Percent-"Free" PSA Improves Predictive Value for Those with Low-Total PSA Levels

Abstract & Commentary

By William B. Ershler, MD

Synopsis: Low PSA levels do not assure the absence of prostate cancer. Among a large number of patients referred for prostate biopsy, 524 had PSA levels of < 2.5 ng/mL and, of these, 125 (23%) were found to have cancer. The ratio of "free" PSA to total PSA was shown to have high predictive value in this subset of patients.

Source: Walz J, et al. Percent free prostate-specific antigen (PSA) is an accurate predictor of prostate cancer risk in men with serum PSA 2.5 ng/mL and lower. Cancer. 2008;113: 2695-2703.

It is now well understood that serum prostate-specific antigen (PSA) is not a perfect screening tool in that low levels are not indicative of absent or negligible risk. For example, the prostate cancer prevention trial (PCPT) demonstrated that 17% of patients with unremarkable digital rectal examination (DRE) and total PSA (tPSA) levels of 1.1-2.0 ng/mL, and 24% of patients with a tPSA of 2.1-3.0 ng/mL, were shown to have biopsy-proven prostate cancer.1 In light of earlier observations that the ratio of free-to-total PSA improved predictive value for patients with higher PSA levels,2,3 Walz et al assessed the ability of percent-free PSA to discriminate between benign and malignant prostate biopsy outcomes in men with PSA � 2.5 ng/mL.

Between 1999 and 2006, 7,880 consecutive men underwent initial prostate biopsy at two major European urology referral centers (Hamburg, Germany and Milan, Italy). Of these, 543 (6.9%) had total PSA levels � 2.5 ng/mL. Of these, 125 (23%) were found to have prostate cancer. Age, total PSA, percent-free PSA, and digital rectal examination findings represented predictors of prostate cancer at biopsy in logistic-regression models. The area under the receiver operating characteristics curve (AUC) quantified the discriminative ability of the predictors. The pathological characteristics of the detected cancers were assessed in individuals treated with radical prostatectomy.

Of those in whom biopsy revealed prostate cancer, 64% underwent a radical prostatectomy and, of these,16.5% were pT3 stage and 35.6% had a pathological Gleason score of 3 + 4 or higher. The most accurate predictor of prostate cancer on biopsy was percent-free PSA (0.68), as compared with age (0.50), total PSA (0.57), or rectal examination findings (0.58). Of patients with percent-free PSA below 14%, 59% had prostate cancer. For those with a percent-free PSA > 28%, prostate cancer occurrence was 9%. In multivariate models, percent-free PSA (p < .001) and rectal examination findings (p = 5 .001) were the only independent predictors of prostate cancer. The combined AUC of all predictors (0.69) was not significantly higher than that of percentage of free PSA alone (0.68).


The risk of prostate cancer is clearly non-negligible in patients with PSA � 2.5 ng/mL. In this study, almost 25% of such patients were found to have prostate cancer, and over one-third had locally-advanced disease. These are striking numbers with significant public health implications. On the one hand, it is important to recognize that these were patients seen at major referral centers and how well they reflect the general population is to be considered. On the other hand, in this cohort, almost 75% had normal rectal exams (in addition to their PSA level of � 2.5 ng/mL). In fact, it is not at all clear what prompted referral for biopsy in the first place. Thus, it is unlikely that 25% of men with PSA levels at or below 2.5 ng/mL and normal DRE have prostate cancer; the prevalence is still likely to be higher than commonly appreciated. This raises the question of whether the prostate cancers discovered in those with low PSA levels are of clinical significance and, unfortunately, in a significant percentage, they are, as witnessed by the locally advanced stage found in over one-third who underwent surgery.

Thus, despite the limitations inherent in retrospective review (patient selection, etc.), this report is notable in that it once again indicates the prevalence of this disease (which will only increase in the next two decades with the aging demographics) and the risk of relying on a low PSA level to indicate absence of disease. Furthermore, and importantly, the data presented indicate the added predictive value of determining the percent-free PSA, as this may ultimately prove useful in developing screening strategies.


1. Thompson IM, et al. Prevalence of prostate cancer among men with a prostate-specific antigen level � 4.0 ng per milliliter. N Engl J Med. 2004;350:2239-2246.

2. Catalona WJ, et al. Comparison of percent free PSA, PSA density, and age-specific PSA cutoffs for prostate cancer detection and staging. Urology. 2000;56:255-260.

3. Roehl KA, et al. Robustness of free prostate specific antigen measurements to reduce unnecessary biopsies in the 2.6 to 4.0 ng/mL range. J Urol. 2002;168:922-925.