What the percentage measures
PSA travels in the blood in two forms. The National Cancer Institute's PDQ summary for health professionals, updated 8 April 2025, describes it this way: serum PSA exists in both free form and complexed to several protease inhibitors, especially alpha-1-antichymotrypsin, and assays for total PSA measure both. Percent free PSA is the free part as a share of the total. A total PSA of 6.0 ng/mL with a free PSA of 0.9 ng/mL is 15% free (our arithmetic).
The direction is the whole point of the test. The FDA-approved labeling for the first percent free PSA assay states that, in patients with elevated PSA, men with prostate cancer tend to have lower percent free PSA values than men with benign disease, and that men with values near and above its cutoff tend to have large glands from benign prostatic hyperplasia. A low percentage leans toward cancer; a high one leans toward benign enlargement. Neither settles it: the same labeling says percent free PSA should not be interpreted as definitive evidence for the presence or absence of prostate cancer, and that prostatic biopsy is required for diagnosis.
Where the 25% number comes from
The cutoff traces to one trial. Catalona and colleagues, publishing in JAMA in May 1998, enrolled 773 men aged 50 to 75 at seven university medical centers, all with a palpably benign prostate and a total PSA of 4.0 to 10.0 ng/mL, and all with a confirmed diagnosis: 379 had prostate cancer and 394 had benign prostatic disease. Free and total PSA were measured with Hybritech's Tandem assays.
From the abstract: the 25% free PSA cutoff detected 95% of cancers while avoiding 20% of unnecessary biopsies. In plain counts from the labeling's Table 4, that is 358 of 379 cancers at or below 25%, and 80 of 394 men with benign disease above it, so 21 cancers sat above the line (our arithmetic). In the authors' multivariate model, percent free PSA was an independent predictor of cancer (odds ratio 3.2) and, in their words, contributed significantly more than age (1.2) or total PSA (1.0) within this PSA range. The authors recommended the 25% cutoff for this group regardless of patient age or prostate size.
The FDA approved the Tandem free PSA assays on 10 March 1998 (PMA P970038). The approved use is narrow: an aid in distinguishing prostate cancer from benign prostatic conditions, used with Tandem total PSA, in men aged 50 years and older with a total PSA between 4 and 10 ng/mL and rectal exam findings not suspicious for cancer.
| Percent free PSA | All ages | Age 50 to 64 | Age 65 to 75 |
|---|---|---|---|
| 10% or less | 56% | 56% | 55% |
| Above 10% to 15% | 28% | 24% | 35% |
| Above 15% to 20% | 20% | 17% | 23% |
| Above 20% to 25% | 16% | 10% | 20% |
| Above 25% | 8% | 5% | 9% |
The same labeling puts the overall probability for a PSA of 4 to 10 ng/mL at 25%; percent free PSA spreads that from 8% to 56%. Two things limit how far the table travels. It counts any prostate cancer found on a 1990s needle biopsy, while the current AUA/SUO guideline aims evaluation at Grade Group 2 or higher cancer. And the cancers above the cutoff were not typical: the labeling's Table 5 gives them a median age of 68 and a median prostate volume of 48 cc, against 63 and 34 cc for cancers at or below 25%, and the trial abstract describes them as generally less threatening in grade and volume.
Your number may not mean the same thing at another lab
The Tandem labeling states that percent free PSA cutoffs and clinical performance differ when free and total PSA assays from different manufacturers are combined: mean values from identical serum samples may be two-fold higher with different assay combinations, and 95% sensitivity cutoffs may vary from 22% to 34%. It says its recommendations apply only to the Tandem assays, that the laboratory report must name the manufacturer of both assays, and that values from different manufacturers' assays cannot be used interchangeably.
A second FDA file shows it in practice. Abbott's AxSYM free PSA assay, approved 4 February 2004 for the same population, reported its own cutoffs from 271 men with a PSA of 4 to 10 ng/mL and a non-suspicious rectal exam: 26.4% gave 96.7% sensitivity with 28.2% specificity, and 23.6% gave 90% sensitivity with 40.3% specificity. Its probability bands are cut differently too.
| AxSYM percent free PSA | Probability of cancer (95% CI) |
|---|---|
| 10% or less | 67.9% (56.4 to 77.6) |
| Above 10% to 15% | 50.0% (41.9 to 58.2) |
| Above 15% to 20% | 32.1% (26.3 to 38.6) |
| Above 20% to 26% | 18.3% (13.0 to 25.0) |
| Above 26% | 9.6% (5.5 to 16.0) |
The Abbott summary adds the caveat that matters for any reader holding one of these tables: probabilities attached to percent free PSA depend on the disease prevalence in the population studied. At an assumed prevalence of 20%, the same 10%-or-less band gives 51.6%; at 45%, it gives 77.7%. On our reading, the honest use of either table is as a direction and a rough size, measured on the assay that produced your result, and not as your personal odds.
What can move the result
- Prostate handling before the draw. The Tandem labeling says the sample should be drawn before a rectal exam, prostatic massage, transrectal ultrasound or biopsy. A rectal exam may cause a transient increase in both free and total PSA, a needle biopsy causes a transient rise in free PSA and a longer one in total PSA, and a six-week wait between biopsy and PSA sampling has been recommended.
- Prostate size. The Tandem labeling says men with values near and above its cutoff tend to have large benign glands, and the cancers the 25% cutoff missed were found primarily in older men with larger glands.
- Finasteride and dutasteride. The NCI PDQ summary says finasteride reduces PSA by approximately 50%. The Tandem labeling says 5-alpha reductase inhibitors, and other drugs used to treat benign enlargement, may affect PSA levels and that care should be taken in interpreting results from men taking them.
- Mouse antibody treatments. Abbott's summary warns that people who have received mouse monoclonal antibody preparations may carry human anti-mouse antibodies that make results falsely high or low.
Where it sits in today's guideline
The AUA/SUO early detection guideline, published in 2023 and amended in 2026 (the edition and access notes are in our PSA between 4 and 10 guide; we read the 2026 web version as archived on 26 August 2026), gives the test one direct paragraph: "Perhaps the most widely available adjunctive test is percent free PSA. Lower percent free PSA is associated with greater likelihood of identifying prostate cancer on biopsy." It adds that the test improves prediction of Grade Group 2 or higher cancer mainly inside combined tests. In the study validating the 4Kscore, removing percent free PSA cut the area under the curve from 0.821 to 0.699, and in a study of the Prostate Health Index, percent free PSA beat total PSA, 0.661 against 0.551.
That paragraph sits under Statement 17: clinicians may use adjunctive urine or serum markers when further risk stratification would influence the decision about biopsy (Conditional Recommendation, Grade C), in patients with a mildly elevated PSA, typically between 2.5 and 10 ng/mL. The guideline says such markers should not be used when the risk is already so low or so high that the result would not change the next step. Percent free PSA is also an input to risk calculators the guideline discusses, including the PCPT calculator, and it is listed among factors linked to finding Grade Group 2 or higher cancer at a repeat biopsy. We searched the guideline text for a percent free PSA cutoff and found none.
The NCI PDQ summary is more cautious: it says the preponderance of evidence on complexed and percent free PSA is not clear and that total PSA remains the standard. Among 108 men who went on to radical prostatectomy, it reports, lower values went with a higher risk of disease extending beyond the prostate capsule.
Sponsored option
Vetted Aug 19, 2026Affiliate link, we may earn a commission at no extra cost to you.
Sesame Our score 4.0/5
Telehealth / primary care
Pricing as of Aug 19, 2026
Telehealth visits from $34; weight-loss program $99/mo or $59/mo annual
What to do next
Bring four questions to whoever ordered the test: which manufacturer's free and total PSA assays produced my number, was the blood drawn before any rectal exam or biopsy, does my finasteride or dutasteride use change how it reads, and how does it shift my estimate in the risk calculator you use. Our PSA above 4.0 page covers the steps the guideline puts after a raised total PSA, the PSA reference page carries the ranges with their sources and dates, and the lab result lookup places a total PSA value against them. Our guide to a PSA between 4 and 10 walks through the repeat test, MRI and biopsy decision, our guide to normal PSA levels by age explains where the 4.0 line came from, and the prostate cancer screening page sets out who recommends screening and when. Your personal screening plan lists the screenings that apply at your age and sex.
Frequently asked questions
What is a normal free PSA percentage?
The AUA/SUO guideline sets none, and neither does the NCI PDQ summary. The figure usually quoted, 25%, is the cutoff from the 1998 trial and the FDA-approved labeling of the Hybritech Tandem assays, for men 50 and older with a total PSA of 4 to 10 ng/mL and a non-suspicious rectal exam. Another FDA-approved assay reported cutoffs of 23.6% and 26.4%, and the Tandem labeling says results from different manufacturers cannot be used interchangeably.
Is a free PSA of 15% bad?
It is below the 25% line, which in the Tandem labeling puts it in a higher-risk band: for men with a PSA of 4 to 10 ng/mL and a non-suspicious rectal exam, the probability of cancer on biopsy was 28% for values above 10% to 15%, against 25% for that PSA range overall. That figure belongs to one assay and counts any cancer, so it is a starting point for a risk discussion, not a result in itself.
Does a high free PSA percentage rule out cancer?
No. The Tandem labeling states that an elevated percent free PSA is not necessarily indicative of the absence of cancer. In the trial behind it, 21 of 379 cancers sat above the 25% cutoff (our arithmetic from the labeling's counts); the authors describe these as more common in older men and generally less threatening in grade and volume.
Can free PSA replace an MRI or a biopsy?
The FDA labeling says prostatic biopsy is required for a diagnosis of cancer, and the AUA/SUO guideline treats percent free PSA as one input to a risk estimate that can inform decisions about MRI and biopsy, made with your clinician. It is most useful, in the guideline's words, when further risk stratification would influence the decision.
