Abstract

INTRODUCTION AND OBJECTIVES: Focal therapy has emerged as an intriguing concept for the treatment of prostate cancer (PCa) with hemiablation of unilateral PCa being the most common approach. However, the incidence of unilateral PCa is not well defined and candidate selection methods are not agreed upon. We assess the incidence of unilateral PCa in two independent cohorts and identify predictors that may aid in candidate selection. METHODS: Retrospective analysis of the Duke Prostate Center (DPC) and Shared Equal Access Cancer Hospital (SEARCH) radical prostatectomy databases. Men with clinically localized PCa were included. Final study cohorts comprised 2616 (DPC) and 1636 (SEARCH) men. Incidence of unilateral (pT1a,b) PCa were described and analyses of predictors were carried out using multivariable regression models. RESULTS: Overall, pathologically unilateral PCa was identified in 13.5 and 13.7% in DPC and SEARCH cohorts, respectively. Among men with unilaterally positive biopsy, the incidence approached 20% in both cohorts. Multivariable analyses showed consistent results. PSA levels, number of positive cores and biopsy Gleason scores were associated with unilateral PCa, when adjusted for cancer laterality on biopsy. Based on these results we developed a scoring system according to the likelihood of unilateral PCa. The scores were categorized in high ( 6 points), intermediate (3-5 points), and low (0-2 points). High, intermediate and low scores had unilateral PCa identified in 22.4-24.7%, 10.5-12.6% and 3.0-3.1%, respectively. Areas under the curve for this score were 0.712 in the DPC and 0.714 in SEARCH cohorts. CONCLUSIONS: In this multicenter study the incidence of unilateral PCa was overall low (13-14%), rising up to 30% in certain subgroups of men. PSA levels, number of positive cores and biopsy Gleason scores were independently associated with unilateral disease when adjusted for biopsy laterality. Based on these findings we propose a score reflecting the likelihood of unilateral disease in men with localized PCa.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call