Real-world (RW) study in patients (pts) with metastatic colorectal cancer (mCRC) with long-term responses to regorafenib in the USA.

Abstract

48 Background: Anecdotal evidence suggests that some pts with mCRC experience significantly longer responses to REG in clinical care settings, including a case exceeding 9 years. However, comprehensive studies on long-term responders to REG are lacking. We evaluated demographic/clinical characteristics of pts with long-term response (LTR) to REG using duration of treatment (DOT) as a surrogate for treatment response. Methods: This was a retrospective cohort study using the US nationwide de-identified Flatiron Health Electronic Health Record-derived database. The study period was January 1, 2013 to May 31, 2023, and adult pts with mCRC who initiated REG monotherapy (index date) from July 1, 2013 to December 31, 2022 were included. DOT for pts with LTR of ≥5 months (LTR5 [primary objective]; mos) or ≥4 mos (LTR4 [secondary objective]) were determined. Results: A total of2,326 eligible pts (median age 64 yrs; 56% male) had initiated REG during the study period. Overall, 346 pts (15%; median age 65 yrs) had LTR5 and 503 pts (22%; median age 65 yrs) had LTR4. Among pts with LTR5, 46% had stage IV disease at initial diagnosis, 68% had ECOG performance status (ECOG PS) 0‒1 at index, 64% received prior bevacizumab (BEV), and of pts tested at index with results, the median carcinoembryonic antigen (CEA) level was 35 ng/mL, and 51% and 5% had a KRAS and BRAF mutation at index, respectively; pts with LTR4 had similar characteristics (Table). Median time from initial CRC diagnosis to index date was 39.2 mos for pts with LTR5 and 38.6 mos for pts with LTR4. Of pts who initiated REG before 2019 (n = 1,070), 14% had LTR5 and 21% had LTR4; of pts who initiated REG from 2019 onwards (n = 1,256), 16% had LTR5 and 22% had LTR4. Among pts with LTR5, 33% and 23% had received REG as third-line (3L) or fourth-line (4L) treatment, respectively; among pts with LTR4, 34% and 23% had received REG as 3L or 4L treatment, respectively. Median time to discontinuation of REG was 7.3 (95% CI 6.9, 7.8) mos in pts with LTR5 and 6.0 (95% CI 5.7, 6.2) mos in pts with LTR4. Median follow-up time (from index date) was 13.3 mos and 11.1 mos in pts with LTR5 and LTR4, respectively. Conclusions: This is the first large-scale RW evidence study to describe demographic/clinical characteristics of pts with LTR to REG. Pts demonstrating LTR typically had favorable ECOG PS at REG initiation, less advanced disease at initial diagnosis, and the majority had received prior BEV. [Table: see text]