A prognostic signature based on methionine metabolism-related genes for cervical cancer: integrated transcriptomic and experimental validation

Abstract

Background Cervical cancer (CC) remains one of the most prevalent malignancies in the female reproductive system. Methionine metabolism (MM) plays a pivotal role in various biological processes and has been implicated in cancer progression. However, its mechanisms in CC remain unclear. Methods Transcriptomic data from 305 patients in The Cancer Genome Atlas (TCGA) (training cohort) and 299 patients from the Gene Expression Omnibus (GEO) (GSE44001) (external validation cohort) were analyzed for differentially expressed MM-related genes (MM-RGs). Prognostic MM-RGs were identified using Cox regression, proportional hazards testing, and Least Absolute Shrinkage and Selection Operator (LASSO) regression. A risk model was constructed and validated. Functional enrichment (Gene Set Enrichment Analysis/Gene Set Variation Analysis (GSEA/GSVA)), in-silico immune infiltration (CIBERSORT) and drug sensitivity (oncoPredict) analyses were performed. Five genes were validated by Quantitative Reverse Transcription Polymerase Chain Reaction (qRT-PCR) on clinical samples. Results Eight MM-RGs (MTHFD1, SMYD2, MSRB3, MTR, ENOPH1, DNMT3B, SLC38A7, PEMT) were identified as prognostic genes. A robust risk score model was developed, stratifying patients into high- and low-risk groups with significant differences in survival outcomes. Functional enrichment revealed pathways such as ECM–receptor interaction and focal adhesion. Immune analysis indicated altered infiltration of Tregs and mast cells. In-silico drug sensitivity analysis predicted 57 agents with significantly different IC50 values between the high- and low-risk groups (p < 0.05). Notably, agents such as Cediranib and BI-2536 exhibited markedly lower IC50 values in the high-risk cohort, suggesting their potential efficacy for advanced-stage treatment. qRT-PCR preliminarily indicated the overexpression of four genes in CC tissues within a small clinical cohort. Conclusion This study establishes a novel MM-based prognostic model for CC and based on in-silico predictions suggests potential therapeutic targets through comprehensive transcriptomic and experimental validation.