External Validation of a Prognostic Model of Overall Survival in Men With Chemotherapy-Naïve Metastatic Castration-Resistant Prostate Cancer.

PURPOSE: We have previously developed and externally validated a prognostic model of overall survival (OS) in men with metastatic, castration-resistant prostate cancer (mCRPC) treated with docetaxel. We sought to externally validate this model in a broader group of men with docetaxel-naïve mCRPC and in specific subgroups (White, Black, Asian patients, different age groups, and specific treatments) and to classify patients into validated two and three prognostic risk groupings on the basis of the model. METHODS: Data from 8,083 docetaxel-naïve mCRPC men randomly assigned on seven phase III trials were used to validate the prognostic model of OS. We assessed the predictive performance of the model by computing the time-dependent area under the receiver operating characteristic curve (tAUC) and validated the two-risk (low and high) and three-risk prognostic groups (low, intermediate, and high). RESULTS: The tAUC was 0.74 (95% CI, 0.73 to 0.75), and when adjusting for the first-line androgen receptor (AR) inhibitor trial status, the tAUC was 0.75 (95% CI, 0.74 to 0.76). Similar results were observed by the different racial, age, and treatment subgroups. In patients enrolled on first-line AR inhibitor trials, the median OS (months) in the low-, intermediate-, and high-prognostic risk groups were 43.3 (95% CI, 40.7 to 45.8), 27.7 (95% CI, 25.8 to 31.3), and 15.4 (95% CI, 14.0 to 17.9), respectively. Compared with the low-risk prognostic group, the hazard ratios for the high- and intermediate-risk groups were 4.3 (95% CI, 3.6 to 5.1; P < .0001) and 1.9 (95% CI, 1.7 to 2.1; P < .0001). CONCLUSION: This prognostic model for OS in docetaxel-naïve men with mCRPC has been validated using data from seven trials and yields similar results overall and across race, age, and different treatment classes. The prognostic risk groups are robust and can be used to identify groups of patients for enrichment designs and for stratification in randomized clinical trials.