Senolytics Cleared 86 Percent of the Damage in Eleven Days. Seven Years Later No Trial Has Found the Benefit.
Cellular senescence has the most legible story in the whole longevity field. Cells stop dividing, refuse to die, and start secreting inflammatory signals that damage the tissue around them. They accumulate with age. Remove them from a mouse and the mouse walks faster, lives longer, and looks younger on almost every measure researchers have tried. That mechanism is not speculative, and in 2019 a Mayo Clinic team showed the same drugs strip senescent cells out of human fat and skin after a single three-day course. Then the field ran the trials that were supposed to convert that biology into an outcome someone can feel. A phase 2 study in 60 postmenopausal women missed its primary endpoint. An Alzheimer's pilot found that one of the two drugs never crossed into the brain. A biotech company took a purpose-built senolytic into 183 patients with knee osteoarthritis and discontinued the program. The gap between what senolytics demonstrably do to cells and what they have been shown to do for people is the entire subject of this article.
Key Takeaways: What the Senolytic Evidence Actually Shows
Intermittent oral dasatinib plus quercetin given to naturally aged mice raised post-treatment survival by 36 percent and reduced the mortality hazard to 65 percent of control, reported in Nature Medicine in 2018 by Xu and colleagues. In the same study, transplanting a small number of senescent cells into young mice produced lasting physical dysfunction within two weeks, which is the strongest evidence that these cells cause aging damage rather than merely marking it. A nine-person open-label pilot in adults with diabetic kidney disease found that a single three-day course of 100 mg dasatinib and 1000 mg quercetin reduced crown-like structures in fat tissue by 86 percent at p equals 0.001, cells staining for senescence-associated beta-galactosidase by 62 percent, p16INK4A-positive cells by 35 percent in fat and 20 percent in skin, and circulating IL-1 alpha, IL-6, MMP-9 and MMP-12 all at p below 0.05. The largest randomized senolytic trial published to date gave the same drug pair to 60 postmenopausal women for three consecutive days every 28 days across 20 weeks, and its primary endpoint of CTx, a bone resorption marker, showed no difference from control. The bone formation marker P1NP did rise significantly at weeks 2 and 4, and by week 20 that difference was gone. Benefit concentrated in the women in the highest tertile for T cell p16 expression, who showed rising formation, falling resorption, and increased bone mineral density at the wrist. In the SToMP-AD pilot, five adults with early Alzheimer's took intermittent dasatinib and quercetin for 12 weeks, dasatinib appeared in cerebrospinal fluid in four of them at a CSF-to-plasma ratio under 1 percent, and quercetin was not detectable in cerebrospinal fluid in anyone. UNITY Biotechnology's UBX0101, a senolytic designed for joint tissue, failed to beat placebo in a phase 2 knee osteoarthritis trial and was discontinued in August 2020. No senolytic trial has ever enrolled healthy middle-aged men as its study population, and no senolytic has been shown to extend lifespan in any human.
What Are Senolytics and What Exactly Do They Remove?
Senolytics are drugs that selectively kill senescent cells, which are cells that have permanently stopped dividing but refuse to undergo normal cell death. A senescent cell is not dormant. It secretes a mix of inflammatory cytokines, growth factors and matrix-degrading enzymes collectively called the senescence-associated secretory phenotype, or SASP, and that secretion is what makes the cell a problem rather than a curiosity. Interleukin-6, interleukin-1 alpha and the matrix metalloproteinases MMP-9 and MMP-12 are the components most often measured. Senescent cells accumulate in fat, skin, bone, joint tissue and blood vessels as people age, and the standard laboratory markers for identifying them are the proteins p16INK4A and p21CIP1 plus an enzyme assay called senescence-associated beta-galactosidase. These cells survive because they upregulate anti-apoptotic pathways, principally the BCL-2 family and certain tyrosine kinase signals, which is precisely the vulnerability senolytics exploit. Dasatinib is an approved tyrosine kinase inhibitor used to treat chronic myeloid leukemia. Quercetin and fisetin are plant flavonoids sold as dietary supplements. The two are given together in most protocols because dasatinib preferentially kills senescent fat cell progenitors while quercetin preferentially kills senescent endothelial cells, and neither covers the full range alone.
How Strong Is the Animal Evidence for Clearing Senescent Cells?
Unusually strong, and stronger on causation than most longevity interventions can claim. The pivotal experiment came in 2018 when Xu and colleagues published in Nature Medicine that transplanting a relatively small number of senescent cells into young healthy mice produced persistent physical dysfunction, measurable within two weeks as reduced maximum walking speed, lower muscle strength, less endurance and reduced daily activity. The transplanted cells also spread senescence into surrounding host tissue. That result reversed the usual problem in aging research, where a marker rises with age and nobody can say whether it is a cause or a consequence. In the same paper, intermittent oral dasatinib plus quercetin given to naturally aged mice raised post-treatment survival by 36 percent and cut the mortality hazard to 65 percent of the control rate, and it worked even when started very late in life. Separately, a 2018 report in EBioMedicine identified fisetin as the most potent senolytic among a panel of flavonoids and found it extended median lifespan in mice by roughly 10 percent when begun in old age. Earlier work using genetically engineered mice, the INK-ATTAC line, had already shown that switching on a suicide gene in p16-expressing cells delayed the onset of age-related decline in multiple tissues.
What Removing Senescent Cells Did to Mice
Percent change in survival and lifespan measures from the animal work that started the field. In the 2018 Nature Medicine study, intermittent oral dasatinib plus quercetin given to naturally aged mice raised post-treatment survival by 36 percent and cut the mortality hazard to 65 percent of the control rate, a reduction of 35 percent. A separate 2018 report in EBioMedicine found fisetin extended median lifespan in mice by roughly 10 percent when started in late life. All three bars are mouse data. No senolytic has been shown to extend lifespan in any human.
Did Senolytics Actually Clear Senescent Cells in People?
Yes, and this is the finding the field is built on. Hickson and colleagues reported in EBioMedicine in 2019 that nine adults with diabetic kidney disease, mean age 69, took 100 mg of dasatinib and 1000 mg of quercetin daily for three consecutive days, with fat and skin biopsies collected before treatment and again 11 days after the last dose. Crown-like structures in fat tissue, which are clusters of macrophages surrounding dying fat cells and a recognized marker of metabolic inflammation, fell 86 percent at p equals 0.001. Cells positive for senescence-associated beta-galactosidase fell 62 percent. Cells positive for p16INK4A fell 35 percent in fat and 20 percent in skin, and p21CIP1-positive cells fell 17 percent in fat and 31 percent in skin. CD68-positive macrophages dropped 28 percent. Circulating IL-1 alpha, IL-6, MMP-9 and MMP-12 were all significantly reduced. Two features of that design matter. The drug was gone from the body long before the biopsies were taken, meaning the effect is a hit-and-run clearance rather than a suppression that requires continuous exposure. And the trial had nine participants, no control group and no clinical endpoint, so what it establishes is target engagement, not benefit.
Three Days of Drug, Measured Eleven Days Later
Percent reduction in senescent cell markers in nine adults with diabetic kidney disease who took 100 mg dasatinib and 1000 mg quercetin daily for three consecutive days, with fat and skin biopsies taken before treatment and 11 days after the last dose. Crown-like structures in fat tissue fell 86 percent at p equals 0.001. Cells staining positive for p16INK4A fell 35 percent in fat and 20 percent in skin. Every reduction shown reached statistical significance. This is the single clearest demonstration that the drugs do in a human what they do in a mouse.
What Happened in the Bone Trial, the Largest Randomized Senolytic Study?
It missed its primary endpoint. Khosla's group at Mayo Clinic ran a phase 2 randomized controlled trial in 60 postmenopausal women, published in Nature Medicine in 2024, giving 100 mg dasatinib plus 1000 mg quercetin daily for three consecutive days repeated every 28 days, for five dosing cycles across 20 weeks. The primary endpoint was CTx, the C-terminal telopeptide of type 1 collagen, which measures how fast bone is being broken down. At 20 weeks there was no significant difference between the treated group and control. The bone formation marker P1NP did rise significantly in the treated group at weeks 2 and 4, then converged with control by week 20. That pattern is worth reading carefully, because it is consistent with a real biological effect that fades rather than with no effect at all. The most interesting result was a subgroup analysis. Women in the highest tertile for p16 expression in T cells, meaning the highest measurable senescent cell burden at baseline, showed larger increases in bone formation, decreases in bone resorption, and an increase in bone mineral density at the radius by 20 weeks. No serious adverse events occurred. The trial is a miss on its own terms and simultaneously the best evidence available that senolytics do something in humans when there is something for them to clear.
The Bone Formation Signal Arrived Early and Was Gone by Week 20
A schematic reading of bone turnover markers across the phase 2 randomized trial of intermittent dasatinib plus quercetin in 60 postmenopausal women, published in Nature Medicine in 2024. P1NP, a marker of bone formation, rose significantly in the treated group at weeks 2 and 4 and was no different from control by week 20. CTx, the bone resorption marker that served as the primary endpoint, never separated from control at any timepoint. Values are relative and directional rather than exact reported percentages, so read the shape rather than the height.
- P1NP, bone formation
- CTx, bone resorption (primary)
Did Senolytics Do Anything for Lung Fibrosis or Alzheimer's Disease?
Neither trial was designed to answer that, and neither produced a clinical result worth citing as one. The idiopathic pulmonary fibrosis study, published in EBioMedicine in 2019, gave 14 participants dasatinib at 100 mg per day and quercetin at 1250 mg per day for three days a week across three weeks, nine dosing days in total. Its stated primary endpoints were retention and completion rates, which is to say it was a feasibility study. Every participant completed the regimen. The most consistent secondary changes appeared in mobility measures including six-minute walk distance, walking speed and sit-to-stand repetitions, but with 14 people, no placebo arm and physical function as a secondary outcome, those changes carry very little weight. The SToMP-AD pilot, published by Gonzales and colleagues in 2023, gave intermittent dasatinib and quercetin to five adults with early-stage Alzheimer's disease for 12 weeks, with cerebrospinal fluid drawn at baseline and at the end. Its primary objective was to determine whether the drugs reach the central nervous system. Dasatinib was detected in four of five participants at concentrations between 0.281 and 0.536 ng/ml, a CSF-to-plasma ratio between 0.42 and 0.92 percent. Quercetin was not detected in cerebrospinal fluid in any participant. No changes occurred in amyloid beta, tau, senescence markers or cognition over 12 weeks. That is a useful negative result: for any brain-directed application, the quercetin half of the standard protocol appears not to arrive.
Which Senolytic Trials Hit What They Set Out to Measure
Every published human senolytic trial arranged by whether it met its own stated primary objective. Emerald marks trials that hit, amber marks a partial or delayed hit, grey marks a miss. The two studies that succeeded set biological or feasibility objectives rather than clinical ones. The bone phase 2 missed on CTx, the SToMP-AD pilot found no quercetin in cerebrospinal fluid at all, and UBX0101 failed against placebo in 183 knee osteoarthritis patients. Bar length is a schematic weight, not a shared unit.
Why Did UNITY Biotechnology's Purpose-Built Senolytic Fail?
UBX0101 failed because clearing senescent cells from an arthritic knee did not reduce pain more than an injection of saline. UNITY Biotechnology was founded specifically to commercialize senolytics and took UBX0101, a small molecule targeting the MDM2-p53 interaction, into a phase 2 trial in patients with painful knee osteoarthritis. In August 2020 the company reported that a single injection failed to show a statistically significant difference from placebo at 12 weeks, and the program was discontinued. That is the most informative failure in the field, because the drug was designed for the job, delivered directly into the affected joint, and tested against a symptom patients report reliably. The company's current lead is UBX1325, also called foselutoclax, an inhibitor of the anti-apoptotic protein BCL-xL delivered by intravitreal injection for diabetic macular edema. In the phase 2b ASPIRE study it did not meet statistical non-inferiority against aflibercept at weeks 20 and 24, then reached non-inferiority on interim 36-week data, with no cases of intraocular inflammation or retinal artery occlusion reported. Read across the whole portfolio, the pattern is that senolytics have never yet beaten an active comparator or a placebo on a hard clinical endpoint in a properly powered trial.
What About Fisetin, the Version People Actually Buy?
Fisetin has the weakest human evidence of the three compounds and the widest consumer use, because it is the only one available without a prescription that is also marketed as a senolytic. It is a flavonoid found in strawberries, apples and persimmons, and it drew attention from the 2018 EBioMedicine screen in which it outperformed quercetin and other flavonoids in killing senescent cells and extended mouse median lifespan by around 10 percent. Human trials have been registered repeatedly and have delivered very little. A Mayo Clinic study of older women with gait disturbance was originally designed with both a fisetin arm and a dasatinib plus quercetin arm, and the fisetin arm was dropped during the COVID-19 pandemic because of recruitment difficulty and cost, leaving the two-arm design that eventually reported. A phase 2 study is testing fisetin alone against dasatinib plus quercetin for frailty in adult survivors of childhood cancer, and a trial in peripheral artery disease is recruiting. As of now there is no published randomized trial showing that oral fisetin reduces senescent cell burden in humans or improves any clinical measure. The doses used in research protocols, typically 20 mg per kilogram of body weight for two or three consecutive days, are far above the amounts in most retail capsules, and oral fisetin has poor bioavailability that the supplement industry addresses with liposomal formulations of unverified absorption.
What Doses Are Used in Research and What Is Known About Safety?
Research protocols converge on 100 mg of dasatinib plus 1000 to 1250 mg of quercetin taken for two or three consecutive days, repeated at intervals of one to four weeks, and the published safety record across those studies is clean at that exposure. The intermittent hit-and-run schedule is deliberate. Senescent cells take weeks to accumulate, so a short pulse that kills them is expected to work without the continuous exposure that drives most drug toxicity. The phase 2 bone trial reported no serious adverse events across 20 weeks and five dosing cycles. The IPF pilot recorded no discontinuations. The clearest limitation is not toxicity in these trials but who has been studied. Every participant enrolled so far has had an existing disease or has been over 60, usually both, and the majority in most of these studies have been women. Dasatinib itself is a chemotherapy agent approved for chronic myeloid leukemia, where it is taken daily and carries documented risks including pleural effusion, cytopenias, bleeding, QT prolongation and pulmonary arterial hypertension. Those risks are established at continuous therapeutic dosing rather than at three days every four weeks, but they are the reason dasatinib is a prescription drug and the reason intermittent senolytic dosing outside a trial is not a supplement decision. Quercetin at gram doses inhibits several cytochrome P450 enzymes and can raise blood levels of other medications.
How Should the Senolytic Evidence Be Read Right Now?
Senolytics have the cleanest mechanism in longevity biology and almost no clinical evidence, and both of those are true at once. The causal case is genuinely strong: senescent cells transplanted into young mice cause dysfunction, clearing them in aged mice raises survival by 36 percent, and a three-day course in humans measurably strips those cells out of fat and skin with circulating inflammatory markers falling alongside. Very few interventions in this category can point to demonstrated target engagement in human tissue. Against that, the scoreboard on outcomes reads one missed primary endpoint in the only sizeable randomized trial, one discontinued program in knee osteoarthritis, one pilot showing that half the standard protocol never reaches the brain, and nothing at all in healthy people. Three things follow for a man weighing this. The subgroup finding in the bone trial suggests senolytics may do more for people with a high baseline senescent burden, which points toward older and sicker rather than younger and healthier, and there is currently no way to measure that burden outside a research laboratory. The bone formation signal that appeared at week 2 and vanished by week 20 raises an unresolved question about whether repeated dosing produces diminishing returns. And the compound most people can actually buy, fisetin, is the one with no published human clearance data at all. The trials that would change this picture are running now, including the frailty study in childhood cancer survivors and the peripheral artery disease trial, and they are the first senolytic studies powered to measure function rather than feasibility.
The Senolytic Record on One Axis
Every strand of the senolytic evidence arranged as a single schematic map. Emerald marks what the animal and human data support, amber marks findings that are real but narrower than they sound, and grey marks claims the evidence does not carry. The biology sits at the strong end, and it is unusually well established for this field. The clinical end is nearly empty, and the emptiest cell of all is healthy middle-aged men, who have never been enrolled in a senolytic trial as a study population.
Frequently Asked Questions
Have more questions? Contact us or explore our full peptide intelligence platform.
Related Articles
Does Telomere Length Determine Lifespan? Short Telomeres Track With Earlier Death. The Genetics Say Longer Ones Raise Cancer Risk.
Does telomere length determine lifespan? Short telomeres track with earlier death, yet genetic studies show longer ones raise the risk of several cancers.
Low-Dose Aspirin Added 8 Percent to Male Mouse Lifespan. In 19,114 Healthy Older Adults, It Added No Healthy Years and Deaths Ran Higher.
Does low-dose aspirin help longevity? It extended male mouse lifespan 8%, but the ASPREE trial in 19,114 healthy older adults found no gain and more bleeding.
Is NAD+ a Peptide? No. It Is a Coenzyme Built From Two Nucleotides, and the Peptide Label Comes From the Clinic, Not the Chemistry
Is NAD+ a peptide? No. NAD+ is a coenzyme built from two nucleotides. What NAD+ is used for, what IV NAD+ studies found, and how it differs from SS-31.