To determine the clinical utility of using proenzyme prostate-specific antigen (pPSA) for early detection of prostate cancer in the 2.5 to 4.0 ng/mL total PSA range. pPSA, the precursor form of PSA that contains a 7 amino acid leader peptide, and truncated forms such as [−2]pPSA and [−4]pPSA can be measured in serum by research immunoassay.

Archival serum from 119 men (noncancer, 88; cancer, 31), obtained before biopsy and in the total PSA range of 2.5 to 4.0 ng/mL, were assayed for total PSA, free PSA (fPSA), and pPSA. pPSA was defined as the sum of the [−2], [−4], and [−7] forms, and the percent pPSA (%pPSA) was defined as pPSA/fPSA.

pPSA averaged 4.6% ± 0.4% (SEM) of total PSA and 39.3% ± 3.5% of fPSA. PSA and %fPSA values were similar between the noncancer and cancer groups, and %pPSA tended to be higher in the cancer group (50.1% ± 4.4%) compared with the noncancer group (35.5% ± 6.7%; P = 0.07). Using receiver operating characteristic analysis to assess clinical utility, the area under the curve for %pPSA was 0.688 compared with 0.567 for %fPSA. At a fixed sensitivity of 75%, the specificity was significantly greater for %pPSA at 59% compared with %fPSA at 33% (P <0.0001).

In the 2.5 to 4.0 ng/mL total PSA range, 75% of cancers can potentially be detected with 59% of unnecessary biopsies being spared using %pPSA; use of %fPSA would result in sparing only 33% of unnecessary biopsies. A large prospective clinical trial is needed to confirm these preliminary findings.