Metformin Does Not Significantly Alter Longitudinal Dynamics of Clonal Hematopoiesis

PURPOSE: Early intervention in patients with clonal hematopoiesis (CH) is an area of intense investigation with no currently approved agents. With recent mechanistic data on metformin as a possible therapeutic agent in CH and its availability in clinical practice, we sought to investigate the clonal dynamics of CH mutations in metformin users. EXPERIMENTAL DESIGN: We analyzed longitudinal targeted deep sequencing of 1,104 CH mutations in 863 metformin-treated type 2 diabetic participants in two longitudinal cohorts: the Women's Health Initiative (WHI) and BioVU, with blood collected at a median of 15.8 and 6.1 years apart, respectively. RESULTS: Metformin duration (per 6 months) was not significantly associated with the overall CH growth rate in WHI [β = -0.05%/year; 95% confidence interval (CI), -0.11 to 0.01; P = 0.08; n = 543] and in BioVU (β = -0.09%/year; 95% CI, -0.22 to 0.05; P = 0.20; n = 561). Inverse variance-weighted random-effects meta-analysis demonstrated a small, statistically significant association (β = -0.06%/year; 95% CI, -0.11 to -0.002; P = 0.04; n = 1,104) without significant heterogeneity (P = 0.60). These results were similar when only considering DNMT3A and DNMT3A-R882 clones. CONCLUSIONS: In our cohorts, the duration of metformin use among diabetic users was associated with a small reduction in CH growth rate (-0.06%/year), which is modest compared with typical DNMT3A clonal growth rates of 5% to 7% annually. Metformin's clinical utility for modulating clonal dynamics in real-world settings seems limited, and its clinical use for this indication requires further investigation in prospective studies.