A novel method of analysis that uses standard of care PSA values to help guide optimal treatment decisions in the management of metastatic prostate cancer.

61 Background: Metastatic prostate cancer is commonly treated with androgen deprivation therapy (ADT) combined with abiraterone, enzalutamide or chemotherapy. However, these regimens have not been directly compared in head-to-head trials. This retrospective study leverages data from Kaiser Permanente Northern California (KPNC) to compare these treatment regimens and introduces a novel analysis of a tumor’s growth rate as a metric that can be leveraged to guide optimal treatment decisions in the management of patients with metastatic prostate cancer. Methods: We identified 2,327 patients with a diagnosis of prostate cancer in the KPNC Health Connect database (1/1/2011–12/31/2021) who were prescribed abiraterone, enzalutamide, or docetaxel across different lines of treatment. Serial prostate-specific antigen (PSA) values were collected during each treatment line. Using these PSA values, we applied a tumor growth equation to calculate tumor growth rates (g-rates) for each treatment regimen and correlated these rates with overall survival (OS). Results: Tumor growth rates were calculated using PSA values. As has been shown in multiple diseases the rates tumor growth is inversely correlated with overall survival (OS), an endpoint as important in real-world practice as it is in clinical trials. Kaplan-Meier (K-M) curves for all-cause mortality by tumor growth rate quartiles demonstrated that OS was consistent with tumor growth rates. Survival was significantly inversely corelated with the log of growth rates. This correlation was apparent both overall and across all lines of treatment analyzed. As expected, the rate of tumor growth while receiving first-line abiraterone was slower compared to second-line abiraterone. Similar trends were observed for enzalutamide and docetaxel in both first- and second-line settings. However, first-line abiraterone treatment achieved a greater reduction in tumor growth rate than did first-line enzalutamide treatment. Box plots comparing tumor growth rates among patients on abiraterone, enzalutamide, and docetaxel found statistically indistinguishable rates across racial/ethnic groups, age, and quintiles of Neighborhood Deprivation Indices (NDI). Conclusions: This study demonstrates that tumor growth rates can be estimated using a novel analytical approach in a real-world setting. While both abiraterone and enzalutamide show excellent real-world outcomes as effective and tolerable therapies, our findings suggest that enzalutamide should follow abiraterone in treatment sequencing. Estimates of tumor growth rates using real-world data can help evaluate the efficacy of therapies in prolonging survival and inform optimal treatment selection for patients with metastatic prostate cancer.