Purpose: The determination of pro prostate specific antigen (proPSA) forms has been suggested to be promising in prostate cancer diagnosis. In this multicenter trial we evaluated the diagnostic usefulness of (-5, -7) proPSA.
Materials and methods: A total of 2,055 white men, including 1,046 with and 1,009 without prostate cancer, with total PSA (tPSA) between 0.28 and 81 ng/ml were retrospectively analyzed. Of these men 2,026 and 1,727 had tPSA less than 20 and less than 10 ng/ml, respectively. All subjects were untreated for prostatic disease and underwent multisector needle biopsy of the prostate. An Elecsys 2010 analyzer was used to determine tPSA, free PSA (fPSA) and (-5, -7) proPSA in the 2,055 serum samples. ROC analyses were performed to discriminate men with biopsy positive and negative results in the entire and in select tPSA ranges.
Results: In the select tPSA range 2 to 4 ng/ml the area under the ROC curve for proPSA (0.53) and proPSA/fPSA (0.59) was not significantly larger than that for tPSA (0.60) or the fPSA/tPSA (f/tPSA) ratio (0.64). In the tPSA range 4 to 10 ng/ml the area under the curve for the ratio proPSA/fPSA (0.67) was larger than for tPSA (0.53) but not larger than for f/tPSA (0.69). The f/tPSA ratio demonstrated the best discriminatory power in this tPSA range of 4 to 10 ng/ml.
Conclusions: In this multicenter study no improvement in diagnostic accuracy was shown when comparing (-5, -7) proPSA and the corresponding ratios with tPSA or f/tPSA. Further studies using other proPSA forms or tumor associated proteins should be done.