miR-200 family as new potential prognostic factor of overall survival of patients with WHO G2 and WHO G3 brain gliomas.

Abstract

Gliomas are the predominant cause of cancer-related deaths among the young population. Even after incorporation of IDH1/2 mutations and 1p19q codeletion there are doubts regarding adjuvant treatment in WHO G2/G3 gliomas. miRNA molecules control about 30% of all genes, also many oncogenes, tumor suppressor genes and genes responsible for the response to ionizing radiation and systemic treatment. Patients with brain gliomas exhibit miRNA disorders. We aimed to evaluate the expression of miR-200 family members in relation to selected clinico-pathological factors and their prognostic value. We enrolled 53 patients diagnosed with WHO G2/G3 brain gliomas treated between 2012-2016. RT-qPCR based expression of miR-200 family was assessed in tumor and surrounding non-cancerous tissue. An analysis of selected clinico-pathological features was carried out. A logistic regression model was prepared for the miRNA signature. The predictive potential of the signature was assessed using the ROC curve. A stepwise backward regression model was used to select variables with a significant predictive potential related to OS. It was shown that miR-200a-3p, miR-200a-5p, miR-200c-5p, miR-141-3p and miR-429 can be independent predictors of survival. Better 2- and 5-year OS was associated with higher expression of miR-200a-3p, miR141-3p and lower expression of miR-200a-5p, miR-200c-5p, miR-429. The strongest predictors of survival were miR-200a-5p, miR-200b-3p, miR-200c-5p, miR-141-3p, miR-429, tumor volume and CTV. Members of the miR-200 family exhibit prognostic value for 2- and 5-year OS. Presented predictive models of survival may be clinically useful for treatment optimization.