Research in context
Evidence before this study
We searched PubMed and ScienceDirect on Dec 1, 2011, and again on Nov 19, 2019, using the following search terms in various combinations: “testosterone”, “type 2 diabetes”, “obesity”, “metabolic syndrome”, “weight loss”, “skeletal muscle mass”, and “insulin resistance”, restricted to the English language. Obesity (with or without insulin resistance and abnormal glucose tolerance) is associated with lowered serum testosterone concentration, which is reversed by weight loss. A low serum testosterone concentration is associated with an increased risk of incident type 2 diabetes. Testosterone treatment decreases fat mass and increases muscle mass in men. In randomised controlled trials, testosterone treatment did not lower HbA1c in men with established type 2 diabetes, but it might improve insulin resistance in men with type 2 diabetes or metabolic syndrome. In a registry-based study, testosterone treatment prevented progression of prediabetes to type 2 diabetes and improved glucose metabolism in people with type 2 diabetes. Weight loss through lifestyle intervention (ie, diet and exercise) prevents the progression of prediabetes to diabetes, including reversal of newly diagnosed type 2 diabetes, but it is not known whether testosterone treatment augments the benefits of a lifestyle intervention.
Added value of this study
The findings of the T4DM trial showed that, among men aged 50–74 years with impaired glucose tolerance or newly diagnosed type 2 diabetes and with a waist circumference of 95 cm or higher, treatment with testosterone significantly reduced the risk of type 2 diabetes (relative risk 0·59) at 2 years compared with placebo, with a between-group difference in mean change from baseline in 2-h glucose (from an oral glucose tolerance test) of −0·75 mmol/L (–1·10 to −0·40). The beneficial effects on glucose metabolism were independent of baseline testosterone concentration. Compared with placebo, testosterone was associated with a greater decrease in fat mass, increase in skeletal muscle mass and strength, and improvement in sexual function. Concerning safety, compared with placebo, treatment with testosterone was not associated with excess cardiovascular or prostate cancer adverse events. However, there were increases in haematocrit and prostate-specific antigen associated with testosterone use. An increase in haematocrit to 54% or higher was flagged in 106 (22%) of 491 testosterone-treated participants, with 25 participants ceasing treatment prematurely as a result.
Implications of all the available evidence
Our findings suggest that testosterone treatment for 2 years, as an adjunct to a lifestyle programme, can prevent or revert type 2 diabetes in overweight men without pathological hypogonadism. This effect compares favourably to that of metformin in the Diabetes Prevention Program and was accompanied by increased muscle mass, grip strength, and sexual function. Increases in haematocrit might be treatment limiting. The minimum dose exposure, duration of treatment, durability of effect, and long-term safety and cardiovascular outcomes of testosterone treatment remain to be determined. We consider it premature to advocate for the widespread use of testosterone for diabetes prevention in men without pathological hypogonadism. Identification of subgroups of men more likely to benefit or most at risk of adverse outcomes might aid treatment decisions, but, if testosterone treatment is considered, there must be a concomitant lifestyle programme and careful monitoring of haematocrit, cardiovascular risk factors, and prostate health.