Inflammatory Markers and Advanced Lung Cancer Inflammation Index Predict Recurrence Patterns in Endometrial Cancer: A Retrospective Cohort Study

Background Systemic inflammatory and host‐related markers have demonstrated prognostic utility across multiple malignancies; however, their specific role in predicting recurrence patterns in endometrial cancer remains inadequately characterized. Notably, the Advanced Lung Cancer Inflammation Index (ALI), a novel host‐related biomarker, has not yet been evaluated in this context. Methods We retrospectively analyzed 687 patients with endometrial cancer treated between 2010 and 2020. Pretreatment inflammatory markers, including the neutrophil‐to‐lymphocyte ratio (NLR), platelet‐to‐lymphocyte ratio (PLR), lymphocyte‐to‐monocyte ratio (LMR), neutrophil percentage‐to‐albumin ratio (NPAR), and ALI, were calculated based on routine laboratory data. Receiver operating characteristic (ROC) curve analysis was used to determine the optimal cut‐off values for these markers. Multivariate logistic regression and multinomial regression were employed to assess predictors of recurrence and recurrence patterns, respectively. The C‐index was used to evaluate the discriminative ability of the prediction models. Results Over a median 52‐month follow‐up, 127 patients (18.5%) recurred. Optimal cut‐offs: NLR > 2.45, PLR > 158, LMR ≤ 3.85, NPAR > 11.2, ALI ≤ 45.6. Multivariate analysis identified NLR (OR = 2.18), PLR (OR = 1.67), LMR (OR = 1.75), and ALI (OR = 2.42) as independent predictors (all p < 0.008). NLR was more predictive in early‐stage disease (interaction p = 0.032). Distinct patterns emerged: NLR predicted distant recurrence (RRR = 3.12), PLR was associated with local recurrence (RRR = 2.03), and ALI was associated with recurrence across all sites (local RRR = 1.82; regional RRR = 2.21; distant RRR = 3.45). The host‐related biomarker model demonstrated superior discrimination (C‐index = 0.821) versus FIGO staging alone (0.718, p < 0.001). Conclusions Pre‐treatment inflammatory markers independently predict endometrial cancer recurrence with distinct site‐specific patterns. ALI is a promising composite host‐related biomarker that significantly improves risk stratification, offering a foundation for the design of prospective studies of personalized surveillance strategies.