The amount of tumor in radical prostatectomy specimens can be determined with several techniques. Maximum tumor diameter correlates well with total tumor volume and can readily be obtained in incompletely submitted specimens. Initial results in a small series suggested that this measure also may predict for prostate-specific antigen (PSA) failure. We studied whether maximum tumor diameter was an independent predictor of PSA failure in a series of 434 men who underwent radical prostatectomy because of prostatic adenocarcinoma; 118 (27.2%) had PSA failure. Preoperative PSA, Gleason score, pathologic stage, margin status, and largest tumor diameter were determined, and multivariate logistic modeling was performed on the outcome of PSA failure. Maximum tumor diameter was an independent risk factor for PSA failure, along with preoperative serum PSA level, and Gleason score > or = 8. Only 15% of men with tumor with maximum diameter < 1 cm had PSA failure, compared with 73% of men with tumor with maximum diameter > 2. Maximum tumor diameter of prostatic carcinoma is a simple, inexpensive, and independent predictor of PSA failure that can be obtained readily from partially submitted radical prostatectomy specimens.