AUA 2018: Development and Validation of a Novel Prognostic Model for Predicting Overall Survival in Treatment Naïve Castration-Sensitive Metastatic Prostate Cancer

San Francisco, CA ( The treatment paradigm for the of management castration sensitive prostate cancer (mCSPC) has changed dramatically over the last 5 years with the publication of the CHAARTED, STAMPEDE and LATITUDE trials. These trials have also demonstrated a survival heterogeneity in patients with castrate sensitive metastatic prostate cancer, arguing that some patients may be overtreated by these aggressive regimens.  A risk stratification system for patients with mCSPC is currently lacking. Dr. Akamatsu, from Kyoto University Hospital, presents a novel prognostic model aimed to predict overall survival (OS) for patients with mCSPC treated with hormone deprivation therapy.  

304 patients with mCSPC treated in Kyoto University Hospital from 1996 to 2016 were included in the analysis. Cox proportional hazards model, was used to identify independent prognostic factors for OS. The distribution of the risk estimates from the model was assessed to determine suitable cutoff points for different risk categories. On multivariate analysis, extent of disease (> 6 metastatic sites on bone or liver metastases), primary Gleason score, and LDH were independently and significantly associated with OS. Using these risk factors patients were segregated into three risk categories. 

The median OS for the entire cohort was 72 months. OS for CHAARTED high-volume (N=172 ) and low-volume (N=110 ) cases were 53, and 131 months respectively. The median OS for high (N=84) and intermediate (N=82) risk groups were 31 and 60 months, and not reached for the low-risk group (N=85). 30% and 34% of the CHAARTED high and low volume cases were classified into intermediate risk group with similar OS, and 11.8% of CHAARTED high-volume patients were reclassified into the low-risk group whose median survival was not yet reached.  An independent multi-institutional cohort of 532 Japanese mCSPC patients diagnosed after 2006 was used to validate the model.  In the validation cohort, the median OS for high (N=214) and intermediate (N=200) risk groups were 41 and 70 months, and not reached for the low-risk group (N=118), further demonstrating the utility of the risk model to appropriately risk stratify patients with mCSPC.  

In summary, Dr. Akamatsu and colleagues have developed a simple, reproducible and translatable predictive model for assessing OS in a Japanese population with mCSPC. The model has significant potential for patient counseling, the guidance of treatment protocols and the design of  future clinical trial if the model is able to be validated in patients treated with the new standard of care therapies (Docetaxel and Abiraterone).  

Presented By: Shunsuke Akamatsu, MD, Kyoto University Hospital, Japan 

Written by: Andres F. Correa, Urologic Oncology Fellow , Fox Chase Cancer Center, Philadelphia, PA, Twitter: @UroCorrea, the 2018 AUA Annual Meeting - May 18 - 21, 2018 – San Francisco, CA USA
email news signup