Alternatives to prostate-specific antigen tests for detecting clinically significant prostate cancer

Prostate cancer is the second leading cause of cancer death in men, following lung cancer. It is necessary to balance efforts to identify prostate cancer in the early stages against complications and harms from overtreatment and overdiagnosis. Even though the large-scale European Randomized Study of Screening for Prostate Cancer showed that prostate-specific antigen (PSA) screening reduces prostate cancer mortality, it is still necessary to evaluate new tests for their ability to better predict aggressive prostate cancer and prevent overdiagnosis of slower-growing insignificant forms of the disease. Among several free-PSA subforms that have been studied, proPSA was shown to have higher values in prostate cancer. Of the circulating forms of proPSA identified in serum, [-2]proPSA has greater stability, and Beckman Coulter has developed an automated immunoassay to detect it. The authors conducted a study to evaluate the clinical utility of [-2]proPSA derivatives in detecting clinically significant prostate cancer (csPCa) and define acceptable cutoffs. They also sought to compare the diagnostic accuracy of [-2]proPSA to PSA and percentage of free PSA (%fPSA). The authors enrolled in the study 237 men with a PSA value between 2 and 10 ng/mL who were scheduled for a prostate biopsy. Blood was collected for PSA, free PSA (fPSA), and [-2]proPSA on the day of the biopsy but before the procedure. The investigators applied parametric and nonparametric tests, receiver operating characteristic curves, and logistic regression analysis to show the outcomes for csPCa and prostate cancer overall. The results showed that 118 (49.8 percent) of the 237 patients had prostate cancer, including 100 (42.2 percent) who had csPCa. The [-2]proPSA derivatives were significantly higher in csPCa and prostate cancer as a whole (P<0.001). The areas under the curve for predicting csPCa were higher for the percentage of [-2]proPSA (%[-2]proPSA; 0.781) and the Prostate Health Index (PHI; 0.814) than for PSA (0.651) and %fPSA (0.724). Of interest, there was an 11 percent gain in diagnostic accuracy when %[-2]proPSA or the PHI was added to the base algorithm of PSA and %fPSA. The data showed that 25 to 29 percent of biopsies could have been spared using %[-2]proPSA and the PHI in the testing algorithm, missing only 10 percent of csPCa. The authors obtained the same results using [-2]proPSA as a reflex test when %fPSA was 25 percent or less. The authors concluded that [-2]proPSA derivatives could improve the diagnostic accuracy of csPCa when PSA values were between 2 and 10 ng/mL. This would prevent unnecessary biopsies and identify patients needing active surveillance. The authors noted that the use of [-2]proPSA as a reflex test based on the %fPSA values can be a cost-effective testing approach for csPCA while reducing the potential harm from unnecessary biopsies.