for that asset. The amounts of money that are involved are staggering and you're going to end up in a lot of situations where the companies, you know, part of why Pfizer didn't take funding for its COVID vaccine.

You know, it spent its own money. There's also, when we look at COVID, you know, just a thing to keep in mind is that if you look at stock price, The companies that failed to develop in COVID vaccines actually did better than the ones that succeeded. It's been hard to be modern and hasn't been all that easy to be Pfizer stock price-wise. Merck and Astrozenica, you know, did really well. Asteroenica did develop a COVID vaccine, but was in kind of the position of it was used in some places, but eventually fell by the wayside. So it's really hard to create that kind of incentive for everything. It could really be helpful in areas like antibiotic drug development is the scariest thing to me.

We're not developing new antibiotics. We will need them. There is resistance. The results could be terrible. We don't really have a good mechanism for creating them. They haven't been commercially successful. What I think... You said you like something better.

I do. I think the government should be in the drug development business. And I think it should be in the drug development business, not for every drug. But there are a lot of rare disease drugs where you could see government funding taking a medicine to market. And then you would have a counter to the story the industry tells of it cost us so much. it was there were i i've done studies of i've done work on the cost of drug development and like to say the cost of developing a new drug is somewhere between a hundred million and uh you know 10 billion dollars depending on your failure rates and depending some drugs are really cheap to develop

and i feel like we've got ourselves in a situation where industry does all the development maybe that's not the best thing. And maybe it would be good if we had some more efforts to actually bring drugs to market through the NIH. But I think even that requires, the industry spends a lot more in development than we spend on medical research from a public standpoint. So it would require some serious coin in our part. I also love the idea of a Manhattan project to speed up drug development to focus on clinical trials and making them faster and creating the technologies, you know, kind of a clinical trial superhighway. That's not even really something. It was something that among a lot of ideas I didn't think worked very well. It was something McCarrie was kind of

talking about that was promising, the previous FDA commissioner. So I do think that there are things we could do if we actually focused on these problems and our goal was we want the drugs that emerge in the future to be developed more cheaply than the ones develop now. And then also you have to have just making development cheaper won't bring the price back down. So you're going to have to figure out what the mechanisms are that do that in a way that doesn't mean the companies just abandoned areas like they sort of have for antibiotics. Matt, last question for you. I'm just so struck by the fact that all of this stuff seems to be happening at once. The Red of Trutide News, the gene therapy news, PCSK9 News, the cancer news. It's just really exciting.

And it's funny because, you know, 2026 has been a very exciting year for scientific discoveries, for medical discoveries. Whereas 2025, all of the podcasts I was doing, all the reporting I was doing was about the Trump administration's war on science and the cutting of basic research spending and all of the hanky-panky going on at the NIH and NSF. And I wonder, how should we how should we think about these two things happening side by side, that on the one hand, you have these medical breakthroughs, and on the other hand, you have this administration war against basic research. How can these two things coexist? I mean, they coexist because the outputs we are seeing in even something like gene editing for PCSK9, which we just said is really early, are building

on years and decades of research. I mean, a lot of this stuff, we're talking about all these genes. started covering this stuff, I was really excited about there was a war between a private company, Salarogenomics, and the government, which was trying to sequence the first genome. And they're both trying to sequence the first genome, which is now something that we probably do every day, cost $3 billion and it costs $300, right? Those advances are what allow you to do PCSK9. They would allow you to figure out that Keras is the cancer gene. I was really excited at that time about a guy who was, you know, he was kind of a science cowboy, a guy named Craig Venter who was running Salera. I just wrote an obituary of Craig.

You know, the first antibiotic that I saw come through, that were about coming through, the guy who developed, it died of an antibiotic drug-resistant infection. that these things take lifetimes and so if you're looking at stuff coming out now and saying you're okay you're not and in a lot of this in you know the heel a monster spit example people like to use it comes from having not just the federal government funding research but from having a partnership between research and and basic science that's functional and that leads to new medicines. And so that's why you worry about these things. And you really are seeing with the Trump administration, you know, I've written a lot about the impacts on the FDA.

I wrote that Marty McCarrey was the worst FDA commissioner that I ever wrote about. We've seen it there. that is not great for drug development, that instability. We've seen it at the CDC. And the question of you have people doing cancer research who were worried about where their grants were going to be. Now, what I would also say is some of that has stabilized, and particularly in things like cancer grant funding.

So, you know, it may have been that last year, we can all hope that last year was the worst year, although people are feeling a bit better, but they're not feeling good. right when I go to a meeting like ASCO. I mean, last year at ASCO and ASCO, I think we wrote a lot about people were worried about science getting done.

It was less of a topic this year, which was nice. It was nice to write about some drug data instead. Yeah, and this seems an important piece to land on, which is that, you know, I started this whole episode with a quote from Lenin about how there's decades or nothing happens in weeks when decades happen. But it's important to remember that the weeks, when decades seem to happen,

those weeks take decades, right? Like, it took decades to get us to this month when, for any number of reasons, you happen to have this wonderful constellation of fantastic news across medicine. So thank you for helping us understand it. Matthew Herper. Thank you so much, Derek.