Prediction of postoperative adjuvant therapy efficacy based on perioperative indicators in patients with mid- to late-stage endometrial cancer

Surgery is one of the primary treatments for mid-to-late-stage endometrial cancer. However, patient responses to adjuvant therapy vary significantly, making the prediction of postoperative adjuvant therapy response crucial for these patients. This study aims to analyze the perioperative clinical and laboratory indicators of patients with mid- to late-stage endometrial cancer, identify key factors influencing the efficacy of adjuvant therapy, and construct a predictive model to provide a reference for clinical decision-making. This study retrospectively enrolled 260 mid-to-late-stage endometrial cancer patients who received postoperative adjuvant therapy and categorized them into effective and ineffective groups based on treatment response. Univariate analysis was performed to compare demographic and tumor characteristics between the 2 groups. Significant factors identified in the univariate analysis were classified into 2 categories (clinical variables and laboratory variables) and further analyzed using multivariate logistic regression and random forest analysis to identify the most influential predictors. A response prediction model was then constructed. Patients were stratified into high- and low-response groups based on the model’s risk score for survival analysis. The effective group showed significantly better outcomes than the ineffective group in terms of obstetric history, FIGO stage, tumor grade, surgical approach, hospital stay, and postoperative complications. The final analysis identified the number of positive lymph nodes, high-risk histological type, surgical approach, albumin level, and neutrophil-to-lymphocyte ratio as the most significant predictive factors. The constructed response prediction model achieved an area under the curve of 0.795, with sensitivity and specificity of 0.799 and 0.670, respectively. The low-response group exhibited significantly worse overall survival and progression-free survival than the high-response group. By analyzing clinical and laboratory variables, this study identified key factors influencing the efficacy of postoperative adjuvant therapy in mid-to-late-stage endometrial cancer patients and developed a predictive model with strong discriminative ability. The findings may provide valuable references for individualized clinical treatment and help improve patient prognosis.