A combined tissue microRNA panel is a predictive biomarker for colorectal cancer recurrence

Abstract

Background: Colorectal cancer (CRC) is one of the leading causes of cancer death worldwide and recurrence is a main factor for its poor prognosis. A promising biomarker is warranted for predicting the risk of recurrence especially when deciding the course of treatment. MicroRNA dysregulation has been extensively reported to play important roles in CRC development and progression, whereas emerging studies also demonstrating molecular changes in normal tissue adjacent to the tumor (NAT) is predictive for cancer progression. We and others previously showed that certain tumor or NAT microRNAs levels were significantly different between CRC patients who developed post-operative recurrent disease (recurrent group) and who did not (non-recurrent group). The aim of this study is to identify a panel of combined tumor and NAT microRNA level as a predictive biomarker for CRC recurrence. Methods: This is a retrospective case control study. MicroRNA was extracted from tumor and NAT specimens of 54 patients in recurrent group and non-recurrent group. QPCR was performed to determine the levels of miR-509-3p (our unpublished data) and other reported miRNAs. Statistical analyses were performed to compare the microRNA levels between recurrent and non-recurrent groups. ROC curves were plotted to determine the biomarker potential. Multiple linear regression was applied to formulate a microRNA panel for predicting the recurrent risk in CRC patients. Results: The NAT levels of miR-509-3p and miR-320e and overexpression of miR-21-5p and miR-509-3p in CRC were significantly different between recurrent and non-recurrent group (p<0.05). ROC analyses demonstrated that they were predictive biomarkers for recurrence (p<0.05) with AUC values 0.65 to 0.69. More importantly, the AUC value further increased to 0.80 for the combined panel. Conclusions: These data suggest that combined tumor and NAT microRNA panel is indeed a promising biomarker for identifying CRC patients with high recurrent risk. Copyright © 2021 European Society for Medical Oncology. Published by Elsevier Ltd.