The Drug Nobody Funded

Every drug has a clock on it. The patent runs out, the price collapses, and the molecule becomes a commodity. That is the intended design. The reward for spending years and huge sums getting a compound through trials is a period of exclusive pricing, and when that period ends anybody can make the pill. What ends with the patent is not the molecule. It is the sponsor. Here is the hole. A prevention trial is the most expensive exercise in medicine. Years of enrolment, thousands of participants, imaging, endpoint adjudication, and a result that lands long after the money is spent. Only a company with a price to defend will sign that cheque. A generic at pennies per tablet has no price to defend. The arithmetic inverts: the cheaper and more widely used a drug already is, the less reason anybody has to fund the trial that would confirm it. The literature on this question is not empty. It is oddly shaped. One query about a single drug family and dementia returned 78 works in the file I read, and the top 12 by relevance tell you who is doing the work. Neurologists. Geriatricians. Academics with a grant and a question. Not a sales force. That is the whole difference. Repurposing is not blocked by evidence. It is blocked by who pays.

Somebody did run the trial. In 2022 a study called PASTIS was published in Alzheimer s and Dementia: Translational Research and Clinical Interventions - 24 authors, 43 citations - testing whether a phosphodiesterase-5 inhibitor changes blood flow in the brain. The design was clean. Randomised, double-blind, placebo-controlled, cross-over. Older people with symptomatic small vessel disease, the group at the centre of the question, given the drug and a placebo on separate visits and imaged both times. The primary question was not memory. It was cerebral blood flow, a measurement you can take before anybody forgets a name. Notice what that choice costs and what it buys. It buys a result in months rather than a decade. It costs the endpoint payers actually want, which is a number attached to a person and their function rather than a number attached to a scan. Two years earlier, in 2020, a pilot in Brain Communications - 28 citations - ran the smaller version: a randomised, double-blind, placebo-controlled cross-over in patients with small vessel disease and stroke, looking at blood flow and endothelial function. In 2023 the same question was still being argued in review form, 18 citations, naming sildenafil, vardenafil and tadalafil as approved vasodilators with a plausible route into this disease. So the work exists. It is small, it is early, and it is underpowered by design, because that is what happens when the question is real and nobody with a balance sheet is asking it.

Small vessel disease is not a niche. Autopsy work published in Hypertension in 2023 - 227 citations - found the pathology in more than 50 per cent of people aged 65 and over. It is the main cause of lacunar stroke, and it shows up in the deep white matter, the tissue around the ventricles, and the deep grey nuclei. A 2022 review in the International Journal of Stroke - 392 citations - puts it plainly: this is the most common pathology underlying vascular cognitive impairment. Then it lists what that looks like, and the list is the part that matters. Motor impairment. Impaired balance. Falls. Parkinsonism. Depression. Apathy. Personality change. That is what ageing actually costs a family. Not a memory score on a form. A father who falls, who stops leaving the house, who is no longer himself, years before anybody writes the word dementia into a file. A 2022 review in Pharmacology and Therapeutics - 117 citations - adds the window. Barrier leakage in the brain appears early in Alzheimer disease, present decades before the clinical diagnosis. Decades. That window is the entire opportunity. If the damage is measurable years before any diagnosis, then the years when nothing appears wrong are the only years in which a prevention trial can work. That is the trial a generic cannot afford to fund.

Here is the honest part, the one that separates a hypothesis from a fact. Observational data associate these drugs with lower dementia rates. That is interesting. It is also the weakest evidence we accept, and it has a known failure mode. The men taking a drug for erectile dysfunction or prostate symptoms are, on average, sitting in a doctor office more often than the men who are not. They are being screened. Their blood pressure is being treated. They drink less. They are measured, and measurement is itself an intervention. So the association might be the drug. It might be the visits. The probability that any single observational association survives a properly powered randomised test is low, and it is lowest exactly where the money is thinnest, which is where this question lives. I have watched that confound in the field for years. Equipment that gets maintained performs better, and then the maintenance gets mistaken for superior engineering. You have to ask what else changed. The unglamorous explanation is usually the correct one. A man on one of these drugs reports feeling sharper, and the boring reading is that he stopped waking five times a night. Sleep improved, so he improved. The compound did something. It may not have done the thing on the label.

A patent expiring does not end the research. It ends the sponsor.

The pharmacy shelf and the supplement aisle are priced by different rules. On the supplement side, a mechanism sells. A story about a pathway is cheap to print and needs no trial. That whole category spent the last decade selling a feeling to healthy men at a premium, and the more interesting half of it is not the compounds. It is the marketing. Off the patent, the maths runs the other way. An expired pill earns pennies. Nobody funds a 10 year prevention trial on pennies, however strong the hypothesis, because the party paying for the trial cannot capture the return. The molecule is not the problem. The business model is missing. There is a way out, and it is public money. Non-dilutive capital exists for exactly this shape of work: health research councils, public programmes, foundations, and the European instruments that carry an explicit repurposing or prevention mandate. It does not move at the speed of a shareholder and it does not answer to a patent clock. This is one of the few places in medicine where the state is the natural venture investor, because the state is the one party that books the savings when a cheap drug works. Cardiovascular disease is still the leading cause of premature death and disability in humans, per a 2019 review with 483 citations, and that burden lands on public budgets first. That is the ledger the state is actually managing.

I have spent sixteen years of field work in rural Nigeria, post-conflict Serbia, Kazan, Fukushima. I deployed 210 energy systems across Africa and Asia. The lesson that keeps repeating is never about the technology. It is about who pays to keep the thing running. A cheap, boring, well-understood component that any local technician can service beats a clever one that needs a specialist and a supply chain. Every time. The failure is almost never the design. It is the maintenance. And maintenance has no champion, because no margin depends on the fifth year of a product life. Medicine has the same problem in reverse. The cheap drug is the boring component. It works, or it might, and there is nobody whose revenue depends on finding out. So the question sits, and the cohort ages, and the window described in that 2022 review closes one year at a time. I signed a EUR 20 million personal guarantee on a business that folded on Christmas Day 2008. I lost savings in the 2013 Cyprus bail-in. Neither is a medical credential and I am not presenting one. They are the reason I read the funding structure before the abstract, in every field I touch.

The useful move here is not a pill. It is a habit of reading. Three questions, in order. Was it randomised. What was the endpoint, a scan, a score, or something a person can feel in their own life. Who paid for the trial, and what do they earn if the result is positive. Most of what is sold to men over 45 fails the third question outright. That is not a moral failing. It is structural. The party with an incentive to test a cheap molecule is almost never the party with the money to run the test. What I will not hand you is a dose. Not a protocol, not a stack, not the numbers from my own chart. This is a case study in how medical evidence gets funded and abandoned, not a prescription. The book was published on 8 April 2026, 393 pages, paperback and Kindle, and it argues the same way: the map you were handed is incomplete, and the correction rarely comes from the party that profits from the map. The cheapest wins in medicine have no sales force. That is why so many of them are still open.