Prostate cancer screening has been argued about more bitterly than any other cancer screening, and the argument is not really about whether PSA detects cancer. It does. The argument is about what happens next, and for a long time what happened next was bad: an elevated PSA led straight to a systematic biopsy, the biopsy found a small cancer that would never have caused harm, and the man was treated with surgery or radiation and left with the incontinence and erectile dysfunction that treatment causes. The cancer was real. The benefit was not.
Almost everything about that sequence has changed. Understanding what changed is what allows screening to be done properly now.
What PSA actually measures
Prostate specific antigen is made by prostate tissue, not by prostate cancer specifically. Anything that increases the amount of prostate tissue, or disturbs it, raises the number. An enlarged prostate raises it. Prostatitis raises it, sometimes dramatically. Recent ejaculation raises it modestly. A long bicycle ride raises it. Urinary retention and instrumentation raise it. Finasteride and dutasteride roughly halve it.
This is why a single number in isolation is close to meaningless, and why the reflex of treating 4.0 as a threshold has caused so much trouble. A PSA of 5 in a man with a 90-gram prostate is a different finding from a PSA of 5 in a man with a 30-gram prostate. A PSA that has moved from 1.2 to 3.0 over two years is more concerning than one that has sat at 4.5 for a decade. Density — the PSA relative to gland volume — and the trajectory over time carry far more information than the absolute value, and both require that the test be repeated and the gland measured rather than a single result acted upon.
Before anything is done about a raised PSA
Repeat it, after a few weeks, having avoided ejaculation and cycling for two days, and with any infection treated. A meaningful share of elevated results are lower on repeat, and a man whose first elevated PSA leads directly to a biopsy has skipped the cheapest and safest step in the pathway.
What MRI changed
The old pathway was PSA, then a systematic biopsy of twelve random locations. That approach both missed significant cancers, because it sampled blindly, and found insignificant ones, because random sampling of an organ full of indolent disease finds indolent disease.
Multiparametric MRI before biopsy changed the arithmetic. A high-quality MRI reported on the standard five-point scale identifies the areas that look suspicious, and biopsies can then be targeted there rather than scattered. Trials of this approach found that MRI-first pathways detect more clinically significant cancer and fewer clinically insignificant cancers than systematic biopsy, and allow a proportion of men with reassuring imaging to avoid biopsy altogether. That is an unusual result in medicine: better detection of what matters and less detection of what does not, at the same time.
There are conditions attached. The quality of the MRI and the experience of the radiologist reading it matter enormously, and a negative MRI does not reduce the risk to zero — it reduces it enough to justify continued monitoring instead of immediate biopsy in appropriately selected men. Our approach to imaging and biopsy is set out on prostate MRI and biopsy.
Tests between PSA and biopsy
Several tests exist to refine the decision when PSA is mildly elevated. Free PSA percentage, the Prostate Health Index, 4Kscore, and urine-based tests including ExoDx all attempt to estimate the likelihood that a biopsy would find significant disease. None replaces judgment, and none is free — some are poorly covered by insurance. Used selectively in genuinely borderline cases, they help a man avoid a biopsy he did not need. Used reflexively on every mildly raised PSA, they add cost without changing decisions.
Who should be screened, and when
Screening is a decision, not a default, and the honest version of that conversation includes the fact that the benefit is modest and delayed while the harms are immediate. For most men, a discussion beginning in the mid-forties to fifty is reasonable, with testing every one to two years depending on the baseline. A man whose PSA at forty-five is very low has a low risk over the following decade and does not need annual testing.
Earlier and more careful screening is warranted for Black men, whose incidence and mortality are substantially higher and who are under-represented in the trials that shaped the guidelines; for men with a father or brother diagnosed, particularly if diagnosed young; and for men carrying BRCA2 or other known pathogenic variants, where the disease tends to be more aggressive. Genetic testing is worth discussing where the family history is striking, including breast, ovarian and pancreatic cancer on either side.
At the other end, screening a man whose remaining life expectancy is under ten years will usually cause more trouble than it prevents, because prostate cancer detected at that point is unlikely to become the thing that shortens his life. That is not ageism; it is the same logic that makes screening worthwhile at fifty-five. Our longevity programme exists partly because estimating that horizon properly, rather than by birthday alone, makes these decisions better.
Finding cancer does not mean treating it
This is the part that most reduces the harm of screening, and it is the part patients are least often told in advance. A large share of the prostate cancer found by screening is low-grade disease that can be monitored rather than treated. Active surveillance — repeat PSA, repeat imaging, repeat biopsy on a schedule, with treatment held in reserve for the minority who progress — has excellent long-term outcomes in appropriately selected men and avoids the side effects of treatment entirely.
Knowing this before you are tested changes the meaning of the test. Screening does not commit you to surgery. It gives you information, and a good part of the time the right response to that information is careful observation.
Questions patients ask
What PSA number should worry me?
There is no single number. The same value means different things depending on your prostate size, your age, your previous results and whether you are taking finasteride or dutasteride. A rising trend and a high PSA relative to gland volume matter more than crossing any particular threshold, which is why a repeat test and a measurement of the gland come before anything else.
Do I still need a digital rectal examination?
It adds little to PSA as a screening test on its own, and its usefulness is limited. It remains worth doing as part of an evaluation because an abnormal-feeling gland is a finding in its own right even when PSA is normal.
Can I have an MRI instead of a biopsy?
Sometimes, at least for now. An MRI before biopsy is the better pathway, and men whose MRI is reassuring and whose other risk factors are low can often be monitored rather than biopsied. A reassuring MRI lowers the risk substantially but does not eliminate it, so it leads to continued surveillance rather than discharge.
My father had prostate cancer. When should I start?
Earlier — commonly around forty to forty-five — and the conversation should include whether genetic testing is worthwhile, particularly if he was diagnosed young or if there is breast, ovarian or pancreatic cancer in the family. A BRCA2 variant in particular changes both screening and, if cancer is found, treatment.
If you find cancer, will I need surgery?
Often not. A large proportion of screen-detected prostate cancer is low grade and suitable for active surveillance, where it is monitored with testing and imaging and treated only if it progresses. Knowing that beforehand is part of deciding whether to be screened at all.
Make an appointment
Manhattan (212) 991-9991 · Forest Hills (718) 360-9550 · same-week visits and most major insurance accepted.