Rates of Positive Surgical Margins and Their Effect on Cancer-specific Mortality at Radical Prostatectomy for Patients With Clinically Localized Prostate Cancer.

Abstract

The objective of this study was to investigate positive surgical margin (PSM) rates in patients with prostate cancer treated with radical prostatectomy (RP) and assess PSM impact on cancer-specific mortality (CSM). Within the Surveillance, Epidemiology, and End Results (SEER) database (2004-2015), we identified men who underwent RP with pathologic T2 or T3a stage. Annual trends of PSM rates were plotted. Subgroups focused on geographic regions, namely the North Central, Northeast, South, and West. Cumulative incidence plots depicted other-cause mortality-adjusted CSM rates. Multivariable competing risks regression models tested the relationship between PSM and CSM. Subgroup analyses focused on pathologic stage, Gleason score, and geographic region. Of 153,329 patients treated with RP, 12.3% (n= 18,935) exhibited PSM. Overall, in pathologic T2 stage and pathologic T3a stage, PSM rates decreased during the study period from 18.7% to 9.7% (P<.001), 15.7% to 7.3% (P< .001), and 39.0% to 18.0% (P< .001), respectively. In subgroup analyses focusing on geographic regions, PSM rates universally decreased. However, the magnitude differed. In multivariable competing risks regression models, PSM rates were associated with higher CSM (hazard ratio, 1.45; P< .001). However, geographic regions failed to reach independent predictor status. Insufficient information about PSM focality, length, and associated Gleason score represent important limitations. It is encouraging that PSM rates decreased during the study period, even after stratification according to tumor stage. PSM decreased within the 4examined geographic regions. However, the rate of decrease varied in magnitude, but geographic regions did not represent an independent predictor of PSM.