Prognostic alternative splicing signature in cervical squamous cell carcinoma

Expression Profile of Thymidine Kinase Genes in Cervical Squamous Cell Carcinoma Confirmed by Various Detection Methods

Background

Thymidine kinases (TKs) are key enzymes involved in DNA synthesis and repair, with alterations in their expression associated with various cancers. Thymidine kinase 1 (TK1) and TK2 are cytosolic enzyme proteins that catalyze the addition of a gamma-phosphate group to thymidine. The existing literature on TK1 in cervical squamous cell carcinoma (CESC) fails to address the clinical role of TK1 overexpression and its possible molecular mechanism in CESC. The clinical significance of TK2 in CESC is also unknown. The objective was to explore the differential expression, clinical significance, and molecular mechanisms of TK1 and TK2 in CESC.

Methods

The researchers collected global high-throughput data, extracted the expression levels of TK1 and TK2, and calculated the integrated standardized mean difference (SMD) and summarized receiver's operating characteristics (sROC) of TK1 or TK2 mRNA to investigate the expression profiles of TK genes fully and objectively in 918 CESC tissues and 360 control tissues. In-house tissue microarrays for immunohistochemical testing were used to verify the protein level of TK1 in 62 CESC tissues and control tissues. The growth effect of TK1 and TK2 in CESC cell lines was assessed using Chronos dependency scores derived from CRISPR knockout screen in the Achilles project. We also analyzed the potential mechanism of TK genes by studying the relationship between TK gene expression and immune infiltration, gene alternations as well as the related signal pathways.

Results

The various detection methods employed all confirmed that the TK1 expression is upregulated and TK2 is downregulated in CESC tissues (SMD: 2.44, 95% confidence interval (CI): 1.36 - 3.51, area under curve (AUC): 0.88, 95% CI: 0.85 - 0.90; SMD: -0.69, 95% CI: -1.25 to -0.14, AUC: 0.75, 95% CI: 0.71 - 0.78). Inhibition of TK1 expression by CRISPR knockout had negative influence on the biological functions of 11 CESC cell lines. The expression of TK2 was negatively correlated with the malignant progression of CESC. Expression of TK genes showed significant association with the immune infiltration of macrophages, CD4+ T cells, and neutrophils. Genes related with TK1 or TK2 were involved in pathways related to DNA replication, proteasome, and homologous recombination.

Conclusions

Clinically, these findings suggest that the differential expression of TK1 and TK2 could serve as potential biomarkers, as well as therapeutic targets for personalized treatment strategies in CESC patients.

A Novel Eight-Gene Signature for Lipid Metabolism Predicts the Progression of Cervical Squamous Cell Carcinoma and Endocervical Adenocarcinoma

Recently, studies on the mechanisms underlying lipid metabolic reprogramming in cancer have increased. However, its significance in cervical cancer remains unclear. In the present study, a prognostic signature was constructed for patients with cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC) using the expression profiles of lipid metabolism-related genes (LMRGs). Furthermore, using various bioinformatics methods, a prognostic gene signature was developed for progression-free survival (PFS). This signature was externally validated using a cervical cancer dataset (GSE44001). The characteristics of the molecular subgroups of LMRGs were analyzed, and target LMRGs were identified via differential gene analysis of the expression profiles and weighted gene correlation network analysis. Thereafter, the identified target genes were used to develop the prognostic gene signature using univariate, least absolute shrinkage and selection operator, and multivariate Cox regression analyses. The performance of the LMRG signature was evaluated using Kaplan-Meier curves, time-dependent receiver operating characteristic curves, decision curve analysis, mutation landscapes, gene set enrichment analysis, and immune score calculation. As a result, a novel eight-LMRG signature comprising ALDH3B2, CERS3, FA2H, GLTP, NR1H3, PLIN3, SLC44A3, and SQLE was constructed. Using this gene signature, patients with CESC and significantly distinguished PFS were divided. This eight-LMRG signature exhibited independent prognostic potential and superior predictive performance compared with a previously developed 12-gene signature. Our findings suggest that our novel eight-LMRG signature contributes to the implementation of precision medicine strategies for managing patients with cervical cancer by facilitating CESC prognosis.

Competing Endogenous RNA (ceRNA) Networks and Splicing Switches in Cervical Cancer: HPV Oncogenesis, Clinical Significance and Therapeutic Opportunities

Cervical cancer (CC) is the primary cause of female cancer fatalities in low-middle-income countries (LMICs). Persistent infections from the human papillomavirus (HPV) can result in cervical cancer. However, numerous different factors influence the development and progression of cervical cancer. Transcriptomic knowledge of the mechanisms with which HPV causes cervical cancer pathogenesis is growing. Nonetheless, there is an existing gap hindering the development of therapeutic approaches and the improvement of patient outcomes. Alternative splicing allows for the production of numerous RNA transcripts and protein isoforms from a single gene, increasing the transcriptome and protein diversity in eukaryotes. Cancer cells exhibit astounding transcriptome modifications by expressing cancer-specific splicing isoforms. High-risk HPV uses cellular alternative splicing events to produce viral and host splice variants and proteins that drive cancer progression or contribute to distinct cancer hallmarks. Understanding how viruses utilize alternative splicing to drive pathogenesis and tumorigenesis is essential. Although research into the role of miRNAs in tumorigenesis is advancing, the function of other non-coding RNAs, including lncRNA and circRNA, has been understudied. Through their interaction with mRNA, non-coding RNAs form a network of competing endogenous RNAs (ceRNAs), which regulate gene expression and promote cervical cancer development and advancement. The dysregulated expression of non-coding RNAs is an understudied and tangled process that promotes cervical cancer development. This review will present the role of aberrant alternative splicing and immunosuppression events in HPV-mediated cervical tumorigenesis, and ceRNA network regulation in cervical cancer pathogenesis will also be discussed. Furthermore, the therapeutic potential of splicing disruptor drugs in cervical cancer will be deliberated.

Immunogenomic Identification for Predicting the Prognosis of Cervical Cancer Patients

Cervical cancer is primarily caused by the infection of high-risk human papillomavirus (hrHPV). Moreover, tumor immune microenvironment plays a significant role in the tumorigenesis of cervical cancer. Therefore, it is necessary to comprehensively identify predictive biomarkers from immunogenomics associated with cervical cancer prognosis. The Cancer Genome Atlas (TCGA) public database has stored abundant sequencing or microarray data, and clinical data, offering a feasible and reliable approach for this study. In the present study, gene profile and clinical data were downloaded from TCGA, and the Immunology Database and Analysis Portal (ImmPort) database. Wilcoxon-test was used to compare the difference in gene expression. Univariate analysis was adopted to identify immune-related genes (IRGs) and transcription factors (TFs) correlated with survival. A prognostic prediction model was established by multivariate cox analysis. The regulatory network was constructed and visualized by correlation analysis and Cytoscape, respectively. Gene functional enrichment analysis was performed by Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG). A total of 204 differentially expressed IRGs were identified, and 22 of them were significantly associated with the survival of cervical cancer. These 22 IRGs were actively involved in the JAK-STAT pathway. A prognostic model based on 10 IRGs (APOD, TFRC, GRN, CSK, HDAC1, NFATC4, BMP6, IL17RD, IL3RA, and LEPR) performed moderately and steadily in squamous cell carcinoma (SCC) patients with FIGO stage I, regardless of the age and grade. Taken together, a risk score model consisting of 10 novel genes capable of predicting survival in SCC patients was identified. Moreover, the regulatory network of IRGs associated with survival (SIRGs) and their TFs provided potential molecular targets.