Nonequimolar-response assays for prostate-specific antigen (PSA) are criticized for overestimating total PSA in some men without prostate cancer (PCA), and underestimating total PSA in some men with PCA. We recently studied three nonequimolar-response PSA assays that had undergone modifications. While two of the studied assays achieved equimolar-response characteristics with improved areas under receiver operating characteristic (ROC) curves (AUC), the modification of the Chiron ACS PSA assay (ACS PSA2, Chiron) failed to achieve this. Recently, the ACS assay underwent another modification (ACS PSA, Bayer), which we investigated.Sera from 305 men (155 without and 150 with PCA, PSA ≥ 2 and ≤ 30 µg/l, Tandem-E) were measured using both modifications of the ACS assay and equimolar-response reference methods (Tandem-R free and Tan