Stratifying Risk and Treatment Benefit: A Model Predicting Overall Survival in Men with Metastatic De Novo Hormone-sensitive Prostate Cancer in Trials Investigating Docetaxel (the STOPCAP Collaboration).

BACKGROUND AND OBJECTIVE: This study aimed to develop and validate a prognostic model for overall survival (OS) in men with de novo metastatic hormone-sensitive prostate cancer (mHSPC), using clinical factors from phase 3 trials to improve survival prediction, guide clinical decision-making, and optimize trial design. METHODS: We analyzed individual patient data from three randomized phase 3 trials (four comparisons) of androgen deprivation therapy ± docetaxel: CHAARTED (n = 575), GETUG-15 (n = 272), and STAMPEDE (arm A, n = 689; arm C, n = 347; arm E, n = 351). A proportional hazard model was developed using CHAARTED and GETUG-15 as the training set, and validated using the STAMPEDE comparisons. Model discrimination performance was assessed by the time-dependent area under the receiver operating characteristic curve (tAUC). KEY FINDINGS AND LIMITATIONS: Lower levels of hemoglobin, high disease volume, elevated alkaline phosphatase, and poor Eastern Cooperative Oncology Group performance status were associated with an increased risk of death. The model achieved tAUC values of 0.72 (95% confidence interval [CI]: 0.69-0.76) and 0.70 (95% CI: 0.67-0.72) in STAMPEDE A versus C and STAMPEDE A versus E comparisons, respectively. Risk stratification into two or three groups showed clear differences in OS, with substantially longer OS in low-risk patients. Moreover, patients in the poor-risk group derived the greatest benefit from docetaxel, while low-risk patients derived minimal benefit. A limitation is that the model was built in de novo mHSPC men. CONCLUSIONS AND CLINICAL IMPLICATIONS: This validated model improves OS prediction in de novo mHSPC patients and supports risk-informed treatment selection. Future work should focus on external validation in contemporary settings.

Duke Scholars

Author Susan Halabi Biostatistics & Bioinformatics, Division of Biostatistics

Author David Page Biostatistics & Bioinformatics, Division of Biostatistics