Development of a Novel Immune Infiltration-Based Gene Signature to Predict Prognosis and Immunotherapy Response of Patients With Cervical Cancer.

Sihui Yu,Xi Li,Jiawen Zhang,Sufang Wu

doi:10.3389/fimmu.2021.709493

Abstract

Predictive models could indicate the clinical outcome of patients with carcinoma. Cervical cancer is one of the most frequently diagnosed female malignancies. Herein, we proposed an immune infiltration-related gene signature that predicts prognosis of patients with cervical cancer and depicts the immune landscape as well. We utilized the transcriptome data of The Cancer Genome Atlas (TCGA) and estimated the infiltration level of 28 immune cell types. We screened out four immune cell types conducive to patient survival and recognized their shared differentially expressed genes (DEGs). Four core genes (CHIT1, GTSF1L, PLA2G2D, and GNG8) that composed the ultimate signature were identified via univariate and multivariate Cox regression. The optimal model we built up could distinguish patients with cervical cancer into high-score and low-score subgroups. These two subgroups showed disparity in aspects of patient survival, immune infiltration landscape, and response to immune checkpoint inhibitors. Additionally, we found that GTSF1L was decreased gradually along with the severity of cervical lesions, and its potential role in immune contexture and clinical practice were also demonstrated. Our results suggested that the Immunoscore based on four immune-related genes could serve as a supplementary criterion to effectively foresee the survival outcome, tumor infiltration status, and immunotherapy efficacy of cervical cancer patients.

Highlights

Cervical cancer (CC), as one of the most frequently diagnosed female malignancies, is the fourth leading cause of cancer mortality in females [1]
We annotated these data with gene transfer format (GTF) files of Ensembl, calculated the normalized enrichment score of various immune cells by single-sample GSEA analysis to characterize their expression pattern in CESC cases (Figure 2A), and explored their impact on patient prognosis (Figure 2B)
We explored the impact of this Immunoscore on the progression-free survival (PFS) and relapse-free survival (RFS) of patients from the The Cancer Genome Atlas (TCGA)-CESC cohort

Summary

Introduction

Cervical cancer (CC), as one of the most frequently diagnosed female malignancies, is the fourth leading cause of cancer mortality in females [1]. Current treatment strategies including surgery, chemoradiotherapy, and immunotherapy have tremendously ameliorated the prognosis, the clinical outcome of advanced cervical cancer patients is still not optimistic [2]. The density of peritumoral CD3+ T cells was proven to have the potential for predicting relapse [6], and tumor-infiltrating CD204+ M2 macrophages predicted worse prognosis in patients with cervical adenocarcinoma [7]. Increased CD4, CD8, CD20, and CD56 signals were associated with good responders to neoadjuvant chemotherapy [8], and the number of CD8+ T cells was correlated with treatment outcome in patients treated with radiotherapy [9]. Quantitative molecular signatures closely associated with immune infiltration might display promising capability in predicting the clinical outcome of cervical cancer

Methods

Results

Conclusion