An Immune-Gene-Based Classifier Predicts Prognosis in Patients With Cervical Squamous Cell Carcinoma

A panel of seven immune-related genes can serve as a good predictive biomarker for cervical squamous cell carcinoma

Objective: Cervical cancer is one of the most common gynecological malignancies. The interaction between tumor microenvironment and immune infiltration is closely related to the progression of cervical squamous cell carcinoma (CSCC) and patients' prognosis. Herein, a panel of immune-related genes was established for more accurate prognostic prediction. Methods: The transcriptome information of tumor and normal samples were obtained from TCGA-CSCC and GTEx. Differentially expressed genes (DEGs) were defined from it. Immune-related genes (IRGs) were retrieved from the ImmPort database. After removing the transcriptome data which not mentioned in GSE44001, IR-DEGs were preliminarily identified. Then, TCGA-CSCC samples were divided into training and testing set (3:1) randomly. Univariate Cox analysis, LASSO regression analysis and multivariate Cox analysis were used in turn to construct the signature to predict the overall survival (OS) and disease-free survival (DFS). External validation was performed in GSE44001, and initial clinical validation was performed by qRT-PCR. Function enrichment analysis, immune infiltration analysis and establishment of nomogram were conducted as well. Results: A prognostic prediction signature consisting of seven IR-DEGs was established. High expression of NRP1, IGF2R, SERPINA3, TNF and low expression of ICOS, DES, HCK suggested that CSCC patients had shorter OS (POS<0.001) and DFS (PDFS<0.001). AUC values of 1-, 3-, five- year OS were 0.800, 0.831 and 0.809. Analyses in other validation sets showed good consistency with the results in training set. The signature can serve as an independent prognostic factor for OS (HR = 1.166, p < 0.001). AUC values of 1-, 3-, five- year OS based on the nomogram were 0.769, 0.820 and 0.807. Functional enrichment analysis suggested that these IR-DEGs were associated with receptor interaction and immune cell activity. Immune infiltration analysis indicated that patients in high-risk group had lower immune infiltration, weaker immune function, and were more likely to benefit from immune checkpoint inhibitor therapy. Through qRT-PCR on clinical samples, expression of NRP1, IGF2R, SERPINA3 and TNF were significantly upregulated in tumor tissue, while ICOS and DES were significantly downregulated. Conclusion: To conclude, the immune-related signature can provide strong support for exploration of immune infiltration, prediction of prognosis and response to immunotherapy through stratify CSCC patients into subgroups.