Want to financially support original, independent reporting like this? You can! Donate at the link below, or just click the advertisement at the top of this article - no purchases or personal information sharing required

The FDA has opened a desk for Alzheimer's prevention without finishing the rulebook that would say what counts as proof a prevention drug worked. Until that gets written down, the money stays in treatment, and whoever eventually answers the question will be setting the template for depression, PTSD and addiction drugs too.

A new Food and Drug Administration (FDA) program says it will take on Alzheimer’s prevention. Whoever settles how prevention gets proven also writes the template that lands next on depression, PTSD and addiction drugs.

On October 22 the FDA announced the Neuroscience Advancement and Innovation Nexus, which it calls NAIN. It is a coordination desk, a place where the three parts of the FDA that regulate pills, biological drugs like antibodies, and medical devices sit together and work on questions that cross all three. Its first project is preventing or delaying Alzheimer’s disease.^1,2,3

The Agency said why it built it. “Through NAIN and the Alzheimer’s Prevention Accelerator, we are bringing the full breadth of FDA’s expertise to bear on one of the most urgent public health challenges of our time,” said Acting Commissioner Kyle Diamantas. The same announcement names the obstacle more plainly than the quote does: developing brain drugs runs into “difficulties in demonstrating meaningful clinical benefit, particularly for conditions that develop over many years.”¹

That is the whole story in one line. Nobody has written down what would count as proof that a drug prevented Alzheimer’s. What NAIN cannot do is fix that by itself. A company’s drug still goes to the same reviewers it would have gone to anyway, on the same timeline, and the desk cannot approve anything or settle an argument about anyone’s trial. What it can do is decide which unanswered questions get worked on first.²

WHAT NAIN IS, IN FOUR LINES

Alzheimer’s attacks memory, and it takes the newest things first: the conversation from this morning, the word that will not come, the reason for walking into the kitchen. Once it has finished with the small moments it takes orientation and language, then the ability to cook, to dress, to manage money. It does not stop there. Alzheimer’s killed 119,399 Americans in 2021, which made it the seventh-leading cause of death in the country that year.⁶

About 6.9 million Americans aged 65 and older are living with it now, a number the same report projects will reach 13.8 million by 2060 unless something arrives that prevents or cures the disease.⁶

The FDA divides that long decline into stages, and the first one has no symptoms at all. A person can carry the physical changes of the disease in the brain and be, in the Agency’s own words, “truly asymptomatic.”⁷

Two approved drugs go after the disease itself rather than its symptoms: lecanemab, sold as Leqembi, and donanemab, sold as Kisunla. Both are antibodies aimed at amyloid, a protein that clumps between brain cells as Alzheimer’s advances. Both are approved for people who already have mild cognitive impairment or mild dementia, the stage at which they were tested.^8,9 The most advanced Alzheimer’s drugs ever approved, in other words, are licensed to start only once the disease has announced itself.

Now imagine a pill that keeps you from ever reaching the later stages. How would anyone know it had worked?

You cannot point to the dementia you do not have. You would have to give the pill to thousands of healthy people, give a placebo to thousands more, wait years, and then show that fewer people in the first group slipped. That is slow and expensive, and it only pays off if the company knows one thing before it starts: what would the government accept as proof?

As it turns out, nobody has written that down. A prevention drug would be taken by healthy adults decades before anything would go wrong, which means a company has to commit the ten years and the money up front and find out afterward what counts as success. Most boards will not make that bet.

It is not an abstract problem, and last month it cost a drug its life. On October 29 Eli Lilly, which already sells one of the two approved Alzheimer’s drugs, dropped mevidalen, an experimental pill meant to improve thinking and daily functioning in people who already have the disease. The trial had enrolled 304 patients and finished in May, and a spokesperson said the program “did not meet our high internal bar for success.”^10,11 Weeks earlier the same company had closed a $2.8 billion deal for AtaiBeckley and its pipeline of psychedelic-inspired medicines for depression.¹⁰ Lilly is not leaving brain medicine. It is putting money where the test of success is already written down. For Alzheimer’s prevention, it is not.

The FDA writes documents called guidance. Guidance is not law. It is more like the Agency thinking out loud, and drug companies read it like a rulebook anyway, because it is the only map to an approval there is.¹²

There is a guidance for Alzheimer’s drugs aimed at people in the earliest stages. It has been sitting in draft since 2024, and that draft was itself a revision of a draft from 2018.^7,12

It describes people who already have the biology of Alzheimer’s in their brains but no symptoms at all, and says, correctly, that these are the people you would most want to reach. Then it arrives at the question that decides whether anyone can afford to run the trial: if a drug moves a marker in the blood or on a brain scan in someone who feels completely fine, is that enough to approve it? The guidance says that question “would require detailed discussion with the Agency.”⁷

In plainer words, come and talk to the FDA, one company at a time, in private. Nothing is written down that the next company could lean on.

That is not laziness, and it is worth saying so plainly. The science is genuinely unsettled, and the Agency cannot publish a standard it has not yet worked out. That is exactly the gap NAIN says it wants to fill.¹

Figure 1. Alzheimer’s drug and biologic trials on the public registry, by who sponsors them and what they aim to do.

Drug companies run about two-thirds of the Alzheimer’s trials aimed at treating the disease. They run under a third of the ones aimed at preventing it. Universities, hospitals and nonprofits carry most of the prevention work, and there is far less of it to carry. For every prevention trial on the public registry there are roughly sixteen aimed at treatment.¹³

A company can live with scientific risk, the chance that the molecule simply does not do anything, because it can estimate that and price it. What it cannot price is a finish line nobody has drawn. Drawing that line is the single most useful thing NAIN could produce.

NAIN’s territory is not just neurology. The FDA said in its own announcement that the program covers mental illness, addiction and pain too.¹

That matters because the tools get reused. Whoever settles how you choose people for a prevention trial, how long you watch them, and what counts as working, is writing the template that lands next on depression, PTSD and addiction programs. The Alzheimer’s question gets answered first, and everyone else inherits the answer.

Set that beside the rest of the Agency’s year. In psychedelics it has been doing concrete, dated things. It finalized its rulebook for psychedelic trials in July, held a public hearing in September, and in October asked the public how ibogaine trials should be designed, a question that stays open until November 20.^14,15,16

Figure 2. Dated federal actions on brain and nervous system drugs in 2026, grouped by subject area.

Alzheimer’s prevention, for now, has a desk. There is no scandal in that. It is a choice about sequence, and sequence is most of what an agency controls.

The useful thing about all three is that you can check them yourself. If a year from now there is a meeting notice, a named lead and a finished guidance, then this piece was too skeptical and the field is better off for it. If the three web pages are still three web pages, the program was a statement of priorities that never found its instrument.

Everything here comes from the FDA’s own pages and guidance documents, the Federal Register, the public trial registry and the labels of the approved drugs, all listed in the sources below. What is not established: what the program will do, who staffs or funds it, or why Alzheimer’s prevention was chosen first. The link between unsettled standards and who runs prevention trials is a reasonable reading of the evidence, not a proven cause.

Author disclosure. Odette Hauke is an epidemiologist and regulatory scientist advising sponsors developing oncology, rare disease and psychedelic drugs, including LSD, MDMA, DMT and ibogaine. She was previously Associate Director of Regulatory Affairs at AtaiBeckley, which Eli Lilly acquired in September 2026, and a consultant to its ibogaine, R-MDMA and DMT programs. She has no financial interest in, or advisory relationship with, any sponsor of an Alzheimer’s disease program.

Enjoyed this column? Subscribe to Psychedelic State(s) of America and share this post with someone who should read it.

Until next time,

The Psychedelic State(s) of America Team