Abstract
95 Background: To investigate whether pelvic lymph node dissection (PLND) reduces prostate cancer specific mortality (PCSM) in surgically-treated men with intermediate-risk prostate cancer (Pca). Methods: We identified 44,112 men diagnosed with intermediate-risk Pca from 2004-2009 in Surveillance, Epidemiology and End Results Program (SEER). We used inverse-probability-of-treatment weighting (IPTW) to adjust for baseline characteristics between PLND + radical prostatectomy (RP, N=26,571), versus RP alone (N=17,541) groups; Cox competing-risk model and propensity score-adjusted analyses were used for validation. Gleason scores were based on prostatectomy since biopsy scores were not available for RP-treated patients in SEER from 2004-2009. Results: After a median follow-up of 54 months, there was no survival benefit associated with PLND + RP compared to RP alone (Gray's test, P=0.30). After IPTW adjustment for baseline characteristics, PLND was still not associated with PCSM (AHR: 0.93, 95% CI: 0.65-1.33). This result was consistent with propensity score-adjusted model (AHR=1.05, 95% CI: 0.71-1.55) and the Cox competing-risk model (AHR=1.06, 95% CI: 0.71-1.57). Of men who received RP with PLND, 502 men (1.9%) had pathologically positive lymph nodes, which were associated with a higher risk for PCSM (AHR: 4.02, 95%CI: 1.83-8.84). Conclusions: PLND with RP was not associated with reduced PCSM compared with RP alone in men with intermediate-risk disease, suggesting that PLND is diagnostic but not therapeutic in this patient population. However, a caveat of this study is that risk group was defined by pathologic Gleason score; the 5% of clinically intermediate risk patients who are typically found to have Gleason 8-10 disease at prostatectomy could not be included in this analysis. [Table: see text]
