madman
Super Moderator
TRAVERSING the Mountain of Ignorance: Testosterone and Cardiovascular Safety
Over the past decade, concerns about the adverse effects of testosterone have paralleled the escalating rates of testosterone prescriptions (1). The princi
academic.oup.com
Over the past decade, concerns about the adverse effects of testosterone have paralleled the escalating rates of testosterone prescriptions (1). The principal safety concerns for testosterone therapy have been the potential for increased incident cardiovascular disease and prostate cancer, but the recent concerns have focused on cardiovascular risk (2). The US Food and Drug Administration posted black box warnings about possible increased risk for heart attacks, strokes, and venous thromboembolic disease in 2014 and 2015. Until recently, the evidence for cardiovascular risk with testosterone therapy has been based on findings from epidemiological studies and underpowered, randomized controlled trials that were not designed to determine cardiovascular outcomes (2)
The TRAVERSE study is the first randomized, placebocontrolled trial rigorously designed to determine the cardiovascular risk of testosterone therapy in men (3). A total of 5642 men, aged 45 to 85 years, with symptoms of consistent with hypogonadism, a low serum total testosterone concentration in 2 early-morning samples, and a high risk of incident cardiovascular events were randomly assigned to transdermal testosterone therapy vs placebo for up to 4 years. The mean age of the participants was 63, the mean body mass index was 35, and 70% had diabetes mellitus; the serum free testosterone concentrations were not reported, but they were likely discordant with—and relatively higher than—the total testosterone concentrations for many of the participants. Testosterone dosages were titrated to maintain serum testosterone concentrations continually within the normal range.
The primary outcome was the first occurrence of a composite event during the treatment period: death from any cardiac cause, nonfatal myocardial infarction, or nonfatal stroke. Cardiovascular events were adjudicated by an independent committee whose members were blinded to treatment group assignment. The primary analysis included all men who had received at least one treatment dose. The primary sensitivity analysis that included all adverse cardiovascular events from the first treatment dose to within 365 days of the last dose corroborated the results of the primary analysis.
The mean duration of treatment and follow-up for each group was about 22 months and 33 months, respectively. For the primary outcome, testosterone was noninferior to placebo; there was no statistically significant difference in the rate of composite cardiovascular events (hazard ratio 0.96; 95% CI, 0.78-1.17; P < .001 for noninferiority). In secondary outcome analyses, there were no observed differences in the components of the composite primary end point that included death from cardiovascular causes, nonfatal myocardial infarction, nonfatal stroke, or coronary revascularization or in the secondary and tertiary cardiovascular end points: death from any cause, urgent visit or hospitalization for heart failure, peripheral arterial revascularization, or overall venous thromboembolic events.
There were important additional secondary findings. Testosterone therapy was associated with a significantly high incidence of nonfatal arrhythmias that warranted intervention (5.2% vs 3.3%; P < .001) and atrial fibrillation (3.5% vs 2.3%; P = .02). Although there was no planned analysis for pulmonary emboli as an outcome, 24 (.9%) of the testosterone-treated men had a pulmonary embolus compared to 12 (.5%) of the placebo-treated men. There was no significant difference in the number of prostate cancer diagnoses between the 2 groups, but the event rate was also very low. There were 12 men diagnosed with prostate cancer (5 adjudicated as high-grade, Gleason 4 + 3 or higher) and 11 men diagnosed with prostate cancer (5 adjudicated as high-grade, Gleason 4 + 3 or higher) in the testosterone and placebo groups, respectively (P = NS).
The incidence of diabetes mellitus was 7.3% in the testosterone-treated group and 8.2% in the placebo group (P = NS). Although this finding did not achieve statistical significance, this trend warrants further study. The 1-year, placebo-controlled Testosterone Trials demonstrated that testosterone therapy modestly improved markers of insulin sensitivity in older men with serum testosterone concentrations less than 275 ng/dL (9.5 nmol/L) (4). In theT4DM trial, 1007 men with an impaired glucose or newly diagnosed type 2 diabetes mellitus and a serum testosterone less than 403 ng/dL (14 nmol/L) were randomly assigned to intramuscular testosterone or placebo for 2 years with the primary outcome of prevalent diabetes (defined by oral glucose tolerance test) (5). At the end of treatment, 12% of the testosterone-treated group met the criteria for type 2 diabetes mellitus vs 21% of the placebo-treated group. These 3 trials support conducting larger, longer-term studies of the effect of testosterone therapy on incident diabetes mellitus, an important risk factor for major adverse cardiovascular events.
While the findings of the TRAVERSE study have helped us further ascend the trail of understanding the cardiovascular safety of testosterone therapy, the study design has created a switchback in the management of male hypogonadism; what is the lower limit of normal for serum total testosterone in men? Two of the most important randomized controlled testosterone trials—the TRAVERSE and Testosterone Trials— were conducted by the same investigators. Both trials used the same state-of-the-art total testosterone assay: liquid chromatography–tandem mass spectrometry in a central laboratory certified by the US Hormone Standardized Program for Testosterone (HoST). HoST was the result of a decade-long project to create a systematic method of measurement of serum total testosterone that would create results that are harmonized—that is, uniform, comparable results across different laboratories in a broad geographic range (6). A similar approach was used decades ago to standardize glycated hemoglobin A1c assays that have greatly improved the diagnosis, management, and outcomes of diabetes mellitus. However, TRAVERSE, a study designed to assess the safety of testosterone therapy, and the T Trials, a coordinated set of trials designed to assess the potential benefits of testosterone therapy, defined low total testosterone concentrations differently from HoST: 300 ng/dL (10.4 nmol/L) for TRAVERSE, 275 ng/dL (9.5 nmol/L) for T Trials, and 264 ng/dL (9.2 nmol/L) for HoST (3, 6, 7).
For future clinical outcomes of testosterone therapy, we must use accurate assays of testosterone and a consistent definition of the lower limit of normal (8). Because of the extensive effort for high-quality methodology and standardization in the development of HoST, studies should use testosterone assays that meet HoST standards and use a lower limit of normal of 264 ng/dL (9.2 nmol/L). Likewise, clinical testosterone assays should be standardized and harmonized worldwide.
Then we can consistently translate the results of clinical trials of testosterone therapy into optimal practice
In the meantime, we have traversed the initial cardiovascular gap. Testosterone therapy is generally safe for the first 2 to 4 years when administered at physiological dosages to hypogonadal men.
