Observation Following Radical Prostatectomy (RP) With Positive Margin(s) (PM): Does a Low-Risk Subgroup Exist?

Abstract

Following RP for prostate cancer (PC), PM is a well-established risk factor for prostate-specific antigen (PSA) failure. Despite phase 3 evidence confirming benefit of adjuvant radiation therapy (RT) in this setting, many patients are not referred owing to the observation that not all patients with PM recur. The primary objective of this study is to determine whether expanded pathologic factors may permit substratification of risk within a well-defined cohort of PM RP patients who did not undergo adjuvant RT. Multi-institutional retrospective analysis of patient- and tumor-specific factor associations with disease control. Eligible patients underwent RP for clinically localized PC, without pathologic involvement of seminal vesicles or lymph nodes and >1 positive surgical margin (defined as ink on tumor). Patients were excluded for pre-RP PSA >30 or adjuvant (nonsalvage) RT or hormone therapy. PSA failure was defined as >0.2 ng/mL and rising. Cox regression was used to assess the effects of demographic and clinicopathologic variables on freedom from failure (FFF). Kaplan-Meier method was employed for disease control estimates. Ad hoc substratifications within post-RP PSA and pathologic features were examined using log-rank. From 2002 to 2010, 215 patients were eligible for this analysis. There was no difference in outcome or follow-up between institutions. Median age at diagnosis was 61 years (range, 43-76 years), and pre-RP PSA 5.8 (1.6-26). At median follow-up of 78 months (14-155 months), 85 patients had experienced PSA relapse. On univariate analysis, PSA failure was associated with higher primary Gleason grade (4-5 vs 3) and total Gleason score (7-9 vs 6) at RP, presence of capsule invasion at positive margin site, and higher initial post-RP PSA (<0.1 vs >0.1, performed within 6 months of RP), with initial post-RP PSA remaining significant at multivariate analysis. Multiple comparisons analysis yielded 4 discrete risk groups (Table). Of note, 45 patients underwent salvage RT, of whom 38 remained without evidence of PSA rise at last follow-up. Within the present study, high 5-year FFF rates were observed for the subgroup of PM RP patients with Gleason score 6 and initial post-op PSA <0.1, without further therapy. These findings are hypothesis-generating, and longer follow-up is needed to determine if this group of PC patients may be safely observed, with RT reserved for the salvage setting.Poster Viewing Abstracts 2515 Table 1Group5 Year FFF95% CIUnadjusted P valueInitial post-op PSA <0.1, GS 686%75-92%-Initial post-op PSA <0.1, GS 7-964%53-73%.06Initial post-op PSA ≥0.1, GS 633%6-66%<.01Initial post-op PSA ≥0.1, GS 7-97%<1-25%<.01 Open table in a new tab