Does Testosterone Cause Prostate Cancer? What the Evidence Now Shows

A man of 58 sits in a consulting room with a total testosterone of 7.9 nmol/L, two morning samples, and a symptom list that has been building for three years. He asks about testosterone replacement. The doctor shakes his head and says the thing that ends most of these conversations: we do not want to feed a prostate cancer.

He leaves untreated. No cancer has been found. None was looked for. The word was enough.

That exchange happens every week, and it rests on a belief that has been the settled wisdom of medicine for eight decades. The belief is not built on nothing. It is built on something remarkably small.

Where the fear came from

In 1941, Charles Huggins and Clarence Hodges published work showing that castration caused advanced prostate cancer to regress, and that giving testosterone to men with metastatic disease made it worse. The finding was genuinely important. Huggins received a Nobel Prize in 1966, and androgen deprivation remains a cornerstone of treating advanced prostate cancer to this day.

The part that travelled less well is how thin the second half of that observation was. The evidence that testosterone worsened the disease rested on a very small number of men, most of whom had already been castrated, and in the clearest documented case, on a single patient.

From that, a general rule was built: testosterone feeds prostate cancer, therefore raising testosterone in any man is dangerous. The logic sounds airtight. It is also the wrong shape, and understanding why is the whole of this article.

The saturation model

Here is the observation that broke the old rule.

If testosterone drove prostate growth in a straight line, men with the highest levels would have the most prostate cancer and men with the lowest would have the least. Study after study has failed to find that relationship. Large pooled analyses of prospective data have found no association between a man's natural testosterone level and his subsequent risk of developing prostate cancer.

The explanation is a mechanism that sounds obvious once stated. Prostate tissue depends on androgen, but its androgen receptors become fully occupied at a relatively low concentration. Below that point, adding testosterone does drive growth. Above it, the receptors are already saturated, and adding more changes very little, because there is nothing left to bind to.

This is the saturation model, described most prominently by Abraham Morgentaler and Abdulmaged Traish. The saturation point sits low, somewhere around the bottom of the normal range. Almost every man being considered for replacement therapy is already at or below it, and moving him from deficient to normal moves him across a part of the curve where the effect is real. Moving a normal man higher moves him along a flat line.

It explains both facts at once. Castration causes regression, because it drops a man far below saturation. Yet raising a hypogonadal man to normal does not cause cancer, because he is not being pushed into unexplored territory. He is being returned to where most men already live.

What the largest trial actually found

The saturation model was a good theory that needed a large, deliberate test. It got one.

The TRAVERSE trial randomised 5,246 men aged 45 to 80 with hypogonadism and either existing cardiovascular disease or a high risk of it, to transdermal testosterone gel or placebo, for a mean of 27.1 months. It is the largest and longest randomised trial of testosterone therapy conducted.

Prostate safety was not an afterthought in TRAVERSE. High-grade prostate cancer, meaning a Gleason score of 4+3 or higher, was a formally adjudicated endpoint with its own analysis.

The results:

• Any prostate cancer: 12 cases in the testosterone group, 11 in the placebo group, out of more than 5,200 men
• High-grade prostate cancer: 5 on testosterone, 3 on placebo
• Urinary symptoms: no worsening compared with placebo
• Acute urinary retention and surgery for benign enlargement: low and similar in both groups

Twenty-three cancers across five thousand men over more than two years, split almost evenly. The numbers are small enough that the difference of one or two cases carries no statistical weight, and that is precisely the point. If testosterone were driving prostate cancer in these men, a trial of this size and length was where it would have shown.

It is worth being exact about what this does and does not settle, because overstating it would repeat the original error in the opposite direction.

Four things this does not mean

It does not mean testosterone is safe for a man with active prostate cancer. TRAVERSE screened those men out. Androgen deprivation remains standard treatment for advanced disease, and nothing here changes that.

It does not mean screening can be skipped. The men in TRAVERSE were assessed before entry and monitored throughout. The low event rate is a result achieved with screening, not evidence that screening is unnecessary.

It does not apply to men who are not hypogonadal. Every man in the trial had two fasting testosterone measurements below 10.4 nmol/L. This is evidence about correcting a deficiency, not about pushing normal men higher. That distinction runs through everything on this site, and what actually happens at high testosterone covers the other side of it.

It does not mean testosterone prevents prostate cancer either. Some men with low testosterone appear to have a reduced risk of prostate cancer death, and some studies have found that low testosterone is associated with more aggressive disease at diagnosis. That is interesting, and it is not the same as a protective effect.

What a sensible assessment looks like before starting

The practical question is not whether testosterone causes cancer. It is how to start treatment in a way that finds any cancer that is already there.

A baseline PSA before starting, alongside an assessment of risk factors. Age, family history and ethnicity all matter, and a man's baseline number matters more than a population threshold.

An honest risk conversation. A man with a father and brother diagnosed young is in a different position from a man with no family history, and should be assessed accordingly.

Urinary symptoms recorded. Not because testosterone is expected to worsen them, but so that any change has something to be measured against.

Deferral, not refusal, if something is unclear. An unexplained PSA rise is a reason to investigate before starting, not a reason to close the conversation permanently.

The wider decision about whether to start at all is a separate question, and TRT versus natural optimisation covers it. The TRT guide covers what treatment involves in practice.

Monitoring once treatment has started

PSA rises modestly in the first months of testosterone therapy in many men. This is expected. The prostate has been running on a low androgen supply and is returning to its normal state, and a small rise reflects that rather than a developing cancer.

What matters is the pattern rather than any single reading.

• Repeat PSA within the first three to twelve months, then according to age and risk
• Judge the trajectory, not the isolated number. A rise that continues climbing is different from one that settles
• Investigate confirmed, sustained rises rather than reacting to one reading, which can be raised by infection, recent ejaculation, cycling or examination
• Routine digital rectal examination is no longer considered necessary in the absence of another clinical reason, a position TRAVERSE's prostate data strengthened considerably

None of this is unusual monitoring. It is the same vigilance that any man of that age warrants, applied consistently. TRT side effects covers the rest of what needs watching, most of it more likely to matter day to day than the prostate.

Who genuinely should be cautious

The evidence has moved, but it has not moved to the point where testosterone is appropriate for everyone.

Men with untreated prostate cancer. This remains a clear contraindication outside specialist settings.

Men treated for prostate cancer. A growing body of work suggests testosterone can be considered in selected men after successful treatment, and some men on active surveillance have been treated in specialist centres. This is genuinely specialist territory, requires urological oversight, and is not a decision for a general practice appointment or an online clinic.

Men with an unexplained raised PSA. Investigate first.

Men with a strong family history. Not excluded, but assessed more carefully and monitored more closely.

The distinction that matters is between caution and refusal. Caution means testing, monitoring and shared decisions. Refusal means a man with genuine hypogonadism goes untreated for a cancer he does not have.

The cost of the old belief

It is easy to treat unnecessary caution as harmless. It is not.

Untreated hypogonadism is not a cosmetic problem. It carries measurable effects on bone density, muscle mass, metabolic health, mood and sexual function, sustained over years. A man refused treatment on prostate grounds is not returned to neutral. He is left with a condition that continues.

For four decades the trade seemed reasonable, because the alternative was thought to be cancer. The evidence no longer supports that trade in men without prostate cancer, and the conversation should have changed by now.

If you are having this conversation, the useful thing to bring is not an argument. It is your numbers: two morning testosterone results, LH, a baseline PSA and your family history. The blood test guide covers what to ask for, and how to read the results covers interpreting them.

Common questions

Does testosterone therapy cause prostate cancer?

The largest randomised trial found 12 cases on testosterone and 11 on placebo among more than 5,200 men over an average of 27 months, with no increase in high-grade disease. Current evidence does not support the belief that correcting a deficiency causes prostate cancer.

Why did doctors believe it for so long?

Because of genuinely important 1941 work showing that castration caused advanced prostate cancer to regress and that testosterone worsened it in men who already had metastatic disease. That finding was extrapolated into a general rule it could not support, and the extrapolation went unchallenged for decades.

What is the saturation model?

Prostate androgen receptors become fully occupied at a relatively low testosterone concentration. Below that point, more testosterone drives growth. Above it, adding more has little further effect, because the receptors are already saturated.

Will my PSA go up on TRT?

Often slightly, in the first months, and that is expected. What matters is whether it settles or keeps climbing. A confirmed, sustained rise needs investigating; a single reading does not.

Can I have TRT if I have had prostate cancer?

Sometimes, in selected men after successful treatment, under urological supervision. It is not a routine decision and should not be made outside specialist care.

Should I still have my prostate checked on TRT?

Yes. A baseline PSA before starting and repeat testing during the first year, then according to age and risk. The low event rates in trials were achieved with monitoring in place, not instead of it.

Sources: Huggins C, Hodges CV, Studies on prostatic cancer: the effect of castration, of estrogen and of androgen injection on serum phosphatases in metastatic carcinoma of the prostate, Cancer Res 1941;1:293-7 · Morgentaler A, Traish AM, Shifting the paradigm of testosterone and prostate cancer: the saturation model and the limits of androgen-dependent growth, Eur Urol 2009;55(2):310-20 · Lincoff AM et al., Cardiovascular safety of testosterone-replacement therapy, N Engl J Med 2023;389(2):107-17 · Bhasin S et al., Prostate safety events during testosterone replacement therapy in men with hypogonadism: a randomised clinical trial, JAMA Netw Open 2023;6(12):e2348692 · Roddam AW et al., Endogenous sex hormones and prostate cancer: a collaborative analysis of 18 prospective studies, J Natl Cancer Inst 2008;100(3):170-83 · Bhasin S et al., Testosterone therapy in men with hypogonadism: an Endocrine Society clinical practice guideline, J Clin Endocrinol Metab 2018;103(5):1715-44 · Educational only, not medical advice. Decisions about testosterone therapy and prostate screening should be made with your doctor.

Keep reading: TRT side effects · TRT before and after · The TRT guide · TRT versus natural optimisation · Blood test: what to check · Free 2-minute testosterone test