Baseline characteristics may impact treatment duration of cabazitaxel in patients with mCRPC: a subanalysis of data from a post-marketing surveillance

Objective: Cabazitaxel has demonstrated improvements in overall survival among patients with metastatic castration-resistant prostate cancer (mCRPC) in the pivotal comparison clinical trials TROPIC, PROSELICA and CARD. However, these trials include mCRPC patients with similar characteristics, and there are limited data on how baseline characteristics affect treatment discontinuation in the patient population. Methods: To assess individual factors that may impact the discontinuation rate of cabazitaxel treatment, we conducted a post hoc analysis of data from a nationwide all-case, post-marketing surveillance of cabazitaxel in Japan. Patients were grouped according to the number of cabazitaxel treatment cycles received (1-2 and ≥3 cycles). Predictive factors were identified through multivariate logistic regression analysis. Results: Across 660 patients with metastatic castration-resistant prostate cancer, 70.2% received ≥3 cycles of cabazitaxel treatment. Those receiving 1-2 cycles of cabazitaxel had a greater proportion of patients with poorer Eastern Cooperative Oncology Group Performance Status, presence of lung and liver metastases, higher prostate-specific antigen level and prior radiation therapy at baseline. Regardless of the number of cabazitaxel cycles received, the primary reason for discontinuation was progression of disease rather than adverse events. Compared with those receiving 1-2 cycles, a lower proportion of patients receiving 3-10 and ≥11 cycles of cabazitaxel treatment experienced adverse events. Multivariate analysis showed a significant association between early discontinuation and presence of liver lesions, poorer Eastern Cooperative Oncology Group Performance Status and higher prostate-specific antigen level at baseline. Conclusions: Post-marketing surveillance data suggest physicians should individualize cabazitaxel treatment based on certain patient characteristics at baseline.