Abstract
Background: Cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC) account for a significant proportion of gynecological malignancies and represent a major global health concern. Globally, CESC is ranked as the fourth most common cancer among women. Conventional treatment of this disease has a less favorable prognosis for most patients. However, the discovery of early molecular biomarkers is therefore important for the diagnosis of CESC, as well as for slowing down their progression process. Methods: To identify differentially expressed genes strongly associated with prognosis, univariate Cox proportional hazard analysis and least absolute shrinkage and selection operator (LASSO) regression analysis were used. Using multiple Cox proportional hazard regression, a multifactorial model for prognostic risk assessment was then created. Results: The expression of biological clock-related genes, which varied considerably among distinct subtypes and were associated with significantly diverse prognoses, was used to categorize CESC patients. These findings demonstrate how the nomogram developed based on the 7-CRGs signature may assist physicians in creating more precise, accurate, and successful treatment plans that can aid CESC patients at 1, 3, and 5 years. Conclusions: By using machine learning techniques, we thoroughly investigated the impact of CRGs on the prognosis of CESC patients in this study. By creating a unique nomogram, we were able to accurately predict patient prognosis. At the same time, we showed new perspectives on the development of CESC and its treatment by analyzing the associations of the prognostic model with immunity, enrichment pathways, chemotherapy sensitivity, and so on. This research provides a new direction for clinical treatment.