Development and Validation of a Prognostic Model for Cervical Cancer Based on Chlamydia trachomatis-Associated Transcriptional Signatures.

BACKGROUND

Chlamydia trachomatis (CT) infection has been identified as an independent predictor of risk of cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC); however, its prognostic relevance and molecular mechanisms remain poorly understood. Therefore, we aimed to construct a CESC prognostic model based on genes associated with CT infection.

METHODS

Transcriptomic data were obtained from GEO and TCGA databases. Differentially expressed genes (DEGs) were identified from GSE63514 (24 normal controls, 28 CESE samples) and GSE158814 (2 negative controls, 2 CT-infection HeLa cells). Overlapping genes were defined as CT-related genes. TCGA-CESC cohort (n = 252) was split into training and validation sets. Univariate Cox regression was used to screen candidate genes, and then multivariate Cox regression were performed to construct a prognostic signature and calculate RiskScore. LASSO-Cox regression and bootstrap analysis evaluated model stability. Patients were stratified by median RiskScore. A nomogram integrating clinical factors and RiskScore was established, followed by evaluation using calibration curves, receiver operating characteristic, and decision curve analyses. Gene set enrichment analysis, immune infiltration, and drug sensitivity analyses were performed to explore the biological and therapeutic implications of the high- and low- RiskScore group.

RESULTS

A total of 31 CT-related CESC genes were identified, among which CSF2RB, LAMA4 and HAL were risk factors, and CSF2RB was a protective factor. A three-gene prognostic model (LAMA4, CSF2RB, HAL) was developed, showing robust predictive performance in both training and validation cohorts. Nomogram analysis confirmed that RiskScore, age, and stage were independent prognostic factors. Functional enrichment revealed that DEGs between the high- and low-risk groups were mainly enriched in immune-related pathways. Immune infiltration analysis indicated that low-risk patients exhibited higher infiltration of effector immune cells (e.g. CD4+, B cells, dendritic cells, M1 macrophages), suggesting a more active antitumor immune microenvironment. Drug sensitivity analysis identified BI-2536, SB505124, and Sepantronium bromide as potential therapeutic agents for high-risk patients.

CONCLUSION

We established a novel prognostic model for CESC based on CT-related genes in an HPV-positive CESC background, which provides new insights into CT infection-associated immune regulation in CESC and individualized immunotherapy and targeted treatment strategies.