Comprehensive Summary of Findings

Trial Design

TRAVERSE (Testosterone Replacement Therapy for Assessment of Long-term Vascular Events and Efficacy Response in Hypogonadal Men) was a phase 4, multicentre, randomised, double-blind, placebo-controlled, noninferiority trial. It enrolled 5,246 men aged 45–80 with preexisting CVD or high CV risk, documented hypogonadism (two fasting testosterone levels <300 ng/dL), and at least one hypogonadal symptom. Patients received daily transdermal 1.62% testosterone gel (dose-adjusted to maintain levels between 12–26 nmol/L) or placebo for a mean of ~22 months, with mean follow-up of 33 months.

Importantly, this was a population of mean age 65, ~70% with diabetes, and multiple cardiovascular comorbidities. The authors conceded that 25% of patients failed to reach the normal testosterone range, with a median increase only from 9.3 to 12.9 nmol/L — suggesting patients were undertreated compared to the T Trials and T4DM. Around 60% in both groups discontinued treatment before the study ended.

1. Cardiovascular Disease (Primary Endpoint)

A primary cardiovascular end-point event (composite of CV death, nonfatal MI, or nonfatal stroke) occurred in 182 patients (7.0%) in the testosterone group and 190 patients (7.3%) in the placebo group (HR 0.96; 95% CI 0.78–1.17; P<0.001 for noninferiority).

There was a small non-significant reduction in cardiovascular mortality (HR 0.84; 95% CI 0.63–1.12) in the testosterone group — notably, 16 extra CV deaths occurred in the placebo group, classified as non-significant.

The bottom line: TRT does not increase MACE in high-risk hypogonadal men. The trial was designed for noninferiority (not superiority), so it couldn't definitively prove a CV benefit, but the trend favoured testosterone. The mean follow-up of only 21.7 months was noted as a limitation — comparable statin trials ran for 5 years to demonstrate atherosclerotic benefit.

2. Prostate Cancer

Prostate cancer occurred in 12 patients (0.5%) in the testosterone group and 11 (0.4%) in the placebo group (P = 0.87). No difference whatsoever. This strongly supports the view that routine digital rectal examination is not required unless otherwise clinically indicated.

This is a landmark finding — the largest RCT to date showing no signal for prostate cancer with TRT, effectively putting to rest the old Huggins hypothesis about testosterone "feeding" prostate cancer in the general hypogonadal population.

3. VTE / Pulmonary Embolism

Pulmonary embolism was reported in 24 men receiving testosterone vs. 12 in the placebo group. This translated to PE incidence of 0.9% vs 0.5% in the placebo group.

However, context matters here. The VTE events consisted of 24 pulmonary emboli in testosterone-treated men vs 12 in placebo; and 16 deep venous thromboses and 11 other peripheral venous thromboses — the overall VTE composite did not reach statistical significance. Isolated examples of such differences within a large trial are of questionable significance, as PE represented just one component within the VTE category that did not differ significantly between arms overall.

Critically, time-dependent Cox hazards models did not show an association between change in haematocrit and VTE risk (HR 0.94; 95% CI 0.84–1.05) — meaning the traditional concern about erythrocytosis driving thrombotic events was not supported. Only 6 participants had haematocrit exceed 54% and required dose discontinuation.

4. Anaemia

Anaemia was detected in 15.7% of the cohort (815 of 5,204 men). Testosterone therapy corrected anaemia in a significantly greater proportion: 41.0% of testosterone-treated vs 27.5% of placebo-treated men at 6 months. TRAVERSE was the first study to demonstrate that functional symptom improvement is closely related to correction of anaemia. This aligns with T Trials data where TRT corrected anaemia even in men with a previously identified alternative cause.

5. Diabetes / Metabolic Outcomes

The initial NEJM paper reported diabetes progression as non-significant, but Hackett's reanalysis identified that since ~70% had diabetes at baseline, the denominator should be the ~30% with prediabetes. Recalculated: 189/607 (31%) on TRT vs 213/558 (40%) on placebo progressed to diabetes, equating to a 22.5% reduction (p=0.029) in new-onset diabetes.

This is notable but considerably less impressive than the T4DM trial's 41% reduction — likely because treatment levels in TRAVERSE failed to reach even the pre-treatment levels in T4DM, and the older TRAVERSE population with more comorbidities was undertreated. There's also the confounding issue that TRT-induced increases in haematocrit are known to artificially lower HbA1c, so some of the HbA1c improvement may be artefactual.

6. Depression

Depression was recorded in 50% of the men with hypogonadism and was mild to moderate, with 10% severe. Modest improvement in depression and Aging Male Symptom scores were found in mild to moderate cases.

This is consistent with other TRT literature — meaningful but modest antidepressant effect, most notable in mild-to-moderate depressive symptoms rather than severe clinical depression.

7. Sexual Function (IIEF)

There was a significant increase in sexual desire and sexual frequency but not erectile function. The authors and commentators note this is unsurprising given the population: erectile dysfunction in these patients was most likely related to endothelial dysfunction or diabetic neuropathy rather than hormonal factors.

Only 4% were taking PDE5 inhibitors — compared with 30% in the BLAST study of hypogonadal men with T2DM, where ED improvement was significantly higher. The T Trials (with lower diabetes rates and better compliance) showed a 2.64-point IIEF improvement vs placebo. This underscores that TRT alone won't fix ED when the primary pathology is vascular/neurogenic — it needs to be combined with PDE5i therapy, which links directly back to the Traish data you just reviewed on testosterone's role in maintaining PDE5 substrate.

8. Bone / Fractures

This was the most surprising and counterintuitive finding. There were 154 confirmed fractures in the testosterone group compared with 97 in the placebo group. Men receiving testosterone had a higher risk for clinical fracture (HR 1.43; 95% CI 1.04–1.97; P = 0.03).

However, non-high impact clinical fractures (analogous to fragility fractures) did not increase significantly (HR 1.32; 95% CI 0.94–1.86). The study did not include BMD assessment, and a meta-analysis by Van Gameren et al. describes an early increase in fractures with increased activity in frail populations but a long-term reduction with development of muscular strength. The most plausible explanation is that improved energy and mobility in previously sedentary frail men led to more falls and traumatic fractures before musculoskeletal benefits caught up — not that testosterone worsened bone quality.

9. Other Notable Findings

Atrial fibrillation: AF was reported in 91 men receiving TRT vs 63 on placebo. But ECGs were not conducted at baseline nor as part of the trial protocol — just reported medical events, so undiagnosed pre-existing AF could have been asymmetric between groups.

Acute kidney injury: 60 patients (2.3%) vs 40 (1.5%) on placebo (P = 0.04). However, 418 TRT vs 393 placebo patients had stage 3 CKD at baseline (a difference of 25), yet the authors listed 60 vs 40 AKI events (a difference of 20) — suggesting the baseline imbalance in renal disease may account for most of this signal.

Haematocrit and MACE: No association was found between haematocrit change and MACE risk (HR 0.97; 95% CI 0.92–1.02).

Key Takeaways for Clinical Practice

The TRAVERSE data, when properly interpreted alongside the T Trials and T4DM, supports several conclusions relevant to your Everlab protocols: TRT in appropriately diagnosed hypogonadal men is cardiovascularly safe, does not increase prostate cancer risk, corrects anaemia, modestly improves depression, reduces diabetes progression (especially with adequate dosing), and improves sexual desire though not necessarily erectile function in men with established vascular disease. The PE signal warrants caution in men with thrombotic history, but the overall VTE composite was not significant. The fracture finding is almost certainly confounded by increased activity in previously sedentary men rather than reflecting a bone-weakening effect of testosterone. The biggest limitation is that patients were undertreated — median testosterone only reached 12.9 nmol/L — suggesting that adequately dosed TRT in clinical practice may produce even better efficacy outcomes than TRAVERSE demonstrated.