Prostate-specific antigen density predicts adverse pathology and increased risk of biochemical failure

Objectives To determine whether the prostate-specific antigen (PSA) density (PSAD), measured using either ultrasound (US) or prostatic weight (PW), is an independent predictor of adverse pathologic findings or biochemical-free survival and whether it outperformed PSA. Methods The data were obtained prospectively from 1327 patients undergoing radical prostatectomy from 1990 to 2003. The US PSAD was calculated by dividing the preoperative PSA level in nanograms per milliliter by the US measured prostate volume in cubic centimeters. The PW PSAD was calculated by dividing the PSA value in nanograms per milliliter by the measured PW of the prostatectomy specimen in grams. Logistic regression analysis was performed to determine whether the US or PW PSAD was more accurate than the PSA level in predicting for adverse pathologic findings. A proportional hazards model was used to determine whether PSAD more accurately predicted for biochemical failure (PSA level greater 0.2 ng/mL). Results Multivariate analysis demonstrated that US and PW PSAD were independent predictors of positive margins (odds ratio [OR] 5.00, 95% confidence interval [CI] 2.65 to 9.47 and OR 29.75, 95% CI 10.18 to 86.96, respectively), extracapsular disease (OR 10.89, 95% CI 5.32 to 22.32 and OR 126.62, 95% CI 37.99 to 422.07, respectively), seminal vesical invasion (OR 6.06, 95% CI 2.96 to 12.41 and OR 33.72, 95% CI 9.79 to 116.15, respectively), and biochemical failure (hazard ratio 3.32, 95% CI 2.38 to 4.63 and hazard ratio 8.70, 95% CI 5.21 to 14.52, respectively). The C-index demonstrated that both US and PW PSAD appeared more discriminant for adverse pathologic findings and biochemical failure than did the PSA level. Conclusions The US and PW PSAD are strong predictors of advanced pathologic features and biochemical failure after radical prostatectomy. The incorporation of PSAD into the risk assessment could provide additional prognostic information beyond grade, stage, and PSA level; therefore, the inclusion of PSAD into nomograms should be considered.