Nomogram based on the novel index LANR, composed of preoperative lymphocytes, albumin, and neutrophils, for predicting prognosis in patients with gastric cancer: a retrospective study.

Purpose: This study aims to explore the relationship between the novel index LANR, which is composed of preoperative lymphocytes, neutrophils, and albumin, and prognosis in patients with gastric cancer (GC), and to develop and visualize a new nomogram for predicting overall survival (OS) in GC patients. Methods: A total of 497 patients (346 in the training cohort and 151 in the validation cohort) with GC who underwent radical resection were retrospectively analyzed. The LANR was calculated as the lymphocyte×albumin/neutrophil. Collinearity diagnostic analysis was performed to assess the correlations between variables. Univariate and multivariate Cox regression analyses were used to identify independent prognostic factors for OS, which were then used to construct a nomogram model. The efficacy of the nomogram was subsequently evaluated in the validation cohort. Results: Multivariate Cox regression analysis showed that tumor size (Hazard ratio [HR]=1.653, P = 0.001), T stage (HR = 3.236, P<0.001), N stage (HR = 2.059, P<0.001), chemotherapy (HR = 1.508, P = 0.005), and LANR (HR = 0.586, P<0.001) were independent significant risk factors for OS in patients with GC. The independent prognostic performance of LANR is superior compared to NLR, PNI and PLR. In the training cohort, the area under the curve (AUC) of the nomograms for predicting 3-, 5- and 7-year OS were 0.768(95% CI = 0.718-0.819), 0.832(95% CI = 0.790-0.875) and 0.893(95% CI = 0.830-0.956), respectively. The AUC of the nomogram for predicting 3-, 5- and 7-year OS were 0.795(95% CI = 0.719-0.871), 0.823(95% CI = 0.756-0.890) and 0.833(95% CI = 0.735-0.931), respectively, in the validation cohort. Both in the training and validation cohorts, the calibration curves showed good consistency between the actual survival rates and the predicted values from the nomogram. The Decision curve analysis also indicated that the model has clinical utility. Conclusion: LANR is an independent prognostic factor for GC. The newly developed nomogram demonstrates high accuracy and potential clinical utility in predicting the OS in GC patients.