Development of a novel transcription factor signature for accurate cervical cancer prognosis

Abstract

Cervical cancer (CC) is a leading cause of cancer-related deaths in women. During tumor development, transcriptional factors regulate the transcription of proto-oncogenes and tumor suppressor genes. We examined the possibility of using transcription factors as prognostic biomarkers for patients with cervical cancer. Single-cell RNA-sequencing data were downloaded from the Gene Expression Omnibus database to identify specific activated transcription factors in different types of cells from CC. Publicly available bulk RNA-sequencing and clinical data of CC were obtained to identify associated prognostic transcription factors using survival analysis and the random survival forest methods. Accuracy and effectiveness of the established transcription factor-related predictive random survival forest model were verified using training and test datasets. We identified specific activated transcription factors in tissue cells of cervical cancer. A 3-transcription factors (PBX4 (PBX Homeobox 4), EBF2 (EBF Transcription Factor 2) and ZNF696 (Zinc Finger Protein 696)) prognostic signature for patients with cervical cancer was constructed showing good survival prediction. Gene function enrichment analysis indicated a correlation between the prognostic characteristics and different signaling pathways associated with cancer. Using the random survival forest model based on the 3-transcription factor signature, patients with cervical cancer were stratified into low- and high-risk groups with significant variations in overall survival (p < 0.001). The area under the curve of the time-dependent receiver operator characteristic revealed a strong predictive accuracy for training and test datasets of the corresponding signature. CC has cellular heterogeneity of transcriptional activation. Our analyses provide a novel transcription factor-associated prognostic model for CC. These transcription factors could be used as effective prognostic biomarkers and potential therapeutic targets for patients with cervical cancer.