The effect of BMI at time of surgery on long-term outcome after radical prostatectomy.

e15203 Background: Body mass index (BMI) at time of surgery was determined among 715 radical prostatectomy patients. The association of BMI with a range of treatment outcomes was considered. Methods: The associations of BMI at time of radical prostatectomy (RP) with disease stage and aggressiveness and long-term outcome were evaluated among 715 patients treated with RP at Roswell Park Cancer Institute between 1993 and 2005. Clinical and pathological aggressiveness indicators included clinical Gleason sum and tumor stage (2002 TNM), highest preoperative PSA, pathological Gleason sum and tumor stage (2002 TNM) and surgical margin status. Ten post-RP recurrence definitions were considered: 1) PSA ≥ 0.2 ng/ml; 2) PSA ≥ 0.4 ng/ml (with 1 confirming value); 3) 1 or more post RP treatments (ADT, radiation, chemotherapy); 4) PSA doubling time < 12 months; 5) PSA doubling time < 9 months; 6) PSA doubling time < 6 months; 7) NCCN definition of PSA failure; 8) AUA definition of PSA failure; 9) diagnosis of metastatic CaP; and 10) death from CaP. Results: Of the 715 men, 33 developed metastatic prostate cancer, and 17 died of prostate cancer. 246 men had BMI ≥ 30. BMI was not significantly associated with clinical or pathological aggressiveness criteria. These analyses showed that there is a trend towards higher risk of the development of metastasis or death for men with BMI ≥ 30, although the association with high BMI and these failure types is not significant. With adjustment for the most significant tumor aggressiveness features (clinical Gleason sum, pathological tumor stage, pathological Gleason sum, and surgical margin status) in proportional hazards regression, men with BMI ≥ 30 had consistently lower risk for all definitions of recurrence except metastasis and death, although no hazard ratios were significant. In contrast, men with higher BMIs had higher risk for metastasis and death from prostate cancer, although neither association is statistically significant. Conclusions: Men with higher BMIs show similar to slightly reduced risk for PSA-based recurrence definitions. Men with higher BMIs had slightly higher risk, though not significant, for metastasis and death. These results seem to support theories that PSA is diluted in men with higher BMIs.