Growth differentiation factor 15 is not associated with glycemic control in patients with type 2 diabetes mellitus treated with metformin: a post-hoc analysis of AIM study

BACKGROUND: Growth differentiation factor 15 (GDF15) was newly discovered to be a promising target of metformin. The study was aimed to investigate the relationship between GDF15 and glycemic control after metformin treatment in patients with type 2 diabetes mellitus. METHODS: The study was a post-hoc analysis of AIM (the effect of Acarbose on glycemic variability in patients with type 2 diabetes mellitus using premixed Insulin compared to Metformin) study. The participants were randomly assigned to 12 weeks of metformin (MET) or acarbose (ACA) treatment combined with insulin. Serum GDF15 levels of 51 subjects from MET group and 53 subjects from ACA group were measured at baseline and after a 12-week treatment. Fasting plasma glucose (FPG), 2-h postprandial plasma glucose (2-h PG) and glycated hemoglobin A1c (HbA1c) were measured at baseline and endpoint. RESULTS: After a 12-week treatment, serum GDF15 levels significantly increased in MET group [baseline vs. endpoint, 936.70 (741.00, 1205.40) pg/mL vs. 1265.20 (1027.90, 1634.00) pg/mL, P < 0.001], but not in ACA group [baseline vs. endpoint, 920.60 (701.45, 1332.55) pg/mL vs. 893.80 (663.25, 1284.05) pg/mL, P = 0.944]. However, there were no significant differences of glycemic control parameters (ΔFPG, Δ2-h PG and ΔHbA1c) between subgroups of MET group divided by median of ΔGDF15 (all P > 0.05). Spearman correlation coefficient and analysis of covariance after adjustment for baseline HbA1c levels showed that ΔGDF15 was not correlated with ΔFPG, Δ2-h PG and ΔHbA1c (all P > 0.05). CONCLUSION: Serum GDF15 levels were significantly elevated after metformin treatment in patients with type 2 diabetes mellitus. However, the increase was not an indicator of the glucose-lowering effect of metformin. TRIAL REGISTRATION: Clinicaltrials.gov, NCT02438397. Registered 8 May 2015.