Prostate

What the marker actually is

Prostate-specific antigen is an enzyme, and its purpose has nothing to do with blood. It belongs in semen, where it breaks down the proteins that make the fluid gel on ejaculation, liquefying it so sperm can move. Small amounts leak into the bloodstream normally, and more leaks whenever the gland is enlarged, inflamed or disordered.

So PSA is a marker of prostate activity, not a cancer test. That single sentence prevents most of the distress this number causes.

The most common reason for a raised PSA is benign enlargement, which affects most men eventually and is a consequence of ageing rather than a disease. Prostatitis and urinary infection also raise it, sometimes substantially. A raised PSA far more often reflects one of those than cancer. Several ordinary activities raise it temporarily as well, and the Biomarker Guide lists them.

It is androgen-driven. The gene that makes PSA is switched on by androgens acting through the androgen receptor, which is the thread connecting this section to the hormone one and the reason the next two parts of this section exist at all.

The part the name gets wrong

Prostate-specific antigen is not specific to the prostate, and it is not confined to men.

Women have a set of small glands and ducts alongside the urethra, described by Alexander Skene in the 1880s and named after him. They develop from the same embryonic tissue as the male prostate, they express the same prostate markers used in male pathology, and the term female prostate has gradually become the accepted description rather than a curiosity. A measurable serum PSA has been confirmed in a woman with benign overgrowth of these glands.

It is produced elsewhere too, in normal and abnormal breast tissue, in breast milk, and in amniotic fluid. Serum PSA in women rises in pregnancy, varies across the menstrual cycle, and runs higher where androgens are raised, which follows directly from the androgen-driven switch above.

Nobody should be testing PSA on the basis of any of that. It is here because it is one of the better examples on this page of a marker being named before it was understood. The clinical use came first and the name survived.

What you should understand before ordering it

PSA-based screening is genuinely unresolved, and the reason is specific rather than vague.

The test detects two kinds of cancer at once. Some would go on to cause serious harm. Others would never have caused any symptom or shortened any life, and would have gone unnoticed entirely. At the point of the blood test there is no way to tell which kind has been found.

That matters because the path after a raised PSA carries its own consequences. Specialist assessment, imaging, sometimes biopsy, sometimes treatment. Applying all of that to cancers that were never going to cause harm does damage without benefit. It is why the UK National Screening Committee has consistently concluded that screening the whole population would cause more harm than good. That is a considered position rather than an oversight.

That changed in one narrow respect in 2026. In June the government accepted a recommendation for England's first targeted prostate screening programme, for men aged 45 to 61 who have both a pathogenic BRCA2 variant and a family history of breast, ovarian, pancreatic or prostate cancer. A PSA test every 2 years, with rollout expected from 2027. It is a deliberately small group at genuinely elevated risk, and it does not extend to men with a family history alone, to Black men, or to anyone else. Those exclusions are contested, and further research including the TRANSFORM trial is running specifically to resolve them.

None of that makes it a bad test. It makes it a test to understand before doing rather than after, which is a different thing and a much less common message.

The numbers, and who they are for

This is the part that causes the most confusion because the rules for men with symptoms and the rules for men without them have never been the same thing, and the latter no longer exists.

For men with possible symptoms, meaning urinary symptoms, erectile dysfunction or visible blood in the urine, the National Institute for Health and Care Excellence publishes age-specific referral thresholds. For men aged 50 to 59, that threshold is a PSA above 3.5 micrograms per litre, rising with age from there.

For men without symptoms, there is no national threshold any more. That group used to be covered by the Prostate Cancer Risk Management Programme, which did set a figure. For men aged 50 to 69 it was a PSA of 3.0 micrograms per litre or above. It's important to note that this was simply for referral not a number for where most men sit. The programme was retired in August 2026 and its pages now point to general NHS advice instead, so what happens after a raised result in a man without symptoms is a clinical judgement with no national figure behind it.

The old programme also said the test was available free to any man aged 50 or over who requested it, once the implications had been talked through. That wording has gone with it. The Department of Health and Social Care position since August 2026 is that every man still has the right to ask his GP for a PSA test, but that the final decision rests with the GP, and that age 50 on its own is no longer a qualifying factor. Anyone can ask, but you're not guaranteed to get it.

Knowing which of those two situations you are in changes how any result should be read, and it is not something the number itself can tell you.

Two common medications roughly halve PSA. Finasteride and dutasteride, taken for benign enlargement or hair loss, both do this, and by the same androgen mechanism. They block the conversion of testosterone into the more potent androgen the prostate actually responds to. A result that looks reassuring on either of them may not be.

What happens next, and why it is not what it was

The overdiagnosis argument above was built in an era when a raised PSA led fairly directly to a biopsy, and a biopsy meant taking a dozen or more samples from a gland nobody could see into. That is no longer the pathway in this country, and it changes the calculation.

An MRI now comes first. NICE says to offer multiparametric MRI as the first investigation for suspected localised prostate cancer, before any biopsy, and to report it on a five-point scale. A score of 3 or more leads to a biopsy aimed at what the scan found. A score of 1 or 2 is grounds for considering no biopsy at all, as a shared decision.

That matters twice over. Targeting a biopsy at a visible lesion finds significant cancer more reliably than sampling blindly, and a proportion of men who would once have been biopsied now are not. The scan is not perfect and the guidance is careful about that, which is why omitting the biopsy is a discussion rather than a rule.

There are also blood refinements we do not sell. The free-to-total PSA ratio compares the unbound fraction against the total, and a lower proportion of free PSA shifts the odds towards cancer rather than benign enlargement. PSA density divides the PSA by the size of the gland, which needs the volume from a scan. Both are used to decide whether a borderline PSA justifies going further. A GP or urologist can arrange the free PSA, and private labs sell it.

So a raised PSA in 2026 opens a more careful pathway than the one the screening debate was originally argued over. It has not removed the overdiagnosis problem. It has made the first step after the blood test a picture rather than a needle.

Why it is on a testosterone panel, and what the evidence now says

Prostate tissue responds to androgens, and PSA monitoring is part of standard practice when someone is on testosterone replacement. That is why PSA sits in TRT Monitor and TRT Monitor Complete rather than being an optional extra on them.

Testosterone can move PSA, and the size of that movement is known. A confirmed rise of more than 1.4 micrograms per litre above baseline happened in about 1% of men on testosterone in the first year, against about 0.5% on placebo. So a meaningful rise is uncommon but real, it happens early if it happens, and a baseline taken before starting is the thing that makes any later value readable at all.

The bigger question is whether testosterone causes prostate cancer, and the evidence has moved a long way. The assumption that it does dates from the 1940s and shaped practice for decades. It has not held up.

The largest trial to examine this found no difference in high-grade prostate cancer, or in prostate cancer of any grade, between testosterone and placebo over roughly 3 years. A 2026 systematic review reached the same conclusion. In June 2026 US regulators asked for another revision to testosterone labelling, explicitly on the basis that the evidence shows less cardiovascular and prostate cancer risk than was assumed when the original warnings were written in 2015.

One limitation matters and should not be skipped. That trial excluded men at entry who had a PSA above 3.0, an abnormal examination, or a previous prostate cancer diagnosis. So it speaks to men assessed as low risk before starting, which is exactly why a proper baseline assessment is part of the picture rather than an afterthought. It does not speak to men who already have prostate cancer, although the observational work in that group has also been far more reassuring than the old assumption predicted.

Practice has lagged the evidence, and reviews published in 2026 say so directly. Men are still routinely told that testosterone causes prostate cancer as though it were settled. It is not what the current evidence shows. What has not changed is the case for monitoring. Not because a rise is expected to mean cancer, but because a baseline and a plan are what allow a genuine signal to be separated from the ordinary noise, without anyone being sent for a biopsy they did not need.

What changes when the prostate is under strain

Benign enlargement presents through the tube it surrounds: a weaker stream, hesitancy before starting, dribbling at the end, a sense of not emptying completely, and getting up at night to pass urine. Those are extremely common with age and are not cancer symptoms.

Early prostate cancer usually produces no symptoms at all, which is the entire reason a blood marker was ever of interest. Symptoms that do warrant prompt attention regardless of any blood result include blood in the urine or semen, new bone pain, and unexplained weight loss.

The limit of the test is that a PSA result cannot establish whether cancer is present. It can only say whether the prostate is releasing more of this protein than expected. Everything that answers the actual question happens elsewhere. Examination, imaging, biopsy, and none of it is something we offer.