Objective:

To evaluate whether preoperative systemic inflammatory response index (SIRI) could predict lymph node metastasis (LNM) in endometrial cancer (EC) and to develop a nomogram that combines SIRI with clinicopathological parameters for LNM risk assessment.

Approach:

• Study Design: Retrospective enrollment of 1,336 EC patients who underwent primary surgery, with a training cohort of 947 cases and an external validation cohort of 389 cases.
• SIRI Calculation: Preoperative SIRI was calculated using peripheral neutrophil, monocyte, and lymphocyte counts.
• Statistical Analysis: Logistic regression analyses identified independent predictors of LNM, which were incorporated into a nomogram. Model performance was evaluated using ROC and calibration curves.

Key Findings:

• SIRI had an AUC of 0.773 for predicting LNM, with a sensitivity of 74.2% and specificity of 75.8%.
• The optimal cutoff value of SIRI was 1.115.
• Multivariate analysis identified age, CA125, histological type, molecular classification, and SIRI as independent predictors of LNM (P < 0.001).
• The nomogram achieved an AUC of 0.889 in the training cohort, outperforming models based solely on SIRI or clinicopathological parameters.
• Significant differences in recurrence-free survival (RFS) and overall survival (OS) were observed between high- and low-risk groups (P < 0.05).

Interpretation:

Preoperative SIRI is an independent predictor of LNM in EC patients, and the nomogram integrating SIRI with clinicopathological factors provides improved predictive performance.

Limitations:

• The study is retrospective, which may introduce selection bias.
• External validation was limited to one cohort.

Conclusion:

The nomogram model incorporating SIRI and clinicopathological parameters is a valuable tool for individualized preoperative risk stratification and surgical decision-making in endometrial cancer.