PSA (Prostate-Specific Antigen): What the Test Actually Tells You About Prostate Cancer Risk

PSA is a protein your prostate makes to liquefy semen. Healthy cells make it. Enlarged prostates make more of it. Inflamed prostates dump it into the blood. Cancer cells make it too, often more per cell than normal tissue, but a big benign prostate can easily out-produce a small tumor. That overlap is the whole problem with PSA as a single number.

PSA travels in two forms in your blood: bound (complexed) and unbound (free). Total PSA is the sum. When total PSA lands in the gray zone of 4 to 10 ng/mL, the free PSA / total PSA ratio can sharpen the picture. A low free fraction (under about 10 to 25 percent) tilts toward cancer, because tumors tend to release proportionally more of the bound form.

A few other refinements urologists actually use:

• PSA velocity: how fast the number is climbing. A rise of more than ~0.75 ng/mL per year in the gray zone gets attention even when each value looks borderline.
• PSA density: total PSA divided by prostate volume on imaging. A "borderline" PSA in a small prostate is more concerning than the same value in a large one.
• Age-specific cutoffs: the prostate enlarges with age. The flat 4.0 cutoff misses younger men with disease and over-flags older men without it.

The textbook line is "below 4.0." Modern practice uses age-specific ranges, because baseline PSA climbs with prostate volume:

• 40 to 49: below 2.5 ng/mL.
• 50 to 59: below 3.5 ng/mL.
• 60 to 69: below 4.5 ng/mL.
• 70 to 79: below 6.5 ng/mL.

Free PSA / total PSA ratio (when total PSA is 4 to 10 ng/mL):

• Above 25 percent: low cancer probability, more consistent with BPH.
• 10 to 25 percent: indeterminate. Most urologists move to MRI or repeat testing here.
• Below 10 percent: higher cancer probability. Biopsy or MRI usually recommended.

After prostatectomy: PSA should fall to undetectable (typically below 0.1 ng/mL on standard assays, below 0.01 on ultrasensitive). Any sustained rise from undetectable is a recurrence signal.

An elevated PSA has many possible causes. Cancer is one of them. Before drawing conclusions, rule out the boring stuff first:

• BPH (benign prostate enlargement). The most common reason for a mildly high PSA in older men. The gland gets bigger, the protein output goes up.
• Prostatitis. Acute or chronic inflammation can spike PSA into double or triple digits. Treat the infection, recheck in 6 to 8 weeks.
• Recent ejaculation. Bumps PSA for 24 to 48 hours. Skip it for two days before the draw.
• Recent digital rectal exam, biopsy, or cystoscopy. Get the PSA before any prostate manipulation, or wait at least a week.
• Long bike rides. Yes, really. Perineal pressure in the prior 48 hours can raise the value.
• UTI. Can keep PSA elevated for weeks until cleared.
• Urinary retention or recent catheterization.
• Finasteride or dutasteride. These cut PSA roughly in half. Multiply your reported value by 2 to compare to the cutoffs above.

One mildly elevated PSA almost never warrants an immediate biopsy. The 2026 sequence looks like this:

1. Repeat the PSA in 4 to 8 weeks under controlled conditions (no recent ejaculation, exam, infection, or cycling).
2. Add free PSA / total PSA ratio if total PSA is in the 4 to 10 range.
3. If still elevated, get a multiparametric prostate MRI before any biopsy. MRI flags the lesions actually worth sampling and lets a lot of men skip biopsy entirely.
4. If biopsy is needed, target the MRI-flagged lesions (PI-RADS 4 or 5), not blind systematic sampling.

Here is the part most men do not hear: many cancers caught this way are low-grade (Gleason 6 / Grade Group 1) and would never have hurt them. Active surveillance, regular PSA, MRI, and selective rebiopsy, is now the standard for low-risk disease, not immediate surgery or radiation. The question has shifted from "do I have cancer" to "do I have cancer that needs to be treated."

Low PSA in an untreated prostate is reassuring, but it is not a guarantee. About 15 percent of men with prostate cancer have a PSA below 4.0, and a smaller number have aggressive disease at low PSA. The number is one piece of the picture, not the whole picture.

After surgery, PSA should be undetectable. After radiation, the goal is a stable low nadir; a rise of more than 2.0 ng/mL above that nadir defines biochemical recurrence (Phoenix criteria). On hormone therapy, a low value is expected and not concerning by itself.

Men on finasteride or dutasteride (for BPH or hair loss) need to remember those drugs roughly halve the result. A "normal" PSA on these meds can be hiding a real elevation. Multiply by 2.

PSA is noisy. Two values drawn a month apart in the same man can differ by 20 to 30 percent from biological variation and assay scatter alone, with nothing actually changing in the prostate. That is exactly why a single "elevated" number rarely justifies action on its own.

What matters is direction over time:

• 1.8, 1.9, 2.0, 2.1 over four years. That is stable BPH.
• 1.5, 2.4, 3.6, 5.2 over four years. That velocity warrants MRI and a real conversation, even though every individual value is borderline.
• Post-prostatectomy: undetectable, then 0.05, then 0.12, then 0.30. That is biochemical recurrence and needs imaging plus oncology input.

For screening, the most useful number is rarely the one in front of you. It is the slope through the prior several years. Bring all your old PSA values to the conversation if you have them.

• PSA above the age-specific cutoff confirmed on a repeat draw, with no infection, recent exam, ejaculation, or cycling. Next step: MRI.
• PSA velocity above ~0.75 ng/mL per year in the gray zone, especially with a free PSA fraction below 10 to 15 percent.
• Any rise from undetectable after prostatectomy. Even 0.1 to 0.2 means something here.
• PSA more than 2.0 above the post-radiation nadir (Phoenix criteria for biochemical recurrence).
• New urinary symptoms with elevated PSA. Rule out prostatitis, retention, and UTI before chasing cancer.
• Strong family history (first-degree relatives, BRCA1/2, Lynch syndrome). Start the screening conversation earlier, around age 40 to 45, with a lower threshold for further workup.

What does not warrant immediate biopsy: a single mildly elevated value, a value drawn within days of an exam or ejaculation, or a value during an active urinary infection.

• Free PSA / total PSA ratio. Refines interpretation in the 4 to 10 ng/mL gray zone.
• Multiparametric prostate MRI. The highest-yield next step after a confirmed elevated PSA. Avoids many unnecessary biopsies.
• Digital rectal exam. Independent of PSA and complementary to it. A nodule on exam can warrant biopsy regardless of PSA.
• 4Kscore, PHI (Prostate Health Index), SelectMDx, MyProstateScore. Secondary blood and urine tests that improve specificity in the gray zone before committing to biopsy.
• Alkaline phosphatase. Rises with bone metastases in advanced prostate cancer; useful when monitoring known disease.
• Testosterone. Relevant when interpreting PSA on hormonal therapy or in hypogonadism.