Risk Stratification of Early-Stage Cervical Cancer with Intermediate-Risk Factors: Model Development and Validation...

Area of Science:

• Oncology
• Gynecologic Oncology
• Cancer Prognostics

Background:

• Adjuvant therapy decisions for cervical cancer (CC) with intermediate-risk factors are controversial.
• Current criteria (e.g., Sedlis criteria) may not adequately stratify risk for all patients.

Purpose of the Study:

• To assess prognoses in early-stage CC patients with pathological intermediate-risk factors.
• To develop and validate predictive models for disease-free survival (DFS) and overall survival (OS).
• To provide a better reference for adjuvant therapy selection.

Main Methods:

• Retrospective analysis of 481 patients with stage IB-IIA CC.
• Utilized Cox regression, machine learning (ML) algorithms, Kaplan-Meier analysis, and AUC for model development and validation.
• Developed prediction models for DFS and OS based on clinical and pathological factors.

Main Results:

• Developed two prognostic models for DFS and OS, stratifying patients into high-risk and low-risk groups.
• ML-based models showed better predictive performance (AUCs up to 0.88 for OS) compared to traditional Sedlis criteria.
• Significant differences in DFS and OS were observed between risk groups identified by the new models (p < .05).

Conclusions:

• The developed prognostic models effectively stratify risk in CC patients with intermediate-risk factors.
• These models can aid in clinical decision-making for individualized adjuvant therapy.
• Further evaluation of long-term survival is needed to corroborate findings.