Low Testosterone Linked to 60% Higher Risk of Prostate Cancer Progression on Surveillance

University of Texas MD Anderson Cancer Center

For decades, the relationship between testosterone and prostate cancer has seemed straightforward: testosterone fuels the disease, so less testosterone should mean less cancer growth. That logic underpins androgen deprivation therapy, a cornerstone treatment for advanced prostate cancer. But a new study from MD Anderson Cancer Center suggests the relationship is more complicated than the textbooks imply, at least in early-stage disease.

The counterintuitive finding

Researchers led by Justin Gregg, an associate professor of urology at MD Anderson, examined data from more than 900 men undergoing active surveillance for prostate cancer. Active surveillance is the standard approach for low-risk prostate cancer: rather than immediate treatment, patients undergo regular monitoring with biopsies and imaging, with treatment initiated only if the cancer shows signs of becoming more aggressive.

The study found that men with low baseline testosterone levels (300 ng/dL and lower) had a 60% higher likelihood of their cancer progressing to Grade Group 3 or higher, which represents significantly more aggressive disease. This association held even after the researchers controlled for age, prostate-specific antigen (PSA) levels, body mass index (BMI), and tumor density and size.

That is the opposite of what the simple "testosterone feeds cancer" model would predict.

Why the conventional wisdom may be wrong for early-stage disease

The apparent contradiction dissolves when you consider that advanced prostate cancer and early-stage prostate cancer may have fundamentally different relationships with hormones. Androgen deprivation therapy works in advanced disease because those cancers have become dependent on testosterone signaling for growth. Remove the testosterone, and the cancer slows.

But in early-stage, indolent prostate cancer, the dynamics may be different. Low testosterone is associated with metabolic syndrome, obesity, chronic inflammation, and other systemic conditions that could independently promote cancer progression through non-hormonal pathways. The tumor microenvironment in a man with low testosterone may differ in ways that favor aggressive behavior, even though the traditional view would suggest the opposite.

The study does not resolve the mechanistic question. It shows an association, not a causal pathway. But the size of the association, 60% increased risk after adjusting for confounders, is difficult to dismiss as noise.

Practical implications for surveillance decisions

If confirmed in further studies, baseline testosterone could serve as an additional clinical marker to help physicians stratify risk among patients choosing active surveillance. Currently, decisions about surveillance intensity rely primarily on PSA levels, biopsy results, and imaging. Adding testosterone to that toolkit could help identify patients who warrant closer monitoring or earlier intervention.

"Active surveillance is a safe and effective option for many men with early-stage prostate cancer," Gregg said. "However, identifying which patients may be more likely to experience progression remains a key challenge."

The emphasis on surveillance remaining safe and effective is important context. The study does not suggest that active surveillance is inadequate or that men with low testosterone should be pushed toward immediate treatment. It suggests that testosterone levels could help refine monitoring strategies, not replace them.

What this study cannot tell us

Several significant limitations constrain the conclusions. The study was retrospective, meaning the researchers analyzed existing clinical data rather than prospectively following patients randomized to different monitoring protocols based on testosterone levels. Retrospective designs are susceptible to confounding from unmeasured variables.

The study demonstrates an association between low testosterone and cancer progression but does not establish causation. Men with low testosterone may share other characteristics, metabolic, genetic, or lifestyle-related, that independently contribute to cancer progression. While the researchers controlled for several known confounders, residual confounding is always possible in observational studies.

Testosterone levels fluctuate throughout the day and can be affected by acute illness, medications, sleep, and stress. A single baseline measurement may not accurately represent a patient's typical hormonal state. The threshold of 300 ng/dL, while commonly used to define low testosterone, is somewhat arbitrary and may not represent the biologically meaningful cutpoint for cancer progression.

The study also does not address whether testosterone supplementation in men with low levels would affect cancer progression, a question with obvious clinical relevance but one that would require a very different study design to answer safely.

A shift in how we think about hormones and early cancer

The broader significance lies in challenging a simplistic model. The relationship between testosterone and prostate cancer is not a single-variable equation. Disease stage, tumor biology, systemic metabolic health, and the hormonal environment interact in ways that the conventional "more testosterone equals more cancer growth" framing does not capture.

Future prospective studies will need to confirm whether testosterone levels add meaningful predictive value beyond existing markers and whether incorporating this information into surveillance protocols actually improves patient outcomes.