Fc-γ Receptor Polymorphisms, Cetuximab Therapy, and Survival in the NCIC CTG CO.17 Trial of Colorectal Cancer

Purpose: Two germline Fc-gamma receptor (FCGR) polymorphisms, rs1801274 [FCGR2A; His(H) 131Arg(R)] and rs396991 [FCGR3A; Phe(F) 158Val(V)] produce altered proteins through amino acid substitutions; both are reported to be associated with cetuximab-related outcomes. We performed a validation of these polymorphisms in NCIC CTG CO. 17, a randomized trial of cetuximab monotherapy in refractory, metastatic colorectal cancer expressing EGFR. Experimental Design: DNA extracted from formalin-fixed paraffin-embedded tissue was genotyped. In addition to log-rank tests, Cox proportional hazard models assessed their relationships with overall (OS) and progression-free survival (PFS), adjusting for clinically important prognostic factors, along with a polymorphism-treatment arm interaction term. Results: Somatic KRAS status was wild-type for exon 2 in 153 (52%) of 293 patients, from whom tumor DNA was available. For FCGR2A H/H, a genotype-treatment interaction for KRAS wild-type patients was observed for OS (P = 0.03). In KRAS wild-type patients carrying FCGR2A H/H, cetuximab (vs. no cetuximab) improved survival substantially, with adjusted HRs (aHR) of 0.36 (OS) and 0.19 (PFS) and absolute benefits of 5.5months (OS; P = 0.003) and 3.7 months (PFS; P = 0.02). In contrast, patients carrying FCGR2A R alleles (H/R or R/R) had aHRs of only 0.78 (OS; 2.8-month benefit) and 0.53 (PFS; 1.6-month benefit). No relationships were found for rs396991 (FCGR3A). Conclusions: In the CO. 17 trial, cetuximab worked best for patients with KRAS wild-type colorectal cancers carrying FCGR2A H/H genotypes. Significantly lower benefits were observed in patients carrying germline FCGR2A R alleles. (C) 2016 AACR.