Longevity Treatments Spread Lifespans Rather Than Equalizing Them

A central aspiration in geroscience is not merely to extend average lifespan but to ensure that most individuals reach old age — an ideal captured by the concept of 'squaring the survival curve,' where the survival curve drops steeply at a late age rather than tapering gradually. This new analysis challenges whether current leading interventions actually achieve that goal.

Researchers at the University of Sydney re-analyzed a published meta-analysis of vertebrate longevity data examining three geroprotective interventions: dietary restriction (DR), rapamycin (an mTOR inhibitor and DR-mimetic), and metformin (another DR-mimetic, though less consistently effective). Rather than focusing solely on mean age-at-death, they examined what happened to variance and the coefficient of variation (CV) — standard deviation relative to the mean.

The key finding: all three treatments increased absolute variance in age-at-death on average, while the CV remained largely unchanged. In plain terms, when these interventions extend life, they do not compress the spread of death ages — they expand it proportionally. Some individuals benefit enormously; others gain little. The distribution widens rather than tightens.

Mechanistically, the authors note this pattern is consistent with the interventions proportionally reducing both the age-dependent (Gompertz slope) and age-independent (baseline hazard) mortality parameters simultaneously — a mathematical outcome that preserves relative variation even as the mean shifts upward.

These results carry important implications for both research and clinical expectations. Interventions celebrated for extending average lifespan may be masking highly heterogeneous individual responses. The equitable vision of everyone living a long, similarly-timed life remains unachieved by current tools. Future research should prioritize identifying who benefits most and why, and whether combination strategies or personalized approaches can narrow the outcome gap.