Real-world outcomes of patients with BRAF-mutated mCRC treated in the United States.

Abstract

4030 Background: BRAF mutations portend a poor prognosis in metastatic colorectal cancer (mCRC). Recent trials have hypothesized that using more aggressive triplet-based chemotherapy regimens such as FOLFOXIRI in the frontline setting may improve outcomes in this patient population. In this study, we utilized real-world data to assess whether FOLFOXIRI is being used in the United States (US) and compared survival outcomes in BRAF mutated (BRAFmt) mCRC stratified by first line (1L) therapy. Methods: The nationwide Flatiron Health EHR-derived de-identified database was reviewed for patients diagnosed with mCRC between 2013 and 2018. Patients who had documented BRAF mutation testing and received a standard 1L therapy were included for analysis. Patients who did not have a visit or medication order within 90 days of metastatic diagnosis were excluded to ensure patients were engaged with care at the data-providing institution. Kaplan-Meier and Cox proportional hazard modeling were used to compare survival outcomes stratified by BRAF mutation status and 1L therapy received. Results: A total of 4,454 patients with documented BRAF mutational status were included, of which 3,988 (89.5%) were BRAF wild type (BRAFwt) and 466 (10.5%) were BRAFmt. Median OS was 15.4 months (mo) in the BRAFmt group compared to 28.1 mo in the BRAFwt group (HR 0.48, 95% CI 0.41- 0.56, p < 0.001). Only 3% (n = 16) of BRAFmt patients received 1L FOLFOXIRI +/- bevacizumab with a median OS of 13.8 mo compared to 15.5 mo in patients receiving a chemotherapy doublet (FOLFOX, CAPEOX, or FOLFIRI) +/- bevacizumab (95% CI 4.9 – not reached vs 14.3 – 19.0, p = 0.38). In BRAFmt patients, multivariate analysis (MVA) did not detect a significant improvement in OS with the use of FOLFIRI plus bevacizumab (HR 0.88, 95% CI 0.50-1.56, p = 0.67) or FOLFOX/CAPEOX plus bevacizumab (HR 0.89, 95% CI 0.59 – 1.34, p = 0.58) when compared to chemotherapy doublet alone. A MVA comparing 1L therapies in the BRAFwt group did not detect a significant improvement in OS with bevacizumab plus chemotherapy doublet compared to chemotherapy doublet alone. When stratified by 1L treatment regimen, similar proportions of BRAFmt patients received second line therapy. Conclusions: This analysis of real-world data confirms the negative prognostic impact of BRAF mutations in mCRC and suggests that FOLFOXIRI has not been widely adopted in the management of these patients in the US. We were unable to demonstrate any significant difference in OS of patients with BRAFmt mCRC based on type of 1L therapy received.