Does time from diagnosis to treatment of high- or very-high-risk prostate cancer affect outcome?

Abstract

To determine whether time from diagnosis to treatment impacted outcomes in a multicentre cohort of high- and very-high-risk (VHR) patients with prostate cancer undergoing radical prostatectomy (RP). In all, 1392 patients from three tertiary centres who underwent RP for either high-risk or VHR disease, from 2005 to 2015, were identified. The cohort was divided into tertiles based on time from diagnostic biopsy to RP. Cumulative incidence of biochemical recurrence (BCR), metastasis, and prostate cancer-specific mortality (PCSM) were calculated for each tertile. The Kaplan-Meier method was used to evaluate for differences in all-cause mortality (ACM) amongst tertiles. Competing risks regression models, as well as Cox proportional hazards regression models, were fitted to assess the association between time-to-event outcomes and patient characteristics. The median (interquartile range [IQR]) time from biopsy to RP was 68 (50-94) days. The median (IQR) follow-up was 31 (12.1-55.7) months. The cumulative incidence of BCR (P = 0.14), metastasis (P = 0.15), and PCSM (P = 0.69) did not differ amongst time-to-treatment tertiles of VHR patients. Also, Kaplan-Meier estimates of ACM (P = 0.53) did not differ amongst time-to-treatment tertiles. Similarly, BCR, metastasis, PCSM, and ACM did not significantly differ amongst time-to-treatment tertiles in multivariable modelling. In this pooled meta-dataset of patients with high-risk or VHR prostate cancer, time from diagnosis to RP did not appear to significantly contribute to differences in clinical outcomes. This finding supports the safety of enrollment of such patients into neoadjuvant clinical trials.