Circular RNA-related CeRNA network and prognostic signature for patients with oral squamous cell carcinoma

Abstract

Background: Circular RNA (circRNA) has an important influence on oral squamous cell carcinoma (OSCC) progression as competing endogenous RNAs (ceRNAs). However, the link between ceRNAs and the OSCC immune microenvironment is unknown. The research aimed to find circRNAs implicated in OSCC carcinogenesis and progression and build a circRNA-based ceRNA network to create a reliable OSCC risk prediction model.Methods: The expression profiles of circRNA in OSCC tumors and normal tissues were assessed through RNA sequencing. From the TCGA database, clinicopathological data and expression patterns of microRNAs (miRNAs) and mRNAs were obtained. A network of circRNA-miRNA-mRNA ceRNA was prepared according to these differentially expressed RNAs and was analyzed through functional enrichment. Subsequently, based on the mRNA in the ceRNA network, the influence of the model on prognosis was then evaluated using a risk prediction model. Finally, considering survival, tumor-infiltrating immune cells (TICs), clinicopathological features, immunosuppressive molecules, and chemotherapy efficacy were analyzed.Results: Eleven differentially expressed circRNAs were found in cancer tissues relative to healthy tissues. We established a network of circRNA-miRNA-mRNA ceRNA, and the ceRNA network includes 123 mRNAs, six miRNAs, and four circRNAs. By the assessment of Genomes pathway and Kyoto Encyclopedia of Genes, it is found that in the cellular senescence, PI3K-AKT and mTOR signaling pathway mRNAs were mainly enrichment. An immune-related signature was created utilizing seven immune-related genes in the ceRNA network after univariate and multivariate analysis. The receiver operating characteristic of the nomogram exhibited satisfactory accuracy and predictive potential. According to a Kaplan-Meier analysis, the high-risk group’s survival rate was signally lower than the group with low-risk. In addition, risk models were linked to clinicopathological characteristics, TICs, immune checkpoints, and antitumor drug susceptibility.Conclusion: The profiles of circRNAs expression of OSCC tissues differ significantly from normal tissues. Our study established a circRNA-associated ceRNA network associated with OSCC and identified essential prognostic genes. Furthermore, our proposed immune-based signature aims to help research OSCC etiology, prognostic marker screening, and immune response evaluation.

Comprehensive analysis of circRNAs for N7‐methylguanosine methylation modification in human oral squamous cell carcinoma

“…Previous studies have shown that cellular senescence and senescence‐associated cytokines are major modulators in head and neck squamous cell carcinoma development 59,60 . He et al found circRNAs implicated in OSCC carcinogenesis and progression through RNA sequencing and assessed differentially expressed circRNAs by GO and KEGG analysis 61 . In their study, genes were mainly enriched in cellular senescence, which was consistent with our results.…”

“… 59 , 60 He et al found circRNAs implicated in OSCC carcinogenesis and progression through RNA sequencing and assessed differentially expressed circRNAs by GO and KEGG analysis. 61 In their study, genes were mainly enriched in cellular senescence, which was consistent with our results. Moreover, homologous recombination is known as a DNA repair pathway of clinical interest in cancer.…”

Construction of a ceRNA regulatory network to explore potential pathogenesis mechanisms involved in human hepatocellular carcinoma

“…Therefore, in 2011, Salmena et al proposed the ceRNA hypothesis 8 , which has attracted widespread attention from researchers and stated that coding and ncRNA molecules with common MREs can compete for miRNA binding at these loci and thus indirectly reciprocally regulate expression by acting as miRNA sponges 9 . CircRNAs are a novel class of noncoding ceRNAs and have a circular structure formed by the covalent closure of precursor messenger RNAs (pre-mRNAs) through backsplicing 10 . CircRNAs have no 5′ cap or 3′ tail and are thus not easily degraded by ribonucleases.…”