The impact of solitary and multiple positive surgical margins on hard clinical end points in 1712 adjuvant treatment-naive pT2-4 N0 radical prostatectomy patients.

ABSTRACT

BACKGROUND: Positive surgical margins (PSMs) increase the risk of biochemical recurrence (BCR) after radical prostatectomy (RP), but their impact on hard clinical end points is a topic of ongoing discussion. OBJECTIVE: To evaluate the influence of solitary PSMs (sPSMs) and multiple PSMs (mPSMs) on important clinical end points. DESIGN, SETTING, AND PARTICIPANTS: Data from 1712 patients from the Centre Hospitalier Universitaire de Québec with pT2-4 N0 prostate cancer (PCa) and undetectable prostate-specific antigen after RP were analyzed. INTERVENTION RP without neoadjuvant or adjuvant treatment. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Kaplan-Meier analysis estimated survival functions, and Cox proportional hazards models addressed predictors of clinical end points. RESULTS AND LIMITATIONS: Median follow-up was 74.9 mo. A total of 1121 patients (65.5%) were margin-negative, 281 patients (16.4%) had sPSMs, and 310 patients (18.1%) had mPSMs. A total of 280 patients (16.4%) experienced BCR, and 197 patients (11.5%) were treated with salvage radiotherapy (SRT). Sixty-eight patients (4.0%) received definitive androgen deprivation therapy, 19 patients (1.1%) developed metastatic disease, and 15 patients (0.9%) had castration-resistant PCa (CRPC). Thirteen patients (0.8%) died from PCa, and 194 patients (11.3%) died from other causes. Ten-year Kaplan-Meier estimates for BCR-free survival were 82% for margin-negative patients, 72% for patients with sPSMs, and 59% for patients with mPSMs (p<0.0001). Time to metastatic disease, CRPC, PCa-specific mortality (PCSM), or all-cause mortality did not differ significantly among the three groups (p=0.991, p=0.988, p=0.889, and p=0.218, respectively). On multivariable analysis, sPSMs and mPSMs were associated with BCR (hazard ratio [HR] 1.711; p=0.001 and HR 2.075; p<0.0001), but sPSMs and mPSMs could not predict metastatic disease (p=0.705 and p=0.242), CRPC (p=0.705 and p=0.224), PCSM (p=0.972 and p=0.260), or all-cause death (p=0.102 and p=0.067). The major limitation was the retrospective design. CONCLUSIONS: In a cohort of patients who received early SRT in 70% of cases upon BCR, sPSMs and mPSMs predicted BCR but not long-term clinical end points. Adjuvant radiotherapy for margin-positive patients might not be justified, as only a minority of patients progressed to end points other than BCR. PCSM was exceeded 15-fold by competing risk mortality.