Objectives: We determine whether the different molecular forms of prostate-specific antigen (PSA) and other PSA variables can predict prostate cancer in men undergoing repeat prostate needle biopsy.
Methods: Between 1997 and 2001, repeat biopsy was performed in 97 patients who had undergone prior negative prostate biopsy. The ability of total PSA (tPSA), complexed PSA (cPSA), free PSA (fPSA), free-to-total PSA (fPSA/tPSA), free-to-complexed PSA (fPSA/cPSA), complexed-to-total PSA (cPSA/tPSA), tPSA density (tPSAD), cPSA density (cPSAD), transition zone tPSA density (tPSATZ) and transition zone cPSA density (cPSATZ) was assessed by univariate and multivariate analyzes as well as receiver operating characteristics (ROC) curves.
Results: Prostate cancer on repeat biopsy was detected in 24% of subjects (23 of 97) who had a negative initial biopsy. The PSA parameters cut-off to ensure a 96% sensitivity of cancer detection, were 29% using fPSA/tPSA, 32% using fPSA/cPSA, 0.18 ng/mL/cc using tPSATZ and 0.16 ng/mL/cc using cPSATZ. The fPSA/tPSA would have prevented 32% of negative biopsies, the fPSA/cPSA 28%, the tPSATZ 23% and the cPSATZ 30%. ROC curve analysis fPSA/tPSA, fPSA/cPSA ratios, tPSATZ and cPSATZ were significantly better predictors of repeat biopsy results than tPSA or cPSA, but there was no significant difference in the ROC curves among these four PSA parameters. In the multivariate logistic regression analysis these four PSA parameters were significant predictors for cancer detection in the repeat biopsy group (P < 0.001).
Conclusion: fPSA/tPSA ratio, fPSA/cPSA ratio, tPSATZ and cPSATZ enhance the specificity of PSA testing compared to tPSA or cPSA when determining which patients should undergo repeat biopsy.