Inhibition of 5α-reductase type 1 with dutasteride in men increases body fat and impairs insulin sensitivity

Background: Inhibition of 5α-reductase type 2 (5αR2) with finasteride lowers prostatic dihydrotestosterone (DHT) in the treatment of benign prostatic hyperplasia (BPH). In addition to inhibiting 5αR2 in prostate, the 'dual' isozyme inhibitor dutasteride also inhibits 5αR1 in liver and adipose, thus lowering local DHT, increasing local glucocorticoids, and potentially impacting fuel metabolism. Hypothesis: 5αR1 inhibition with dutasteride induces metabolic dyshomeostasis. Methods: In a double-blind randomised controlled design, we studied 46 men (20-85 years) before and after 3 months of either dutasteride (0.5 mg daily; n=16), finasteride (5 mg daily; n=16) or control (tamsulosin MR; 0.4 mg daily; n=14). The primary outcome was insulin sensitivity, measured during a two-step (10; 40 mU/m2/min) hyperinsulinaemic euglycaemic clamp, with deuterated glucose and glycerol tracers. Data are mean (95% CI) change from baseline, compared by one-way ANOVA. Results: Dutasteride, but not finasteride or tamsulosin, increased body fat (by 1.6% (0.2; 3.0), p=0.02) and reduced insulin-stimulated glucose uptake (glucose rate of disposal during high dose insulin decreased with dutasteride-5.7 (-12.5; 1.1) Âμmol/kg fat-free mass/min, versus finasteride +7.2 (0.9; 13.6) and tamsulosin +7.0 (2.7; 11.3), p=0.002). Glucose production and lipolysis during low dose insulin infusion were unchanged. Consistent with impaired peripheral (likely skeletal muscle) insulin sensitivity after dutasteride, but not finasteride or tamsulosin, tracer infusion alone induced hyperinsulinaemia (by 5.6 (-1.5; 12.8) pmol/L, p=0.03) and there were increases in HOMA-IR (+0.39 (0.08; 0.7), p=0.03) and fasting C-peptide (+76 (19.2; 132.8) pmol/L, p=0.04). There were no changes in fasting glucose, cholesterol, body mass index, waist:hip ratio or blood pressure. There were also no differences in post-treatment ratio of visceral:subcutaneous adipose volumes (magnetic resonance imaging; L4/5), hepatic fat (1H spectroscopy), nor in serum adipokine profile or subcutaneous adipose adipokine mRNAs. Conclusion: 5αR inhibition with dutasteride, but not finasteride, impairs peripheral insulin sensitivity, and increases body fat. This highlights a novel determinant of metabolism and suggests that dual 5αR inhibitors should be re-evaluated in men with prostatic disease.