To test prostate-specific antigen mass density (PSAMD) as a predictor of total tumor volume (TTV) at radical prostatectomy (RP).

We conducted a detailed pathologic analysis of 469 RP from men with NCCN low-risk prostate cancer who had Gleason score of 3 + 3 = 6 (grade group 1) at RP. We then compared the ability of PSA, PSA density (PSAD), PSA mass (PSAM-absolute amount of PSA in patient's circulation), and PSAM density (PSAM divided by prostate weight without seminal vesicles) to predict TTV at RP. PSAM was calculated by multiplying plasma volume (estimated body surface [weight, kg^0.425 × height, m^0.72 × 0.007184] × 1.67) by PSA. Performance of the above measures in different BMI categories was assessed. Kruskal-Wallis test was used to compare the means and Spearman's rank correlation coefficient to assess the correlations.

The 469 men were normal weight (n = 129), overweight (n = 253), and obese (n = 87). Mean age of the patients' was 57.4 years and PSA of 4.53 ng/ml. Increase of prostate weight with body mass index (BMI) was reflected in PSAM (both P <.001) but not in other measures. BMI did not correlate with TTV and PSA. Among PSA, PSAD, PSAM, and PSAMD, PSAMD had the highest correlation with TTV (r = 0.336; P <.001). Prostate weight had stronger (negative) association with PSAMD (r = −0.394; <.001) than TTV.

PSAMD is the biochemical measure with the best correlation with TTV at RP. Unlike other measures, it is not affected by BMI-related hemodilution. Thresholds should be established to use this more objective measure clinically in surveillance algorithms and in planning radical prostatectomy.