# Epithelial Cell–Specific Prognostic Signature (FTH1, RIT1, WASL, NDRG2, KIFC3) Stratifies Cervical Cancer Patients and Correlates With Immune Infiltration

## TL;DR

This study identifies a five-gene signature in cervical cancer epithelial cells that predicts patient survival and immune response, offering potential for improved clinical outcomes.

## Contribution

The study introduces a novel epithelial cell-specific prognostic model for cervical cancer linked to immune infiltration and drug sensitivity.

## Key findings

- A five-gene signature (FTH1, RIT1, WASL, NDRG2, KIFC3) stratifies cervical cancer patients into high- and low-risk groups with distinct survival outcomes.
- The high-risk group shows reduced immune/stromal scores and unique immune cell infiltration patterns.
- FTH1 knockdown inhibits cancer cell proliferation, migration, and invasion while increasing apoptosis.

## Abstract

Cervical cancer (CC) remains one of the leading female malignancies. Epithelial cells (EpCs), primarily derived from the cervical squamous and glandular epithelium, are targeted by human papillomavirus to drive CC. Herein, we aimed to develop an EpC‐specific risk model to improve clinical outcomes and unravel tumor immune microenvironment alterations in CC.

scRNA‐seq data from GSE208653 were processed using Seurat (including SCTransform for normalization and Harmony for batch correction). EpC heterogeneity was analyzed via subclustering, pseudotime trajectory analysis with monocle2, and cell–cell communication inference with CellChat. The hdWGCNA package identified EpC‐specific coexpression modules. Prognostic genes were screened by univariate Cox and LASSO regression, and a Riskscore model was built using multivariate Cox regression. Immune infiltration was assessed by ssGSEA, MCPCounter, and ESTIMATE algorithms. Drug sensitivity correlation was analyzed using pRRophetic. In vitro functional assays validated key gene roles in CC cells.

Forty thousand four hundred fifty‐seven cells were annotated into eight cell populations with a lower percentage of EpCs. Thereafter, EpCs were categorized into three subclusters with specifically highly expressed genes in peculiar biological pathways and with distinct trajectories of fate. A strong cell–cell communication network was observed, particularly involving Ep C3 and immune cells, via ligand–receptor pairs such as LGALS9‐CD44 and HBEGF‐EGFR. The hdWGCNA analysis revealed Ep C3–specific gene modules, from which a five‐gene prognostic signature (FTH1, RIT1, WASL, NDRG2, and KIFC3) was constructed. The resulting risk model effectively stratified patients into high‐ and low‐risk groups with significantly different overall survival in both TCGA‐CESC and GSE52903 cohorts, supported by time‐dependent ROC curves. The high‐risk group exhibited lower immune/stromal scores and distinct immune cell infiltration patterns. The risk score significantly correlated with sensitivity to several chemotherapeutic agents. Crucially, in vitro experiments confirmed that FTH1 knockdown inhibited the proliferation, migration, and invasion of CC cells while enhancing the level of apoptosis in cancer cells.

A proposed EpC‐specific gene signature for CC may be applicable to support clinical decision‐making.