Biogen had been told by the FDA three times that their new monoclonal antibody therapy did, paraphrasing, absofuckinglutely nothing. Then Aduhelm was mysteriously granted approval anyway in June of 2021. In an apparent breakthrough, it was the first drug that could scrub away those pesky amyloid plaques in the brain.
But drugs need to be proven safe and effective at treating something in order to gain approval. Brain plaques have never been proven as the cause of Alzheimer’s. Researchers raised alarms that amyloid plaque removal was being used as a proxy for treating the disease. Biogen was like “don’t worry, we’ll fix it in post,” and said they’d study it more after it hit market.
Two months later Dr. Matthew Schrag, one of the more vocal researchers with objections to Aduhelm’s approval, got a phone call that would eventually throw the field of Alzheimer’s research into turmoil.
Today’s Moment of Science… The Alzheimer’s Data Manipulation Scandal.
In 1901, a 51 year old woman named Auguste Deter started having jealous fits at her husband seemingly out of nowhere. Confusion about how to navigate their home followed. She would yell for hours sometimes in a “horrible” voice and rave about her fear that she would imminently be murdered. When reading she would slip between lines and eventually read one character at a time. Some questions became entirely incomprehensible to her; eventually all of them were. Unresponsive and bedridden, she passed away after four and a half years of illness.
German psychiatrist Alois Alzheimer diagnosed Auguste with the first case of what would become his namesake disease. He published “About A Peculiar Disease Of The Cerebral Cortex” in 1907, with observations he’d made through the course of her decline. In the autopsy, Alzheimer noted the plaques and tangles afflicting her neurons, hallmark signs of the disease to this day.
A neurodegenerative illness, Alzheimer’s diagnoses account for 60-70% of dementia cases. With 50 million Alzheimer’s patients worldwide, that’s a goddamn payday if a new drug even kinda works.
Ahem. I mean that’s a lot of people for whom pharmaceutical companies only have altruistic intentions.
But it’s hard enough to treat some illnesses for which we have every possible disease course mapped. We’re not even sure how advanced Alzheimer’s is when symptoms first appear, and the cause, if there is just one? That’s a future Nobel Prize.
It’s not entirely clear what role amyloid beta plaque has in Alzheimer’s disease. The amyloid hypothesis suggests that amyloid beta cozying up in your glitchy computer could be toxic to nerve cells, triggering neurodegeneration. There are more hypotheses involving sleep disturbances, genetics, inflammation, and other super annoying proteins. The plurality of research dollars are currently going to this one thing that’s still hypothetical and we’re getting fucking nowhere with it. So.
Dr. Matthew Schrag was asked to look at yet another Alzheimer’s drug in the pipeline, Simufilam. The evidence that it worked seemed shaky, and a couple of short sellers wanted to see if their hunch on betting against the company behind it, Cassava Sciences, was solid.
Schrag was the type of researcher who didn’t just look at data for the one drug, no. He looked at all the underlying science from the last two decades. As someone who has fallen down a few rabbit holes in my life, respect.
The report submitted to the FDA on Simufilam was fair but unkind. At least 34 previous studies from two of the main researchers showed clear signs of image manipulation in western blot analysis. Data analysis from phase II trials conducted by an outside lab suggested that the drug- hold onto your tuchus here- did absofuckinglutely nothing. Then after a reanalysis by another lab, presto, “robust” results.
Annoyingly familiar cycle we’ve got here, huh?
The story doesn’t end with consulting on one drug. Schrag went looking on the website Pubpeer for examples of western blot manipulation to help him better understand the look of doctored results. He stumbled across multiple articles with problems in the images all from the same author, Sylvain Lesné, and went looking for more.
Lesné’s seminal paper published in Nature in 2006, “A Specific Amyloid Beta Protein Assembly In The Brain Impairs Memory,” provided real evidence for the amyloid hypothesis. A𝛃*56, pronounced as amyloid beta star fifty-six, when injected into rats produced memory impairment. It was cited over 2,000 times, one of the most cited Alzheimer’s papers in the last twenty years.
It was never able to be reproduced. A𝛃*56 has been damn near impossible to confirm in any other lab. And the manipulation of the western blot images has been described as “shockingly blatant.”
After corroborating analysis from image manipulation experts and Alzheimer’s researchers (including some who ‘support’ the amyloid hypothesis), hundreds of published western blot results from Dr. Schrag’s investigation remain under scrutiny. Over seventy images from twenty of Lesné’s papers are under some degree of suspicion, ten of which involve A𝛃*56.
Though there have already been retractions and papers are being re-evaluated, it’s such a recent development that it’s unclear what impact this will have on the field as a whole (Note- a year and a half later, it’s still unclear how this has changed the direction of research). Have the last sixteen years of research been lost? That’s difficult to say. Amyloid plaque may very well be the explanation but by the time symptoms show up, it’s too advanced to treat. With evidence pointing in every direction, it continues to warrant investigation.
41% of patients enrolled in clinical trials of Aduhelm experienced brain bleeds. Simufilam hasn’t been approved for market, and Cassava Sciences is under investigation by the DOJ, the SEC, and the NIH.
This has been your Moment of Science, planning on getting more sleep (eventually) just in case lack of sleep is the cause.
To get the MOS delivered to your inbox and support my efforts to be a part-time rocket scientist via google, head to patreon.com/scibabe.
Join the discussion!