Relationship of phosphodiesterase type 5 inhibitor to biochemical recurrence after definitive therapy for prostate cancer.

Abstract

119 Background: Phosphodiesterase type 5 inhibitors (PDE5i) are commonly utilized among men receiving definitive treatment for prostate cancer (PCa). Recently, contradictory clinical evidence regarding the association between biochemical recurrence (BCR) and receipt of PDE5i has been reported. We aimed to clarify this question among a multi-center, cohort of men with PCa. Methods: Participants enrolled in Cancer of the Prostate Strategic Urologic Research Endeavor (CaPSURE) at diagnosis and received radical prostatectomy, RP, or radiation therapy, RT (external beam radiation therapy or brachytherapy) without androgen deprivation therapy ≤ 6 months. We examined associations between clinical, Prostate Cancer Index Sexual Function score, SF (0-100 (best)), and pathological characteristics as well as baseline, time-dependent and ever-use of PDE5i (patient or physician reported) with risk of BCR using descriptive analyses, Kaplan-Meier method, and multivariate Cox proportional hazards models. Results: A total of 4,844 men were identified (3,847 RP, 997 RT.) Men received RP at median age 61 years (IQR 56-66), had a mean baseline SF score 62.4 (SD 26.7), were followed for a median of 77 months (IQR 36-117), and 3,089 (80%) were prescribed or reported use of PDE5i post-operatively. Patients treated with RT were older with a median age 69 years (IQR 63-74) and mean baseline SF score 46.6 (SD 29.5), were followed for a median of 91 months (IQR 53-125), and had lower rates of reported PDE5i usage (421 men, 42%). In Cox regression models baseline use (HR = 0.9 (95% CI 0.3-2.4) p = 0.83) and ever-usage (HR = 0.7 (0.5-1.1) p = 0.73) of PDE5i were not independently associated with BCR after RP. Time-dependent use (HR 1.3 (0.6, 2.7) p = 0.45) also was not associated with BCR after RP among men taking PDE5i medications. Baseline (HR = 4.6 (0.6-37.3) p = 0.14) and ever-usage (HR 0.6 (0.3-1.5) p = 0.30) also were not associated with BCR after RT. This study was limited by incomplete data on prescription compliance. Conclusions: PDE5i usage following RP or RT for PCa was not associated with BCR. PDE5i therapy should not be withheld due to concerns that it increases the risk of disease recurrence.