Read the Full Video Transcript

Rashid Sayyid: Hello everyone, this is Rashid Sayyid. I'm a urologic oncology fellow at the University of Toronto, and along with Zach Klaassen, associate professor, program director at Augusta University. We'll be discussing the recent publication looking at cardiovascular safety of testosterone replacement therapy. This article was recently published in the New England Journal of Medicine by Lincoff, et al.

So the cardiovascular effects of testosterone replacement therapy in middle-aged and older men with hypogonadism or low testosterone levels remain unknown. And there's a lot of conflicting data from retrospective studies. Some suggest there's an increased cardiovascular risk and others suggest that perhaps there's decreased cardiovascular risk. There's also some small RCTs in this space, but they're limited by inconsistent results. They've not been powered to assess cardiovascular outcomes, and they also have short follow-up periods.

In March 2015, based on the available literature, the FDA issued a guidance statement in response to concerns and conflicting data regarding the cardiovascular safety of testosterone replacement therapy. And the manufacturers of approved testosterone products will require to conduct clinical trials to determine whether testosterone replacement therapy is associated with an increased risk of cardiovascular events. And in accordance with this, the TRAVERSE or the testosterone replacement therapy for assessment of long-term vascular events and efficacy response in hypogonadal men trial was designed to determine the effects of testosterone replacement therapy on the incidence of major adverse cardiovascular events among middle-aged and older men with hypogonadism and either preexisting or high risk of cardiovascular disease.

And so, this was a Phase 4 randomized, double-blind, placebo controlled, non-inferiority trial that was run across 316 centers in the United States between May of 2018 and February of 2022. And this trial specifically included men aged 45 to 80 years, and all men had to have at least one hypogonadal symptom, which included potentially decreased sexual desire and libido, decreased erections, fatigue, low or depressed mood, loss of auxiliary pubic hair or decreased frequency of shaving, or hot flashes. All patients needed to have a fasting serum testosterone less than 300, and it had to be measured in the morning between 5:00 and 11:00 AM, because we know that testosterone levels are highest during that time. And so, to allow for consistency in comparison and inclusion, that was one of the criteria applied.

Next, if we continue the inclusion criteria, all patients, again, not only did they have to have one hypogonadal symptom, but they also had to have either cardiovascular disease or an increased risk of cardiovascular disease. And this was defined as such. So, cardiovascular disease could be clinical or angiographic evidence of coronary artery disease, cerebrovascular disease, or peripheral arterial disease. So, not only the heart, the heart, the brain, or the peripheral vessels.

Now if they didn't have cardiovascular disease, they had to have increased risk of cardiovascular disease, defined as at least three of the following risk factors, hypertension, high blood pressure, dyslipidemia, they had to be current smokers Stage 3 Chronic Kidney Disease, diabetes, elevated C-reactive protein, age of at least 65, or an Agatston calcium score that was higher than the 75th percentile for their age and race.

What about exclusion criteria? Severe hypogonadism, they're out meaning testosterone, less than a hundred. History of prostate cancer, prostate nodules or an elevated PSA, for obvious reasons, thrombophilia, because of the inherent higher risk of blood clots, uncontrolled heart failure, if they had an acute coronary syndrome stroke or a revascularization of the coronary peripheral vasculature within the prior four months, and if they were on testosterone or androgenic steroids within the prior six months.

Patients in this trial were randomized one-to-one to either daily transdermal testosterone gel or placebo gel. Some people may ask, "Well, what about injections?" It's probably ethically questionable to inject patients with a sham injection. And so for that reason, to minimize potential ethical issues, the authors went in favor of testosterone gel. The randomization was stratified by presence or absence of cardiovascular disease. It's important to note that post-baseline testosterone level measurements were concealed from patient trial to minimize risk of unblinding. If testosterone levels were available to either the practitioners or the patients, obviously, patients who have an increase in the testosterone levels, that kind of would give a strong indication that they're in the testosterone gel arm as opposed to placebo gel arm. So to minimize that risk of unblinding, everybody was unaware of the results.

The dose adjustments, now, for testosterone level or the side effects, mainly the elevated hematocrit with greater than 45% were centrally managed by automated algorithm. And again, to minimize the risk of inadvertent unblinding, the placebo group as well underwent sham adjustments, just to mix it up and make sure there were no consistent patterns. Now, treatment was discontinued if patients have a testosterone greater than 750 or hematocrit greater than 54%. Even after the adjustments, or if they had new prostate cancer diagnoses or were deemed at higher suicide risk, they were out of the trial. Patients with elevated PSAs were referred to urologists per a pre-determined protocol, again, to make sure that no high grade cancer went into this.

The study endpoints were adjudicated by an independent clinical-events committee, and the primary study endpoint was the occurrence of any component of the following composite endpoint. So if patients had death from cardiovascular causes, or they had a non-fatal myocardial infarction, or non-fatal stroke, that was considered an event and that was the primary outcome. Now, for the secondary outcome, it's the same as the primary outcome, but they added coronary revascularization, so kind of a sensitivity analysis, adding this outcome to the composite outcome. Looking at tertiary events, they looked at death from any cause, hospitalization or urgent visit for heart failure, peripheral arterial revascularization, or a venous thromboembolic event.

Given this was a time-to-event analysis, the authors used survival analysis with Cox proportional hazards analysis to evaluate the association of testosterone replacement therapy versus placebo with the study events adjusting for presence or absence of underlying cardiovascular disease. This was a non-inferiority trial. So essentially, we're asking the question, "Is testosterone replacement therapy not worse compared to placebo, with regards to cardiovascular events?" Obviously, we would expect it to be better for the hypogonadal symptoms of testosterone levels. For this trial, testosterone replacement therapy would be deemed non-inferior for the adverse outcomes, if the upper boundary of the 95% confidence interval did not cross 1.5. And this was an event driven analysis with 256 events providing 90% power at a type 1 error rate of 5% to evaluate non-inferiority, so 90% higher than the typical 80%. So, there was sufficient power to assess whether testosterone replacement was inferior with regards to the cardiovascular outcomes.

The safety population included any patient who received at least one dose of testosterone or placebo. And the intent to treat approach and sensitivity analysis was performed looking at major adverse cardiovascular events from the time of randomization to the end of available follow-up. And then, they also looked at it between randomization and a year following treatment with a thought that any event that occurs post 365 days or one year is unlikely to be related to the testosterone replacement. And at this point, I'll turn it over to Zach to go over the results and discussion for this patient.

Zach Klaassen: Thanks so much for that introduction, Rashid. So this is the baseline characteristics of patients in the full analysis population. You can see that they're delineated by the testosterone group versus the placebo group. Generally just looking at this Table 1, these are well-balanced populations. We can see the mean age was roughly 63 years. There was 80% of patients that were white, and 16% to 17% of patients that were black. These were somewhat overweight patients with a median BMI of 35. The median testosterone at the time of randomization was 227. Roughly 50% of patients had pre-existing cardiovascular disease and 50% of patients were at increased cardiovascular risk. In terms of patients that had a history of coronary artery disease, 45% of patients, 12% had a history of cerebral vascular disease, and roughly 6% of patients had a history of peripheral artery disease.

With regards to cardiovascular risk factors, roughly two thirds of patients had diabetes. The majority of patients had hypertension and dyslipidemia, and about 20% of patients were current smokers. 15% had Stage Three Chronic Kidney Disease. And the majority of patients had never had prior testosterone use. For medications, the majority were on lipid lowering therapy. Roughly just over half were taking aspirin. Median PSA was low in these patients, less than one. Hematocrit was 42, between these two groups of patients.

These figures look at the median serum testosterone levels and changes in serum testosterone levels over time. The testosterone group is in orange and the placebo group is in blue. And we can see that the median increase from baseline in serum testosterone levels was 148 for the testosterone group compared to 14 for the placebo group. This is the primary outcome. This is the primary Cardiovascular Composite Safety Endpoint in the safety population. We can see, again, placebo in blue and testosterone in orange. Hazard ratio was 0.96, 95% confidence interval, 0.78 to 1.17. So completely not statistically significant. And as Rashid mentioned, the upper limit of the 95% confidence interval at 1.17 is less than 1.50, so this is testosterone being non-inferior to placebo with regards to this Primary Cardiovascular Composite Safety Endpoint.

This is the Primary Cardiovascular Composite Safety Endpoint in the principal sensitivity analysis population. And just as a reminder, this is data on endpoint events that occurred more than one year after discontinuation of testosterone or placebo. These events were censored. And again, we see a very similar curve, hazard ratio of 1.02, 95% confidence interval, 0.81 to 1.27. So, no difference between testosterone and placebo.

Moving to the secondary endpoints, this is the Secondary Cardiovascular Composite Safety Endpoint. And just as a reminder, this is the primary endpoint with the added outcome of coronary revascularization. Again, no difference between placebo and testosterone, hazard ratio of 1.02.

Looking at some of the more specific secondary endpoints, this is death from cardiovascular causes, no difference between placebo and testosterone, hazard ratio of 0.84, 95% confidence interval of 0.63 to 1.12.

Nonfatal Myocardial Infarction, no difference between these groups, hazard ratio of 1.10. We also see no difference in nonfatal stroke, hazard ratio of 0.94, 95% confidence interval, 0.60 to 1.49.

And finally, we move over to the Adjusted Cardiovascular Endpoints in the Safety Population. The top half of this table is what we've just covered in the Kaplan-Meier curve, so we'll focus on the bottom half of this table. This is the tertiary endpoints. And we see death from any cause, no difference between these groups, hazard ratio of 0.98. Hospitalization or urgent visit for heart failure, no difference between these two groups, hazard ratio of 1.10. Peripheral artery revascularization, hazard ratio of 0.92, no difference between testosterone and placebo groups. And finally, venous thromboembolic events, statistically no different, hazard ratio of 1.46, 95% confidence interval of 0.92 to 2.32. Although, we do see 1.7% of patients having VTE in the testosterone group compared to 1.2% in the placebo group. And when the authors broke this down into the components of VTEs, we do see that pulmonary embolism was slightly higher in the testosterone group with a percentage of 0.9% compared to 0.5%.

With regards to adverse events, we see any adverse event occurring in just under half of these patients, serious adverse events in just over one quarter of these patients in both groups. I've highlighted several interesting adverse events that were statistically significant in terms of comparing the testosterone group to the placebo group. We did see higher nonfatal arrhythmias warranting intervention at 5.2% in the testosterone group compared to 3.3% in the placebo group. Atrial fibrillation was higher in the testosterone group, 3.5% compared to 2.4% in the placebo group. And finally, acute kidney injury was higher in the testosterone group at 2.3% compared to 1.5% in the placebo group.

So by way of discussion, the TRAVERSE trial showed that among almost 5,200 patients who received testosterone or placebo for a mean of 22 months, testosterone replacement therapy was noninferior to placebo. With respect to the incidence of major adverse cardiovascular events. This was associated with a hazard ratio of 0.96 and a non-inferiority finding that was confirmed on sensitivity analyses. As I just mentioned on the previous slides, the incidence of pulmonary embolism was higher with the testosterone group than the placebo group, although meta-analyses of RCTs previously have not shown an association between VTE and testosterone use. So the findings from TRAVERSE support current guidelines that testosterone should be used with caution in men who have had previous thromboembolic events.

And finally, these findings regarding cardiovascular safety of testosterone may facilitate a more informed consideration of the potential benefits and risks of testosterone replacement therapy among middle aged and older men with hypogonadism.

So in conclusion, among men with hypogonadism and established cardiovascular disease or multiple risk factors for incident cardiac events, testosterone replacement therapy was noninferior to placebo with respect to the occurrence of major adverse cardiac events during a mean 22-month follow-up and the overall incidence of adverse events was low. We thank you very much for your attention. We hope you enjoyed this UroToday Journal Club discussion of the TRAVERSE trial, recently published in the New England Journal of Medicine.