Predictive value of preoperative L3-SMI, AGR, and PNI for overall survival in patients undergoing radical gastrectomy for gastric cancer.
Objective: Gastric cancer (GC) is the fifth most common cancer and the third leading cause of cancer-related death worldwide, posing a serious threat to public health. Prognostication of overall survival (OS) in patients undergoing radical gastrectomy remains a clinical priority. Evidence suggests that preoperative nutritional and inflammatory status correlated with postoperative outcomes. This study aims to evaluate the prognostic value of the skeletal muscle index at the third lumbar vertebra (L3-SMI) as a trichotomous variable and to compare the performance of commonly used nutritional and inflammation-related indicators in predicting postoperative survival in GC patients.
Methods: This retrospective study analyzed clinical data of patients who underwent radical gastrectomy with neoadjuvant chemotherapy between 2011 and 2018 at the Department of Gastrointestinal Surgery, Xiangya Hospital of Central South University. L3-SMI was measured by preoperative CT, and 8 preoperative nutritional/inflammatory indices were calculated from the latest laboratory tests before surgery: neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), lymphocyte-to-monocyte ratio (LMR), systemic immune-inflammation index (SII), systemic inflammation response index (SIRI), pan-immune-inflammation value (PIV), albumin-globulin ratio (AGR), and prognostic nutritional index (PNI). L3-SMI was categorized into 3 groups using X-tiler software. ROC curves were used to determine optimal cut-off values for the other eight indices. Kaplan-Meier curves and univariate/multivariate Cox proportional hazards models were used to analyze the association between variables and OS. Concordance index (C-index) and subgroup analysis assessed predictive performance and consistency across patient subgroups.
Results: A total of 546 patients were included, with a minimum follow-up time of 36 months. Kaplan-Meier and univariate analysis showed that L3-SMI and the 8 indicators were significantly associated with OS (all P<0.01). After adjusting for age, gender, tumor site, differentiation, pTNM stage, type of surgery, anemia, CEA, and AFP, multifactorial Cox analysis revealed that L3-SMI (HR=0.676, 95% CI 0.523 to 0.872), AGR (HR=0.611, 95% CI 0.452 to 0.827), and PNI (HR=0.590, 95% CI 0.418 to 0.833) were independent predictors of OS. The full model confirmed the independent prognostic roles of L3-SMI, AGR, and PNI. Among all indicators, PNI had the highest C-index for 1-year OS prediction (0.632, 95% CI 0.568 to 0.695), while AGR showed the best performance at 3 years (0.585, 95% CI 0.548 to 0.622) and 5 years (0.578, 95% CI 0.542 to 0.613). Subgroup analysis indicated that higher L3-SMI, AGR, and PNI were associated with lower mortality risk in patients aged<65 years, with lower gastric tumors, poor differentiation, stage III pTNM, or who underwent subtotal gastrectomy.
Conclusions: Compared with other indicators, preoperative nutritional markers such as L3-SMI, AGR, and PNI demonstrated superior prognostic value for OS in gastric cancer patients undergoing radical gastrectomy. Assessing these indices can help identify patients at high risk of poor prognosis, thereby guiding targeted nutritional interventions and potentially improving survival outcomes.