[Is it possible to define a threshold for free PSA index that is useful in the daily practice of urology?]

Retrospective evaluation of the use of the free PSA index before prostatic biopsies.

The authors retrospectively studied the values for total PSA, free PSA, and free PSA index (ratio of free PSA over total PSA expressed as a %) in men with a total PSA between 2 and 10 ng/ml, from a population of 391 men prior to prostatic biopsies. They also isolated a subgroup of patients in whom the free PSA index could have been used as a first-line marker to decide whether or not to perform prostatic biopsies.

The mean values for total PSA, free PSA, and free PSA index were compared as a function of the diagnosis, age, and ultrasound prostatic volume. The yields of the various cut-off values for the free PSA index for PSA between 2 and 4 ng/ml, 4 and 10 ng/ml, and 2 and 10 ng/ml with a normal digital rectal examination are reported. Between 2 to 10 ng/ml, at a cut-off value of 30%, 94.1% of cancers would have been detected (sensitivity) and 22% of biopsies would have been avoided, 10 of which would have been useless, i.e. a 30.3% economy of useless biopsies not performed (specificity). At the cut-off value of 15%, less than half of cancers would have been detected (47.1%) and 90.9% of useless biopsies would have been avoided. Biases creating difficulties of interpretation were the assay kits, the reference population, age, storage of sera, and prostatic volume.

The free PSA index would be a useful first-line parameter in only 12.7% of candidates for prostatic biopsies. The cut-off value of 30%, validated for our assay method, would be able to detect the majority of cancers in men aged 50 to 65 years, while avoiding biopsies in the third of men with no detectable cancer.