In the ‘Testosterone for the prevention of type 2 diabetes’ (T4DM) Study, treatment with testosterone vs. placebo reduced the OGTT diagnosis of T2D by 40% after 2 years. The mechanism(s) by which testosterone prevents glycaemia and T2D diagnosis remains unclear. Direct effects on beta-cell function, augmentation of the glucagon-like peptide-1 (GLP-1) action and reduction in insulin resistance have all been proposed as mechanisms.
We aimed to assess the mechanism of testosterone protection against T2D by measuring glycaemic traits in a randomised control trial and determining in vitro effects of testosterone on glucose stimulated insulin secretion (GSIS) and GLP-1 action in a human beta-cell line (EndoC-βH1).
Men from T4DM were included if they had fasting bloods at baseline, 18, 66, and 102 weeks (N=743). We compared the change in glycaemia (GA-glycated albumin), insulin secretion (HOMA2-B) and resistance (HOMA-IR) with testosterone or placebo. The effect of acute (20 min) and chronic (2-day) testosterone exposure on GSIS and GLP-1 action was assessed in EndoC-βH1 at 3 testosterone concentrations (5, 10, 15nM).
GA decreased in both groups (p<0.001) and was lower in the testosterone compared with placebo (-3.4 umol/L; 95%CI:1.4-5.4, p<0.001). C-peptide, HOMA-IR and HOMA-B initially decreased (time effect, all P<0.001) but no difference due to testosterone treatment was detected. Exploratory analysis at maximal glycaemic difference between testosterone and placebo (week 66) demonstrated a non-significant reduction of C-peptide (-0.10, p=0.2), HOMA-B (-6.1, p=0.2) and HOMA-IR (-0.20, p=0.3). Acute exposure of EndoC-βH1 to 5nM, 10nM and 15nM testosterone did not increase GSIS compared with control. Testosterone, at 15nM for 2 days did not alter GSIS or GLP-1 mediated insulin secretion.
In men with pre-diabetes or early T2D, testosterone treatment reduces glycaemia in vivo without in vitro evidence of increased insulin secretion or augmentation of GLP-1 action.