Support: U10CA180821, U10CA180882, CA31946, Bristol-Myers Squibb, Genentech, Pfizer Background: Irinotecan/5-FU (FOLFIRI) or oxaliplatin/5-FU (FOLFOX), combined with bevacizumab or cetuximab, are first-line treatments for mCRC. We aimed to identify germline variants associated with survival in mCRC patients treated with these regimens. Methods: In CALGB 80405, patients with wild-type KRAS (codons 12 and 13) mCRC received either FOLFOX ( > 80% of patients) or FOLFIRI and were randomized to either cetuximab or bevacizumab. The primary endpoint of the study was OS.DNA was extracted from peripheral blood and genotyped for ~700,000 single-nucleotide polymorphisms (SNPs). The association between SNPs and OS in 609 patients (both arms combined) of European genetic ancestry was tested by a Cox proportional hazards model. Results: Median OS in genotyped patients was 29.6 months, and was comparable to that of the parent study (Venook et al., ASCO 2014). The three most significant SNPs associated with OS were in the genes RDH14 (hazard ratio, HR 1.63, p < 1.12x10^-6), TMEM16J (HR 1.52, p < 2.03x10^-6), and AXIN1 (HR 1.40, p < 4.26x10^-6). Among these genes, the most compelling evidence for a link to the biology of CRC is for AXIN1. The AXIN1 protein functions as a negative regulator of WNT signaling by interacting with APC in CRC (Li et al., Cell 2012). In the present GWAS, rs11644916 (G to A) in AXIN1 is a common germline intronic variant (30% allele frequency). Median OS for patients with the AA, AG or GG genotypes of rs11644916 was 18.4 (95% CI 14.2-27.6), 25.6 (23.6-30.4) or 36.6 (32.9-41.1) months, respectively. Testing of arm and extended RAS analysis as covariates did not alter the association with the rs11644916 in AXIN1. Conclusions: This is the first, large GWAS ever conducted in mCRC patients treated with standard of care in a randomized phase III trial. A common SNP in the AXIN1 gene confers worse OS. This study selects AXIN1 as a new putative determinant of CRC progression. Further replication in additional patient cohorts and functional studies in CRC experimental models are required. Clinical trial information: NCT00265850