Testosterone Therapy and Healthy Aging

Testosterone therapy is marketed to men as a way to get back energy, strength and drive. Large trials now show what it does and doesn’t do for older men, and the results are narrower than the advertising. Humanic Blue puts the trial results beside those promises, including the questions the trials haven’t settled.

In the Testosterone Trials, testosterone gel modestly improved sexual function and slightly improved mood in men 65 or older with symptoms and repeatedly low testosterone. It did not meet the main goals for energy or physical function. A separate trial, TRAVERSE, met its planned safety test for major heart and blood-vessel events in selected men with low testosterone and elevated cardiovascular risk. That does not mean it found no harms: fractures and several other adverse events occurred more often with testosterone. Longer-term prostate and overall safety remain uncertain. No trial shows that testosterone therapy helps men live longer.

What has testosterone therapy actually improved?

The Testosterone Trials, published in 2016 and funded primarily by the National Institutes of Health, assigned men age 65 or older with symptoms and an average of two testosterone readings below 275 ng/dL to testosterone gel or placebo for one year. The unit ng/dL means nanograms per deciliter, a measure of testosterone concentration in blood. Participants had to qualify for at least one of the sexual-function, physical-function or vitality trials, so these findings do not apply to asymptomatic men or to anyone with a single low reading.

The clearest finding was a modest sexual-function benefit. Mood and depressive-symptom scores improved slightly. Energy—called vitality in the trial—and walking were less tidy. The main energy outcome counted how many men reached a predefined improvement on a fatigue questionnaire; testosterone did not significantly increase that proportion. A separate analysis compared average changes in questionnaire scores. That change was small and not statistically significant within the vitality trial, but was significant when participants from all three trials were combined. A pooled walking analysis also favored testosterone. Those secondary findings are useful clues, not replacements for the missed primary outcomes. The one-year study was too small and short to establish major safety outcomes. See the outcome-by-outcome ledger.

What the safety trial found

The TRAVERSE trial enrolled men ages 45 to 80 with symptoms, two fasting testosterone levels below 300 ng/dL, and cardiovascular disease or elevated cardiovascular risk. Participants were assigned to testosterone gel or placebo. Mean treatment lasted about 22 months and mean follow-up about 33 months. AbbVie and other organizations funded the trial, and several authors were company employees.

Cardiovascular outcome. The prespecified composite of cardiovascular death, nonfatal heart attack or nonfatal stroke occurred in 7.0% of the testosterone group and 7.3% of the placebo group. The trial asked whether testosterone increased that combined event rate beyond a safety margin set in advance. It stayed within the margin—the finding called noninferiority. That does not establish benefit, rule out smaller risk increases, or establish safety in lower-risk men, with other formulations or doses, or over longer periods. Atrial fibrillation (an irregular heart rhythm), acute kidney injury (a sudden decline in kidney function) and pulmonary embolism (a clot blocking a blood vessel in the lungs) were observed more often with testosterone. Those were possible safety concerns outside the trial’s main cardiovascular comparison. See Testosterone and Aging.

Fractures. Better bone density did not translate into fewer broken bones. In a prespecified TRAVERSE fracture substudy of about 5,200 participants followed for a little over three years, clinical fractures occurred in 3.5% with testosterone and 2.5% with placebo. The difference was statistically significant, but the trial did not determine why.

Prostate. In a TRAVERSE prostate analysis, high-grade prostate cancer—cancer with features associated with more aggressive behavior—was uncommon in both groups. The estimate was imprecise and not statistically significant. Other measured prostate events also did not differ significantly, while prostate-specific antigen (PSA), a protein measured in blood as one part of prostate assessment, rose more with testosterone. The trial excluded men whose PSA was above 3.0 nanograms per milliliter (ng/mL) or who had severe urinary symptoms. Its duration was too short to settle prostate-cancer risks that may take years to emerge.

How does cardiovascular noninferiority differ from a finding of benefit or absence of all harms?

Who do the guidelines say qualifies for treatment?

The American College of Physicians’ 2020 guideline, endorsed by the American Academy of Family Physicians, addresses men with age-related low testosterone. It advises clinicians to:

• discuss treatment only when sexual dysfunction is present and the person wants to improve sexual or erectile function;
• reassess within 12 months and periodically thereafter, stopping if sexual function does not improve; and
• not start testosterone to improve energy, vitality, physical function or cognition.

The Endocrine Society reiterated in July 2026 that diagnosis requires compatible symptoms plus consistently low, accurately measured testosterone, generally confirmed with at least two early-morning fasting tests. Symptoms alone are not diagnostic, and reversible contributors should be considered.

The FDA’s 2025 class-wide changes removed boxed cardiovascular-risk language after TRAVERSE, retained the limitation for age-related hypogonadism (low testosterone production) and added blood-pressure warnings. In April 2026, the FDA invited manufacturers to seek a possible new indication for low libido with idiopathic hypogonadism (low testosterone without an identified cause), but emphasized that a supplemental approval would still require adequate evidence. As of that announcement, approved products remained indicated for specific forms of hypogonadism tied to known structural or genetic causes. See What Are the Symptoms of Low Testosterone?.

Which eligibility limits belong beside a claim about testosterone therapy's benefits?