Evidence.zone is a free resource for men on active surveillance for prostate cancer. Every intervention rated by the strength of the published research — so you know what's proven, what's promising, and what's noise.
What's new in the evidence database this month.
1 New Intervention Added
Matrine — Watch, don't take (Tier 3 — Needs Evidence)
A quinolizidine alkaloid from Sophora flavescens (Kushen), and the principal active ingredient in Compound Kushen Injection — an intravenous herbal medicine approved in China as adjunctive cancer therapy. The preclinical prostate signal is real: anti-proliferative and pro-apoptotic in both androgen-sensitive and castration-resistant cell lines, working through ER stress and reversal of epithelial-to-mesenchymal transition. Unusually for a botanical, it pushes against the androgen pathway rather than boosting it — the opposite of the testosterone-adjacent supplements marketed to men our age.
Three gaps separate that from anything useful on surveillance. The human evidence is intravenous, not the oral capsule you'd buy. It's in advanced disease, not early localised disease. And there is no human prostate data at all. Over the top of which sits dose-dependent liver, nerve and reproductive toxicity — described in the review literature as a limit on the compound's own clinical use.
Tier 3 has a new name
"Insufficient Evidence" has become Needs Evidence.
Small change, deliberate one. The old label described a shortage. The new one says who has the problem. These compounds aren't waiting on better readers — they're waiting on someone to run the trial.
The action line is unchanged: Watch, don't take.
Nine compounds, no trials
Matrine is the ninth entry in the library with the same shape: genuine preclinical activity against prostate cancer cells, and not one human prostate trial. It joins Ivermectin, Fenbendazole, Mebendazole, Berberine, Apigenin, Artemisinin, Taurine and Bitter Melon.
That cluster is worth pausing on, because it's easy to read the wrong way.
Nearly everything kills cancer cells in a dish. That isn't cynicism, it's a fact about the method. In a petri dish you can hold a compound at a concentration you will never reach in prostate tissue by swallowing a capsule. The mouse gets injected, often straight into the tumour. So "inhibited proliferation in PC-3 cells" tells you a mechanism exists. It tells you very little about a man on surveillance taking it for three years.
Matrine is the sharpest version of the problem, because it does have human data. Pooled trials show Compound Kushen Injection improved tumour response and one-year survival when added to chemoembolisation. For liver cancer. Given intravenously. In advanced disease. Three substitutions away from a man on surveillance swallowing a capsule — and every one of them matters. This is how evidence gets quietly transposed onto a situation it was never about.
So the question that counts isn't whether the biology looks plausible. It's whether anyone has spent the money to test it in men with early prostate cancer. For these nine, nobody has — and there's a structural reason. They're cheap, off-patent, or unpatentable. No one owns the upside of proving they work, so no one funds the trial that would prove it. Not a conspiracy, just how drug development is paid for.
It's also why Tier 3 is the largest tier on this site, and why it will stay that way.
Needs Evidence doesn't mean no. It means nobody knows, and the people who could find out have no reason to. If you decide to take something from that tier, do it with your eyes open: tell your consultant, get liver function checked, and don't let it displace the things in Tiers 1 and 2 that have actually been tested in humans.
The library exists so you can tell those two situations apart.
@paulsweeney The interesting thing is that the economic incentives *do* exist.
Even in the U.S., the government is the primary purchaser of medicine and medical services; outside the U.S., it is often almost the *sole* purchaser. So if investing millions in trials for less-expensive remedies means saving billions in taxpayer dollars, the math is pretty simple.
And it does happen: there are huge trials for practices like active surveillance, different surgical interventions, ADT holidays, etc, that don't have any commercial potential. STAMPEDE is an excellent example of a long-running, multi-faceted trial that has had a big impact on us as prostate-cancer patients.
So if a natural remedy hasn't had a major trial, it's not just because it has no commercial potential for pharma companies; it's because data from early, small-scale academic studies hasn't been sufficient, consistent, or persuasive enough to take it to the next level.