Sialyltransferase-related genes as predictive factors for therapeutic response and prognosis in cervical cancer.

Abstract

Background: Cancer-associated hypersialylation is believed to be related to the metastatic cell phenotype and the suppression of sialyltransferases (SiaTs) has been suggested to be a potent preventive strategy against metastasis. The present research discovered SiaTs-related genes for cervical cancer (CC).

Methods: The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases were applied to obtain the relevant samples. Mutation dataset were processed using mutect2 software. The gene modules were obtained via weighted gene co-expression network analysis (WGCNA), and the enrichment analysis on the genes within the modules was implemented. Cox regression analysis and "glmnet" R package were applied to establish the relevant risk model. "MCPcounter" R package, ESTIMATE algorithm and TIMER online tools were used to depict the tumor immune microenvironment in CC. The mutation landscape was additionally plotted, and the response to immunotherapy in different cohorts were compared. Further reverse-transcription quantitative PCR and Transwell assays were performed to verify the expression and potential function of the screened key genes.

Results: Mutation of 14 SiaTs was seen in CC. Subsequently, WGCNA-based identification of SiaTs-related gene modules was significantly enriched in metabolism-related pathways. The established RiskScore model manifested excellent prognostic classification efficiency. A poorer prognosis and occurrence of both immune evasion and reduced immunoreactivity may be seen in high-risk patients yet relatively higher immune cell scores were noticeable in low-risk patients. Angiogenesis and MYC target V2 may be the differentially activated pathways in high-risk patients, while those in low-risk patients were KRAS Signaling DN and Interferon alpha response. In addition, most immune checkpoint-correlated genes were identified to express higher in low-risk patients, while higher sensitivities to chemotherapy drugs was discovered in high-risk patients. Cellular assays have revealed that KCNK15, LIF, TCN2, SERPINF2, and CXCL3 were highly expressed yet PIH1D2, DTX1 and GCNT2 were low-expressed in Hela cells and that silencing CXCL3 diminished the number of migrated and invaded Hela cells.

Conclusion: In this study, we systematically constructed and validated a risk scoring model based on SiaTs-related genes, which can effectively predict the prognosis and potential response to immunotherapy and chemotherapy in CC patients. This provides a new molecular basis and clinical reference for achieving individualized treatment.