Section Structure
Highlights
Background and Unmet Clinical Need
Study Design and Methods
Key Results
Clinical Interpretation
Limitations and Generalizability
Implications for Practice and Research
Funding and Trial Registration
References
Highlights
In a randomized, placebo-controlled, double-blind phase 2 trial, 12 weeks of testosterone cypionate improved sexual activity in prostate cancer survivors with symptomatic hypogonadism after radical prostatectomy.
Testosterone therapy also improved sexual desire, negative affect, body composition, loaded stair-climbing power, and peak aerobic performance, while erectile function did not significantly change.
No participant in either study group developed biochemical recurrence, defined as PSA 0.2 ng/mL or greater, during the 12-week intervention period.
The findings are clinically provocative but should be interpreted as proof of concept. The trial was not large enough or long enough to determine long-term oncologic safety.
Background and Unmet Clinical Need
Testosterone replacement therapy has long occupied a difficult space in prostate cancer survivorship care. On one hand, symptomatic hypogonadism is common in aging men and can substantially impair libido, energy, mood, body composition, exercise capacity, and overall quality of life. On the other hand, testosterone has historically been viewed through the lens of prostate cancer biology, and many clinicians have hesitated to prescribe it to men with a prior history of prostate cancer because of concern about stimulating residual malignant cells.
This tension is especially relevant for men with low-risk, organ-confined prostate cancer who undergo radical prostatectomy and then have durable biochemical remission. Many of these patients live for years or decades after treatment. A subset develops or continues to experience biochemical hypogonadism with symptoms such as reduced libido, erectile difficulties, fatigue, and diminished well-being. Yet guideline recommendations have remained cautious, largely because randomized trial data have been lacking. Most available evidence has consisted of retrospective series, small prospective cohorts, and expert opinion.
The trial by Bhasin and colleagues directly addresses this evidence gap. Rather than asking whether testosterone is universally safe after prostate cancer, the investigators studied a highly selected population: men with organ-confined, low-grade disease, treated with radical prostatectomy, with undetectable PSA for at least 2 years, and with documented symptomatic hypogonadism. That narrow design is important because it aligns the scientific question with a realistic clinical scenario in which the potential quality-of-life benefits of treatment may justify careful re-evaluation of long-standing assumptions.
Study Design and Methods
This was a randomized, placebo-controlled, double-blind, parallel-group phase 2 clinical trial conducted at 2 academic medical centers. Men aged 40 years and older were eligible if they had previously undergone radical prostatectomy for organ-confined, low-grade prostate cancer, specifically Gleason score 6 (3 + 3) or 7 (3 + 4), had an undetectable PSA for at least 2 years, and had a mean of 2 testosterone values below 275 ng/dL along with symptoms of hypogonadism such as low libido, erectile dysfunction, or fatigue.
Participants were assigned through concealed block randomization, stratified by age group (40 to 60 years or older than 60 years) and by phosphodiesterase-5 inhibitor use. The intervention was testosterone cypionate 100 mg administered intramuscularly once weekly for 12 weeks, compared with placebo injections on the same schedule.
The first participant was randomized on May 13, 2019, and the last study visit was completed on May 16, 2025. A total of 136 men underwent randomization, with 68 assigned to testosterone cypionate and 68 assigned to placebo. Overall, 125 completed the study.
The primary efficacy endpoint was sexual activity. Secondary endpoints included sexual desire, erectile function, well-being, body composition, aerobic capacity, and physical function. The prespecified safety endpoint was biochemical recurrence, defined as PSA 0.2 ng/mL or greater.
The trial design deserves emphasis for several reasons. First, placebo control and blinding reduce expectancy effects, which are particularly relevant when studying symptoms such as libido, fatigue, and subjective well-being. Second, stratification by age and PDE5 inhibitor use helped limit confounding in sexual function outcomes. Third, by focusing on men with at least 2 years of undetectable PSA after surgery, the investigators enriched for a population at lower short-term risk of recurrence, which was necessary for an ethically acceptable first trial.
Key Results
Baseline Characteristics
The groups were balanced at baseline. Mean age was 68.6 years, with a standard deviation of 6.5 years. Fifty-two participants, or 38%, had Gleason 6 disease, and 84, or 62%, had Gleason 7 disease. This distribution reflects a clinically common post-prostatectomy survivor population rather than an ultra-low-risk cohort alone.
Primary Efficacy Outcome: Sexual Activity
The primary endpoint favored testosterone therapy. Testosterone cypionate significantly increased sexual activity more than placebo after adjustment for PDE5 inhibitor use and age. The between-group difference was 0.91 daily events, with a 95% confidence interval of 0.56 to 1.26, and a P value below .001.
This is a meaningful signal. In hypogonadism studies, improvements in laboratory testosterone values do not always translate into changes in real-world behavior. Here, the primary endpoint captured functional lived experience rather than a biomarker alone. That distinction strengthens the clinical relevance of the findings.
Secondary Sexual and Psychosocial Outcomes
Testosterone treatment improved sexual desire compared with placebo. It also improved the sexual domain score on a prostate cancer quality-of-life measure and reduced negative affect more than placebo. These findings suggest that the benefit was not limited to sexual frequency alone, but extended to motivational and emotional dimensions that often matter most to patients.
However, erectile function did not significantly change. This is clinically plausible. Erectile dysfunction after radical prostatectomy is multifactorial and often largely neurovascular in origin. Even if testosterone improves desire and overall sexual engagement, it may not overcome post-surgical structural or nerve-related limitations, especially over only 12 weeks. This is a useful counseling point: in this population, testosterone may improve libido and sexual activity without necessarily restoring erectile mechanics.
Body Composition and Physical Performance
Beyond sexual outcomes, testosterone therapy improved body composition, loaded stair-climbing power, and peak aerobic performance measured as VO2 peak. These effects broaden the potential relevance of treatment, because symptomatic hypogonadism often presents as a whole-body syndrome rather than a purely sexual complaint. Gains in physical function and aerobic capacity may be especially important in older cancer survivors, in whom sarcopenia, fatigue, and reduced exercise tolerance can accelerate functional decline.
Although the abstract does not provide numeric effect sizes for all secondary endpoints, the directional consistency across body composition and performance measures supports a biologically coherent treatment effect. Testosterone is known to influence lean mass, muscle performance, and exercise capacity in hypogonadal men, and the results observed here align with that mechanistic expectation.
Safety: Biochemical Recurrence
No participant in either group experienced biochemical recurrence during the trial. This was defined as PSA 0.2 ng/mL or greater. This result will understandably attract the greatest attention, because safety concerns have been the major barrier to broader clinical acceptance of testosterone use after prostate cancer treatment.
Still, the safety finding must be interpreted with restraint. A zero-event result over 12 weeks in 136 randomized participants is reassuring in the short term, but it does not establish long-term oncologic safety. Biochemical recurrence can occur years after surgery, and the trial was explicitly not powered or designed to detect differences in clinical recurrence, metastasis, cancer-specific mortality, or overall survival.
Clinical Interpretation
This study is important because it replaces speculation with randomized evidence in a field previously dominated by retrospective reassurance and theoretical concern. The results support the view that, in carefully selected men with low-grade prostate cancer successfully treated by radical prostatectomy and with sustained undetectable PSA, short-term testosterone replacement can improve symptoms and function without an immediate PSA signal.
The study also fits with evolving biological thinking. The traditional fear that any increase in testosterone will necessarily stimulate prostate cancer growth has been challenged by the saturation model, which proposes that prostate tissue response to androgens plateaus once androgen receptors are sufficiently occupied. While that model remains debated and should not be overextended, it offers a framework that helps explain why restoring testosterone from hypogonadal to physiologic ranges may not inevitably produce aggressive prostate stimulation in all contexts.
For the practicing clinician, the main message is not that testosterone is now broadly safe after any prostate cancer. Rather, it is that blanket contraindications may be increasingly difficult to justify in highly selected post-prostatectomy survivors with low-risk disease, persistent biochemical remission, and clinically significant hypogonadal symptoms. Shared decision-making remains essential. Patients should be counseled that the symptomatic benefits appear real, but long-term cancer safety is still uncertain.
The absence of effect on erectile function is also instructive. Men considering therapy after prostatectomy should not be told to expect testosterone to reverse surgery-related erectile dysfunction. More realistic goals would include improved libido, greater sexual engagement, better mood, improved energy or physical performance, and favorable body composition changes. In men with persistent erectile dysfunction, concomitant strategies such as PDE5 inhibitors, vacuum erection devices, intracavernosal therapy, or penile rehabilitation may still be needed.
Limitations and Generalizability
The limitations are substantial and should frame any interpretation.
First, follow-up was short. Twelve weeks is adequate for a proof-of-concept efficacy signal but inadequate for oncologic reassurance. Prostate cancer recurrence risk unfolds over years, not months.
Second, the sample size was modest. Even if no biochemical recurrences were observed, the trial cannot exclude a clinically important increase in rare or delayed events.
Third, the population was highly selected. The findings do not apply to men with high-grade prostate cancer, those treated with radiation therapy, those with prior androgen deprivation therapy, or those without sustained undetectable PSA after radical prostatectomy. The authors explicitly note these boundaries.
Fourth, the intervention was intramuscular testosterone cypionate at a fixed dose of 100 mg weekly. The results should not automatically be generalized to other formulations, dosing schedules, or treatment durations.
Fifth, the abstract does not detail adverse events beyond biochemical recurrence. Broader testosterone safety considerations, including erythrocytosis, fluid retention, sleep apnea exacerbation, or cardiovascular events, require attention in any real-world treatment program, especially in older men.
Finally, the primary endpoint, while clinically meaningful, focused on sexual activity over a relatively brief period. Longer studies are needed to determine whether benefits are sustained, whether treatment satisfaction remains high, and whether symptomatic gains justify ongoing monitoring burdens and uncertain long-term risk.
Implications for Practice and Research
This trial is best viewed as a rational starting point for a larger, longer-term safety and effectiveness program. It establishes feasibility, provides a signal of symptomatic benefit, and suggests that a more definitive trial is ethically and scientifically justified.
In current practice, the study may encourage experienced clinicians to consider testosterone treatment in exceptional cases after radical prostatectomy, but only with careful patient selection, explicit discussion of evidence gaps, and close PSA monitoring. It should not be used to support indiscriminate prescribing.
Future studies should address several unanswered questions. Longer follow-up is essential to evaluate biochemical recurrence trajectories, imaging recurrence, metastasis, and survival. Broader populations should be studied, while still respecting risk stratification. Comparative data across testosterone formulations would be helpful. It will also be important to identify which symptomatic domains predict greatest benefit, and whether baseline factors such as time since prostatectomy, nerve-sparing status, obesity, or severity of hypogonadism modify response.
From a survivorship perspective, the study also highlights an important principle: cancer cure and symptom recovery are not the same. As more men live long after treatment for localized prostate cancer, survivorship care must address endocrine, sexual, metabolic, and functional outcomes with the same seriousness given to recurrence surveillance.
Funding and Trial Registration
ClinicalTrials.gov identifier: NCT03716739.
The abstract provided does not specify funding details. Readers should consult the full JAMA Internal Medicine publication for funding source, sponsor role, conflict-of-interest disclosures, and protocol details.
Conclusion
Among men with symptomatic hypogonadism who had undergone radical prostatectomy for organ-confined, low-grade prostate cancer and maintained undetectable PSA for at least 2 years, 12 weeks of testosterone cypionate improved sexual activity, sexual desire, mood-related outcomes, body composition, physical function, and aerobic performance compared with placebo. No biochemical recurrences occurred during the study period.
These findings are clinically important because they challenge a longstanding therapeutic taboo with randomized evidence. At the same time, they should be interpreted exactly as the investigators intended: as proof of concept, not as definitive reassurance about long-term cancer safety. The study moves the field forward, but it does not close the case. For now, testosterone therapy after prostatectomy remains a highly individualized decision requiring careful selection, informed consent, and vigilant monitoring.
References
1. Bhasin S, Burnett AL, Gagliano-Jucá T, Storer TW, Reid KF, Shang YV, Privat F, Chandra MS, Weiss M, Memish-Beleva Y, Lam H, Kibel AS, Pencina KM. Testosterone Treatment in Prostate Cancer Survivors With Hypogonadism: A Randomized Clinical Trial. JAMA Internal Medicine. 2026-05-11. PMID: 42113507. https://pubmed.ncbi.nlm.nih.gov/42113507/
2. Mulhall JP, Trost LW, Brannigan RE, et al. Evaluation and Management of Testosterone Deficiency: AUA Guideline. Journal of Urology. 2018;200(2):423-432.
3. Bhasin S, Brito JP, Cunningham GR, et al. Testosterone Therapy in Men With Hypogonadism: An Endocrine Society Clinical Practice Guideline. Journal of Clinical Endocrinology and Metabolism. 2018;103(5):1715-1744.
4. Morgentaler A, Traish AM. Shifting the Paradigm of Testosterone and Prostate Cancer: The Saturation Model and the Limits of Androgen-Dependent Growth. European Urology. 2009;55(2):310-320.
