A lncRNA and radiomics-based model for predicting the response of non-small cell lung cancer to chemo- and radio-therapy.

This study aimed to identify a novel plasma lncRNA biomarker and establish a model based on lncRNAs and radiomics for predicting the response of non-small cell lung cancer (NSCLC) to chemo- and radio-therapy. Next-generation sequencing and integrated bioinformatics analysis were used to identify lncRNAs associated with the response of NSCLC to chemo- and radio-therapy. RT-qPCR was utilized to detect MIF-AS1 expression in the plasma of NSCLC patients. Radiomics analysis was performed on CT images of NSCLC patients. The model was constructed by multiple logistic regression. The expression of the lncRNA MIF-AS1 was up-regulated in the plasma of patients with chemo- and radio-resistant NSCLC, as validated by RT-qPCR in 124 NSCLC patients. Furthermore, in vitro experiments demonstrated that knockdown of MIF-AS1 expression significantly reduced cellular proliferation and invasion, and increased the sensitivity of NSCLC cells to the chemotherapeutic drug cisplatin. Using the ceRNA network, a DNA-damage repair related protein RAD21 was identified as a target gene of MIF-AS1. Finally, two radiomic features were found to be associated with the response of NSCLC to chemo- and radio-therapy. Combining the MIF-AS1 level and the two radiomic features, a model was established to predict the response of NSCLC to chemo- and radio-therapy, with a high AUC of 0.808. MIF-AS1 could be a novel biomarker for predicting the response of NSCLC to chemo- and radio-therapy. This model, which uses both CT radiomics and MIF-AS1 levels, increases the accuracy of predicting therapeutic response in NSCLC patients.