What Did the Testosterone Trials Actually Prove About Treating Low T in Older Men?

At a glance

• Trial name: Testosterone Trials (TTrials)
• N: 790 men aged 65 or older
• Intervention: 1% testosterone gel (AndroGel), dose-adjusted to mid-normal range
• Comparator: Matching placebo gel
• Duration: 12 months
• Primary endpoints: Sexual function (PDQ-Q4), vitality (FACIT-Fatigue), physical function (6-minute walk test)
• Key result: Significant improvement in sexual function and 6-minute walk distance; modest, mixed vitality benefit
• Year / Journal: 2016, New England Journal of Medicine

Why This Trial Exists

Before 2016, the clinical evidence supporting testosterone replacement in older men was surprisingly thin. Dozens of small studies had hinted at benefits, but none had the size or rigor to produce definitive answers. The FDA and the Endocrine Society both flagged this gap. Testosterone prescriptions had surged through the 2000s, driven largely by direct-to-consumer marketing, and the medical community was prescribing a hormone without strong randomized data showing it actually helped the symptoms it was supposed to treat.

The Testosterone Trials were designed by the National Institute on Aging specifically to fill that gap. The coordinating structure was unusual: seven sub-trials running in parallel under one enrollment umbrella, each addressing a different symptom domain. The 2016 NEJM publication reported results from the three trials with the largest enrollment, covering sexual function, physical function, and vitality.

Who Got In (and Who Didn't)

Enrollment criteria were strict. Men had to be 65 or older with an average of two morning serum testosterone levels below 275 ng/dL. They also had to have symptoms in at least one of the three domains: reduced libido or erectile function, difficulty walking or climbing stairs, or low energy and fatigue.

Exclusion criteria filtered out men with prostate cancer, PSA above 4 ng/mL, severe lower urinary tract symptoms, recent cardiovascular events (MI or stroke within 3 months), uncontrolled heart failure, or hematocrit above 48%. Men already on testosterone or medications known to affect testosterone were also excluded.

This is an important detail. The enrolled population had genuine, confirmed hypogonadism with real symptoms, not the "low T" marketing demographic of middle-aged men feeling vaguely tired. The mean baseline testosterone was approximately 232 ng/dL, well below any standard threshold for deficiency.

What They Were Given

Participants were randomized 1:1 to either 1% testosterone gel (AndroGel) or an identical placebo gel applied daily to the shoulders and upper arms. Doses were titrated at months 1, 2, 3, 6, and 9 to keep testosterone in the range of 400 to 798 ng/dL, which the investigators defined as the mid-normal range for young men.

The dose adjustment protocol was well-executed. By month 3, testosterone levels in the treatment group had risen to a mean of approximately 565 ng/dL, closely matching the target. Placebo group levels stayed flat. Blinding was maintained throughout, and adherence rates exceeded 85% in both groups.

What They Measured

Each of the three primary endpoints used a validated instrument:

• Sexual function: The Psychosexual Daily Questionnaire item 4 (PDQ-Q4), which captures sexual desire, and the Derogatis Interview for Sexual Functioning (DISF-SR-II), which covers a broader range of sexual activity and satisfaction.
• Physical function: The 6-minute walk test, a standardized measure of functional exercise capacity. Secondary physical measures included stair-climbing power and lean body mass.
• Vitality: The FACIT-Fatigue scale, a 13-item questionnaire measuring fatigue intensity and impact on daily life.

The trial also collected safety data across all domains, including hematocrit, PSA, cardiovascular events, and mood.

Results: What Actually Changed

Sexual Function

This was the clearest win. Men receiving testosterone reported significantly increased sexual desire, with a treatment effect on PDQ-Q4 of 0.58 points on a 0-to-3 scale (P <0.001). Erectile function also improved. The DISF-SR-II composite showed a between-group difference of 2.64 points (P <0.001). Sexual activity frequency increased as well.

The effect sizes were moderate by clinical standards, but they were consistent across subgroups and sustained over the full 12 months. For a population in which many men had essentially given up on sexual activity, these changes were clinically meaningful.

| Outcome | Testosterone (change) | Placebo (change) | P-value | |---|---|---|---| | PDQ-Q4 (sexual desire) | +0.45 | −0.13 | <0.001 | | DISF-SR-II (sexual function composite) | +2.93 | +0.29 | <0.001 | | Sexual activity (any, %) | +5.6% | −1.2% | <0.001 |

Physical Function

The 6-minute walk test showed a statistically significant improvement. Men on testosterone walked an average of 33 meters farther at 12 months compared to baseline, while placebo-treated men improved by 7 meters. The between-group difference of roughly 26 meters was modest but reached significance.

Stair-climbing power did not show a significant difference between groups. Neither did self-reported physical function questionnaires such as the PF-10. This disconnect, where an objective test improves but subjective perception does not, is common in exercise physiology studies and suggests the walking improvement may not have been large enough for men to notice in daily life.

Vitality

Results here were the weakest. The FACIT-Fatigue score improved more in the testosterone group than placebo, but the difference was small (approximately 2 points on a 52-point scale) and the clinical significance was debatable. A secondary vitality measure, the SF-36 vitality subscale, did show a significant benefit, while the FACIT-Fatigue primary endpoint met statistical significance only marginally.

The investigators acknowledged that vitality is a diffuse symptom, harder to capture with any single instrument than sexual desire or walking capacity.

The Companion Trials Most Summaries Skip

Beyond the three primary symptom domains, four additional TTrials sub-studies reported results in subsequent publications:

• Bone Trial: Testosterone increased bone mineral density at the spine and hip, with particular improvements in estimated bone strength. Published in JAMA Internal Medicine, 2017.
• Cognitive Function Trial: No significant improvement in verbal memory or executive function. This was a negative result and an important one. Published in JAMA, 2017.
• Anemia Trial: Testosterone corrected unexplained anemia in a significant proportion of men and improved hemoglobin in men with anemia of known cause. Published in JAMA Internal Medicine, 2017.
• Cardiovascular Trial: Coronary artery plaque volume, measured by CT angiography, increased more in the testosterone group, raising a safety signal rather than demonstrating benefit. Published in JAMA, 2017.

These companion results are essential context. Testosterone helped sexual function and bones but did nothing for cognition and potentially worsened coronary atherosclerosis.

Safety Signals and What the Trial Could Not Answer

The TTrials were explicitly not designed as a safety study. With 790 men followed for one year, the trial lacked statistical power to detect differences in rare events such as heart attacks, strokes, or prostate cancer. The FDA label for testosterone products continues to carry warnings about cardiovascular risk, and the TTrials did not resolve that uncertainty.

Specific safety observations from the primary publication:

• Hematocrit: Rose above 54% in some testosterone-treated men, requiring dose reduction. Polycythemia remains the most common dose-limiting adverse effect of any testosterone formulation.
• PSA: Increased modestly in the testosterone group. No prostate cancer diagnoses occurred during the trial, but the 12-month window was far too short to be reassuring.
• Cardiovascular events: Numerically similar between groups. The coronary plaque finding from the cardiovascular sub-trial was concerning but based on a surrogate marker, not clinical events.

The larger TRAVERSE trial, published in 2023 in the NEJM, was specifically designed to answer the cardiovascular question. It enrolled over 5,000 men and found that testosterone did not increase the rate of major adverse cardiovascular events over a median follow-up of 33 months. TRAVERSE provided the reassurance that the TTrials could not.

Limitations the Authors Acknowledged

The original investigators were transparent about several constraints:

1. Duration. Twelve months is sufficient to detect symptom changes but tells us nothing about durability. Do benefits persist at year three? Year five? Unknown from this dataset.
2. Population specificity. All participants were 65 or older with confirmed low testosterone. Results should not be extrapolated to younger men, men with borderline levels, or men without symptoms.
3. Single formulation. Only transdermal gel was studied. Whether injectable testosterone enanthate, cypionate, or other formulations produce equivalent results is not answered here.
4. Effect sizes. While statistically significant, the physical function and vitality improvements were small enough that individual patients might not perceive a difference.
5. No long-term safety conclusions. The study was underpowered for cardiovascular events, prostate outcomes, and mortality.

What This Means for Clinical Practice Today

The TTrials shifted the evidence base in a specific and limited way. Before 2016, prescribing testosterone to older men was based on extrapolation from younger populations and small trials. After the TTrials, clinicians could point to a well-designed RCT confirming that testosterone gel improves sexual function in symptomatic older men with genuinely low levels.

The Endocrine Society's 2018 guidelines incorporated TTrials data when recommending testosterone therapy for men with symptomatic hypogonadism, while emphasizing shared decision-making about cardiovascular uncertainty. The subsequent TRAVERSE results have further clarified the risk-benefit conversation.

For a patient sitting in clinic, the practical message is: if you are 65 or older, have documented low testosterone, and your primary complaint is reduced sexual function, the TTrials provide solid evidence that testosterone gel is likely to help. If your main complaint is fatigue or physical weakness, the evidence for benefit is weaker. And regardless of the symptom, the decision should include monitoring for polycythemia and a frank discussion about what remains unknown.

Frequently asked questions

›

›

›

›

›

›

›

›

›

›

References

• Snyder PJ, Bhasin S, Cunningham GR, et al. Effects of Testosterone Treatment in Older Men. N Engl J Med. 2016;374(7):611-624. PubMed
• Lincoff AM, Bhasin S, Flevaris P, et al. Cardiovascular Safety of Testosterone-Replacement Therapy. N Engl J Med. 2023;389(2):107-117. PubMed
• Bhasin S, Brito JP, Cunningham GR, et al. Testosterone Therapy in Men With Hypogonadism: An Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2018;103(5):1715-1744. PubMed
• Snyder PJ, Kopperdahl DL, Stephens-Shields AJ, et al. Effect of Testosterone Treatment on Volumetric Bone Density and Strength in Older Men With Low Testosterone. JAMA Intern Med. 2017;177(4):471-479. PubMed
• Resnick SM, Matsumoto AM, Stephens-Shields AJ, et al. Testosterone Treatment and Cognitive Function in Older Men With Low Testosterone and Age-Associated Memory Impairment. JAMA. 2017;317(7):717-727. PubMed
• FDA. AndroGel (testosterone gel) prescribing information. FDA Label