In a 145-person, two-year randomized double-blind placebo-controlled trial, metformin produced a steady improvement in a 95-deficit frailty index (−0.006 per year vs. placebo, p=0.022) and reduced epigenetic biological age by approximately 0.4 years annually on two independent DNA methylation clocks (PC-Horvath2 and PC-Hannum). Participants averaged 72 years old with glucose intolerance. A key nuance: the drug's apparent worsening of the Fried frailty phenotype score in year one was entirely attributable to metformin-driven weight loss (−5.7 kg vs. −2.3 kg), not genuine functional decline.
This trial lands at a critical moment in geroscience. Metformin is already under investigation in the landmark TAME trial as a broad longevity intervention, but evidence from controlled human trials targeting frailty specifically has been scarce. The dual signal here — deficit accumulation and epigenetic clocks moving in the right direction — adds meaningful mechanistic texture, since epigenetic clocks like PC-Horvath2 track biological rather than chronological aging. That said, the cohort was 94% White with only 145 participants, limiting generalizability. The two-year window may also be too short to detect functional endpoints that matter most to patients. The weight-loss confounding in the Fried score underscores how metformin's metabolic effects complicate frailty measurement. Crucially, this is a preprint not yet peer-reviewed — effect sizes and conclusions may shift after independent scrutiny. Incrementally confirmatory but clinically promising.
