Several things, addressing issues arising at different points in this thread:
Protocol: most people are defaulting to the Mayo Clinic's study protocol of 20 mg/kg/day, orally for 2 consecutive days (often with some effort to enhance bioavailability). The dose seems about right or a little bit high based on either the HED method or true allometric scaling, but on reflection the time period does not: in the rodents, the acute clearance protocol was 5 days, and if anything the idea of "biological time" might suggest a longer period for a longer-lived mammal.
Fisetin metabolites: someone suggested that these might have senolytic activity. Remember, the original basis for looking at fisetin was in vitro evidence that the unmetabolized parent molecule had substantial senolytic activity in vitro: it would be very suprising if the detoxification products — which are formed, exactly, to chemically castrate them and ensure their rapid removal from the circualtion — had similar activity.
NAC or other GSH precursors/boosters: Someone suggested taking these to protect the liver. I'm not clear on why anyone's worried about this in the first place: fisetin has shown no acute toxicity at much higher doses (in rodents) than used in this study, and in this study “Amylase and alanine aminotransferase (ALT) were significantly lower in serum of aged WT mice fed the diet supplemented with fisetin, consistent with improved pancreatic and liver homeostasis (Fig. 5C). … Finally, fisetin reduced oxidative stress in the liver of old WT mice (Fig. 5L-M)."
In any case, the notion that NAC or other GSH precursors/boosters against any putative toxicity is an overextrapolation from the fact that these protect hte liver against acetaminophen poisoning. The latter is a quite specific effect, which results from the fact that GSH is required to detoxify acetaminophen, and depletes the liver of same: it doesn't apply generically to hepatotoxicity, and I'm not aware of any reason to think that it would apply to fisetin.
Thymic rejuvenation: OP2040 asked if thymic rejuvenation would be a powerful way to deal with age-related accumulation of senescent cells (SEN). Unfortunately, this won't work: the thymus produces T-cells, whereas the cells responsible for immunosurveillance of SEN are natural killer cells. As you may know, SENS Research Foundation has launched a collaboration with Dr. Judith Campisi to develop rejuvenation biotechnology to augment the ability of NK immunosurveillance of senescent cells.
Curcumin: a number of people have expressed enthusiasm for using curcumin as a senolytic, based on the in vitro studies in the recent fisetin paper. I am highly skeptical that curcumin has senolytic effects in vivo. People following the literature will know that long-term studies of senolytic drug or gene therapies have consistently reported some degree of at least median life expectancy, with credible maximum life extension reported for fisetin (!). If curcumin had senolytic effects (and wasn't otherwise terribly toxic), you'd expect a similar result, even with no attention paid to effects on SEN. But there have been several well-done lifespan studies using curcumin in otherwise-healthy aging mice, and not one of them has found a significant effect on even median life expectancy.
Hokinol: a couple of people have suggested this as a possible senolytic. I'm not clear on where this idea comes from: I can find no evidence for it at all, even in vitro. Can anyone elucidate?
IAC, I don't think it's going to fly, even if there's a mouse study our there showing such an effect: I'm prepared to stand corrected if there are counterexamples, but every in vivo study of hokinol for any purpose that I've seen has delivered the stuff in a lipid emulsion via injection — a crazy thing to try at home, and it's unlikely they'd be doing it if oral or even regular i.p. injections would do the trick.
Fasting as senolytic: I have seen a number of posts in this thread and elsewhere (on Longecity and beyond) suggesting that fasting destroys senescent cells. I am aware of zero evidence for this, and lots of reasons to believe that it doesn't happen. It seems to have come about from wild overextrapolation from Valter Longo's studies showing that fasting can trigger apoptosis in autoreactive T-cells, which are a different kind of "bad cells" altogether. AFAICS, neither Longo nor anyone else has done any studies showing that fasting triggers the death of SEN.
In fact, a prominent feature of senescent cells is exactly that they have high, not low, activation of the autophagy-lysosomal system. This is exactly the basis for the widely-used sanescence-associated beta-galactosidase (SA-β-gal) assay for SEN: senescent cells have far more active lysosomes than nonsenescent cells do, with the result that while SA-β-gal — which is a lysosomal enzyme — is undetectable in non-senescent cells under the standard assay conditions, it is prominent in senescent cells. In fact, activation of autophagy is necessary for the conversion of cells to senescence (PMID 19279323, and rapamycin — whose principle mechanism of action is to suppress mTOR and thereby activate autophagy — reduces the rate at which cells under stress become senescent and inhibits the SASP, but has no effect on SEN survival once they're formed (eg. PMIDs 21715679, 28371119, 19471117 ). (This is fully consistent with the fact that animals on CR per se accumulate fewer senescent cells as they age (PMIDs 26983960, 20844316, 29575469) — including, apparently, CR humans (PMID 29575469 ).
By contrast, inhibiting autophagy has been shown to trigger the death of in senescent cells with exceptionally high SASP engagement (PMID 23945590).
Piperlongumine: forget this until we have in vivo evidence of senolytic activity, long-term safety, and a reliable standardized supply, folks: see here and subsequent posts.
