'Are We Sure Fisetin Actually Clears Senescent Cells?' by Jay Kordich—guest blogger
Full article:
I've been taking fisetin as a monthly “hit-and-run” senolytic. The idea is simple: senescent cells are sometimes described as “zombie cells”—cells that have stopped dividing but haven't died. Instead, they remain and pump out inflammatory substances known collectively as the senescence-associated secretory phenotype, or SASP.
Fisetin is one of the natural compounds that has been identified as a potential senolytic: something that selectively kills these troublesome cells. It sounds great, but recently I started asking a rather awkward question on some forums: “How do we actually know that fisetin removes senescent cells in humans?” The answer turns out to be less satisfying than I expected.
There is no shortage of evidence that fisetin can affect senescent cells. In laboratory experiments, fisetin has been shown to kill various types of senescent cells. In animal models, fisetin can reduce the number of cells carrying senescence-associated markers and improve a variety of age-related problems. This is where the excitement about fisetin comes from.
The senolytic research showed that fisetin could reduce senescent-cell burden in old mice and improve several measures of health and ageing. Subsequent animal studies have produced further encouraging results. So this isn't a completely invented mechanism created by the supplements industry. There is a scientific foundation underlying it. But when we move from mice to people things become less certain.
You might imagine that scientists can simply take a blood sample, count the senescent cells, give someone fisetin, take another blood sample and see whether the number has fallen. Unfortunately, biology hasn't made things that easy. There isn't a single definitive test for senescent cells, and so researchers use a collection of characteristics. They might look for proteins such as p16 or p21, changes in SA-β-gal activity, loss of proliferation, DNA-damage responses, changes in Lamin B1 or various SASP proteins.
The problem is that none of these is a perfect “senescent cell detector”. Different kinds of senescent cells can have different characteristics. A cell can express p16 without necessarily displaying every other feature of senescence. SASP molecules can increase for reasons that have nothing to do with senescent cells. This means that measuring one marker and seeing it fall after fisetin doesn't automatically prove that fisetin has gone around the body killing senescent cells.
Researchers themselves acknowledge this problem. A major 2024 review of fisetin concluded that reliable, sensitive and accessible measures of senescence and treatment response still need to be validated. The authors specifically pointed out that the lack of consensus on validated senescence markers makes it difficult to assess whether fisetin is actually producing senolysis in humans. That's quite an admission.
“But haven't humans taking fisetin shown reductions in senescence markers?” you might ask. And the answer is yes. One human study looked at people who reported taking fisetin at around 100 mg per day. After treatment, researchers found reductions in certain circulating SASP-related markers and a reduction in a population of peripheral blood mononuclear cells displaying a particular senescence-associated phenotype. That's encouraging, but the researchers themselves were careful about what they claimed. The study wasn't the standard high-dose, two-day “hit-and-run” senolytic protocol. It was observational, the participants were self-dosing and the study didn't establish that the changes resulted from actual removal of senescent cells.
The 2024 review makes exactly this point: these human observations suggest beneficial effects, but further clinical studies are needed to establish whether the effects are actually mediated by senescent-cell removal.
Imagine I take fisetin and my blood level of IL-6 falls. That's potentially good news, but IL-6 isn't a senescent-cell counter. It can change for many reasons. Likewise, suppose the number of cells carrying a particular p16-associated marker falls. That's more interesting, but even then, we would ideally want to know whether those cells actually disappeared because fisetin selectively killed them, or whether something else changed their phenotype or their marker expression. This distinction matters because fisetin has other biological properties.
It can influence inflammatory signalling, oxidative stress and other cellular pathways. In other words, fisetin doesn't come with a molecular label saying: “I am only here to kill senescent cells”. It can have other effects too. So if inflammation goes down after taking fisetin, we can't automatically say: “We have just cleared lots of senescent cells”. We need stronger evidence.
This is the part that surprised me. Surely the obvious thing would be to take people with a measurable senescence-associated profile, give them fisetin and measure them before and after treatment. And researchers are beginning to do exactly that.
There are now clinical trials specifically designed to investigate fisetin's effects on senescent cells and to work out better ways of measuring those effects. One trial, for example, is giving participants a short two-day fisetin treatment and then following them for up to three months, measuring pharmacokinetics, safety, inflammation, senescent cells and general health.
Another human study found that a particular senescence-associated cell population in blood, measured using C12FDG, decreased in people reporting fisetin use. This is a useful piece of evidence because it moves beyond simply measuring a generic inflammatory molecule. But even this isn't the same as having a universally accepted measurement of total senescent-cell burden throughout the body. Your liver, muscles, blood vessels, skin, lungs, fat tissue and other organs can contain different populations of senescent cells. A blood test can't simply tell us what is happening in every one of them.
In other words, we haven't yet reached the point where we can confidently say that taking a fisetin capsule causes a measurable reduction in the total number of senescent cells throughout your body.
Does that mean taking fisetin is pointless? Not necessarily. It just means there is enough evidence to make fisetin scientifically interesting. And enough animal evidence to make the senolytic hypothesis compelling.
But there isn't yet enough evidence to say: “We know exactly how much fisetin you need to remove X% of your senescent cells”. And that has consequences for all the other questions we've been asking. How much should we take? How often? Should it be two days or three? Is 1,000 mg enough? Is 1,500 mg better? Does taking it with fat improve the result? Does liposomal fisetin work better? Do we need a monthly pulse? Nobody has properly nailed all of those questions down in humans.
Which makes the liposomal debate rather amusing. This was actually what led me down this rabbit hole. I was looking at plain fisetin versus liposomal fisetin and thinking about whether I should be trying to maximise absorption. But then I realised something rather obvious. We know that liposomal or otherwise enhanced formulations can increase fisetin exposure. Fine. But we don't yet know what level of fisetin exposure is required to clear senescent cells in humans. So if a liposomal product gives me ten times as much fisetin in my bloodstream, what have I actually gained? Maybe it's better. Maybe considerably better. Maybe the extra exposure makes no difference whatsoever to senescent-cell clearance. We just don't know.
The sensible experiment would be remarkably straightforward: Plain fisetin vs enhanced fisetin vs placebo. Measure senescence-associated biomarkers before treatment. Give the treatments. Measure them again. And then, ideally, measure actual clinical outcomes as well. That would tell us whether improved fisetin absorption actually translates into improved senolytic activity. Until then, “better absorbed” and “better senolytic” are two different claims.
Sometimes the most useful discovery isn't finding the answer. It's discovering which question hasn't actually been answered yet. And with fisetin, that question is surprisingly fundamental: Does taking fisetin actually clear a meaningful number of senescent cells in humans? We have good reasons to think it might, but we don't yet have good enough evidence to say that it definitely does.
(of yeah, this one upsets me quite a bit... )
, while DOPC liposomes were completely stable for four weeks at the same temperature, and the authors attributed the difference directly to oxidative susceptibility (PMID 24646435). The underlying chemistry is not new either, oxidation and hydrolysis are the two degradation routes for phospholipid dispersions and they are the reason liposomal stability is a formulation problem rather than a given (PMID 8242840).
