Predictive Value of Peripheral Blood Inflammatory Markers and Nutritional Indices for Survival in Young Patients with Advanced Non-Small Cell Lung Cancer: Construction of a Nomogram.

BACKGROUND

This study aimed to investigate the prognostic value of peripheral blood inflammatory markers and nutritional indices for overall survival (OS) in young patients with advanced non-small cell lung cancer (NSCLC) at initial diagnosis. Additionally, we sought to develop a survival prediction model based on combined indicators.

METHODS

We retrospectively analyzed the clinicopathological characteristics, inflammatory markers, and nutritional indices of young patients with advanced NSCLC initially diagnosed at the Fourth Hospital of Hebei Medical University between January 2013 and March 2025. Univariate and multivariate Cox regression analyses were performed to identify independent prognostic factors for OS. Three prognostic models were constructed: a clinical-inflammation model, a clinical-nutrition model, and a clinical-inflammation-nutrition model. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), concordance index (C-index), and decision curve analysis (DCA), with TNM staging as the reference.

RESULTS

A total of 514 patients were included, with a median follow-up of 56.6 months and a median survival time of 27.2 months. Multivariate analysis identified sex, liver metastasis, gene mutation, targeted therapy, white blood cell count, and serum albumin level as independent prognostic factors for OS. Among the three models, the clinical-inflammation-nutrition model showed the best predictive performance, with 1-, 3-, and 5-year AUCs of 0.788, 0.756, and 0.704, respectively, and a C-index of 0.711 (95% CI: 0.696-0.726). DCA further demonstrated its clinical net benefit.

CONCLUSIONS

Peripheral blood inflammatory markers and nutritional indices are closely associated with survival in young patients with advanced NSCLC. In this exploratory single-center study, a prognostic model integrating clinicopathological factors, inflammatory markers, and nutritional indices showed better apparent predictive performance than models based on single categories of variables or TNM staging, warranting further validation in independent cohorts.