Senolytics Shift Cognition and Mobility in Alzheimer's-Risk Older Adults, STAMINA Pilot Finds
When gait slows and memory starts to slip, that combination carries a specific weight in Alzheimer’s research. These two signals appearing together mark an elevated risk threshold. A new pilot study, the first to test senolytic drugs directly in this group, has now generated human data on what clearing senescent cells might do for both the brain and the body.
Clearing Zombie Cells to Protect the Brain
Published in eBioMedicine (a Lancet journal) in February 2025, the STAMINA trial (Senolytics To Alleviate Mobility Issues and Neurological Impairments in Ageing) is the first clinical study to apply senolytic therapy to older adults at risk for Alzheimer’s disease. The research was led by the Hinda and Arthur Marcus Institute at Hebrew SeniorLife, funded by the U.S. National Institute on Aging.
Senescent cells, sometimes called “zombie cells,” stop dividing but don’t die. They persist in tissues, releasing a chronic stream of inflammatory molecules that degrade local function. As they accumulate with age, this low-grade inflammation becomes a background noise that undermines tissue health, including the brain. Removing them is the premise of senolytic therapy.
The Protocol: Two Days On, Twelve Days Off
Participants were 12 older adults (mean age 77) who met two specific criteria simultaneously: gait speed below 1 m/s and mild cognitive impairment (MCI). This combination is an established marker of elevated Alzheimer’s risk in aging research.
The dosing protocol was intermittent. Participants took 100 mg of dasatinib and 1,250 mg of quercetin on two consecutive days, then rested for 12 days, repeating this cycle six times over 12 weeks. Dasatinib, a cancer drug, clears senescent preadipocytes; quercetin targets senescent endothelial cells. Together, they cover a broader range of cell types than either agent alone, an approach validated in preclinical studies and earlier human work on diabetic kidney disease.
MoCA Scores and the TNF-α Connection
Across all participants, the Montreal Cognitive Assessment (MoCA) score increased by 1.0 point on average, a non-significant result. In the subgroup with the lowest baseline scores, where cognitive decline was most pronounced, the improvement reached 2.0 points and crossed statistical significance (p=0.04). A clinically meaningful change threshold is typically considered 1 to 2 points.
Inflammatory biomarkers shifted in a consistent direction. TNF-α decreased by 3.0% and IL-6 by 15.5%. The finding that gave the researchers greatest confidence was the correlation between TNF-α reduction and MoCA improvement (r=−0.65, p=0.02). The more inflammation came down, the better the cognitive scores moved. That correlation links mechanism to outcome in a way that a single-number result alone cannot.
On mobility, dual task gait speed improved by 0.03 m/s and Timed Up and Go time shortened by approximately one second, but neither reached significance in a sample this small.
Why Dual Task Gait Speed
The inclusion of dual task gait speed as a primary mobility measure was deliberate. Walking while performing a secondary cognitive task demands executive function. When the brain’s frontal processing is under strain, walking while simultaneously counting or talking slows down before other gait measures do. By tracking this specific metric, the STAMINA team was explicitly monitoring the cognition-mobility interface, not just physical function.
A Clean Safety Profile
No serious adverse events related to the intervention occurred over 12 weeks. Six instances of mildly reduced white blood cell counts were reported, all transient. Adherence was 99% for dasatinib and 100% for quercetin. For a drug combination that includes a cancer chemotherapy agent, this tolerability profile is notable.
Lead author Dr. Courtney L. Millar offered a calibrated read of the data: “Our findings suggest that senolytic treatment with Dasatinib and Quercetin is well-tolerated in older adults at risk for Alzheimer’s disease and may improve cognition by targeting the harmful effects of cellular senescence.” She immediately followed with the necessary caveat: “These results are promising, but this is a very small pilot study and we do not know if these findings occurred just by chance. We need additional research.”
The study’s central structural limitation is the absence of a control group. Single-arm, open-label designs cannot separate drug effects from placebo response or natural variation over 12 weeks.
The First Door Into Alzheimer’s Territory
What STAMINA opens is not a conclusion but a direction. Senolytic research to date has concentrated on joint health, kidney function, and physical frailty. This is the first time the therapy has been aimed at Alzheimer’s risk, and the first time human data has suggested that clearing senescent cells might shift cognitive markers in this population.
The TNF-α to MoCA correlation is the piece that makes this more than a feasibility study. It proposes a mechanism: reduce inflammation from senescent cells, and the brain environment becomes measurably less hostile to cognitive function. Whether that mechanism holds in a randomized, placebo-controlled trial with hundreds of participants is the question that follows. STAMINA’s contribution is establishing that the question is worth asking with the full weight of a large trial behind it.