Freakonomics Radio - 671. Why Has There Been So Little Progress on Alzheimer’s Disease?

Episode Date: April 17, 2026

One possibility: a leading hypothesis pursued by researchers (and funders) was built on science that now appears to be fraudulent. Stephen Dubner speaks with the scientist and the journalist who blew the whistle.   SOURCES: Charles Piller, investigative journalist for Science, author of Doctored. Matthew Schrag, associate professor of neurology at Vanderbilt University Medical Center.   RESOURCES: Doctored: Fraud, Arrogance, and Tragedy in the Quest to Cure Alzheimer's, by Charles Piller (2025). "The brain makes a lot of waste. Now scientists think they know where it goes," by Jon Hamilton (NPR, 2024). "The history of Alzheimer’s disease," by Lisa Kiani and Richard Hodson (Nature, 2024).   EXTRAS: "Can Marty Makary Fix the F.D.A.?" by Freakonomics Radio (2026). "Are You Ready for the Elder Swell?" by Freakonomics Radio (2025). Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

You may have seen the news recently about two new FDA-proofed blood tests that may detect Alzheimer's disease in the early stages. Early detection is important for any disease, but especially for Alzheimer's, which can take root for 20 years before symptoms develop. These symptoms, as you probably know, include memory loss and other cognitive, physiological, and behavioral issues. The reason I say that you probably know these symptoms is because Alzheimer's affects more than seven, million people in the U.S., most of them over 65. Age-related memory loss has been seen throughout human history, but the disease was not formally documented until 1906 by the German physician Alois Alzheimer. When Alzheimer autopsied the brain of a woman who had had memory loss and hallucinations, he found that her brain had shrunk and withered with numerous tangles and

what he called peculiar deposits. Scientists have been trying to figure out those deposits and tangles ever since, the National Institutes of Health spends around $4 billion a year on Alzheimer's and dementia research that's up from around $1 billion a decade ago, and that puts it second only to cancer spending. Which makes sense because the elderly population in the U.S. is big and getting bigger. Much of this Alzheimer's research is centered around one dominant theory of the disease. But what if that theory is flawed? No one's getting better with these drugs.

Every scientist who works with them, every clinician, will say the same if they don't, they're lying. So is flawed even the right word, or should it be fraud? Cheaters tend to cheat. When we find clear overt problems in a paper or in a group of papers, often it propagates through somebody's entire work. Today, on Freakonomics Radio,

we follow an investigation that found decades of problematic Alzheimer's research, and we ask, with some sense of earned optimism, where does Alzheimer's treatment go from here? This is Freakonomics Radio, the podcast that explores the hidden side of everything, with your host, Stephen Dubner. The story we are telling today is a sobering one and an important one. We'll hear from two people. We had hoped there would be more, but in the end, the others did not want to talk.

You'll understand why as we go. Let's start here. My name is Charles Pillar. I am an investigative journalist with science magazine and author of the book, Doctored. The full title of Pillars book is Doctored, Fraud, Arrogance and Tragedy in the quest to Cure Alzheimer's. Charles Pillar has been a journalist pretty much his entire adult life. I grew up in the Chicago area. I became interested in reporting.

Kind of the way that so many people, boomers like me, did, being inspired by the Watergate investigations. I was too young to be a journalist then, but not too young to be inspired and thinking about things, particularly during the Vietnam War period and all of the societal challenges there. Pillar is 71 years old. Before coming to Science magazine, he wrote for the LA Times, the Sacramento B, and the life sciences news outlet stats. For the past several years, his reporting has been focused on Alzheimer's research. I asked him why. Well, the answer is to me pretty clear, and that is that unlike cancer, diabetes, heart disease, and myriad other ailments that afflict us,

Alzheimer's disease is something that the nation and the world has spent tens of billions of dollars on in recent decades. And yet, we have no remedy that arrests or reverse. the terrible cognitive decline of the disease. And so even though there may be corruption in cancer research or in heart disease research, we've seen enormous advances in those diseases and sometimes even real cures for specific ailments that have benefited society dramatically. We have not seen that in Alzheimer's disease, and that makes it not just more tragic, but also terrifying for people who are facing those possible symptoms. What is the thinking on potential environmental drivers of Alzheimer's disease?

Yeah, it's a great question, and I would say that care and prevention is one of the key things that I think our society should be increasingly concerned about. It's well known that people who are subject to highly polluted environments are experiencing Alzheimer's at a higher rate and also at a more severe age of onset, earlier age of onset. Secondly, I would say that, like so many ailments in our society, Alzheimer's is in part a disease of inequality. Now, of course, anyone from any economic level can get the disease, but it's more prevalent among people who do not have the same level of economic opportunities. To what degree would that be caused by less or worse medical care itself? In other words, how much advantage is there to early or constant monitor? and then whatever kind of treatment there may be? Or are you saying that low-income people may have, in their lives, drivers that are responsible for Alzheimer's? I think it's both things. Clearly, problems like obesity are important in Alzheimer's disease. The other thing is that they've found over and over that the higher educational attainment that people have, it has a benefit that can reduce the severity or the age of onset.

for the disease. Okay, so that was Charles Pillar on a few potential drivers of Alzheimer's. So how does the disease itself work? For that, we need a neuroscientist. My name is Matthew Shrag. I'm an associate professor of neurology at Vanderbilt University Medical Center. I take care of patients who have strokes and memory problems, and I run a laboratory trying to understand how blood vessel diseases contribute to memory problems like Alzheimer's. disease. Describe for me how much of your work is what a layperson like me might consider Alzheimer's related? Most of it. Our perspective is that blood vessel diseases contribute to Alzheimer's disease. We don't think of Alzheimer's disease as a blood vessel disease, but that's one of

the issues that we've been working on is trying to understand the degree to which these other factors that have been largely overlooked may play a key role in how this disease works. So just walk me through quickly all the schools you attended and all the degrees you got. Yeah, I was at the University of North Dakota for my undergraduate work. And then I went to Loma Linda University and I got an MD and PhD there. And then I went to Yale for my residency work. Were you wanting to be this kind of scientist since the time you were a kid? Did your family or school or background encourage you? Not really. You know, this happened partly because I sat down with somebody when I started getting serious about trying to get into medical school and said, what are the things I

should do to improve my odds of getting in? And one of the things they said is, well, you should do some research. I went and knocked on my professor's office and said, can I do some research? And he said to me, son, they didn't mean botany. And so he very graciously didn't take advantage of me and shuttled me to a more appropriate laboratory. The lab that Shrag got shuttled to was run by a neuropharmacologist named Othman Greby, who studied Alzheimer's. I really enjoyed the atmosphere in the laboratory. I really loved the ways that you could try to study things that were too small to see and try to infer how complicated systems worked.

The research that Shrag worked on with Greby looked at the potential impact that diet could have on the onset and progression of Alzheimer's. In 2006, they and a few other co-authors published a paper in the Journal of Neurochemistry called deposition of iron and beta amyloid plaques is associated with cortical cellular damage in rabbits fed with long-term cholesterol-enriched diets. So what was the idea of this paper? The idea of this paper was that cholesterol accumulates in the brain. We had raised a group of rabbits and fed them a high cholesterol diet, which was thought in some

ways to mimic Alzheimer's disease. And the thought with this paper was that cholesterol, levels would accumulate within the neurons themselves and that that could have a toxic effect on the neurons. Pretend that I know nothing about the neuroscience here, which, you know, would be close to accurate. Can you just summarize the state of Alzheimer's research, what's known, what's not known, what's disputed? Alzheimer's disease is a complicated disease, but our current formulation of what happens in the brain is that there is an accumulation of these proteins. that clump together. And two in particular,

one is called beta amyloid, or just amyloid, clumps together outside of the cells of the brain, sort of in between the cells, and another protein called tau clumps inside of the cells of the brain, and that leads to the neurons eventually not functioning properly and eventually dying, and the brain shrinks, and people begin to lose their memory. Where did the tau cells come in in that scenario?

Tao is inside the neurons. It's a protein that is involved in the structure, of a neuron. And if you think back to high school biology, you know, the neuron is this big, complicated cell that has a very long tail on it. It's one of the biggest cells in the body, and they have to live for your entire life. And the tau protein is part of how the cell gets that unusual structure with the long tail that allows it to function like an electrical wire. So if you have tau clumped together and no longer participate in maintaining the skeleton of the cell, the cell doesn't function properly. So that's the second key piece of what's going on in Alzheimer's disease,

that the neurons are structurally destabilized. So for at least since, I would say, about 1990 in round numbers, the main thought in this field is that the plaques, this little beta amyloid protein that clumps outside of the cells, is the primary driver of the disease. And a huge amount of the effort related to drug development has been dedicated to preventing it from accumulating. or breaking it up and clearing it from the brain. The general idea is that plaque is the first domino, and then there's a complicated series of events leading up to the neurons dying, but the plaque is the primary driver of Alzheimer's disease.

This domino effect is known as the amyloid cascade hypothesis, and that is the hypothesis toward which many billions of research dollars have been directed. So what's the problem with that? The problem is, one, the idea may not be. be correct. And two, the biology of this is likely much more complicated. And I think that this is the primary issue. It's not that beta amyloid has nothing to do with what's going on, but most diseases at the end of your life are not simple. I can't think of another disease where there's one single first domino that falls that causes the rest of the cascade. By targeting just one

very specific factor, I think we're not having a big enough impact on the disease. Shrag's own research, which looks at the connections between Alzheimer's and blood vessel diseases, is a pushback against the amyloid cascade hypothesis. And Shrag has also been pushing back as a whistleblower. I understand that outside of your work at Vanderbilt and your work within the academic community generally, that you also work as an independent research integrity consultant, are all four of those words correct and in the right order? That's a good description of what I do. When you first started pursuing a career in academic research, is this how you envisioned it was going to work out? Not in the slightest. No. So how do you feel about it at this moment? If you could start again, would you pursue the whistleblower detour or no?

I think I probably would simply because of how I'm wired. But I think that initially when I started down this road, I did it knowing it might have been a pretty colossal lapse in judgment. For yourself, for your own career. That's right. But my viewpoint, is that it's time for a direction change in this field. And I just think that sometimes you need to make a decision and just go with it. Matthew Schragge's first exercise in validating scientific research was unofficial. It came during his Ph.D. training when he tried to reproduce the findings of a paper that had parallels to the rabbit paper that he had worked on as an undergrad with Offman Greby. He found that the work was not reproducible. In 2021, this work took on a new scale.

I was approached by an attorney and asked to help a group of whistleblowers investigate their concerns about the integrity of the data underlying this drug semifilum, which was being developed by a little biotech company out of Texas called Casava Sciences. Casabas sciences had been known earlier as pain therapeutics. They spent years trying to break into the opioid market, and then they switched their focus to semufilum, an experimental drug for Alzheimer's. The company's research collaborator was a medical school professor at City University of New York named Hau Yan Wang. His research had found that semufilum could reverse the misfolding of the protein filament A, which was hypothesized as a cause of Alzheimer's. Consistent with the amyloid hypothesis, semufilum also promised to reduce inflammation by targeting the buildup of cow and beta amyloid in the brain. They had produced some data from their phase two trials that raised alarm bells for these whistleblowers. Who were the whistleblowers?

They were a pair of scientists, a cardiologist and a neuroscientist who had some experience in this space and basically said they didn't think the pharmacology of this drug made sense that there were some problems in the basic science that were just hard to interpret. And then when they looked at the image data, a lot of those photographs looked like they had been retouched. So I was paid a fee to help them research this and produce a document cataloging their concerns, but also to provide due diligence to the attorneys that were pursuing this. Was the case about these scientists whistleblowers trying to prevent this drug from going forward because they thought it would be ineffective and was based on misrepresented data? Was that the genesis of your involvement in this? It was certainly the driver for me to take the risk of speaking publicly on an issue like this. You take a substantial risk to engage on a topic like this against a multi-billion dollar company.

And where were you in your career at this point? Oh, I was very junior. I was an assistant professor pre-tenure just getting started. How did you then get on the list of people to be called by an attorney representing these whistleblower scientists? I had given some interviews in the national media, and I think they were looking for somebody who was a bit outspoken, perhaps, but also who had some credibility. Where did your credibility come from to that point? I had worked as an Alzheimer's researcher for quite a period of time at this point, and at the time, one particular drug, aducanamab, was getting a lot of oxygen, and I had been talking publicly

about some of the weaknesses of this approach to treating Alzheimer's disease. Okay, getting back to the Casabasian's case, the attorney. For the whistleblowers reached out, and you said the images looked like they had been retouched. Give me some detail there. The images just didn't fit what a scientific expectation would be. Did they look visually manipulated? It looked visually manipulated. Some of these results had copy and paste type changes where it looked like things had been moved around or duplicated,

or where the intensity of one part of the image had been adjusted. And ultimately, a lot of these results, how dark a splotch on the photograph is, is what we ultimately measure to turn it into data. So changing the intensity of those values changes the underlying data. To a layperson, that sounds like a sort of Mickey Mouse misrepresentation. And I don't mean to insult Mickey Mouse, but it seems kind of amateurish that to the naked eye, the images in a scientific paper could be seen to be manipulated. I don't know if you've had the experience of somebody showing you something and saying, what do you see, and you don't see anything, and then somebody explains to you what you're supposed to see, and now all of a sudden you couldn't possibly turn the clock back and not see that. I think there's some of that. I think once you've seen the patterns, you're right, that they seem quite obvious. But I'll also say there's a difference between seeing something suspicious and being able to prove it. And that's part of what we were working on, was trying to do this in a very systematic and thorough way, and then also show that these are not random things, that these are not accidents or slips, but that there is a coherent pattern to it that changes the interpretation

of the data. Tell me about who you worked with and what that work was like to look at these data and what your conclusions were. I worked independently, and once I settled in my mind that there was some legitimacy to these complaints, the next step was then to do some additional research and forensics on the material, on the scientific publications that were supportive of developing this drug and see if there were additional problems. And so I helped them build a dossier that was eventually sent to the FDA to contest the continued development of this drug. And now it's in the FDA's hands. Maybe you're thinking that, of course, the FDA will follow your prescription because you had a lot of confidence in your data. Right. But I don't know. Did you have any

skepticism that the FDA would wind it down? I was shocked that the FDA didn't take more action than they did. I expected within days for this to be a no-brainer that this could not continue. And I was really surprised by the pace and the ultimate decisions. But one of the things that happened was we've noticed this pattern where cheaters cheat. That shouldn't maybe shock people. But when I started investigating this scientist, what we found is that the problems extended well beyond the work that he had done for this company. And so we had to decide what we were going to do with that. Obviously, the attorneys were very interested in this company, but I felt like there was an additional scientific objective here to correct the scientific record more broadly. So we agreed that

I would be permitted to continue to pursue the problems in the rest of this individual's work. I see. This is Hau Yan Wang at City University of New York. Is that right? That's right. how young long was the primary scientist working for Casava sciences. And so I ultimately sent a set of dossiers to the National Institute of Health, describing more broadly the patterns that we saw in his work and in the work of his collaborators. And what did you hear back from the NIH? Well, that's a very opaque process. So they thank you for your input, and then you hear very little.

One of the whistleblowers in the Casabas Sciences case reached out to a reporter at the prestigious journal Science, who connected Matthew Shrag with a colleague of theirs, the investigative journalist Charles Pillar. And it became clear to me he was looking to go a lot deeper than most of the reporters that I had talked to. Up to that point, my comments in the media had been mostly on background, and my role had been as a consultant for this attorney

and the whistleblowers. working for. But obviously the story was bigger at that point. It was more than just Casava sciences. It involved a lot of scientists and, you know, the problem was expanding. Coming up after the break, we will hear how the problem expands. I'm Stephen Dubner, and this is Freakonomics Radio. Before the break, the Vanderbilt neuroscientist Matthew Schrag described what he considered doctored research coming out of the biotech firm Casava Sciences. The investigative journalist Charles Pillar heard about Shrag's work and suggested that they collaborate. Here's Matthew Shrag.

I agreed to show Charles the dossiers that I had prepared and had sent to the NIH. And I had sent an additional dossier to the NIH about the issues at the University of Minnesota, which were incidental to the work that I was doing on Kasava Sciences. But I had become aware of this problem and had been dabbling with some image analysis techniques and basically used this set of problem papers that I had found at the University of Minnesota as a test case to see how this technique would perform. One of these papers was a hugely influential study that had been published in the journal Nature in 2006. It was called a specific beta amyloid protein assembly in the brain impairs memory.

Among the co-authors were two University of Minnesota researchers, Silvan Lesne and Karen Ash. The research used genetically engineered mice whose brains produced excess amounts of amyloid protein. The researchers extracted that protein from the mice and injected it into the brains of rats, who soon after showed signs of cognitive decline. This paper came along at a moment where the amyloid hypothesis had a big stumble, a clinical trial looking to remove beta amyloid from the brain with a vaccine, which frankly is a really clever approach. But the trial failed, and sometimes really good, really clever ideas don't work. And this was one of them. And people started questioning whether this was the right hypothesis. So this reformulation

came along that perhaps it wasn't the big clumps, the plaques of beta amyloid that was the primary driver of the disease. Perhaps somewhere in between a single molecule of beta amyloid floating around and where it was millions of these molecules in a giant clump, perhaps there's some soluble small clump of beta amyloid that was the toxic component. And they call those oligmer. which is just Greek for small clump, I guess. Sylvain Lesnay proposed that he had found that silver bullet, that a clump of a certain number of strands of beta amyloid was the primary cause. And his paper was cited and continued to be cited right up until we challenged it.

This experiment gave new confidence to people in the field and became one of the most cited experiments in the history of the disease through 2022. That again is the journalist Charles Pillar. So what happened next? Shrag and I started to look carefully at this, and ultimately I wrote a story about it in Science magazine. The major issue they found with the paper by Lesnay and Ash had to do with the images that were used as evidence.

Primarily in this case, it involved something called Western Blot images. Now, these are a very, very common technique used in science, particularly often in Alzheimer's research and in other fields as well. And what they are are a photographic representation of proteins. So you can test a protein sample for the type and amount of different proteins within it. This is very important in Alzheimer's research because you're trying to determine what proteins are in an animal brain, for example, that might be behind the disease in some way. And so what Shragg found from his work, and this was validated by other forensic image experts by me, going to them, and also by leaders in the field of Alzheimer's research, what was found was that

these images apparently were manipulated in severe ways that tended to support the hypothesis of the experiment when the actual data did not support it. So I assume that you contact Les Ney. What's that conversation like? That conversation is one that regrettably never occurred because he wouldn't speak to me. Okay. And what about Matthew Shrag, your co-investigator, Did Lesnay or anyone else from his lab, including, I guess, Karen Ash, did they speak with Shrague or respond to any comments? Is he sending them information about his charges of manipulation? I mean, Shrag's a scientist. I was the one who, as a journalist writing the article, of course, I did everything I could possibly do to get Lesnay and his co-investigator, who at the time of the experiment was his boss, Karen Ash, to speak to me.

prior to the article appearing, they both refused to speak with me. This is detailed in my book and also in the article that I wrote about it. And it was, I think, not that surprising in a way I found over many years of experience doing these sorts of articles that people don't have a great story to tell often are unwilling or unable by their institutions, restricting them from saying anything to a reporter. Ash, I did have a dialogue with, and she very interestingly replied that she wanted to talk with me but her university advised for not to, and she decided that she would go along with their advice.

I explained to her at the time that I thought it was terrible advice because this was her opportunity to get her perspective into the most red, deepest look at the subject. The thing that frustrates me about that is universities are also part of the public trust, and they get a lot of research funding from taxes. And so I feel like it should be incumbent upon them to at least respond, at least take the charges seriously. when you found that you weren't getting a serious response from her, what was your next step then?

In this case, what happens is the National Institutes of Health, which is the funder of this work, take a look at it. They agreed that there was a possible problem there. They referred it to an obscure government agency called the Office of Research Integrity at the Health and Human Services Department. This is a tiny little group. They don't have the manpower to examine every possible example of fraud. or misconduct in the scientific world. And so what they do is they give it back to the university where the alleged misconduct took place,

in this case the University of Minnesota. To me, this is exactly the wrong approach because you're entrusting the organization that has the most to lose and the least to gain from a robust, complete, and publicly accessible examination of possible misconduct to do that examination. So they took years to assess Lesne.

In the meantime, he was still collecting many hundreds of thousands of dollars in federal funds to continue the same lines of research that had already been proved questionable. And I might add, published a paper after my initial story came out that we found had apparently manipulated scientific images in it. This goes on and on, and the university is going about their business. Finally, in 2025, Lesnei retires from his job. He resigns from his post as a tenured full professor. This is a pretty young guy. So what you're talking about is a guy who ended his scientific career, in effect. What was the extent of that university investigation?

Well, we'll never know because they never released any of the particulars, except to say that they had asked for journals to consider attracting other Lesnay articles. Now, the initial story, this seminal paper in nature that was influential in the field of Alzheimer's disease, and I would argue, and I think many people agree, had a very, very high degree of importance in moving the field forward. This paper was retracted, ultimately. It was co-author Karen Ash, who formally requested that the paper be retracted from the journal Nature. In a note, she published elsewhere, Ash acknowledged problems with the data but said she had had no knowledge of it until it was brought to her attention.

We reached out to both Karen Ash and Silvan Lesne. Lesne declined to speak with us, and we didn't hear back from Ash. And what about Kasava Sciences and their experimental Alzheimer's drug, some euphilum? Charles Pillar and Matthew Shrag kept digging into that as well. Pillar published an article in science about an internal report from the City University of New York, which found numerous signs of image manipulation in How Yan Wang's work and bad data management. In 2024, the Department of Justice indicted Wang for allegedly fabricating and falsifying data in NIH grant applications related to semufilum.

Wong faced multiple charges, including major fraud against the United States, wire fraud, and false statements. A few months later, Kasava Sciences agreed to pay $40 million to the securities and exchange commission to settle charges that it misled investors on the effectiveness of semufilum. Casava stopped development of the drug after publicly stating that it had failed in human clinical trials showing no cognitive advantage over a placebo. Kassava president and CEO Remy Barbee A and senior vice president and neuroscientist Lindsay Burns both resigned.

And last month, Kassava Sciences announced that it was once again changing its name, this time to Philana Therapeutics. How Yan Wang also agreed to pay a fine related to the SEC charges and in 2025, all charges of fraud made by the DOJ were dropped. We tried to reach Wang and got no response directly, but his lawyers said they are pleased that the case against Wang was dropped in light of new evidence and maintain that Wang is innocent of the accusations made against him. We also reached out to Falana for comment. A spokesperson for the company said they are currently focused on researching and developing a treatment for a rare form of epilepsy. Meanwhile, Charles Pillar, as part of his book research, had been looking into the track record of many Alzheimer's researchers.

I focused in on about 50 who were both influential and who seemed to be people who needed a closer look. Probably the most interesting one is this guy, Eliyzer Moslia, who was previously the head of the Division of Neuroscience for the National Institute on Aging. This guy was overseeing a budget of about $2.7 billion per year, and this made him one of the most influential people in the world by far. to steer the agenda of neuroscience research, because this is the biggest funder worldwide for this work. He, for decades, had been a researcher at the University of California in San Diego. He became an eminent scientist in both Parkinson's and Alzheimer's research. So when he was hired to be the head of the neuroscience division at NIA, it made him a logical choice, someone who had respect, had contacts, had prestige, had himself.

demonstrated a propensity for influential research, and he operated his own lab within NIA as well as directing their programs. And at that time, were there any suspicions of research misbehavior or image doctoring? No, there were no suspicions of him, according to the NIH. So with the cooperation and assistance of four leading forensic image experts, and that included Matt Tragg, a woman by the name of Mu Yang, who is at Columbia University, who's becoming a pretty well-known forensic image expert. Also, Kevin Patrick, who is not a scientist, but is a independent researcher who is excellent and very seasoned at looking

at these images. And Elizabeth Bick, who is a world-renowned expert in scientific image analysis. And the four of them took a look at Mosley's papers. We found 132 papers out of the 800. We didn't look at all 800 because it was too massive a task. at 132 where image irregularities, including apparently doctored images, apparently misused or reused images, apparently images that were published separately in different papers in an improper way

to make different scientific points, what came out of it was a 300-page dossier detailing all of the concerns. And this traces back to Masley's work going back 30 years, including at UC San Diego, and also the work coming out of his lab at the National Institute on Aging. Work that he's directing personally?

That he's directing personally in his own lab. So you compiled this dossier that identifies 132 papers with signs of apparent image doctoring. How do you present this information to him and what's his response? Well, in the most straightforward way possible, I wrote to him, I presented the dossier to him. He had 100% of the information.

I asked him a wide range of questions about the problems that we found, and, of course, we offered him the opportunity to present evidence that would suggest that we had it wrong in some way. And so the response was that he didn't respond. The response from the agency was fascinating. At first, they would refuse to respond.

And then as the date of publication became closer and closer, literally three days before the story was published, they said, wait, wait, we might have something for you. And then they timed their announcement about this to land the moment my story went live on the Science Magazine website. And that announcement was that Maslia was no longer in his job. Eliezer Maslia declined to speak with us. And the NIH sent us a statement to say that they do not comment on personnel matters.

Coming up after the break, Matthew Shrag has a reckoning with his. former mentor. I'm Stephen Dubner. This is Freakonomics Radio. We'll be right back. You said you were directed to a lab where you were an undergraduate at University of North Dakota. And there was someone, I guess, running the lab, right? Offman Grieby, who became really a mentor of yours, correct? Yeah, absolutely. He taught me an awful lot and became a very good friend. Turns out that he was one of the people that you blew the whistle on the hardest, yes? Yeah, I mean, it was a very disappointing end to our relationship.

This again is Matthew Shrag. He told us earlier about this research with Greby, looking at plaque buildup in the brains of rabbits who'd been fed a high cholesterol diet. I asked Shrag what his role had been in this research. Well, I helped raise the animals, and I helped to stain the tissue. After the work with Casava sciences,

I very much expected that every paper I'd ever looked at was going to get analyzed and dissected. But I wasn't aware of problems. What do you mean by that? You expected that every paper would get dissected? I expected that people I was blowing the whistle on would have a good, hard look at my research and make sure that my work was clean. I see. And I was quite astonished when somebody reached out to me and said, have you looked at these images from the first couple of papers that you ever published? What were the images?

They were microscope images, so we were looking at a field of neurons. The suggestion was that the neurons looked too much alike. It looked like some of them might have been copied and pasted multiple times. So who was it that, in your view, took and manipulated those images before they were published in the paper? Well, that happened at the hands of my mentor, Othman Greby. And I know this partly because he was responsible for the paper and put the figures together, but also partly because when I asked him about it, he told me. This is years later.

This is when we discovered the problem. And I have to say that once these images were brought to my attention, I looked at everything. In fact, I recruited a couple of other experts in this area to help me so that we could do this fairly quickly. I also felt it was important to have an unbiased set of eyes or two because I certainly felt that I was partial to Othman. We very rapidly went through everything we published together, but also everything else that he had published. And how much of it had problems? Unfortunately, just an enormous portion. And, you know, the types of problems were the sorts of things that were very hard to dismiss. Meaning manipulated imagery and so on? That's right. And in a way that was coherent and changing the interpretation of the results.

I have to say when I talked to Othman about it, he said that he thought the underlying science was right and that he was trying to make it a little bit clearer and said it was more exaggeration than manipulation. You hear these kinds of ways of sort of shrinking the problem. But when you looked at the pattern, it was very hard to ignore. I have to say it took me a little bit bit of time to settle that this was going to be a huge problem because I knew

Othman, I just assumed that this was a mistake. So my initial reaction was, hey, we're seeing these problems in these initial couple of papers, and I didn't know the scale of it at the time. I basically said, hey, let's track down the source data and fix it. I reached out to him to try to do that. I reached out to the university to see what sorts of things might still be around. But the problem was that at some point, Othman had gotten rid of all of his records. I think he knew that this was a problem, and he told me he was very ashamed of what had happened. And so at some point, he closed his laboratory and disposed of everything, in part out of shame of knowing that this data was not at the standards that he expected of himself. So when you called him ultimately,

what was that call like? It was tough. It was a very tough call. Initially, he said that he wanted to speak out and exonerate me from being part of the problem, which I appreciated, but also I don't think most people would look at the middle author of the paper who's an undergraduate and say, that's the bad guy. I don't want to say that never happens, but it's unlikely. So did he speak out? He had some initial conversations with people, and he sent me some things in writing to say what he said, but ultimately he decided not to speak publicly, which I just thought was a really sad mistake. I told him at the time, and I still think that he was a wonderful teacher. A lot of times people who have trouble in one domain can still have a lot of value to

bring to the table. I told him that if he could find it in himself to speak out about this and talk, not just about what happened, but why it happened and why other people might experience this and the importance of really hanging on to the basic principles that we as scientists all sign up for, that his role as a teacher could just be exponentially more important. And I still sort of hope that he would do that someday. I understand that there had been an earlier misconduct inquiry into his work back in 2006 or 2007. Do you know anything about that? I do now. I didn't at the time. Apparently, it actually involved this paper that we're talking about. Some of the problems had been picked up in a grant review is the story that I was told. Othman told me this story eventually,

that they had picked up on some anomalies in the data, and they had. had referred it both to the university and to the journal. And the journal asked him to send original images so that they could inspect them and insisted on a correction to the paper. They did publish a correction. The correction was fairly vague in terms of what had happened. And so it was presented as a simple accident. Did you ask him why he did what he did? You know, I did. And his point of view at that time was that he wasn't making anything up. that he was trying to make the images more convincing or easier to understand,

but he equated it with a bit of an exaggeration rather than whole-scale fabrication, which I must say in practice is no different. But that was his perspective. His perspective was also that everybody did this, which is not true. And we have to say that. We have to say that. It's easy to erode the standards, but we have to consistently say, no, not everybody does this. Okay, so you were a kid when you worked with him,

and now you're an actual academic. I have to say, though, he remained an important person in my life. When I moved universities, he stayed on my dissertation committee, so he remained an advisor. He came to my graduation and went through all the regalia putting the hood on me. You know, he came to my wedding. And so this was somebody who was more than just a passing academic acquaintance. And I think that's important when you think about these kinds of stories, is that there's also underneath all this, often a very personal story. These are real human beings who are often very attached to each other.

So it's easy for people to say, why don't people take a stronger stand? But I know how hard it is to take that stand. That it often has a real human cost. The 2006 Rabbit Paper by Greby, Shrag, and three other co-authors was retracted in 2023. Other joint papers from Shrag and Greby were retracted soon after, along with a few more Greby papers. Since 2022, Greby has been teaching at the University of Texas Rio Grande Valley. Of all the Alzheimer's researchers that we contacted who had been accused of wrongdoing, Othman Greby came closest to sitting for an interview, but ultimately he declined.

He did send us an email. He wrote, Research is teamwork and I stand by the data, at least most of it, of my research. No fabrication was made, but some polishing could have been done. However, I take full responsibility as I am the laboratory director. He also wrote, I insist that the imaging figures in Dr. Shrag's two manuscripts from my lab are the images he handed to me. He had many years to question the pictures if he believed they were altered. Grieby also told us that he lost access to his lab and his data after retiring from the University of North Dakota in 2020.

The journalist Charles Pillar told us that he tried to speak with Greby before Pillar published an article on these allegations in science. Pillar said that Greby agreed to an in-person interview, but backed out a few days prior. I went back to Charles Pillar with some more questions. I do write that Alzheimer's research has essentially been hijacked by a cabal of self-interested researchers, government accomplices, and corporate greed. That's pretty dramatic in Charles. I can imagine that some ears hearing that immediately go to conspiracy theory. territory. Persuade me that this is not conspiracy theory that you're floating here, that the

researchers, government accomplices, and corporate greed are spoiling the pool for everyone here, and that this is based on research and reporting and actual empirical arguments. Just to put it plainly, nearly all of them have built their careers on explaining, exploring and expanding the benefits and the importance of the amylite hypothesis. Their money, their prestige, their career aspirations are all tied up with that. You write that 11 of 16 FDA medical examiners who worked on drug approvals left to become employees or consultants for the companies that they just regulated. Public citizens concluded that regulatory capture has now infiltrated the FDA.

Tell me a little bit more about that. Well, let me give you an example in the Alzheimer's field that I think is very instructive. The first of the anti-amylate antibodies that was approved by FDA was a drug called aducanamab. That's the scientific name. The trade name was adjuhelm. It's been withdrawn from the market. The reason it was withdrawn is that it was dangerous and ineffective. This drug was essentially the product of regulatory capture.

The guy who was overseeing the shepherding of it through the FDA, his name is. Billy Dunn was found in a journalistic investigation by the website stat and then later by a monumental U.S. Congress investigation to have been working hand-in-glove with the company that was promoting and trying to get the drug approved. Billy Dunn left the FDA shortly after this episode, and what did he do? He took a big job as a member of the board of a biotech company that developing Alzheimer's and Parkinson's disease remedy. There's many, many other examples, and this has been described as a revolving door at the FDA. This is one element of the Trump administration program that is being championed by Marty McCarrie, who's the FDA commissioner, that I think is really smart, which is to try to reduce this revolving door phenomenon.

There's too many conflicts of interest in drug development. So, yes, I think that's a real problem that needs to be corrected. How do you think about your role at a time when there is a growing distrust of science in some places, including some places within the federal government? A lot of your reporting has been shining a light on misbehavior, which you could argue might destroy trust even more. So how does that change the way you work? Yeah, thanks for the questions. Stephen. I think about this a lot. I just want to make it clear that I really disagree strenuously with a lot of the moves being made by the Trump administration, regarding science and research. I think it's important for universities to have freedom to pursue science and scientific research in a vigorous way without feeling that they have to knuckle under to federal demands for certain kinds of political issues. I also disagree with the way in which attacks have been made on federal research funding at the NIH. the administration is not wrong about everything.

And critics of the scientific enterprise are not wrong that there has been arrogance and complacency. And in my view, condescension on the part of many people in the scientific research enterprise towards members of the public. Condescension along the lines of trust us for scientists? Yeah, I mean, trust must be earned. Now, let me make it clear that two things can be true at once. we can be highly critical of attacks on science and attacks on the scientific enterprise by people who I think are misguided. And at the same time, we can insist that the institutional authorities of science, and by that I mean the gatekeepers of knowledge like the universities and journals like the one I work for, and the funders like the National Institutes of Health and the regulators like FDA, we can insist that they do better at their jobs, that they be more vigilant.

and more careful and also be more responsible for protecting scientific knowledge from pollution by bad ideas and corruption. We spoke with a few people on background who have been the focus of your research. One person said, sorry, I am tired of this. It is a non-story. The anti-amloid treatments are working now. Tens of independent confirmations. Pillar you is merely on a book tour. response to that? Is it indeed true that anti-ameloid treatments are working, that there are tens of independent confirmations of that research? So anti-ameloid antibodies now, there are two approved drugs by the FDA. And these are drugs that remove amyloid proteins with regard to how effective the drugs are. I think there's evidence that they can very slightly slow the cognitive decline associated with

Alzheimer's disease, although there are many clinicians who say that the degree of slowing compared to people who are not on the drugs is so subtle that it can be imperceptible to clinicians, patients, and family members. So no one's getting better with these drugs. On the other hand, if I'm a scientist, whether working on this particular research or other research, I might say, well, Charles, you know better than most people that science is really hard. Alzheimer's is a disease it's been with us for, it sounds like pretty much all of human civilization. And you need to take these incremental steps to try to find provable directions, at least, of better clinical treatment. What would your response to that be?

Is it that this is just not the viable path towards success or that it may be, but after all these years and all this money, that if there hasn't been more credit, improvement, then we should really be looking elsewhere? I don't think that things are mutually exclusive, Stephen. What I would say, and contrary to what some critics of my book have described it as, in which I would regard to strawman arguments, I have never said that amylate proteins have nothing to do with Alzheimer's disease. I have always said that they don't have everything to do with it, and that it's important for people to understand that Alzheimer's is almost certainly

a disease of many factors, and it will almost certainly require many different approaches, perhaps in combination, to result in any sort of remedy that are rest or versus the symptoms of the disease. Now, what concerns me about the dominance of anti-emoloid antibodies within this field is that, one, they're very expensive, two, they are dangerous. They can cause brain swelling and bleeding that has on occasion resulted in death. Third, they're not effective in arresting the symptoms of the disease. And fourth, it's very important to understand that when this idea, with its incredible dominance over decades, soaks up a lot of the mind share and a lot of the ideas behind drug development and a lot of the big money behind Alzheimer's drugs. In other words,

if you look at the range of clinical trials, you'll see that a lot of things are being tried. But if you look where the big money is being spent, it's predominantly in anti-ameloid remedies that we know from historical experience are not that great. I went back to Matthew Shrag to ask what he thinks the future looks like for the research and treatment of Alzheimer's disease. I would reformulate Alzheimer's disease and perhaps a number of other neurodegenerative diseases too as diseases of failed waste clearance in the brain, that you have an accumulating. of waste products in the brain. Beta amyloid plaques are one of them, and tau tangles are one of them. And I think that thinking about this in a more complete way helps,

that saying that this isn't one specific thing causing the problem, but that we have a problem with waste management and that the accumulation of all of these waste products is adding up to a toxic sphere. And if you reformulated that way, you can explain the clinical trial results that we've been getting with the amyloid therapy. big, which basically says this is one of the waste products. And perhaps there's some sort of minimal

benefit from clearing one of the waste products. But we need to take a bigger approach. And there's a lot of different ways of getting rid of waste products. One of them is through the blood vessels. You can actually wash it out of the brain by washing it through the blood vessels. And so our and a number of other labs, too, have theorized that degeneration of the blood vessels slows down that process of clearing waste products out of the brain. And so if we could focus on blood vessel health, we may get somewhere. And actually, we have a hint of that in the clinical trials. The one real bright spot in the clinical trial space is that aggressive blood pressure control seems to protect memory, especially early on.

Early on meaning what? Like, before people's memory has deteriorated. So if you get patients at the very first sign of memory problems, or likely even before that, we see some benefit in memory. And I think it's pointing at blood vessel health being a big part of keeping the brain healthy. Is Alzheimer's curable? I don't know. I would like to think so.

It's certainly a complicated disease. And I think that there had been a certain nihilism about neurological diseases in general. They used to say about neurologists that we could diagnose the disease but not treat them. They would say that we were the diagnose and audios doctors. But over the last few decades, we've seen a real renaissance in neurology and neuroscience. A lot of diseases that have been hard to treat. We're seeing very good treatments developed. We see that in a pretty dramatic way with epilepsy.

We've got quite a long list of treatments now, including many patients being able to be surgically cured. Stroke has completely morphed in my career. This was one of the most depressing service lines in the hospital. Somebody would come in with stroke and be horribly disabled. And now if patients can get to us in time, we're often able to do huge amounts for them. And so I think we've seen a bit of a renaissance in neurology and neuroscience, but it's missing in the neurodegenerative space. And I have to ask the question to what degree research integrity issues have held back development here. People described how hard it was to get funding if your ideas didn't conform with this hypothesis.

I have to tell you, Matthew, when I first started doing this for economics work, I signed up for a whole bunch of academic journals, mostly within the realm of econ, and I got some neuroscience journals because I thought, the brain's pretty important. I'm interested in this kind of stuff. I should really read these. I couldn't. It was so complicated. I felt like if you didn't have at least a PhD to start with, it was just totally indecipherable. On the other hand, everybody listening to this has one, a brain that is. And so I do wonder if you, as an ex-executive, have any advice for how to think about the brain and its centrality to who we are as individuals, but especially how we should think about brain science as a topic that we should all be a little bit

invested in somehow. You know, the brain is a wonderful, beautiful organ. And I have to say all of that complexity of the way we talk about it aside, if you sit somebody down on a micro, and let them see these neurons with their own eyes. It is just one of these really moving experiences. It's an incredibly beautifully organized organ that allows us to access who we are. And it's an organ that we have a lot to learn about. But we do know a few things. One is that we need our memories to last lifetime to really get everything out of life that we're trying to get out of life. We've seen incredible progress in other fields of medicine that has. has expanded life expectancy over the last century or century and a half, a lot of the benefit of

those advances are being lost to this disease. If those extra years that you're getting at the end of your life are of poor quality, this is something that affects all fields and a huge number of people. This has affected our political leaders and our musicians and our sports figures and almost every family in this country. I think that it's easy to accept this as just a part of old age and say there's nothing we can do about it, but I think there must be something that we can do about it. This is one of the challenges for our generation is to crack the code on this set of diseases, to make sure that the quality of life at the end of the life is what it should be. So, Matthew, when you look back at your career, it seems from my perspective that you're doing well,

you're at a great research university, Vanderbilt, but you've also devoted a lot of time and attention to this whistleblower function. How would you describe overall your standing in the field as a result of the whistleblower function? Do you feel you've been somewhat ostracized, mostly celebrated, a good bit of both? There's definitely been some degree of backlash, and I think that's expected and almost unavoidable. Some opportunities lost, but I think that's okay. There's also other opportunities gained. The ability to have important conversations about where our field is going is a unique opportunity that I wouldn't give up. Does anyone in your family ever say to you, you know, Matthew, be so much easier if you just stick to your knitting, do your research, shut up a little bit.

Did anybody say that? I will say that I sought the advice of quite a number of advisors before agreeing to go on the record with Charlie. And I don't think anybody thought it was a good idea. He's a very thoughtful person and I think a very, very cautious person. And I think that that's key. There's a huge amount of trust that you put in somebody by doing this. I'd like to thank Matthew Schrag and Charles Piller for putting their trust in us in telling you this story. I do wish that some of the scientists whose work they had exposed had also been willing to talk, but I guess I can see why they chose not to. The book that Charles Pillar wrote is called doctored, fraud, arrogance, and tragedy in the quest to cure Alzheimer's.

If you want to let us know what you thought of this episode, our email is Radio at Freakonomics.com. We are also on all the regular social platforms, even a few of the irregular ones. Coming up next time on the show, we go inside one of the most essential software companies that you have probably never heard of. I had no interest in starting a company. I had no idea how to do it. Judy Faulkner is founder and CEO of Epic Systems, which manages the vast majority of electronic health records in the United States. She doesn't do a lot of interviews, but she agreed to this one, and we cover a lot of ground, including why Epic has never gone public and never will. The shareholders were only interested in return on equity, not what value the company was giving to the world.

And I thought, who wants owners who were like that? That's next time on the show. Until then, take care of yourself. And if you can, someone else, too. Freakonomics Radio is produced by Renbud Radio. You can find our entire archive on any podcast app. It's also at Freakonomics.com, where we publish transcripts and show notes. This episode was produced by Dalvin Abouaji and edited by Ellen Frankman.

It was mixed by Eleanor Osborne with help from Jeremy Johnston. The Freakonomics Radio network staff also includes Augusta Chapman, Elsa Hernandez, Gabriel Roth, Ilaria Montenicourt, Jake Loomis, Mandy Gorenstein, Peter Madden, Teo Jacobs, and Zach Lipinski. Our theme song is Mr. Fortune by the hitchhikers, and our composer is Luis Gera. As always, thanks for listening. I see that you live on what's called a hobby farm outside of Nashville and that you do beekeeping, yes? I did that for a number of years. It was a great pleasure keeping bees.

With the last baby, I let the bees go. The Freakonomics Radio Network, the hidden side of everything.