How Not to Study a Disease: The Story of Alzheimer's

by Karl Herrup

MIT Press 2021, 257 pages
Reviewed by T. Nelson

By now, most people know what happens to scientists who challenge the prevailing dogma: they are ignored and ridiculed, their funding dries up, and eventually they're driven out of science. Decades later they're proven right. They get a Nobel Prize (if they're still alive), our scientific leaders congratulate themselves on the success of the scientific method, and a new unchallengeable dogma is born.

Percentage of AD articles studying beta-amyloid. This only includes papers with ‘beta-amyloid’ in the title or abstract

For almost forty years, the unchal­len­ge­able dogma in Alz­hei­mer's disease, or AD, has been that it's caused by a build-up of a small protein called beta-amyloid, which forms amyloid plaques in the brain. Even today, AD is class­i­fied as a “protein folding disorder.” Despite the fact that over 600 clinical trials of anti-AD drugs have failed—including aducan­umab, an antibody against beta-amyloid that the FDA recently approved—the percentage of AD papers studying beta-amyloid has scarcely changed (see graph).

If the general public knew about this, they'd be outside the doors of our labs with pitchforks.

This popular-style book explains how science got into this mess. It's easily understandable by a layman with no background in science, but it's also important for scientists to read it. It's especially important for my friends and colleagues at the NIH's Center for Scientific Review, because these days NIH funding is the main determinant of what we can study.

Karl Herrup is one of many who have argued against the beta-amyloid theory. But what to replace it with? One possibility is inflammation. Herrup's theory is that unrepaired DNA damage causes the inflammation in AD, and that inflammation, not beta-amyloid, is what kills our neurons. A corollary is that beta-amyloid does not cause inflammation but is actually induced by it as a defense. But good luck getting any of that published, let alone funded. Even Herrup had to be careful in his Journal of Neuroscience article not to mention AD until the end. The dogma is, as Herrup was told, “If you're not studying beta-amyloid, you're not studying Alzheimer's disease.” Every risk factor has to be hammered into the beta-amyloid paradigm. The cholesterol-carrying protein apolipoprotein E, for instance, works by mediating “clearance” of beta-amyloid. TREM2, a protein in microglia that participates in inflammatory signaling, is only relevant because it's activated by beta-amyloid.

This focus on the “bad protein” theory requires us to ignore its limitations. Herrup says that defining AD as a disease of beta-amyloid, as industry and government have done, is circular logic. More than that, in my opinion, it turns it into a non-falsi­fiable hypothesis, or maybe an example of the One True Scotsman fallacy: if a patient doesn't have amyloid plaques, then by definition he's not a true AD patient, but some­thing else. The data are telling us this definition no longer works. Many people have plaques but not AD; many more have AD, as determined by clinical testing, but few plaques. And we now know that getting rid of the plaques and the soluble beta-amyloid has no effect on the disease.

This is an enormous problem for researchers: if there is no objective way of determining whether a patient has the disease, there is no chance of curing it. No non-human animal, not even genetically engineered mice, ever gets AD, so until basic researchers discover what causes it, all that can be done is to run random clinical trials based on hunches.

Unfortunately, that's what the NIH is doing. NIH's response to the brick wall we've rammed into is to redirect funding away from basic research and toward more clinical trials—in other words, to repeat the same failed strategy that gave us 176,969 papers so far on the subject and almost nothing to show for it.

Herrup suggests reorganizing the funding agencies by moving AD from the National Institute of Aging to NINDS. NINDS bureaucrats would love that, but it's just rearranging the deck chairs. If we're going to cure things faster, fundamental changes will be needed in how we fund science.

oct 23 2021