A microRNA expression signature for clinical response in locally advanced cervical cancer

Objective. Nearly 50% of patients who are diagnosed with locally advanced cervical cancer have an unfavorable pathological response to conventional treatment. MicroRNAs (miRNAs) are potential biomarkers in cervical cancer; however, their role in identifying patients who do not respond to conventional treatment remains poorly investigated. Here, we identify a set of miRNAs that can be used as molecular markers to predict the pathological response in locally advanced cervical cancer patients receiving radiation and chemotherapy treatment. Methods. Forty-one patients diagnosed with locally advanced cervical cancer were invited to participate in this study and enrolled after they signed an informed consent. Two patient cohorts were randomized for miRNA expression profiling, a discovery cohort (n = 10) and a validation cohort (n = 31); profiling was performed by means of a miScript miRNA PCR Array. After a median clinical follow-up of 45 months, statistical analysis was performed to identify miRNAs that could discriminate non -responders from complete pathological responders to conventional treatment. Results. miRNA expression profiling identified 101 miRNAs that showed significant differences between non responders and complete pathological responders (p < 0.05). Seven differentially expressed miRNAs were selected, and their expression patterns were confirmed in the validation phase; thus, miR-31-3p, -3676, -125a-5p, -100-5p, -125b-5p, and -200a-5p and miR-342 were significantly associated with clinical response. Expression of this miRNA signature above the median level was a significant predictor of non -response to standard treatment (p <0.001). Conclusions. These seven validated miRNA signatures could be used as molecular biomarkers of chemo- and radio-resistance in locally advanced cervical cancer patients. (C) 2016 The Authors. Published by Elsevier Inc.