Metformin and Aging Research: Observational Data and the TAME Trial

Observational studies: an intriguing signal with known pitfalls

• UK database study. Researchers compared 78,241 people with type 2 diabetes who started metformin, 12,222 who started a sulphonylurea (another diabetes drug) and 90,463 matched people without diabetes. Median survival was 15% lower in the people without diabetes than in the metformin group. [2] This was retrospective record data, so the groups may differ in ways the analysis did not capture.
• Meta-analysis. A 2017 systematic review pooled 53 studies and found lower all-cause death in people with diabetes taking metformin than in people without diabetes (hazard ratio 0.93). The authors noted that differences in baseline characteristics could bias the findings, despite statistical adjustment. [3]
• A known bias. "Immortal time bias" happens when exposure is defined by something after the study clock starts, which can make a drug look protective. In a gastric cancer analysis, studies with this bias showed a 33% lower risk with metformin, while studies without it showed no association. [4]

Label: promising but limited, with a high risk of bias. People who take metformin differ from those who do not, and association is not causation. See Reading longevity studies.

Randomized trials in older adults: not a longevity result

• Muscle (MASTERS trial). In 94 healthy adults aged 65 and older doing supervised resistance training for 14 weeks, those on placebo gained more lean body mass and thigh muscle than those on metformin (a study amount of 1,700 mg a day). [5]
• Walking speed (MET-PREVENT). In 72 people with a mean age of 80 and probable sarcopenia (age-related muscle loss) and frailty, metformin did not improve 4-metre walk speed and was poorly tolerated. [6]
• Diabetes Prevention Program. This long trial in people at high diabetes risk was about preventing diabetes, not aging, but it gives long-term safety data (see below). [7]

Label: no good evidence of benefit for healthy aging outcomes so far; insufficient evidence on lifespan. Lifespan and healthspan outcomes need much larger and longer trials.

The TAME trial

TAME (Targeting Aging with Metformin) is the planned large trial in which metformin was to be the first drug tested for an age-targeting effect. [8][9] We could not read the sponsor's page or a trial record this session, so we do not state its size, funding or current status. Proper evidence not found on those points. For how aging-focused trials are designed, see Aging clinical trials: TAME and others.

How it compares and where to go next

Metformin and rapamycin are the two drugs most often discussed in aging research; see metformin vs rapamycin. Animal findings are covered in aging in mice, worms and humans. What "off-label" means is explained in off-label prescribing for longevity.