Analysis of KRAS and BRAF mutant colorectal cancers in a multiracial population.

Abstract

1599 Background: We previously demonstrated racial disparities in colorectal cancer (CRC) survival despite adjustments for tumor and socioeconomic factors. Molecular differences in breast and lung tumors contribute to racial survival inequality, yet in CRC, molecular tumor characteristics have not been studied extensively in a racially diverse cohort. We performed a comprehensive evaluation of KRAS and BRAF mutations in a multiracial population to determine the prevalence in CRC and the degree to which these molecular traits impact survival. Methods: A retrospective cohort study of patients diagnosed with CRC between 2008 and 2011 from hospital tumor registries was performed. The prevalence of KRAS and BRAF mutations was determined for the study population and individual racial groups. Multivariable Cox proportional hazards regression models for survival were built for KRAS and BRAF mutation status while adjusting for age at diagnosis, race, and stage of disease. Results: Of 706 patients diagnosed with CRC, KRAS mutational analysis was performed on 148 subjects. 14% of subjects were white (W), 64% Asian (A), and 21% Native Hawaiian/Pacific Islander (NH). KRAS mutation was identified in 48 subjects (32%). The prevalence of mutant tumors among racial groups was W 33%, A 36%, and NH 30%. Analysis for KRAS G13D mutations revealed a prevalence of W 11%, A 9%, and NH 7%. When compared to published datasets of predominantly white patients, our multiracial cohort had a significantly higher rate of KRAS G13D mutant tumors, p=0.039. Of 74 subjects tested for the BRAF mutation, two mutant tumors were detected (3%). The prevalence of the BRAF mutation by race was 10% W, 3% A, and 0% NH (p=0.18). BRAF and KRAS G13D mutations were adverse prognostic factors in a multivariate analysis, although the odds ratios failed to meet statistical significance. Conclusions: The prevalence of BRAF and KRAS mutations in this multiracial cohort is similar to what has been previously reported. However the rates of KRAS G13D and BRAF mutant tumors in our cohort are higher than prior reports. Furthermore, KRAS G13D which has been postulated to be a favorable prognostic factor for CRC, may adversely impact survival of minority patients.