Constructing a Novel Signature Based on Immune-Related lncRNA to Improve Prognosis Prediction of Cervical Squamous Cell Carcinoma Patients

We downloaded gene expression data, clinical data, and somatic mutation data of cervical squamous cell carcinoma (CSCC) patients from The Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression (GTEx) databases. Predictive lncRNAs were screened using univariate analysis and least absolute shrinkage and selection operator (LASSO) regression, and risk scores were calculated for each patient according to the expression levels of lncRNAs and regression coefficients to establish a risk model that could be a novel signature. We assessed the correlation between immune infiltration status, chemotherapeutics sensitivity, immune checkpoint proteins (ICP), and the signature. Therefore, we selected 11 immune-related lncRNAs (WWC2,AS2, STXBP5.AS1, ERICH6.AS1, USP30.AS1, LINC02073, RBAKDN, IL21R.AS1, LINC02078, DLEU1, LINC00426, BOLA3.AS1) to construct the risk model. Patients who were in the high-risk group had a shorter survival time than those in the low-risk group ( p < 0.001). Risk scores in the signature were negatively correlated with macrophage M1, macrophage M2, and T cell CD8 + ; what’s more, T cell CD8 + was higher in the low-risk group. The expression levels of ICP such as PD-L1, PD-1, CTLA-4, TIGIT, LAG-3, and TIM-3 were substantially higher in the low-risk group. For chemotherapeutic agents, high-risk scores were associated with higher half-inhibitory concentrations (IC50) of cisplatin. These findings suggested that the risk model can be a novel signature for predicting CSCC patients’ prognosis, and it also can be used to formulate clinical treatment plans for CSCC patients.