Geroprotectors

Summary

This site already covers seven compounds that the community calls longevity compounds: epitalon, MOTS-c, humanin, SS-31, NAD+, FOXO4-DRI, and glutathione. What was missing was the rest of the shelf — the one that mixes a transplant drug, an over-the-counter supplement, a phase 1 antibody, and a hospital procedure under the same geroprotector label. There are seventeen rows here, and they don't belong to the same category of thing. Rapamycin has 62,814 articles, 2,382 registered studies, and 259 phase 3 trials — but the conditions declared in the three largest registrations I opened are pediatric heart transplant, coronary stent, and breast cancer, not aging. 17-alpha-estradiol, which appears on every geroprotector list, has 20 articles crossing the term with lifespan and no registered study.

Seventeen different things under the same label

“Geroprotector” is a word that lumps together, in the same sentence, sirolimus — a transplant immunosuppressant, with a package insert — and urolithin A powder bought online. This page's job isn't to rank that list by promise, it's to measure how much human outcome evidence exists for each row, and show that the answer changes by two orders of magnitude from one row to the next.

Three method notes, written before the numbers:

First. The search rapamycin OR sirolimus returns the drug's entire literature, almost all of it transplant and oncology. The number is the size of the field, not evidence of use in longevity — which is exactly what doesn't exist in phase 3.

Second. Where the compound only makes sense on this page if it's linked to aging, the query brings the written cross-reference: metformin AND (aging OR longevity), acarbose AND (aging OR lifespan), "therapeutic plasma exchange" AND aging. Without the cross-reference, metformin would bring back the diabetes literature and the number wouldn't say anything about the subject of this page.

Third. The anti-IL-11 row measures the class, not a product: "interleukin-11" AND (antibody OR monoclonal). There's no INN to search for.

No number on this page came from the site's secondary source.

How much evidence exists, compound by compound

The numbers below were compiled by me on September 12, 2026, on PubMed and ClinicalTrials.gov. Each row's query is written out in full, next to the number, so anyone can repeat it and contradict me — and so the script that rechecks this site's counts can run all of them.

The condition column is not a claim about mechanism: it's the Condition field copied from the phase 3 records that the search itself returned. When there is no phase 3 record, the cell says so.

What exists in phase 3, and who it belongs to

Article count measures academic attention, not outcome evidence. The table below separates what has a registered phase 3 trial — and shows the largest registration the search returned in each row, with the declared number of participants and the sponsor.

A registration is not a result. A recruiting phase 3 trial is declared intent with money behind it, and nothing more than that.

Where the numbers don't add up

• 6 of the 17 rows have no registered phase 3 trial. They are Anti-IL-11, GlyNAC, recombinant klotho, plasma exchange, acarbose, 17-alpha-estradiol. Among them are the two that appear most often in longevity sales material: recombinant klotho and plasma exchange.
• Rapamycin has 259 phase 3 trials and none of the ones I opened is about aging — pediatric heart transplant, coronary stent, breast cancer. It's the compound with the most serious biology on the list and, at the same time, the one that relies most on extrapolation when the subject is age.
• NAD+ precursors have more registrations than senolytics. NMN and NR add up to 155 registered studies; fisetin and the dasatinib + quercetin pair add up to 58. The site already has a NAD+ page that says the route of administration changes everything; the count here doesn't contradict that — it only shows where the clinical effort went.
• The 2 phase 3 trials of fisetin declare zero participants enrolled. Both are from an individual investigator, not an industry sponsor, and both are registered with the enrollment field at zero. Registration isn't evidence — and here there wasn't even anyone enrolled.
• 8 of the 13 active ingredients searched have no active registration. The ones that do — sirolimus, metformin, canakinumab, dasatinib, acarbose — are for transplant, diabetes, autoinflammatory disease, and leukemia. None are for aging.

In Brazil

I searched every active ingredient in the open data on registered medicines, downloaded on September 12, 2026 — 43,508 rows, counting only Active status.

Two rows from the evidence table don't belong here because they aren't an active ingredient: plasma exchange is a procedure and anti-IL-11 is a class without a product. The rest were searched by the Portuguese name of the active ingredient, including the ones sold as supplements — and that's where the table shows what matters: a supplement doesn't show up in the drug database because it isn't a drug, and that isn't the same as being safe.

References

1. Public ClinicalTrials.gov API, version 2 — the source for all the study and phase 3 counts on this page.
2. PubMed E-utilities (esearch) — the basis for all article counts on this page.
3. NCT05653258 registration on ClinicalTrials.gov — the only phase 3 trial that the search for dasatinib with quercetin returned, sponsored by Cedars-Sinai Medical Center.
4. NCT06739720 registration on ClinicalTrials.gov — the only phase 3 registration for nicotinamide mononucleotide that the search returned, sponsored by the University of Hong Kong.

← Back to all compounds