Supervised exercise tracked better function and a shift in frailty markers near age 80

What the research found

Study (human exercise intervention). Diego Marcos-Perez and colleagues report in the Journal of Cachexia, Sarcopenia and Muscle (doi 10.1002/jcsm.70362; 24 September 2026) on a multicentre comparison of a control group and a supervised exercise group in older adults, plus a separate testosterone-suppressed group of older men on androgen deprivation for prostate cancer. The control group had 24 people (14 women, 10 men; median age 80 plus or minus 5). The exercise group had 44 people (38 women, 6 men; median age 79 plus or minus 4). In the exercise group, Short Physical Performance Battery rose from 9.1 to 10.1, grip strength from 16.9 to 18.3 kg, and gait speed from 0.78 to 0.85 m/s. A sarcopenia score fell from 3.0 to 2.1, and Fried frailty criteria fell from 2.1 to 1.6 (all p at or below 0.004 within that before-after comparison). ANCOVA attributed better SPPB (p = 0.04) and grip strength (p < 0.005) to exercise versus control. Four of six FRAILOMIC markers (miR125b, miR194, RAGE, troponin) and six of seven BIOFRAIL markers (EGR1, CXCL8, GOS2, NSF, DDX11L1, miR454) reversed after training. Those marker shifts were not seen in 21 testosterone-suppressed men, who lost muscle. Latent class analysis called 36 people responders (32 from the exercise group) and 29 non-responders. miR454, EGR1, DDX11L1, and troponin best separated responders, and the BIOFRAIL trio miR454, EGR1, and DDX11L1 had an AUC of 0.83 for response. The public record does not state program length here. Mostly women, with men included. Association of a delivered program with function and blood or tissue markers, not a promise that any class reverses frailty. Educational exercise notes.

Evidence level: Human study. A study in people, but an association isn't proof of cause.

Read the original at Journal of Cachexia, Sarcopenia and Muscle