Fisetin

updated

Fisetin is a plant flavonol sold as a senolytic that clears senescent cells and supports healthy aging, and no FDA-approved fisetin drug or label turned up in searches on 11 October 2026. The senolytic result is from cells and mice. Human evidence is small and mostly null or biomarker-only: a knee osteoarthritis trial with posted results showed no secondary-outcome difference, a COVID-19 nursing-home trial was stopped for futility, and no record shows senescent cell clearance in people.

Plant flavonol, promoted as a senolytic

  • Fisetin was the most potent of 10 flavonoids screened for senolytic activity in senescent fibroblasts, and it extended median and maximum lifespan in wild-type mice given it late in life; no living humans were studied.
  • In 15 healthy volunteers, 1000 mg of unformulated fisetin reached a plasma Cmax of 9.97 ng/ml versus 238.2 ng/ml for 1000 mg of the hydrogel formulation FF-20, a 26.9-fold AUC difference.
  • In a registry-posted trial in 74 adults with knee osteoarthritis, the 6-minute walk at month 12 was 544.2 m on fisetin and 543.6 m on placebo (p = 0.9572); the results are sponsor-posted and not peer reviewed.
  • The COVID-19 nursing-home trial (20 enrolled) was stopped early due to futility, according to its registry limitations statement.
  • In a 10-adult aging-clock pilot, biological age by the TruAge test was lower in 4, higher in 5 and unchanged in 1.
  • A PubMed search on 11 October 2026 for fisetin, senescence or senolytic terms and human trials returned 1 record, which was not a fisetin trial.

What is fisetin, and is it approved?

Fisetin is a plant flavonol sold as a supplement for senolytic and healthy-aging claims. On 11 October 2026, Drugs@FDA and DailyMed searches found no fisetin drug or label, so no FDA-approved fisetin product turned up. Products differ, so no trial result transfers to another product or dose.

Fisetin (PubChem CID 5281614, C15H10O6, 286.24 g/mol, CAS 528-48-3; synonyms include 5-Desoxyquercetin and 3,3',4',7-Tetrahydroxyflavone) is a plant flavonol and a single defined molecule.

The products differ. The human pharmacokinetic study below used unformulated fisetin and a hydrogel formulation called FF-20, one registered trial names a branded capsule (Novusetin), the Gulf War trial used fisetin sourced from Rhus succedanea, and registered trial arms use either 100 mg capsules or about 20 mg/kg/day short courses. No trial result transfers to a different product or dose.

On 11 October 2026, an openFDA Drugs@FDA query for fisetin returned no match and a DailyMed drug-name search returned no labels, so no FDA-approved fisetin drug or label was found in those two sources. FDA dietary-ingredient or GRAS status is not established here.

PubChem: fisetin identity; no FDA drug or label found, 11 October 2026

Chemical identity record

National Library of Medicine. PubChem compound record CID 5281614 (fisetin). openFDA Drugs@FDA and DailyMed drug-name searches for fisetin run on 11 October 2026.

Fisetin: PubChem CID 5281614, C15H10O6, 286.24 g/mol, CAS 528-48-3, with 5-Desoxyquercetin and 3,3',4',7-Tetrahydroxyflavone among the listed synonyms. It is a plant flavonol and a single defined molecule, but the products differ: the human pharmacokinetic study used unformulated fisetin and a hydrogel formulation called FF-20, one registered trial names a branded capsule (Novusetin), the Gulf War trial used fisetin sourced from Rhus succedanea, and registered trial arms use either 100 mg capsules or about 20 mg/kg/day short courses. On 11 October 2026 an openFDA Drugs@FDA query for fisetin returned no match and a DailyMed drug-name search returned no labels.

Participants / model
Chemical identity, not a clinical population
Treatment
Defined molecule; several product forms
Follow-up
Searches run on 11 October 2026
Study design
NLM chemical record plus dated US approval-record search
Funding
not stated in the source

No trial result transfers to a different product or dose. FDA dietary-ingredient or GRAS status is not established here.

Read the original source

Where does the senolytic claim come from?

From cell, mouse and tissue-explant work. Fisetin was the most potent of 10 flavonoids screened in senescent fibroblasts, lowered senescence markers in old and progeroid mice, and extended median and maximum lifespan in wild-type mice given it late in life. No living humans were studied.

The authors screened 10 flavonoids for senolytic activity in senescent murine and human fibroblasts and found fisetin the most potent. Acute or intermittent fisetin lowered senescence markers in multiple tissues of progeroid and old mice, and it lowered senescence in a subset of cells in murine and human adipose tissue explants.

Given to wild-type mice late in life, fisetin restored tissue homeostasis, reduced age-related pathology and extended median and maximum lifespan. The authors say this suggests human trials are feasible.

The work used cultured cells, mice and tissue explants. No living humans were studied, and the route and dose given to the mice are not stated in the source, so the mouse result is not a human dose or a human effect.

Origin of the claim: cells, mice and tissue explants, no living humans

Animal study

Yousefzadeh MJ et al. EBioMedicine. 2018;36:18-28. PMID 30279143.

The authors screened 10 flavonoids for senolytic activity in senescent murine and human fibroblasts and found fisetin the most potent. Acute or intermittent fisetin lowered senescence markers in multiple tissues of progeroid and old mice, lowered senescence in a subset of cells in murine and human adipose tissue explants, and, given to wild-type mice late in life, restored tissue homeostasis, reduced age-related pathology, and extended median and maximum lifespan. The authors say this suggests human trials are feasible.

Participants / model
Cultured cells, progeroid and old wild-type mice, and human adipose tissue explants; no living humans
Treatment
Fisetin given to mice (acute or intermittent), with cell and explant exposure; route and dose not stated in the source
Follow-up
Acute or intermittent exposure; schedule not stated in the source
Study design
Animal and ex vivo study
Funding
not stated in the source

Cell, mouse and explant findings only. They cannot be read as an effect in people.

Read the original source

How much fisetin reaches the blood?

Little, unless it is formulated. In 15 healthy volunteers, a single 1000 mg dose of unformulated fisetin reached a plasma Cmax of 9.97 ng/ml against 238.2 ng/ml for 1000 mg of the hydrogel formulation FF-20, a 26.9-fold AUC difference.

In 15 healthy volunteers in a double-blind crossover study, a single 1000 mg dose of unformulated fisetin reached a plasma Cmax of 9.97 ng/ml (AUC 0 to 12 hours, 12.67), against 238.2 ng/ml (AUC 341.4) for 1000 mg of the hydrogel formulation FF-20, which delivers 192 mg fisetin. That is a 26.9-fold AUC difference and a Cmax more than 23 times higher.

The authors call this the first human pharmacokinetic study of fisetin and describe unformulated fisetin as poorly bioavailable. No adverse events were reported. The abstract gives no units for AUC and does not state funding or conflicts, and no half-life appears in the sources reviewed here.

A result from one formulation cannot be applied to another, so trial results with capsules, branded products or a formulated hydrogel do not transfer to each other.

First human PK study: 15 volunteers, 26.9-fold AUC difference between formulations

Human study

Krishnakumar IM et al. J Nutr Sci. 2022;11:e74. PMID 36304817.

In a double-blind crossover study, a single 1000 mg dose of unformulated fisetin reached a plasma Cmax of 9.97 ng/ml (AUC 0 to 12 hours, 12.67), against 238.2 ng/ml (AUC 341.4) for 1000 mg of the hydrogel formulation FF-20, which delivers 192 mg fisetin. That is a 26.9-fold AUC difference and a Cmax more than 23 times higher. The authors call this the first human pharmacokinetic study of fisetin and describe unformulated fisetin as poorly bioavailable. No adverse events were reported.

Participants / model
15 healthy volunteers
Treatment
Single oral dose of 1000 mg unformulated fisetin or 1000 mg FF-20 (192 mg fisetin)
Follow-up
Single dose; exposure measured over 0 to 12 hours
Study design
Double-blind randomized crossover pharmacokinetic study, formulation comparison
Funding
not stated in the source

Plasma fisetin and the metabolite geraldol were measured. The abstract gives no units for AUC and does not state funding or conflicts, and no half-life or Tmax appears in the sources reviewed.

Read the original source
PubChem: fisetin identity; no FDA drug or label found, 11 October 2026

Chemical identity record

National Library of Medicine. PubChem compound record CID 5281614 (fisetin). openFDA Drugs@FDA and DailyMed drug-name searches for fisetin run on 11 October 2026.

Fisetin: PubChem CID 5281614, C15H10O6, 286.24 g/mol, CAS 528-48-3, with 5-Desoxyquercetin and 3,3',4',7-Tetrahydroxyflavone among the listed synonyms. It is a plant flavonol and a single defined molecule, but the products differ: the human pharmacokinetic study used unformulated fisetin and a hydrogel formulation called FF-20, one registered trial names a branded capsule (Novusetin), the Gulf War trial used fisetin sourced from Rhus succedanea, and registered trial arms use either 100 mg capsules or about 20 mg/kg/day short courses. On 11 October 2026 an openFDA Drugs@FDA query for fisetin returned no match and a DailyMed drug-name search returned no labels.

Participants / model
Chemical identity, not a clinical population
Treatment
Defined molecule; several product forms
Follow-up
Searches run on 11 October 2026
Study design
NLM chemical record plus dated US approval-record search
Funding
not stated in the source

No trial result transfers to a different product or dose. FDA dietary-ingredient or GRAS status is not established here.

Read the original source

What did the placebo-controlled trials with posted results find?

No benefit appeared on the registered secondary outcomes. In knee osteoarthritis (74 randomized), adverse events were not higher than on placebo and the 6-minute walk at month 12 was 544.2 m versus 543.6 m. A COVID-19 nursing-home trial was stopped for futility. These results are sponsor-posted on ClinicalTrials.gov and have not been peer reviewed.

NCT04210986 (Steadman Philippon Research Institute, completed 2023) randomized 74 adults aged 40 to 80 with knee osteoarthritis (34 fisetin, 40 placebo) to oral fisetin, about 20 mg/kg for two consecutive days, 28 days off, then two more consecutive days, or placebo. Treatment-emergent adverse events, the primary outcome, occurred in 28 of 34 on fisetin and 33 of 40 on placebo (Fisher exact odds ratio 0.99, 95% CI 0.25 to 4.02), and serious adverse events were 2 of 34 and 2 of 40. Among registered secondary outcomes, the 6-minute walk at month 12 was 544.2 m on fisetin and 543.6 m on placebo (p = 0.9572), and posted contrasts for C-reactive protein, the cartilage marker COMP, timed-up-and-go and quadriceps torque were not significant. 25 of 34 fisetin and 35 of 40 placebo participants completed.

COVFIS-HOME (NCT04771611, sponsored by investigator James L. Kirkland of Mayo Clinic) randomized 55 adults with COVID-19, 29 to fisetin and 26 to placebo. At 60 days, 26 of 27 analyzed on fisetin and 22 of 24 on placebo had no limitations on the WHO ordinal scale, and serious adverse events were 0 of 29 versus 2 of 26. COVID-FIS in nursing-home residents (NCT04537299) enrolled 20, 10 per arm, and was stopped early: the registry limitations statement says the DSMB and NIA determined the study should be stopped due to futility, and the posted severity scores were nearly identical in both arms.

NCT04313634 (Mayo Clinic) randomized 74 women aged 60 or older to dasatinib plus quercetin (30), fisetin (14) or no intervention (30), with no placebo. The CTX bone-resorption outcome at 20 weeks was a median change of +13.7% on fisetin, -4.1% on dasatinib plus quercetin and -7.7% for controls. The single posted test (p = 0.611) matches the dasatinib plus quercetin versus control comparison in the published paper, and no separate test for fisetin is posted. Plasma SASP at 2 weeks changed a median -0.9% on fisetin versus -2.5% for controls (p = 0.953 across arms). The published paper reports the dasatinib plus quercetin arm, and no publication of the fisetin arm was found.

These results are posted by the sponsors on ClinicalTrials.gov and have not been through peer review. A PubMed search on 11 October 2026 found no peer-reviewed report of the knee osteoarthritis trial or of the two COVID-19 trials; the only COVID-19 trial paper was a 2021 design paper (PMID 34375437).

Knee osteoarthritis: 74 randomized, adverse events not higher than placebo, registry-posted results

Trial registry record

ClinicalTrials.gov record NCT04210986 (Steadman Philippon Research Institute, completed 2023). Results are posted by the sponsor and have not been through peer review.

The trial randomized 74 adults aged 40 to 80 (34 fisetin, 40 placebo). Treatment-emergent adverse events, the primary outcome, occurred in 28 of 34 on fisetin and 33 of 40 on placebo (Fisher exact odds ratio 0.99, 95% CI 0.25 to 4.02). Serious adverse events were 2 of 34 and 2 of 40. Among registered secondary outcomes, the 6-minute walk at month 12 was 544.2 m on fisetin and 543.6 m on placebo (p = 0.9572), and posted contrasts for C-reactive protein, the cartilage marker COMP, timed-up-and-go and quadriceps torque were not significant. 25 of 34 fisetin and 35 of 40 placebo participants completed. A PubMed search on 11 October 2026 (fisetin AND (osteoarthritis OR knee) AND (senolytic OR senescence) AND (trial OR randomized)) returned 3 records, none a report of this trial.

Participants / model
74 randomized adults aged 40 to 80 with knee osteoarthritis
Treatment
Oral fisetin about 20 mg/kg for 2 consecutive days, 28 days off, then 2 more consecutive days, or placebo
Follow-up
Two short courses 28 days apart; outcomes through month 12
Study design
Registry-posted results of a randomized placebo-controlled trial; not peer reviewed
Funding
Sponsor: Steadman Philippon Research Institute; funding source not stated in the source

Registry results are sponsor-posted and have not been through peer review. No peer-reviewed report of this trial was found in PubMed on 11 October 2026.

Read the original source
COVID-19: 55 outpatients, and 20 nursing-home residents with the trial stopped for futility

Trial registry record

ClinicalTrials.gov records NCT04537299 (COVID-FIS) and NCT04771611 (COVFIS-HOME). Results are posted by the sponsor and have not been through peer review.

COVFIS-HOME (NCT04771611, sponsored by investigator James L. Kirkland of Mayo Clinic) randomized 55 adults, 29 to fisetin and 26 to placebo, to about 20 mg/kg/day for two days at days 0 to 1 and 8 to 9. At 60 days 26 of 27 analyzed on fisetin and 22 of 24 on placebo had no limitations on the WHO ordinal scale, and serious adverse events were 0 of 29 versus 2 of 26. COVID-FIS in nursing-home residents (NCT04537299) enrolled 20, 10 per arm, and was stopped early: the registry limitations statement says the DSMB and NIA determined the study should be stopped due to futility, and the posted severity scores were nearly identical in both arms (a mean of 2 with SD 0 at most time points). A PubMed search on 11 October 2026 (fisetin[tiab] AND (COVID* OR SARS-CoV-2)[tiab] AND (randomized OR trial OR results)[tiab]) returned 11 records; the only trial paper was a 2021 design paper (PMID 34375437).

Participants / model
55 adults with COVID-19 (age 18 or older); 20 older nursing-home residents with COVID-19
Treatment
Oral fisetin about 20 mg/kg/day for 2 days, repeated at days 8 to 9, or placebo
Follow-up
60 days (COVFIS-HOME); stopped early (COVID-FIS)
Study design
Registry-posted results of two randomized placebo-controlled pilots; not peer reviewed
Funding
COVFIS-HOME is sponsored by investigator James L. Kirkland of Mayo Clinic; funding source not stated in the source for either trial

Registry results are sponsor-posted and have not been through peer review, and neither trial has a peer-reviewed report in PubMed as of 11 October 2026. One trial was stopped for futility.

Read the original source
Older women: 14 on fisetin versus 30 untreated, open-label, registry-posted results

Trial registry record

ClinicalTrials.gov record NCT04313634 (Mayo Clinic). Results are posted by the sponsor and have not been through peer review. The dasatinib plus quercetin arm is published separately (PMID 38956196).

The trial randomized 74 women aged 60 or older to dasatinib plus quercetin (30), fisetin (14, about 20 mg/kg/day for three consecutive days, every 28 days over 20 weeks), or no intervention (30). The CTX bone-resorption outcome at 20 weeks was a median change of +13.7% on fisetin, -4.1% on dasatinib plus quercetin and -7.7% for controls; the single posted test (p = 0.611) matches the dasatinib plus quercetin versus control comparison in the published paper, and no separate test for fisetin is posted. Plasma SASP (senescence-associated secretory phenotype) at 2 weeks changed a median -0.9% on fisetin versus -2.5% for controls (p = 0.953 across arms). One serious adverse event, a gastric hemorrhage, occurred in the fisetin arm (1 of 14), and 12 of 14 on fisetin reported other adverse events versus 5 of 30 controls.

Participants / model
14 women aged 60 or older on fisetin (11 analyzed for CTX), versus 30 untreated controls
Treatment
Oral fisetin about 20 mg/kg/day for 3 days every 28 days over 20 weeks
Follow-up
20 weeks
Study design
Registry-posted results of a randomized open-label trial with a no-intervention control; not peer reviewed
Funding
Sponsor: Mayo Clinic; funding source not stated in the source

The control group got no placebo, so the adverse-event comparison is open-label. The published paper reports the dasatinib plus quercetin arm, and no publication of the fisetin arm was found.

Read the original source

What did the aging-clock pilots and small disease trials find?

The aging-clock pilots were small and uncontrolled: in 10 healthy adults over 50, biological age was lower in 4, higher in 5 and unchanged in 1. Small randomized trials in Gulf War illness, colorectal cancer and obesity used inflammation markers, and none measured senolytic or aging outcomes.

In 10 healthy adults over 50 taking 500 mg fisetin daily for one week per month for six months, biological age by the TruAge test was lower in 4 of 10, higher in 5 of 10 and unchanged in 1 of 10. Telomere length did not change significantly and no adverse effects were noted. The authors write that fisetin is not recommended as an anti-aging agent until more extensive studies are done, and the abstract describes no placebo group.

A separate longitudinal pilot of 19 participants given dasatinib and quercetin plus fisetin for 6 months (10 had been in an earlier dasatinib plus quercetin study) found non-significant increases in epigenetic age acceleration. Several of its authors are TruDiagnostic employees. These are biomarker clocks in small open studies, not aging outcomes.

In 21 male veterans with Gulf War illness, in a placebo-controlled pseudo-randomized crossover trial, fisetin did not reduce symptom severity at the lower (p = 0.504) or higher (p = 0.616) dose, with no conflict of interest declared. In 37 patients with colorectal cancer on chemotherapy, 100 mg fisetin (n = 18) or placebo (n = 19) for seven weeks lowered IL-8 versus placebo (p < 0.03), while hs-CRP and MMP-7 fell only within the fisetin group.

In 44 obese men in a four-arm 12-week trial (fisetin 200 mg/day, training, both, or placebo), IL-6 fell in the fisetin-alone arm (p = 0.013) and TNF-alpha fell in every active arm, with the largest drops in glucose, insulin and HOMA-IR in the fisetin plus training arm; no conflict of interest was reported. These are surrogate markers in disease or obesity populations, not senolytic or aging outcomes.

Aging-clock pilots: 10 adults on fisetin (5 of 10 had a higher biological age) and 19 on a combined regimen

Human study

Lee E et al. Altern Ther Health Med. 2024;30(9):6-10. PMID 39269340.

In 10 healthy adults over 50 taking 500 mg fisetin daily for one week per month for six months, biological age by the TruAge test was lower in 4 of 10, higher in 5 of 10 and unchanged in 1 of 10, telomere length did not change significantly, and no adverse effects were noted. The authors write that fisetin is not recommended as an anti-aging agent until more extensive studies are done. The abstract describes no placebo group. A separate longitudinal pilot of 19 participants given dasatinib and quercetin plus fisetin for 6 months (10 had been in an earlier dasatinib plus quercetin study) found non-significant increases in epigenetic age acceleration.

Participants / model
10 healthy adults over 50; 19 participants on dasatinib and quercetin plus fisetin (health status not stated in the abstract)
Treatment
Oral fisetin 500 mg daily, 1 week a month for 6 months; dasatinib and quercetin plus fisetin for 6 months
Follow-up
6 months in each pilot
Study design
Two small uncontrolled pilot studies
Funding
not stated in the source; several authors of the 19-participant pilot are TruDiagnostic employees

These are biomarker clocks in small open studies, not aging outcomes.

Read the original source
Small randomized trials: Gulf War illness (21), colorectal cancer (37), obesity (44)

Randomized controlled trial

Hodgin KS et al. Int J Environ Res Public Health. 2021;18(5). PMID 33802381.

In 21 male veterans with Gulf War illness in a placebo-controlled pseudo-randomized crossover trial, fisetin did not reduce symptom severity at the lower (p = 0.504) or higher (p = 0.616) dose, with no conflict of interest declared. In 37 patients with colorectal cancer on chemotherapy, 100 mg fisetin (n = 18) or placebo (n = 19) for seven weeks lowered IL-8 versus placebo (p < 0.03), while hs-CRP and MMP-7 fell only within the fisetin group. In 44 obese men in a four-arm 12-week trial (fisetin 200 mg/day, training, both, or placebo), IL-6 fell in the fisetin-alone arm (p = 0.013) and TNF-alpha fell in every active arm, with the largest drops in glucose, insulin and HOMA-IR in the fisetin plus training arm; no conflict of interest was reported.

Participants / model
21 veterans with Gulf War illness; 37 patients with colorectal cancer on chemotherapy; 44 men with obesity
Treatment
Oral fisetin; doses as stated per study
Follow-up
Not stated in the source for the Gulf War trial; 7 weeks (colorectal cancer); 12 weeks (obesity)
Study design
Three small randomized trials; the Gulf War trial was pseudo-randomized and a crossover
Funding
No conflict of interest was declared in the Gulf War and obesity trials; funding is not stated in the source

These are surrogate markers in disease or obesity populations, not senolytic or aging outcomes.

Read the original source

Does fisetin clear senescent cells and support healthy aging in people?

No supporting study in people was found. A PubMed search on 11 October 2026 returned 1 record, which was not a fisetin trial, and the one senescence-related outcome with posted values, plasma SASP at 2 weeks, showed no difference.

Sellers describe fisetin as a senolytic that clears senescent cells and supports healthy aging in people. The senolytic part is supported in mice. The human part has no supporting evidence in the records reviewed.

A PubMed search on 11 October 2026 for fisetin with senescence or senolytic terms, limited to humans and to randomized or clinical trial publication types, returned 1 record, a small cell lung cancer chemotherapy-resistance paper that is not a fisetin trial. The one senescence-related outcome with posted values among the registry results, plasma SASP at 2 weeks in NCT04313634, showed no difference (median -0.9% on fisetin versus -2.5% for controls, p = 0.953 across arms). The carpal tunnel record lists secondary outcomes on senescence blood markers but posts no values. No record reviewed shows fisetin clearing senescent cells in people, and no record reports a hard aging outcome.

A ClinicalTrials.gov search the same day with fisetin as an intervention returned 36 studies: 10 recruiting, 5 active not recruiting, 2 enrolling by invitation, 1 not yet recruiting, 7 completed, 2 terminated, 2 suspended, 4 withdrawn and 3 unknown status. Six have posted results. A 2026 review counted 34 registered trials and 4 completed trials with available results and judged the clinical evidence limited and heterogeneous (PMID 42796982).

Two Mayo frailty trials (NCT03430037 in older women and NCT03675724 in older adults) were first posted in 2018, are listed as enrolling by invitation with primary completion in November 2026, and have no results. NCT07195318 (Denmark, 100 mg daily, an estimated 120 adults aged 50 or older, primary outcome suPAR, primary completion listed as 31 October 2026) is the registered healthy-aging trial, and its design paper is PMID 42633989. Registered studies without results show that a study was registered, not that it works.

PubMed search, 11 October 2026: 1 record, not a fisetin trial; no human senescent cell clearance

PubMed search record

PubMed search run on 11 October 2026: fisetin[tiab] AND (senescen*[tiab] OR senolytic*[tiab]) AND humans[mh] AND (randomized controlled trial[pt] OR clinical trial[pt]). 1 record returned.

The search returned 1 record, a small cell lung cancer chemotherapy-resistance paper that is not a fisetin trial. Among the registry results reviewed, the one senescence-related outcome with posted values, plasma SASP at 2 weeks in NCT04313634, showed no difference (median -0.9% on fisetin versus -2.5% for controls, p = 0.953 across arms). The carpal tunnel record lists secondary outcomes on senescence blood markers but posts no values. No record reviewed shows fisetin clearing senescent cells in people, and no record reports a hard aging outcome.

Participants / model
Records reviewed, not a population
Treatment
Not applicable
Follow-up
Search run on 11 October 2026
Study design
Literature search and registry results
Funding
not stated in the source

A search result shows what PubMed returned on that date. It is not proof that no unindexed study exists.

Read the original source
Registry search, 11 October 2026: 36 studies, 6 with posted results

Trial registry record

ClinicalTrials.gov search with fisetin as an intervention, run on 11 October 2026. 36 studies returned. A 2026 review (PMID 42796982) counted 34 registered trials.

A ClinicalTrials.gov search run on 11 October 2026 with fisetin as an intervention returned 36 studies: 10 recruiting, 5 active not recruiting, 2 enrolling by invitation, 1 not yet recruiting, 7 completed, 2 terminated, 2 suspended, 4 withdrawn and 3 unknown status. Six have posted results: the knee osteoarthritis, older women and two COVID-19 records, plus a Gulf War record (NCT02909686) and a carpal tunnel record (NCT05416515; single-arm, 40 participants, mean BCTQ change of -0.5). A 2026 review (PMID 42796982) counted 34 registered trials and 4 completed trials with available results and judged the clinical evidence limited and heterogeneous. The two Mayo frailty trials, NCT03430037 (older women) and NCT03675724 (older adults), were first posted in 2018, are listed as enrolling by invitation with primary completion in November 2026, and have no results. NCT07195318 (Denmark, 100 mg daily, an estimated 120 adults aged 50 or older, primary outcome suPAR, primary completion listed as 31 October 2026) is the registered healthy-aging trial, and its design paper is PMID 42633989.

Participants / model
Registered trials, not a population
Treatment
Not applicable
Follow-up
Search run on 11 October 2026
Study design
Trial-registry search plus a review
Funding
not stated in the source

Registered studies without results show that a study was registered, not that it works. Registry results are sponsor-posted and not peer reviewed.

Read the original source
Older women: 14 on fisetin versus 30 untreated, open-label, registry-posted results

Trial registry record

ClinicalTrials.gov record NCT04313634 (Mayo Clinic). Results are posted by the sponsor and have not been through peer review. The dasatinib plus quercetin arm is published separately (PMID 38956196).

The trial randomized 74 women aged 60 or older to dasatinib plus quercetin (30), fisetin (14, about 20 mg/kg/day for three consecutive days, every 28 days over 20 weeks), or no intervention (30). The CTX bone-resorption outcome at 20 weeks was a median change of +13.7% on fisetin, -4.1% on dasatinib plus quercetin and -7.7% for controls; the single posted test (p = 0.611) matches the dasatinib plus quercetin versus control comparison in the published paper, and no separate test for fisetin is posted. Plasma SASP (senescence-associated secretory phenotype) at 2 weeks changed a median -0.9% on fisetin versus -2.5% for controls (p = 0.953 across arms). One serious adverse event, a gastric hemorrhage, occurred in the fisetin arm (1 of 14), and 12 of 14 on fisetin reported other adverse events versus 5 of 30 controls.

Participants / model
14 women aged 60 or older on fisetin (11 analyzed for CTX), versus 30 untreated controls
Treatment
Oral fisetin about 20 mg/kg/day for 3 days every 28 days over 20 weeks
Follow-up
20 weeks
Study design
Registry-posted results of a randomized open-label trial with a no-intervention control; not peer reviewed
Funding
Sponsor: Mayo Clinic; funding source not stated in the source

The control group got no placebo, so the adverse-event comparison is open-label. The published paper reports the dasatinib plus quercetin arm, and no publication of the fisetin arm was found.

Read the original source

How is fisetin supposed to work?

Fisetin is described as a senolytic, a compound that clears senescent cells. That is supported in cells and mice. In people, unformulated fisetin was poorly bioavailable, and no record reviewed shows senescent cell clearance.

Fisetin is described as a senolytic, a compound that clears senescent cells. That description comes from the cell, mouse and explant work: fisetin was the most potent of 10 flavonoids screened in senescent fibroblasts and lowered senescence markers in progeroid and old mice and in adipose tissue explants.

Exposure is a constraint. Unformulated fisetin was poorly bioavailable in 15 healthy volunteers, and a hydrogel formulation raised plasma exposure 26.9-fold by AUC at the same 1000 mg product dose.

In people, the one senescence-related outcome with posted values, plasma SASP at 2 weeks, showed no difference, and no record reviewed shows senescent cell clearance.

Origin of the claim: cells, mice and tissue explants, no living humans

Animal study

Yousefzadeh MJ et al. EBioMedicine. 2018;36:18-28. PMID 30279143.

The authors screened 10 flavonoids for senolytic activity in senescent murine and human fibroblasts and found fisetin the most potent. Acute or intermittent fisetin lowered senescence markers in multiple tissues of progeroid and old mice, lowered senescence in a subset of cells in murine and human adipose tissue explants, and, given to wild-type mice late in life, restored tissue homeostasis, reduced age-related pathology, and extended median and maximum lifespan. The authors say this suggests human trials are feasible.

Participants / model
Cultured cells, progeroid and old wild-type mice, and human adipose tissue explants; no living humans
Treatment
Fisetin given to mice (acute or intermittent), with cell and explant exposure; route and dose not stated in the source
Follow-up
Acute or intermittent exposure; schedule not stated in the source
Study design
Animal and ex vivo study
Funding
not stated in the source

Cell, mouse and explant findings only. They cannot be read as an effect in people.

Read the original source
First human PK study: 15 volunteers, 26.9-fold AUC difference between formulations

Human study

Krishnakumar IM et al. J Nutr Sci. 2022;11:e74. PMID 36304817.

In a double-blind crossover study, a single 1000 mg dose of unformulated fisetin reached a plasma Cmax of 9.97 ng/ml (AUC 0 to 12 hours, 12.67), against 238.2 ng/ml (AUC 341.4) for 1000 mg of the hydrogel formulation FF-20, which delivers 192 mg fisetin. That is a 26.9-fold AUC difference and a Cmax more than 23 times higher. The authors call this the first human pharmacokinetic study of fisetin and describe unformulated fisetin as poorly bioavailable. No adverse events were reported.

Participants / model
15 healthy volunteers
Treatment
Single oral dose of 1000 mg unformulated fisetin or 1000 mg FF-20 (192 mg fisetin)
Follow-up
Single dose; exposure measured over 0 to 12 hours
Study design
Double-blind randomized crossover pharmacokinetic study, formulation comparison
Funding
not stated in the source

Plasma fisetin and the metabolite geraldol were measured. The abstract gives no units for AUC and does not state funding or conflicts, and no half-life or Tmax appears in the sources reviewed.

Read the original source
Older women: 14 on fisetin versus 30 untreated, open-label, registry-posted results

Trial registry record

ClinicalTrials.gov record NCT04313634 (Mayo Clinic). Results are posted by the sponsor and have not been through peer review. The dasatinib plus quercetin arm is published separately (PMID 38956196).

The trial randomized 74 women aged 60 or older to dasatinib plus quercetin (30), fisetin (14, about 20 mg/kg/day for three consecutive days, every 28 days over 20 weeks), or no intervention (30). The CTX bone-resorption outcome at 20 weeks was a median change of +13.7% on fisetin, -4.1% on dasatinib plus quercetin and -7.7% for controls; the single posted test (p = 0.611) matches the dasatinib plus quercetin versus control comparison in the published paper, and no separate test for fisetin is posted. Plasma SASP (senescence-associated secretory phenotype) at 2 weeks changed a median -0.9% on fisetin versus -2.5% for controls (p = 0.953 across arms). One serious adverse event, a gastric hemorrhage, occurred in the fisetin arm (1 of 14), and 12 of 14 on fisetin reported other adverse events versus 5 of 30 controls.

Participants / model
14 women aged 60 or older on fisetin (11 analyzed for CTX), versus 30 untreated controls
Treatment
Oral fisetin about 20 mg/kg/day for 3 days every 28 days over 20 weeks
Follow-up
20 weeks
Study design
Registry-posted results of a randomized open-label trial with a no-intervention control; not peer reviewed
Funding
Sponsor: Mayo Clinic; funding source not stated in the source

The control group got no placebo, so the adverse-event comparison is open-label. The published paper reports the dasatinib plus quercetin arm, and no publication of the fisetin arm was found.

Read the original source
PubMed search, 11 October 2026: 1 record, not a fisetin trial; no human senescent cell clearance

PubMed search record

PubMed search run on 11 October 2026: fisetin[tiab] AND (senescen*[tiab] OR senolytic*[tiab]) AND humans[mh] AND (randomized controlled trial[pt] OR clinical trial[pt]). 1 record returned.

The search returned 1 record, a small cell lung cancer chemotherapy-resistance paper that is not a fisetin trial. Among the registry results reviewed, the one senescence-related outcome with posted values, plasma SASP at 2 weeks in NCT04313634, showed no difference (median -0.9% on fisetin versus -2.5% for controls, p = 0.953 across arms). The carpal tunnel record lists secondary outcomes on senescence blood markers but posts no values. No record reviewed shows fisetin clearing senescent cells in people, and no record reports a hard aging outcome.

Participants / model
Records reviewed, not a population
Treatment
Not applicable
Follow-up
Search run on 11 October 2026
Study design
Literature search and registry results
Funding
not stated in the source

A search result shows what PubMed returned on that date. It is not proof that no unindexed study exists.

Read the original source

What risks do the studies report?

Adverse events were not higher than on placebo in the placebo-controlled trials with posted results, but one gastric hemorrhage occurred among 14 women in an open-label arm and 12 of 14 reported other adverse events. Long-term human safety is unknown, because all exposures found lasted 7 weeks to 6 months or were short courses.

In the placebo-controlled trials with posted results, adverse events were not higher than on placebo: treatment-emergent adverse events occurred in 28 of 34 on fisetin and 33 of 40 on placebo in knee osteoarthritis, with serious adverse events in 2 of 34 and 2 of 40, and serious adverse events were 0 of 29 versus 2 of 26 in COVFIS-HOME. The pharmacokinetic study reported no adverse events, and the 10-adult aging-clock pilot noted none.

In the open-label arm of NCT04313634, one serious adverse event, a gastric hemorrhage, occurred in the fisetin arm (1 of 14), and 12 of 14 on fisetin reported other adverse events versus 5 of 30 untreated controls. The control group got no placebo, so that comparison is open-label.

The nursing-home COVID-19 trial was stopped early due to futility, according to its registry limitations statement.

No oral animal toxicology study giving a no-observed-adverse-effect level was identified, and long-term human safety is unknown: all human exposures found were short courses or lasted 7 weeks to 6 months. Drug interactions were not searched. Results do not transfer between products, since unformulated fisetin and the formulated hydrogel differed 26.9-fold in exposure.

Knee osteoarthritis: 74 randomized, adverse events not higher than placebo, registry-posted results

Trial registry record

ClinicalTrials.gov record NCT04210986 (Steadman Philippon Research Institute, completed 2023). Results are posted by the sponsor and have not been through peer review.

The trial randomized 74 adults aged 40 to 80 (34 fisetin, 40 placebo). Treatment-emergent adverse events, the primary outcome, occurred in 28 of 34 on fisetin and 33 of 40 on placebo (Fisher exact odds ratio 0.99, 95% CI 0.25 to 4.02). Serious adverse events were 2 of 34 and 2 of 40. Among registered secondary outcomes, the 6-minute walk at month 12 was 544.2 m on fisetin and 543.6 m on placebo (p = 0.9572), and posted contrasts for C-reactive protein, the cartilage marker COMP, timed-up-and-go and quadriceps torque were not significant. 25 of 34 fisetin and 35 of 40 placebo participants completed. A PubMed search on 11 October 2026 (fisetin AND (osteoarthritis OR knee) AND (senolytic OR senescence) AND (trial OR randomized)) returned 3 records, none a report of this trial.

Participants / model
74 randomized adults aged 40 to 80 with knee osteoarthritis
Treatment
Oral fisetin about 20 mg/kg for 2 consecutive days, 28 days off, then 2 more consecutive days, or placebo
Follow-up
Two short courses 28 days apart; outcomes through month 12
Study design
Registry-posted results of a randomized placebo-controlled trial; not peer reviewed
Funding
Sponsor: Steadman Philippon Research Institute; funding source not stated in the source

Registry results are sponsor-posted and have not been through peer review. No peer-reviewed report of this trial was found in PubMed on 11 October 2026.

Read the original source
COVID-19: 55 outpatients, and 20 nursing-home residents with the trial stopped for futility

Trial registry record

ClinicalTrials.gov records NCT04537299 (COVID-FIS) and NCT04771611 (COVFIS-HOME). Results are posted by the sponsor and have not been through peer review.

COVFIS-HOME (NCT04771611, sponsored by investigator James L. Kirkland of Mayo Clinic) randomized 55 adults, 29 to fisetin and 26 to placebo, to about 20 mg/kg/day for two days at days 0 to 1 and 8 to 9. At 60 days 26 of 27 analyzed on fisetin and 22 of 24 on placebo had no limitations on the WHO ordinal scale, and serious adverse events were 0 of 29 versus 2 of 26. COVID-FIS in nursing-home residents (NCT04537299) enrolled 20, 10 per arm, and was stopped early: the registry limitations statement says the DSMB and NIA determined the study should be stopped due to futility, and the posted severity scores were nearly identical in both arms (a mean of 2 with SD 0 at most time points). A PubMed search on 11 October 2026 (fisetin[tiab] AND (COVID* OR SARS-CoV-2)[tiab] AND (randomized OR trial OR results)[tiab]) returned 11 records; the only trial paper was a 2021 design paper (PMID 34375437).

Participants / model
55 adults with COVID-19 (age 18 or older); 20 older nursing-home residents with COVID-19
Treatment
Oral fisetin about 20 mg/kg/day for 2 days, repeated at days 8 to 9, or placebo
Follow-up
60 days (COVFIS-HOME); stopped early (COVID-FIS)
Study design
Registry-posted results of two randomized placebo-controlled pilots; not peer reviewed
Funding
COVFIS-HOME is sponsored by investigator James L. Kirkland of Mayo Clinic; funding source not stated in the source for either trial

Registry results are sponsor-posted and have not been through peer review, and neither trial has a peer-reviewed report in PubMed as of 11 October 2026. One trial was stopped for futility.

Read the original source
Older women: 14 on fisetin versus 30 untreated, open-label, registry-posted results

Trial registry record

ClinicalTrials.gov record NCT04313634 (Mayo Clinic). Results are posted by the sponsor and have not been through peer review. The dasatinib plus quercetin arm is published separately (PMID 38956196).

The trial randomized 74 women aged 60 or older to dasatinib plus quercetin (30), fisetin (14, about 20 mg/kg/day for three consecutive days, every 28 days over 20 weeks), or no intervention (30). The CTX bone-resorption outcome at 20 weeks was a median change of +13.7% on fisetin, -4.1% on dasatinib plus quercetin and -7.7% for controls; the single posted test (p = 0.611) matches the dasatinib plus quercetin versus control comparison in the published paper, and no separate test for fisetin is posted. Plasma SASP (senescence-associated secretory phenotype) at 2 weeks changed a median -0.9% on fisetin versus -2.5% for controls (p = 0.953 across arms). One serious adverse event, a gastric hemorrhage, occurred in the fisetin arm (1 of 14), and 12 of 14 on fisetin reported other adverse events versus 5 of 30 controls.

Participants / model
14 women aged 60 or older on fisetin (11 analyzed for CTX), versus 30 untreated controls
Treatment
Oral fisetin about 20 mg/kg/day for 3 days every 28 days over 20 weeks
Follow-up
20 weeks
Study design
Registry-posted results of a randomized open-label trial with a no-intervention control; not peer reviewed
Funding
Sponsor: Mayo Clinic; funding source not stated in the source

The control group got no placebo, so the adverse-event comparison is open-label. The published paper reports the dasatinib plus quercetin arm, and no publication of the fisetin arm was found.

Read the original source
First human PK study: 15 volunteers, 26.9-fold AUC difference between formulations

Human study

Krishnakumar IM et al. J Nutr Sci. 2022;11:e74. PMID 36304817.

In a double-blind crossover study, a single 1000 mg dose of unformulated fisetin reached a plasma Cmax of 9.97 ng/ml (AUC 0 to 12 hours, 12.67), against 238.2 ng/ml (AUC 341.4) for 1000 mg of the hydrogel formulation FF-20, which delivers 192 mg fisetin. That is a 26.9-fold AUC difference and a Cmax more than 23 times higher. The authors call this the first human pharmacokinetic study of fisetin and describe unformulated fisetin as poorly bioavailable. No adverse events were reported.

Participants / model
15 healthy volunteers
Treatment
Single oral dose of 1000 mg unformulated fisetin or 1000 mg FF-20 (192 mg fisetin)
Follow-up
Single dose; exposure measured over 0 to 12 hours
Study design
Double-blind randomized crossover pharmacokinetic study, formulation comparison
Funding
not stated in the source

Plasma fisetin and the metabolite geraldol were measured. The abstract gives no units for AUC and does not state funding or conflicts, and no half-life or Tmax appears in the sources reviewed.

Read the original source
Aging-clock pilots: 10 adults on fisetin (5 of 10 had a higher biological age) and 19 on a combined regimen

Human study

Lee E et al. Altern Ther Health Med. 2024;30(9):6-10. PMID 39269340.

In 10 healthy adults over 50 taking 500 mg fisetin daily for one week per month for six months, biological age by the TruAge test was lower in 4 of 10, higher in 5 of 10 and unchanged in 1 of 10, telomere length did not change significantly, and no adverse effects were noted. The authors write that fisetin is not recommended as an anti-aging agent until more extensive studies are done. The abstract describes no placebo group. A separate longitudinal pilot of 19 participants given dasatinib and quercetin plus fisetin for 6 months (10 had been in an earlier dasatinib plus quercetin study) found non-significant increases in epigenetic age acceleration.

Participants / model
10 healthy adults over 50; 19 participants on dasatinib and quercetin plus fisetin (health status not stated in the abstract)
Treatment
Oral fisetin 500 mg daily, 1 week a month for 6 months; dasatinib and quercetin plus fisetin for 6 months
Follow-up
6 months in each pilot
Study design
Two small uncontrolled pilot studies
Funding
not stated in the source; several authors of the 19-participant pilot are TruDiagnostic employees

These are biomarker clocks in small open studies, not aging outcomes.

Read the original source
Small randomized trials: Gulf War illness (21), colorectal cancer (37), obesity (44)

Randomized controlled trial

Hodgin KS et al. Int J Environ Res Public Health. 2021;18(5). PMID 33802381.

In 21 male veterans with Gulf War illness in a placebo-controlled pseudo-randomized crossover trial, fisetin did not reduce symptom severity at the lower (p = 0.504) or higher (p = 0.616) dose, with no conflict of interest declared. In 37 patients with colorectal cancer on chemotherapy, 100 mg fisetin (n = 18) or placebo (n = 19) for seven weeks lowered IL-8 versus placebo (p < 0.03), while hs-CRP and MMP-7 fell only within the fisetin group. In 44 obese men in a four-arm 12-week trial (fisetin 200 mg/day, training, both, or placebo), IL-6 fell in the fisetin-alone arm (p = 0.013) and TNF-alpha fell in every active arm, with the largest drops in glucose, insulin and HOMA-IR in the fisetin plus training arm; no conflict of interest was reported.

Participants / model
21 veterans with Gulf War illness; 37 patients with colorectal cancer on chemotherapy; 44 men with obesity
Treatment
Oral fisetin; doses as stated per study
Follow-up
Not stated in the source for the Gulf War trial; 7 weeks (colorectal cancer); 12 weeks (obesity)
Study design
Three small randomized trials; the Gulf War trial was pseudo-randomized and a crossover
Funding
No conflict of interest was declared in the Gulf War and obesity trials; funding is not stated in the source

These are surrogate markers in disease or obesity populations, not senolytic or aging outcomes.

Read the original source

Why does fisetin sit in C tier?

C fits a defined molecule with real mouse senolytic and lifespan data, an exposure problem unless it is formulated, and human evidence that is small and mostly null or biomarker-only. No human record shows senescent cell clearance.

The mouse lifespan and senolytic data are real, and the placebo-controlled human studies found adverse events no higher than on placebo.

The human evidence is small and mostly null or biomarker-only: a posted knee osteoarthritis trial with no secondary-outcome differences, a COVID-19 nursing-home trial stopped for futility, an uncontrolled pilot in which 5 of 10 people had a higher biological age, and no human record showing senescent cell clearance. The posted results are sponsor-reported and not peer reviewed.

The grade covers fisetin as sold for senolytic and healthy-aging claims. Two Mayo frailty trials have no results, and a Denmark healthy-aging trial lists primary completion as 31 October 2026.

Origin of the claim: cells, mice and tissue explants, no living humans

Animal study

Yousefzadeh MJ et al. EBioMedicine. 2018;36:18-28. PMID 30279143.

The authors screened 10 flavonoids for senolytic activity in senescent murine and human fibroblasts and found fisetin the most potent. Acute or intermittent fisetin lowered senescence markers in multiple tissues of progeroid and old mice, lowered senescence in a subset of cells in murine and human adipose tissue explants, and, given to wild-type mice late in life, restored tissue homeostasis, reduced age-related pathology, and extended median and maximum lifespan. The authors say this suggests human trials are feasible.

Participants / model
Cultured cells, progeroid and old wild-type mice, and human adipose tissue explants; no living humans
Treatment
Fisetin given to mice (acute or intermittent), with cell and explant exposure; route and dose not stated in the source
Follow-up
Acute or intermittent exposure; schedule not stated in the source
Study design
Animal and ex vivo study
Funding
not stated in the source

Cell, mouse and explant findings only. They cannot be read as an effect in people.

Read the original source
First human PK study: 15 volunteers, 26.9-fold AUC difference between formulations

Human study

Krishnakumar IM et al. J Nutr Sci. 2022;11:e74. PMID 36304817.

In a double-blind crossover study, a single 1000 mg dose of unformulated fisetin reached a plasma Cmax of 9.97 ng/ml (AUC 0 to 12 hours, 12.67), against 238.2 ng/ml (AUC 341.4) for 1000 mg of the hydrogel formulation FF-20, which delivers 192 mg fisetin. That is a 26.9-fold AUC difference and a Cmax more than 23 times higher. The authors call this the first human pharmacokinetic study of fisetin and describe unformulated fisetin as poorly bioavailable. No adverse events were reported.

Participants / model
15 healthy volunteers
Treatment
Single oral dose of 1000 mg unformulated fisetin or 1000 mg FF-20 (192 mg fisetin)
Follow-up
Single dose; exposure measured over 0 to 12 hours
Study design
Double-blind randomized crossover pharmacokinetic study, formulation comparison
Funding
not stated in the source

Plasma fisetin and the metabolite geraldol were measured. The abstract gives no units for AUC and does not state funding or conflicts, and no half-life or Tmax appears in the sources reviewed.

Read the original source
Knee osteoarthritis: 74 randomized, adverse events not higher than placebo, registry-posted results

Trial registry record

ClinicalTrials.gov record NCT04210986 (Steadman Philippon Research Institute, completed 2023). Results are posted by the sponsor and have not been through peer review.

The trial randomized 74 adults aged 40 to 80 (34 fisetin, 40 placebo). Treatment-emergent adverse events, the primary outcome, occurred in 28 of 34 on fisetin and 33 of 40 on placebo (Fisher exact odds ratio 0.99, 95% CI 0.25 to 4.02). Serious adverse events were 2 of 34 and 2 of 40. Among registered secondary outcomes, the 6-minute walk at month 12 was 544.2 m on fisetin and 543.6 m on placebo (p = 0.9572), and posted contrasts for C-reactive protein, the cartilage marker COMP, timed-up-and-go and quadriceps torque were not significant. 25 of 34 fisetin and 35 of 40 placebo participants completed. A PubMed search on 11 October 2026 (fisetin AND (osteoarthritis OR knee) AND (senolytic OR senescence) AND (trial OR randomized)) returned 3 records, none a report of this trial.

Participants / model
74 randomized adults aged 40 to 80 with knee osteoarthritis
Treatment
Oral fisetin about 20 mg/kg for 2 consecutive days, 28 days off, then 2 more consecutive days, or placebo
Follow-up
Two short courses 28 days apart; outcomes through month 12
Study design
Registry-posted results of a randomized placebo-controlled trial; not peer reviewed
Funding
Sponsor: Steadman Philippon Research Institute; funding source not stated in the source

Registry results are sponsor-posted and have not been through peer review. No peer-reviewed report of this trial was found in PubMed on 11 October 2026.

Read the original source
COVID-19: 55 outpatients, and 20 nursing-home residents with the trial stopped for futility

Trial registry record

ClinicalTrials.gov records NCT04537299 (COVID-FIS) and NCT04771611 (COVFIS-HOME). Results are posted by the sponsor and have not been through peer review.

COVFIS-HOME (NCT04771611, sponsored by investigator James L. Kirkland of Mayo Clinic) randomized 55 adults, 29 to fisetin and 26 to placebo, to about 20 mg/kg/day for two days at days 0 to 1 and 8 to 9. At 60 days 26 of 27 analyzed on fisetin and 22 of 24 on placebo had no limitations on the WHO ordinal scale, and serious adverse events were 0 of 29 versus 2 of 26. COVID-FIS in nursing-home residents (NCT04537299) enrolled 20, 10 per arm, and was stopped early: the registry limitations statement says the DSMB and NIA determined the study should be stopped due to futility, and the posted severity scores were nearly identical in both arms (a mean of 2 with SD 0 at most time points). A PubMed search on 11 October 2026 (fisetin[tiab] AND (COVID* OR SARS-CoV-2)[tiab] AND (randomized OR trial OR results)[tiab]) returned 11 records; the only trial paper was a 2021 design paper (PMID 34375437).

Participants / model
55 adults with COVID-19 (age 18 or older); 20 older nursing-home residents with COVID-19
Treatment
Oral fisetin about 20 mg/kg/day for 2 days, repeated at days 8 to 9, or placebo
Follow-up
60 days (COVFIS-HOME); stopped early (COVID-FIS)
Study design
Registry-posted results of two randomized placebo-controlled pilots; not peer reviewed
Funding
COVFIS-HOME is sponsored by investigator James L. Kirkland of Mayo Clinic; funding source not stated in the source for either trial

Registry results are sponsor-posted and have not been through peer review, and neither trial has a peer-reviewed report in PubMed as of 11 October 2026. One trial was stopped for futility.

Read the original source
Aging-clock pilots: 10 adults on fisetin (5 of 10 had a higher biological age) and 19 on a combined regimen

Human study

Lee E et al. Altern Ther Health Med. 2024;30(9):6-10. PMID 39269340.

In 10 healthy adults over 50 taking 500 mg fisetin daily for one week per month for six months, biological age by the TruAge test was lower in 4 of 10, higher in 5 of 10 and unchanged in 1 of 10, telomere length did not change significantly, and no adverse effects were noted. The authors write that fisetin is not recommended as an anti-aging agent until more extensive studies are done. The abstract describes no placebo group. A separate longitudinal pilot of 19 participants given dasatinib and quercetin plus fisetin for 6 months (10 had been in an earlier dasatinib plus quercetin study) found non-significant increases in epigenetic age acceleration.

Participants / model
10 healthy adults over 50; 19 participants on dasatinib and quercetin plus fisetin (health status not stated in the abstract)
Treatment
Oral fisetin 500 mg daily, 1 week a month for 6 months; dasatinib and quercetin plus fisetin for 6 months
Follow-up
6 months in each pilot
Study design
Two small uncontrolled pilot studies
Funding
not stated in the source; several authors of the 19-participant pilot are TruDiagnostic employees

These are biomarker clocks in small open studies, not aging outcomes.

Read the original source
PubMed search, 11 October 2026: 1 record, not a fisetin trial; no human senescent cell clearance

PubMed search record

PubMed search run on 11 October 2026: fisetin[tiab] AND (senescen*[tiab] OR senolytic*[tiab]) AND humans[mh] AND (randomized controlled trial[pt] OR clinical trial[pt]). 1 record returned.

The search returned 1 record, a small cell lung cancer chemotherapy-resistance paper that is not a fisetin trial. Among the registry results reviewed, the one senescence-related outcome with posted values, plasma SASP at 2 weeks in NCT04313634, showed no difference (median -0.9% on fisetin versus -2.5% for controls, p = 0.953 across arms). The carpal tunnel record lists secondary outcomes on senescence blood markers but posts no values. No record reviewed shows fisetin clearing senescent cells in people, and no record reports a hard aging outcome.

Participants / model
Records reviewed, not a population
Treatment
Not applicable
Follow-up
Search run on 11 October 2026
Study design
Literature search and registry results
Funding
not stated in the source

A search result shows what PubMed returned on that date. It is not proof that no unindexed study exists.

Read the original source
Registry search, 11 October 2026: 36 studies, 6 with posted results

Trial registry record

ClinicalTrials.gov search with fisetin as an intervention, run on 11 October 2026. 36 studies returned. A 2026 review (PMID 42796982) counted 34 registered trials.

A ClinicalTrials.gov search run on 11 October 2026 with fisetin as an intervention returned 36 studies: 10 recruiting, 5 active not recruiting, 2 enrolling by invitation, 1 not yet recruiting, 7 completed, 2 terminated, 2 suspended, 4 withdrawn and 3 unknown status. Six have posted results: the knee osteoarthritis, older women and two COVID-19 records, plus a Gulf War record (NCT02909686) and a carpal tunnel record (NCT05416515; single-arm, 40 participants, mean BCTQ change of -0.5). A 2026 review (PMID 42796982) counted 34 registered trials and 4 completed trials with available results and judged the clinical evidence limited and heterogeneous. The two Mayo frailty trials, NCT03430037 (older women) and NCT03675724 (older adults), were first posted in 2018, are listed as enrolling by invitation with primary completion in November 2026, and have no results. NCT07195318 (Denmark, 100 mg daily, an estimated 120 adults aged 50 or older, primary outcome suPAR, primary completion listed as 31 October 2026) is the registered healthy-aging trial, and its design paper is PMID 42633989.

Participants / model
Registered trials, not a population
Treatment
Not applicable
Follow-up
Search run on 11 October 2026
Study design
Trial-registry search plus a review
Funding
not stated in the source

Registered studies without results show that a study was registered, not that it works. Registry results are sponsor-posted and not peer reviewed.

Read the original source

Studies and sources

PubChem: fisetin identity; no FDA drug or label found, 11 October 2026

Chemical identity record

National Library of Medicine. PubChem compound record CID 5281614 (fisetin). openFDA Drugs@FDA and DailyMed drug-name searches for fisetin run on 11 October 2026.

Fisetin: PubChem CID 5281614, C15H10O6, 286.24 g/mol, CAS 528-48-3, with 5-Desoxyquercetin and 3,3',4',7-Tetrahydroxyflavone among the listed synonyms. It is a plant flavonol and a single defined molecule, but the products differ: the human pharmacokinetic study used unformulated fisetin and a hydrogel formulation called FF-20, one registered trial names a branded capsule (Novusetin), the Gulf War trial used fisetin sourced from Rhus succedanea, and registered trial arms use either 100 mg capsules or about 20 mg/kg/day short courses. On 11 October 2026 an openFDA Drugs@FDA query for fisetin returned no match and a DailyMed drug-name search returned no labels.

Participants / model
Chemical identity, not a clinical population
Treatment
Defined molecule; several product forms
Follow-up
Searches run on 11 October 2026
Study design
NLM chemical record plus dated US approval-record search
Funding
not stated in the source

No trial result transfers to a different product or dose. FDA dietary-ingredient or GRAS status is not established here.

Read the original source
Origin of the claim: cells, mice and tissue explants, no living humans

Animal study

Yousefzadeh MJ et al. EBioMedicine. 2018;36:18-28. PMID 30279143.

The authors screened 10 flavonoids for senolytic activity in senescent murine and human fibroblasts and found fisetin the most potent. Acute or intermittent fisetin lowered senescence markers in multiple tissues of progeroid and old mice, lowered senescence in a subset of cells in murine and human adipose tissue explants, and, given to wild-type mice late in life, restored tissue homeostasis, reduced age-related pathology, and extended median and maximum lifespan. The authors say this suggests human trials are feasible.

Participants / model
Cultured cells, progeroid and old wild-type mice, and human adipose tissue explants; no living humans
Treatment
Fisetin given to mice (acute or intermittent), with cell and explant exposure; route and dose not stated in the source
Follow-up
Acute or intermittent exposure; schedule not stated in the source
Study design
Animal and ex vivo study
Funding
not stated in the source

Cell, mouse and explant findings only. They cannot be read as an effect in people.

Read the original source
First human PK study: 15 volunteers, 26.9-fold AUC difference between formulations

Human study

Krishnakumar IM et al. J Nutr Sci. 2022;11:e74. PMID 36304817.

In a double-blind crossover study, a single 1000 mg dose of unformulated fisetin reached a plasma Cmax of 9.97 ng/ml (AUC 0 to 12 hours, 12.67), against 238.2 ng/ml (AUC 341.4) for 1000 mg of the hydrogel formulation FF-20, which delivers 192 mg fisetin. That is a 26.9-fold AUC difference and a Cmax more than 23 times higher. The authors call this the first human pharmacokinetic study of fisetin and describe unformulated fisetin as poorly bioavailable. No adverse events were reported.

Participants / model
15 healthy volunteers
Treatment
Single oral dose of 1000 mg unformulated fisetin or 1000 mg FF-20 (192 mg fisetin)
Follow-up
Single dose; exposure measured over 0 to 12 hours
Study design
Double-blind randomized crossover pharmacokinetic study, formulation comparison
Funding
not stated in the source

Plasma fisetin and the metabolite geraldol were measured. The abstract gives no units for AUC and does not state funding or conflicts, and no half-life or Tmax appears in the sources reviewed.

Read the original source
Knee osteoarthritis: 74 randomized, adverse events not higher than placebo, registry-posted results

Trial registry record

ClinicalTrials.gov record NCT04210986 (Steadman Philippon Research Institute, completed 2023). Results are posted by the sponsor and have not been through peer review.

The trial randomized 74 adults aged 40 to 80 (34 fisetin, 40 placebo). Treatment-emergent adverse events, the primary outcome, occurred in 28 of 34 on fisetin and 33 of 40 on placebo (Fisher exact odds ratio 0.99, 95% CI 0.25 to 4.02). Serious adverse events were 2 of 34 and 2 of 40. Among registered secondary outcomes, the 6-minute walk at month 12 was 544.2 m on fisetin and 543.6 m on placebo (p = 0.9572), and posted contrasts for C-reactive protein, the cartilage marker COMP, timed-up-and-go and quadriceps torque were not significant. 25 of 34 fisetin and 35 of 40 placebo participants completed. A PubMed search on 11 October 2026 (fisetin AND (osteoarthritis OR knee) AND (senolytic OR senescence) AND (trial OR randomized)) returned 3 records, none a report of this trial.

Participants / model
74 randomized adults aged 40 to 80 with knee osteoarthritis
Treatment
Oral fisetin about 20 mg/kg for 2 consecutive days, 28 days off, then 2 more consecutive days, or placebo
Follow-up
Two short courses 28 days apart; outcomes through month 12
Study design
Registry-posted results of a randomized placebo-controlled trial; not peer reviewed
Funding
Sponsor: Steadman Philippon Research Institute; funding source not stated in the source

Registry results are sponsor-posted and have not been through peer review. No peer-reviewed report of this trial was found in PubMed on 11 October 2026.

Read the original source
Older women: 14 on fisetin versus 30 untreated, open-label, registry-posted results

Trial registry record

ClinicalTrials.gov record NCT04313634 (Mayo Clinic). Results are posted by the sponsor and have not been through peer review. The dasatinib plus quercetin arm is published separately (PMID 38956196).

The trial randomized 74 women aged 60 or older to dasatinib plus quercetin (30), fisetin (14, about 20 mg/kg/day for three consecutive days, every 28 days over 20 weeks), or no intervention (30). The CTX bone-resorption outcome at 20 weeks was a median change of +13.7% on fisetin, -4.1% on dasatinib plus quercetin and -7.7% for controls; the single posted test (p = 0.611) matches the dasatinib plus quercetin versus control comparison in the published paper, and no separate test for fisetin is posted. Plasma SASP (senescence-associated secretory phenotype) at 2 weeks changed a median -0.9% on fisetin versus -2.5% for controls (p = 0.953 across arms). One serious adverse event, a gastric hemorrhage, occurred in the fisetin arm (1 of 14), and 12 of 14 on fisetin reported other adverse events versus 5 of 30 controls.

Participants / model
14 women aged 60 or older on fisetin (11 analyzed for CTX), versus 30 untreated controls
Treatment
Oral fisetin about 20 mg/kg/day for 3 days every 28 days over 20 weeks
Follow-up
20 weeks
Study design
Registry-posted results of a randomized open-label trial with a no-intervention control; not peer reviewed
Funding
Sponsor: Mayo Clinic; funding source not stated in the source

The control group got no placebo, so the adverse-event comparison is open-label. The published paper reports the dasatinib plus quercetin arm, and no publication of the fisetin arm was found.

Read the original source
COVID-19: 55 outpatients, and 20 nursing-home residents with the trial stopped for futility

Trial registry record

ClinicalTrials.gov records NCT04537299 (COVID-FIS) and NCT04771611 (COVFIS-HOME). Results are posted by the sponsor and have not been through peer review.

COVFIS-HOME (NCT04771611, sponsored by investigator James L. Kirkland of Mayo Clinic) randomized 55 adults, 29 to fisetin and 26 to placebo, to about 20 mg/kg/day for two days at days 0 to 1 and 8 to 9. At 60 days 26 of 27 analyzed on fisetin and 22 of 24 on placebo had no limitations on the WHO ordinal scale, and serious adverse events were 0 of 29 versus 2 of 26. COVID-FIS in nursing-home residents (NCT04537299) enrolled 20, 10 per arm, and was stopped early: the registry limitations statement says the DSMB and NIA determined the study should be stopped due to futility, and the posted severity scores were nearly identical in both arms (a mean of 2 with SD 0 at most time points). A PubMed search on 11 October 2026 (fisetin[tiab] AND (COVID* OR SARS-CoV-2)[tiab] AND (randomized OR trial OR results)[tiab]) returned 11 records; the only trial paper was a 2021 design paper (PMID 34375437).

Participants / model
55 adults with COVID-19 (age 18 or older); 20 older nursing-home residents with COVID-19
Treatment
Oral fisetin about 20 mg/kg/day for 2 days, repeated at days 8 to 9, or placebo
Follow-up
60 days (COVFIS-HOME); stopped early (COVID-FIS)
Study design
Registry-posted results of two randomized placebo-controlled pilots; not peer reviewed
Funding
COVFIS-HOME is sponsored by investigator James L. Kirkland of Mayo Clinic; funding source not stated in the source for either trial

Registry results are sponsor-posted and have not been through peer review, and neither trial has a peer-reviewed report in PubMed as of 11 October 2026. One trial was stopped for futility.

Read the original source
Aging-clock pilots: 10 adults on fisetin (5 of 10 had a higher biological age) and 19 on a combined regimen

Human study

Lee E et al. Altern Ther Health Med. 2024;30(9):6-10. PMID 39269340.

In 10 healthy adults over 50 taking 500 mg fisetin daily for one week per month for six months, biological age by the TruAge test was lower in 4 of 10, higher in 5 of 10 and unchanged in 1 of 10, telomere length did not change significantly, and no adverse effects were noted. The authors write that fisetin is not recommended as an anti-aging agent until more extensive studies are done. The abstract describes no placebo group. A separate longitudinal pilot of 19 participants given dasatinib and quercetin plus fisetin for 6 months (10 had been in an earlier dasatinib plus quercetin study) found non-significant increases in epigenetic age acceleration.

Participants / model
10 healthy adults over 50; 19 participants on dasatinib and quercetin plus fisetin (health status not stated in the abstract)
Treatment
Oral fisetin 500 mg daily, 1 week a month for 6 months; dasatinib and quercetin plus fisetin for 6 months
Follow-up
6 months in each pilot
Study design
Two small uncontrolled pilot studies
Funding
not stated in the source; several authors of the 19-participant pilot are TruDiagnostic employees

These are biomarker clocks in small open studies, not aging outcomes.

Read the original source
Small randomized trials: Gulf War illness (21), colorectal cancer (37), obesity (44)

Randomized controlled trial

Hodgin KS et al. Int J Environ Res Public Health. 2021;18(5). PMID 33802381.

In 21 male veterans with Gulf War illness in a placebo-controlled pseudo-randomized crossover trial, fisetin did not reduce symptom severity at the lower (p = 0.504) or higher (p = 0.616) dose, with no conflict of interest declared. In 37 patients with colorectal cancer on chemotherapy, 100 mg fisetin (n = 18) or placebo (n = 19) for seven weeks lowered IL-8 versus placebo (p < 0.03), while hs-CRP and MMP-7 fell only within the fisetin group. In 44 obese men in a four-arm 12-week trial (fisetin 200 mg/day, training, both, or placebo), IL-6 fell in the fisetin-alone arm (p = 0.013) and TNF-alpha fell in every active arm, with the largest drops in glucose, insulin and HOMA-IR in the fisetin plus training arm; no conflict of interest was reported.

Participants / model
21 veterans with Gulf War illness; 37 patients with colorectal cancer on chemotherapy; 44 men with obesity
Treatment
Oral fisetin; doses as stated per study
Follow-up
Not stated in the source for the Gulf War trial; 7 weeks (colorectal cancer); 12 weeks (obesity)
Study design
Three small randomized trials; the Gulf War trial was pseudo-randomized and a crossover
Funding
No conflict of interest was declared in the Gulf War and obesity trials; funding is not stated in the source

These are surrogate markers in disease or obesity populations, not senolytic or aging outcomes.

Read the original source
Registry search, 11 October 2026: 36 studies, 6 with posted results

Trial registry record

ClinicalTrials.gov search with fisetin as an intervention, run on 11 October 2026. 36 studies returned. A 2026 review (PMID 42796982) counted 34 registered trials.

A ClinicalTrials.gov search run on 11 October 2026 with fisetin as an intervention returned 36 studies: 10 recruiting, 5 active not recruiting, 2 enrolling by invitation, 1 not yet recruiting, 7 completed, 2 terminated, 2 suspended, 4 withdrawn and 3 unknown status. Six have posted results: the knee osteoarthritis, older women and two COVID-19 records, plus a Gulf War record (NCT02909686) and a carpal tunnel record (NCT05416515; single-arm, 40 participants, mean BCTQ change of -0.5). A 2026 review (PMID 42796982) counted 34 registered trials and 4 completed trials with available results and judged the clinical evidence limited and heterogeneous. The two Mayo frailty trials, NCT03430037 (older women) and NCT03675724 (older adults), were first posted in 2018, are listed as enrolling by invitation with primary completion in November 2026, and have no results. NCT07195318 (Denmark, 100 mg daily, an estimated 120 adults aged 50 or older, primary outcome suPAR, primary completion listed as 31 October 2026) is the registered healthy-aging trial, and its design paper is PMID 42633989.

Participants / model
Registered trials, not a population
Treatment
Not applicable
Follow-up
Search run on 11 October 2026
Study design
Trial-registry search plus a review
Funding
not stated in the source

Registered studies without results show that a study was registered, not that it works. Registry results are sponsor-posted and not peer reviewed.

Read the original source

Why is Fisetin in C tier?

C fits a defined molecule with real mouse senolytic and lifespan data, a clear exposure problem unless it is formulated, and small human studies that are mostly null or biomarker-only. A posted knee osteoarthritis trial showed no secondary-outcome differences, a COVID-19 nursing-home trial was stopped for futility, and no human record shows senescent cell clearance. The grade covers the supplier's senolytic and healthy-aging claims.