PSA: What a Prostate-Specific Antigen Result Actually Tells You

These rules are worth following before any PSA test, because a falsely high result can lead to more testing. Other test-specific instructions are collected in the guide to preparing for a blood test.

These common prostate and hair-loss drugs push PSA the other way. The prescribing information for finasteride reports that it lowers PSA by about 50 percent within six months, and instructs that a treated man's PSA be doubled before comparing it with the normal range, with a new baseline set at least six months into treatment. Dutasteride behaves the same way. A man on either drug with a PSA of 2.4 has something closer to 4.8 in untreated terms, and a confirmed rise above his own lowest value is taken seriously even if the printed number still looks normal.

Usually a repeat test first. PSA fluctuates, MedlinePlus lists repeating the test as a common next step, and assay guidance recommends confirming a borderline rise before acting on it. A second test a few weeks later, with the preparation rules followed, settles many borderline results.

Several measures then help separate an enlarged prostate from cancer:

• Free PSA splits the total into PSA floating freely and PSA bound to proteins; a higher free percentage leans toward a benign cause. Where the cutoff is set matters: across reported cutoffs from 8 to 25 percent, the share of cancers detected ranges from about 45 to 95 percent, and the settings that catch the most also send the most men with benign glands to biopsy. Finasteride does not change the percentage, so it needs no adjustment.
• PSA density divides PSA by the prostate's volume on imaging, so a high PSA from a large gland counts for less. It is also used to choose men for active surveillance.
• PSA velocity, how fast PSA rises, carries less weight than it once did. In the Prostate Cancer Prevention Trial analysis it added no independent value in predicting cancer. It matters more when following a known cancer over time, as described in the guide to tracking lab trends.

An MRI scan and, if needed, a biopsy come next. As a guide to expectations, about 1 in 4 men biopsied for a raised PSA are found to have cancer.

The 4.0 ng/mL cutoff comes from the original Hybritech assay and the study behind it, which biopsied 6,374 men whenever PSA passed 4.0. In that study, 23 percent of men with a PSA above 4.0 and a normal rectal exam had cancer, rising to 49 percent when the exam was also abnormal. Assays calibrated to the newer WHO 96/670 standard read about 20 percent lower, where 3.1 ng/mL is the equivalent cutoff.

So PSA values from different assays are not interchangeable. When a lab changes its method, a new baseline is set rather than treating the change as real, and a PSA that seemed to jump the year a hospital changed analyzers may not have changed at all. Like every marker on the tumor markers panel, PSA results compare reliably only within one lab.

For men aged 55 to 69, the U.S. Preventive Services Task Force's 2018 recommendation calls PSA screening an individual decision to make with a clinician (grade C). For men 70 and older it recommends against screening (grade D). The recommendation has been listed as under update since late 2023, so the advice may change.

The careful wording reflects the trade-off. Per 1,000 men aged 55 to 69 screened over about 13 years, the Task Force estimates roughly 100 prostate cancer diagnoses, 80 men treated, 50 with sexual problems and 15 with urinary incontinence as a result. Screening finds some aggressive cancers early and saves lives; it also finds slow-growing cancers that would never have caused harm, and treating those carries the same side effects. In a given screening round, about 6 to 7 percent of men get a false positive result.