Background
Testosterone replacement therapy (TRT) improves health in some but not all men with type 2 diabetes (T2DM) and adult‐onset testosterone deficiency (TD). Such heterogeneity is compatible with the concept of patient subgroups that respond differently to therapy.
Objectives
Use baseline SHBG and age to identify putative subgroups that demonstrate different responses in variables such as waist circumference and HbA1c following TRT.
Materials and methods
A randomized double‐blind trial approach was used to recruit and randomize men with T2DM and adult‐onset TD into placebo and TRT‐treated groups. Multiple regression was used to study differences between groups.
Results
Baseline SHBG and change in SHBG (∆SHBG) were inversely related in the TRT group. Both median values of SHBG and age mediated the effect of TRT on ∆SHBG depending on whether baseline values were ≤ or>median (28.1 nmol/L, 63 years, respectively). In men with both SHBG ≤ 28.1 nmol/L and age ≤ 63 years (subgroup 1), TRT was positively associated with ∆SHBG (c = 4.67, 95%CI 1.17‐8.16, P = .010) while in those with SHBG > 28.1 nmol/L and age > 63.1 years (subgroup 4) the association was inverse (c = −7.07, 95%CI −11.64 to −2.49, P = .003). The association between TRT and change (∆) in waist circumference, HbA1c and International Index of Erectile Function (IIEF) score differed between subgroups; in subgroup 4 but not subgroup 1, the therapy was significantly associated with ∆waist circumference, ∆HbA1c and ∆IIEF.
Discussion
Though the mechanism remains unclear, our finding of different responses to TRT in terms of change in waist circumference, HbA1c and IIEF score supports the concept of subgroups in men with T2DM and adult‐onset TD.
Conclusion
Our approach may provide a basis for identifying men who will or will not derive benefit from TRT though a larger study is required.