Combined inflammatory-lipid index and tumor markers for predicting the spatial localization of lesions in early-stage non-small cell lung cancer

TL;DR

This study developed a model combining blood markers and tumor indicators to predict where lung cancer tumors are located in early-stage patients.

Contribution

The novelty lies in combining inflammatory-lipid indices and tumor markers to predict lesion localization in early-stage non-small cell lung cancer.

Findings

WBC, α-HBDH, HDL, CEA, SF, CA153, and CA199 were identified as independent predictors of lower lobe tumor localization.

The developed nomogram achieved an AUC of 0.806, showing good predictive performance and calibration.

The model showed favorable clinical utility based on decision curve analysis.

Abstract

This study evaluated the predictive value of combined inflammatory-lipid indices and tumor markers in determining lesion localization in early-stage non-small cell lung cancer (NSCLC) and developed a predictive model.

A retrospective analysis of 206 early-stage NSCLC patients was conducted from December 1, 2023, to September 30, 2024. Patients were grouped based on tumor location: upper lobe and lower lobe. Significant predictors were identified through univariate and multivariate logistic regression analyses, leading to the development of a nomogram. Predictive performance was assessed using the receiver operating characteristic (ROC) curve and area under the curve (AUC). Model calibration was evaluated with a calibration plot, and decision curve analysis (DCA) was utilized to assess the model’s relevance in clinical settings.

Among the 206 patients, 135 (65.53%) had upper lobe tumors, and 71 (34.47%) had lower lobe tumors. Significant differences were found in white blood cell (WBC) count, lymphocyte count, α-hydroxybutyrate dehydrogenase (α-HBDH), high density lipoprotein cholesterol (HDL) triglycerides, low-density lipoprotein cholesterol (LDL), total cholesterol, carcinoembryonic antigen (CEA), serum ferritin (SF), carbohydrate antigen 125 (CA125), carbohydrate antigen 153 (CA153), and carbohydrate antigen 199 (CA199) (all p < 0.05). Multivariate logistic regression identified WBC (OR: 1.46, 95% CI: 1.13–1.95, p = 0.007), a-HBDH (OR: 1.01, 95% CI: 1.00–1.03, p = 0.041), HDL (OR: 7.08, 95% CI: 1.50–36.16, p = 0.015), CEA (OR: 1.12, 95% CI: 1.02–1.23, p = 0.021), SF (OR: 1.01, 95% CI: 1.00–1.02, p = 0.020), CA153 (OR: 1.08, 95% CI: 1.00–1.16, p = 0.037), and CA199 (OR: 1.16, 95% CI: 1.07–1.27, p < 0.001) as independent risk factors for lower lobe tumor localization. An AUC of 0.806 was obtained for the nomogram (95% CI: 0.743–0.868), indicating good calibration, and showed favorable clinical utility based on decision curve analysis (DCA).

WBC count, lymphocyte count, α-HBDH, HDL, CEA, SF, CA153, and CA199 are significant predictors of lesion localization in early-stage NSCLC. The developed nomogram, based on readily available clinical parameters, demonstrated strong predictive performance and may aid in individualized diagnosis and treatment planning. Further large-scale external validation is needed.

Figures