დასაწყისი · კატალოგი · პროფილი · ცხრილი
სკრინინგი BODY HANDBOOK
სკრინინგი · §136
PSA Screening for Prostate Cancer
A PSA blood test catches some prostate cancers early enough to matter — and finds many more that would never have hurt you. That's why every serious guideline says talk it through with a doctor before just running it: the harms are concrete (incontinence and impotence after treating a cancer that wouldn't have killed you) and the benefit is real but small. Done well — multi-round testing, an MRI before any biopsy, active monitoring for low-risk findings — it cuts prostate-cancer deaths by roughly a fifth in men 55 to 69. The math runs differently for Black men and BRCA2 carriers, who start earlier and gain more.
Decide · Yearly Evidence Moderate თავი სკრინინგი

The strongest signal: a modest cut in prostate-cancer deaths — about one prevented per 600 men screened over 13 years — with no measurable bump in overall survival. The cost is one blood draw a year or two and a real conversation. The field is split, and what you do with an elevated number matters more than whether you test.

Your prostate makes a protein called PSA. A small amount leaks into your blood normally; more leaks in when something in there is off — could be cancer, but could also be a benign enlargement that comes with age, an infection, last Saturday's bike ride, or sex the night before the draw. That's why no single PSA number is a verdict on its own. A reading above 4 ng/mL has been the rule-of-thumb cut-off for decades, but a large prevention-trial cohort biopsied men below that threshold and still found prostate cancer in roughly 15% of them, including a meaningful share of higher-grade disease (Thompson 2004). The number is information, not a diagnosis.

What changed the test over the last decade is what happens after a high reading. The old pathway went straight from elevated PSA to biopsy — twelve random needle sticks aimed at the prostate, hoping to land on cancer if one was hiding. The new pathway puts an MRI of the prostate in between: lesions get scored, and only suspicious ones get biopsied, with the needle aimed at the lesion. In a trial where men referred for biopsy were randomized to either approach, the MRI-first arm caught more dangerous cancer and far less of the indolent kind, and a real fraction of men avoided biopsy entirely on a clean scan (Kasivisvanathan 2018). A second trial confirmed the same pattern when MRI was added to a population screening program instead of a clinical referral (Eklund 2021). PSA is the first step; the MRI is what keeps the next step proportional.

Does it actually work?

Strip the trial details for a second. If you and 600 men your age all start getting PSA tested today, the math says one of you will avoid a prostate-cancer death over the next 16 years that you'd have otherwise had. Three or four of the 600 will be diagnosed and treated for a cancer that never would have killed them. That's the modest, real, contested benefit — and the modest, real, contested cost.

The European trial (ERSPC) is what the modern read mostly leans on: 162,000 men aged 55 to 69, invited to a PSA every 2 to 4 years versus an un-invited control group, followed for over a decade and a half.

The American trial (PLCO) randomized 76,000 men, found no mortality difference (Andriole 2009), and for a decade was read as the killing blow against screening — except more than 80% of the un-invited "control" arm got PSA-tested anyway in usual care. PLCO actually compared organized screening to ad-hoc screening. When analysts modeled out the contamination, both trials are consistent with a 25 to 30% reduction in prostate-cancer mortality from screening (Tsodikov 2017). A UK trial offered a single PSA invitation to 419,000 men aged 50 to 69 and saw no mortality benefit at 10 years (Martin 2018); the honest read of that result is mostly that one screen isn't enough — the ERSPC benefit was built across multiple rounds.

The harder, honest number: pooled meta-analyses find a small reduction in prostate-cancer deaths and no detectable shift in overall survival (Ilic 2018). Screening rearranges the cause-of-death column without (so far) lengthening lifespan in total. That doesn't make the benefit unreal — a prostate-cancer death prevented is a real win — but it's why "screening saves lives" is a louder claim than the data supports.

Three things to unlearn first

A high PSA does not mean cancer. A reading above 4 ng/mL has roughly a one-in-four shot of revealing any prostate cancer and closer to one in seven for clinically significant cancer (Thompson 2004). Most elevated PSAs come from a prostate that's gotten bigger with age, from infection or chronic prostatitis, or from transient causes — sex, a long bike ride, a recent catheter. A screen-positive is a call to investigate, not a diagnosis.

Not all prostate cancer is dangerous. Most screen-detected prostate cancer is the lowest-grade kind (Grade Group 1, formerly called Gleason 6), and in long-term active-surveillance cohorts that disease has near-zero prostate-cancer mortality at 15 years (Klotz 2015). The UK ProtecT trial randomized men with clinically localized cancer to active monitoring, surgery, or radiation and found about 3% prostate-cancer mortality across all three groups at 15 years — no significant difference (Hamdy 2023). A cancer label is not the same as a death sentence, and aggressive treatment of low-risk disease is the harm screening's critics warned about.

Early detection doesn't automatically save lives. ERSPC's number-needed-to-detect is around 18 to 27 — for every one prostate-cancer death prevented, that many men go through diagnosis and probably treatment (Hugosson 2019). The modal screened-and-treated man avoids no death and gains some morbidity. That's the trade-off the shared-decision framing exists to name out loud.

What's in each room

Both rooms have a worst case. Walk into the late-presentation room first: bone pain that doesn't quit, a back ache that turns out to be a vertebral metastasis, the doctor's tone changing at the second visit because the imaging looks different from what they expected. Your wife notices you're tired before you do. The treatment plan from that point is about controlling something instead of curing it. About 34,000 American men die of prostate cancer every year; lifetime risk of dying from it is around 2.5%. The screen-detection benefit exists because some of those men were a few years of earlier action away from a different ending.

The other room: a man six years past a prostatectomy that was supposed to be the safe move, carrying a small bag of pads in case the leak gets unpredictable, sexual function decline his partner has noticed for years now — all for a cancer the long-term follow-up later showed mostly wouldn't have killed him (Donovan 2016; Hamdy 2023). The trade he made — continence and intimacy for cancer-free reassurance — looks different in hindsight than it did the day he signed the consent form.

Neither room is the typical reader. The typical reader is the 60-year-old whose first PSA is 2.8 and whose second one a year later is 3.1: numbers that say nothing in particular and start a conversation. The stakes are about which room you'd be sorrier to end up in — and whether the decisions in between (when to start, what to do with a borderline result, whether to get an MRI before any biopsy) close one door or the other.

The decision and the pathway

The "shared" part isn't ceremonial. The actual pieces of a real PSA decision are: do you want to know the number given the trade-offs above; if yes, when do you start, how often, and what will you do with a borderline result before any of that happens.

For an average-risk man, current guidelines converge on a window. The US Preventive Services Task Force frames ages 55 to 69 as a discretion call (a "C" recommendation — neither encouraged nor discouraged by default) and recommends against routine screening at 70 and older (USPSTF 2018). The urology bodies start the conversation earlier: the AUA's 2023 guideline suggests opening the discussion at 45 to 50 in average-risk men, screening every 2 to 4 years where pursued, and stopping when life expectancy falls below about a decade (Wei 2023). One useful midlife signal: a PSA below 1.0 ng/mL at age 60 predicts very low future prostate-cancer mortality and may justify long intervals or stopping.

The blood test itself is cheap — usually $20 to $60 cash, typically covered by US insurance under preventive screening in the recommended age window. The downstream MRI runs a few hundred to a few thousand dollars and is increasingly covered for elevated-PSA work-up. The biopsy, if you get to one, is a 15- to 30-minute procedure under local anaesthesia (or sometimes a brief general), with manageable but real complications — blood in the urine for days, blood in semen for weeks, and a small infection risk that's driven the field toward a perineal approach over a rectal one. What you do with the number is the larger half of the decision; almost every unhappy ending in a PSA program traces back to skipping a step in the list above.

Where the math runs differently

The average-risk 55-year-old isn't the only reader. Three groups face a high enough baseline risk that the screening trade-off shifts further toward testing — earlier, more often, with a lower threshold for the MRI step.

Black men of African ancestry. Population data put prostate-cancer incidence roughly 70% higher and mortality roughly double that of White men, with earlier onset and more aggressive disease at presentation (Mahal 2022). The major screening trials enrolled almost entirely European-ancestry men, so the direct evidence is thinner — but the higher pre-test probability of clinically significant cancer means the same diagnostic cascade produces a larger absolute mortality benefit. The AUA 2023 guideline recommends opening the conversation at 40 to 45 in this group, not 50 (Wei 2023).

Family history. One first-degree relative diagnosed with prostate cancer roughly doubles your lifetime risk; two or more, higher still. Start the conversation at 40 to 45, especially if the relative was diagnosed under 65.

BRCA2 carriers. Inherited mutations in BRCA2 raise lifetime prostate-cancer risk five- to eight-fold and produce more aggressive, earlier-onset disease. Annual screening starting at 40 is the standard recommendation. BRCA1 carriers and men with the HOXB13 G84E variant also face elevated risk, though less dramatically.

If two or more of these apply to you — a Black man with a brother diagnosed at 55, a BRCA2 carrier with family history — the recommendation is to treat the screening question as already mostly answered and to bring the same care to the downstream pathway.

When the test misleads — and when it shouldn't run at all

Two flavors of don't-screen. The first is people whose readings will be transiently wrong; the second is people for whom the benefit can't realistically arrive in time.

Transient false alarms. Sex in the 48 hours before the draw, a recent urinary-tract infection (effects last weeks), active inflammation in the gland, a recent catheter or biopsy, and serious cycling can all push PSA up without any cancer changing. Any elevated PSA should be repeated 4 to 6 weeks later before triggering an MRI or biopsy. Men taking finasteride or dutasteride for benign prostate enlargement — or finasteride for hair loss — have roughly halved PSA values — the lab number needs to be doubled for clinical interpretation, and most labs don't auto-correct for it.

Where this goes wrong in practice

Each of these is recoverable; together they're most of the bad outcomes from PSA programs.

Biopsy without an MRI first. In 2024 this is the largest avoidable harm in the pathway. Trial evidence is now strong that MRI-targeted biopsy detects more clinically significant cancer and substantially less indolent cancer than the old systematic-needle approach, and a clean MRI can spare biopsy entirely (Kasivisvanathan 2018; Eklund 2021). If a urologist quotes you a biopsy on a single elevated PSA without first sending you for an MRI, that's a question to ask out loud.

Treating the lowest-grade cancer aggressively. Grade Group 1 disease has near-zero metastatic potential at 15 years and is now standardly managed with active surveillance — serial PSA, repeat MRI, repeat biopsy on change. Surgery or radiation for this disease buys incontinence and erectile dysfunction without any survival benefit (Klotz 2015; Hamdy 2023). Surveillance adoption has risen sharply but isn't universal, especially outside academic centres.

One-and-done screening. A single PSA invitation has not shown a mortality benefit (Martin 2018); ERSPC's benefit came from screening every 2 to 4 years across more than a decade. If you're going to screen at all, plan to screen across rounds.

Letting a rising PSA drift. An elevated reading that isn't acted on — repeated, then MRI'd, then biopsied if warranted — is the worst of both worlds: anxiety without resolution. Velocity over 0.75 ng/mL per year, or any reading over 10 ng/mL, deserves work-up even if a single number looks borderline.

Screening past the cliff. Continuing routine PSA testing past the life-expectancy threshold turns the test into pure overdiagnosis machinery (USPSTF 2018). The hardest version of this conversation is the man whose father died of prostate cancer and who therefore wants to keep screening into his late 70s — the family history that justified earlier and more intense screening doesn't justify screening past the runway.

What changes if you do this thoughtfully

Most of the payoff is the absence of bad endings: not the slow back-pain story, not the prostatectomy at 62 for a cancer that wouldn't have killed you at 80. Both rooms get skipped.

Year one looks like nothing — a blood draw with your annual physical, a number in your chart, a brief conversation. Year five is the same. Most decades of being a thoughtful screener pass uneventfully; the test is doing low-grade probability work in the background while the rest of your life happens.

The version that earns the program is the year your PSA drifts from 1.2 to 3.4, you repeat at 3.2, your doctor orders an MRI, the MRI flags a suspicious zone, the targeted biopsy comes back Grade Group 3 (the not-low-not-high kind), and you and your urologist talk through whether to operate, irradiate, or watch closely. At 15 years you're still here, and so are the guys you went to college with. The ProtecT data say the operate-vs-radiate-vs-monitor question matters less than the field once thought for cancers caught at this stage (Hamdy 2023); the screen was what mattered.

The other version is the year your PSA goes to 4.5, you and your doctor agree on the MRI, the scan is clean, and nothing happens. No biopsy. No cancer label. You stay on cadence. That's payoff too — the modern pathway is what made it possible, and the man who got that pathway in 2008 instead of 2024 would have been on a biopsy table.

Adjacent rabbit holes

A few related threads worth pulling on separately:

  • Benign prostate enlargement (BPH) — the most common cause of an elevated PSA in an aging man, and the source of overlapping symptoms (slower stream, more night trips to the bathroom). Different condition, different management.
  • Testosterone therapy — historically considered a prostate-cancer accelerator; the modern read is less alarmist, but the topic deserves its own treatment alongside any decision about PSA monitoring on therapy.
  • BRCA-related cancer screening as a whole — a BRCA2 carrier's elevated prostate risk sits inside a broader panel that includes breast and pancreatic risk for the carrier and their relatives. The right framing is the panel, not the prostate alone.
  • Active surveillance protocols. What happens after a low-risk diagnosis is its own decision branch with its own evidence base — when to repeat PSA, when to repeat MRI, when a change forces a treatment decision.
  • The rest of the midlife checkup. If you're sitting in your doctor's office at 50 talking PSA, lipid screening, blood-pressure work, colon cancer screening, and a hemoglobin A1c share the room — and each has a sharper benefit-cost case than this one.
Related in the handbook
Related
  • — Prostate screening is the one cancer test that's a conversation, not an automatic — it sits apart from the rest.
  • — BRCA2 carriers face higher, earlier prostate-cancer risk — a reason to start this conversation sooner and weigh it differently.
  • — An inflamed prostate pushes PSA up on its own, so prostatitis can throw a scary number that has nothing to do with cancer.
  • — The concrete harm of over-treating prostate cancer is incontinence and impotence — weigh that before chasing every elevated number.
  • — Finasteride and dutasteride cut PSA by about half; note the drug before calling a prostate result reassuring.
  • — Before and during testosterone therapy, PSA gets watched — the hormone can stir up an existing prostate cancer.
  • — A PSA test answers a cancer question, not a urinary-symptom one — many night-time trips have nothing to do with the prostate.
  • — A blood-based multi-cancer test is the new alternative being pitched against single-cancer screens like PSA.
  • — PSA is the textbook case for reading the actual numbers — one death prevented per 600 men, harms up front.
  • — The incontinence after prostate treatment is exactly what pelvic-floor training helps recover.
  • — Men reaching for a prostate supplement should also sort out the actual screening question, which it can't answer.
·
136