Primary tumor location as a predictor of survival in patients with RAS wild-type colorectal cancer who receive molecularly targeted drugs as first-line therapy: a multicenter real-world observational study by the Japanese Society for Cancer of the Colon and Rectum

Background Primary tumor location is considered a predictor of overall survival (OS) in RAS wild-type (WT) metastatic colorectal cancer (mCRC) treated with bevacizumab (BEV) or an anti-epidermal growth factor antibody (cetuximab or panitumumab [CET/PAN]) as first-line molecularly targeted therapy. BEV is recommended for right-sided mCRC and CET/PAN for left-sided mCRC based on post-hoc analyses of clinical trial data, but real-world evidence is lacking. Methods We retrospectively collected data of patients who started BEV or CET/PAN plus 5-fluorouracil-based doublet chemotherapy between January 2013 and December 2016 as first-line treatment for RAS WT mCRC at any of 24 Japanese institutions. OS was compared between the BEV and CET/PAN groups according to primary tumor location by Cox multivariate regression analysis in the full cohort and in a propensity score-matched cohort. Results In total, 935 patients were enrolled. Median OS was 24.6 months with BEV and 20.9 months with CET/PAN in right-sided mCRC ( n = 213; adjusted hazard ratio [HR] 0.73, 95% confidence interval [CI] 0.50–1.06) and 35.7 months and 30.0 months, respectively, in left-sided mCRC ( n = 722; adjusted HR 0.92, 95% CI 0.74–1.13). In the propensity score-matched cohort, OS was significantly better in the BEV group than in the CET/PAN group in right-sided mCRC (HR 0.52, 95% CI 0.28–0.96) but was not significantly different in left-sided mCRC (HR 0.78, 95% CI 0.53–1.07). Conclusion Real-world data showed that OS was better with BEV than with CET/PAN in right-sided mCRC. However, there was no significant difference in OS in left-sided mCRC.