Introduction Endometrial adenocarcinoma, the sixth most prevalent women’s cancer worldwide, with about 9,700 new UK cases annually. Machine Learning (ML) has been widely used across many fields, especially in medicine in predicting prognosis. This study aims to recognise key prognostic factors influencing 5-year progression-free (PFS) and overall survival (OS) rates in patients with endometrial cancer in West Yorkshire and select features that will be used in ML models to predict outcomes in endometrial cancers patients.

Methods This is a retrospective analysis of 253 cases between 2017-2019. Cox regression was used to evaluate OS and PFS. Hazard Ratios (HR) with 95% Confidence Intervals (CI) were used to quantify risk.

Results For PFS, significant HR increases were observed for FIGO Stages 3 and 4, G3 Cancer (HR 7.51), Outer half myometrial involvement, Adjuvant Chemotherapy, Radiotherapy, Chemoradiotherapy, Serous type and Sarcoma type (HR 7.51). Higher BMI exhibited a correlation with improved survival. For OS, elevated risks were associated with FIGO Stages 3 and 4, G3 Cancer, Outer half myometrial involvement, Adjuvant Chemotherapy, Serous type, Sarcoma type (HR 9.77), Length of stay, and Age. Higher BMI displayed protective effects.

Conclusion Our study provides a basis for ML model for prognosis estimation in patient with endometrial cancers.