MicroRNA profiling predicts survival in anti-EGFR treated chemorefractory metastatic colorectal cancer patients with wild-type KRAS and BRAF.

Anti-EGFR monoclonal antibodies (anti-EGFRmAb) serve in the treatment of metastatic colorectal cancer (mCRC), but patients with a mutation in KRAS/BRAF and nearly one-half of those without the mutation fail to respond. We performed microRNA (miRNA) analysis to find miRNAs predicting anti-EGFRmAb efficacy. Of the 99 mCRC patients, we studied differential miRNA expression by microarrays from primary tumors of 33 patients who had wild-type KRAS/BRAF and third- to sixth-line anti-EGFRmAb treatment, with/without irinotecan. We tested the association of each miRNA with overall survival (OS) by the Cox proportional hazards regression model. Significant miR-31* up-regulation and miR-592 down-regulation appeared in progressive disease versus disease control. miR-31* expression and down-regulation of its target genes SLC26A3 and ATN1 were verified by quantitative reverse transcriptase polymerase chain reaction. Clustering of patients based on miRNA expression revealed a significant difference in OS between patient clusters. Members of the let-7 family showed significant up-regulation in the patient cluster with poor OS. Additionally, miR-140-5p up-regulation and miR-1224-5p down-regulation were significantly associated with poor OS in both cluster analysis and the Cox proportional hazards regression model. In mCRC patients with wild-type KRAS/BRAF, miRNA profiling can efficiently predict the benefits of anti-EGFRmAb treatment. Larger series of patients are necessary for application of these miRNAs as predictive/prognostic markers.