Real-world treatment (Tx) sequences and time to discontinuation (rwTTD) in the first-line (1L) metastatic castration-resistant prostate cancer (mCRPC) setting.

Abstract

55 Background: There is limited real-world data on tx sequencing in metastatic prostate cancer. Additionally, while guidelines support adding novel hormonal agents (NHA) to androgen deprivation tx (ADT) for metastatic hormone-sensitive prostate cancer, use is limited because of clinical and/or pt-driven factors. We evaluated tx patterns, rwTTD, and time-to-chemotherapy (rwTTC) for 1L mCRPC pts, including Black/African Americans (BAA) treated predominantly in a community oncology setting. Additionally, we analyzed systemic treatments prior to 1L mCRPC as well as progression to 2L mCRPC. Methods: This retrospective study included pts with mCRPC from the nationwide Flatiron Health EHR-derived, deidentified database who initiated 1L tx between 1/2013 and 1/2023. Study end was 4/2023. Tx patterns were described by time period (TP) of 1L initiation: TP1 - 01/2013–12/2017 and TP2 - 01/2018–01/2023; Kaplan-Meier analyses were conducted for rwTTD/TTC and a Cox proportional hazards regression model was used to determine the association between pt characteristics and rwTTD/TTC for 1L mCRPC. Subgroup analyses were performed for BAAs. Results: Of the 8915 1L-tx pts (median age 74 yrs), 871 (10%) were BAA (median age 70 yrs) and 5870 (66%) were NHA-naïve at 1L mCRPC. Overall, the percent of pts who were NHA-naïve decreased from 2018 (72%) to 2023 (39%) and a similar trend was observed for BAA (74% [2018] to 25% [2023]). 1L NHA use in mCRPC increased from TP1 to TP2 (overall: 70% to 80%; BAA: 73% to 83%), primarily driven by enzalutamide (Enz) (overall: 30% to 39%; BAA: 33% to 38%) and decrease in chemotherapy (overall: 16% to 12%; BAA: 17% to 13%). The top 3 1L mCRPC regimens for the overall population and BAA were abiraterone (Abi) (35%; 38%), Enz (32%; 33%), and docetaxel (10%; 11%). Among pts who progressed to 2L mCRPC, the most common sequences for patients tx with Abi or Enz for 1L mCRPC were ADT monotherapy in the prior line of tx and a switch to Abi (19%) or Enz (24%) upon progression from 1L to 2L. 1L rwTTD for the overall cohort is in the Table. Age ≥80 years, ECOG PS ≥1, socioeconomic status = 5 (highest), and median time ≤39.8 months from initial dx to mCRPC were associated with a higher hazard of discontinuation (all p ≤0.05). Conclusions: Overall, most pts received their first NHA in 1L mCRPC; NHA use increased and chemotherapy use decreased from TP1 to TP2. There was heterogeneity in tx sequences in patients treated with an NHA in 1L mCRPC; some pts were rechallenged with a different NHA in 2L despite lack of robust prospective data to support use. Overall, 1L rwTTD was short with a small numerical increase in TP2. Factors contributing to discontinuation, such as treatment resistance, need to be better understood to further improve real-world outcomes. [Table: see text]