Background: To test the hypothesis that patients with pT2 and positive surgical margins (SM) have a similar biochemical-recurrence (BCR) risk to patients with pT3a, and negative SM. Moreover, we examined the effect of incorporating positive SM as a higher stage on the discrimination accuracy of the current TNM staging system.
Materials and methods: We evaluated 1,503 prostate cancer patients treated with radical prostatectomy, between 1998 and 2010. Only individuals with pT2N0 or pT3aN0, without neoadjuvant and/or adjuvant therapy, were included. Cox regression analyses tested the relationship between SM status (negative [R0] vs. positive [R1]) and BCR rate, after stratification according to T stage. Predictive accuracy of the current T stage and of a novel T stage, which consider positive SM as a higher stage, was quantified with Harrell's concordance index.
Results: Positive SM rate was 20.3%. The 5-year BCR rates were 96%, 82%, 78%, and 62% for patients with, respectively, pT2R0, pT2R1, pT3aR0, and pT3a1 (all P ≤ 0.03). In multivariable analyses, the BCR rate was 3.6-, 2.5-, and 6.0-fold higher (all P < 0.001) in patients with, respectively, pT2R1, pT3aR0, and pT3aR1 stage relative to patients with pT2R0 stage. The maximum univariable (14.1%) and multivariable (6.9%) discrimination accuracy gains were observed, when tumor stage was stratified into pT2R0 vs. pT2R1/pT3R0 vs. pT3R1.
Conclusions: The presence of positive SM at radical prostatectomy (RP) specimen substantially increases the BCR risk. Patients with pT2R1 have similar BCR risk to patients with pT3aR0. Considering these patients as 1 category substantially improves the discrimination accuracy of the current TNM staging system.
Keywords: Local; Neoplasm recurrence; Pathology; Predictive value of tests; Prostatectomy; Prostatic neoplasms; Surgery.
Copyright © 2013 Elsevier Inc. All rights reserved.