Background: Capecitabine and Oxaliplatin (CAPOX) combined with the humanised anti-vascular endothelial growth factor Bevacizumab represents a standard first-line treatment for metastatic colorectal cancer (mCRC). However, the response rate is only approximately 50% and currently there is no biomarker to predict treatment response. This study aims to correlate KRAS status with response to CAPOX and Bevacizumab (CAPOX-Bev). Methods: Forty-five patients with mCRC were retrospectively screened between January 2012 and December 2015 at Broomfield Hospital, UK. DNA was extracted from formalin-fixed paraffin-embedded (FFPE) tissue samples (Qiagen FFPE tissue kit). Twelve types of mutations in Exons 2, 3 and 4 of the KRAS gene were analysed using a real-time quantitative polymerase chain reaction (qPCR) assay (EntroGen). Treatment response was assessed according to RECIST criteria version 1.1 by comparing pre-treatment and post-treatment radiological CT scans. The KRAS status was correlated with CT tumour response (responders: partial response and complete response; non-responders: progressive disease and stable disease). A Chi-square test was used to determine the correlation between KRAS status and tumour response. Results: 19/45 patients were KRAS wild type (WT, 42%) and 26 were KRAS mutant (MT, 58%). 8/19 (42%) KRAS WT patients were responders, compared to 3/26 (12%) KRAS MT patients. Conversely 11/19 (58%) of WT patients were non-responders, compared to 23/26 (88%) MT patients. The correlation of treatment response and KRAS status was statistically significant (p = 0.018), with a 31% difference in response rate between KRAS WT (42%) and KRAS MT (12%) groups. Conclusions: Within this pilot retrospective analysis, KRAS mutations demonstrated clinical value in identifying patients who are more likely to respond to first-line CAPOX-Bev in advanced colorectal cancer. This finding requires prospective evaluation within a large patient cohort powered to further detect potential differences in overall and progression free survival.

Chi-square without Yates correction χ²analysisχ² = 5.554, p = 0.0184.