Construction of a postoperative disease-free survival prediction model for non-small cell lung cancer patients based on dual-energy computed tomography parameters and blood inflammatory indicators.

BACKGROUND

Non-small cell lung cancer (NSCLC) remains a leading cause of cancer-related death worldwide, and a significant proportion of patients undergoing curative surgery experience recurrence. The tumor node metastasis (TNM) staging system, while foundational, has limited precision for individualized recurrence risk prediction. Emerging biomarkers, including quantitative parameters from dual-energy computed tomography (DECT) and systemic inflammatory indices, have shown promising yet isolated prognostic value. However, a comprehensive model integrating these multimodal data to personalize recurrence risk assessment is lacking. This study aimed to construct a nomogram that integrates DECT quantitative parameters, blood inflammatory indicators, and clinical characteristics to predict disease-free survival (DFS) at 1, 2, and 3 years after surgery in patients with resectable NSCLC.

METHODS

A retrospective study included 140 patients with pathologically confirmed NSCLC who underwent DECT examination within 2 weeks before surgery, randomly assigned to a training set of 98 cases and a test set of 42 cases in a 7:3 ratio. Clinical characteristics, DECT quantitative parameters, and blood test indicators were collected. Potential predictive variables were identified through least absolute shrinkage and selection operator (LASSO) regression analysis, while independent risk factors were established using both univariate and multivariate Cox proportional hazards regression models, leading to the construction of the comprehensive nomogram model. The model's efficacy was evaluated using receiver operating characteristic (ROC) curves, calibration curves, decision curve analysis (DCA), and C-index, with stratified survival analysis performed using the Kaplan-Meier method.

RESULTS

Multivariate Cox regression showed that venous phase normalized iodine concentration (VNIC), effective atomic number (Zeff) observed during plain scan, and neutrophil-lymphocyte ratio (NLR) were identified as independent prognostic indicators for postoperative recurrence and metastasis in NSCLC patients (all P<0.05). The comprehensive nomogram model incorporating these three indicators along with TNM staging, lymph node metastasis, and pathological staging showed area under the curve (AUC) values of 0.896, 0.926, and 0.948 for DFS prediction at 1, 2, and 3 years in the training set, and 0.882, 0.915, and 0.934 in the test set, significantly higher than those of the clinical, DECT, and blood models used alone (all P<0.05). Calibration curves demonstrated good consistency between predicted and actual values, and DCA confirmed a high clinical net benefit of the model. Stratified survival analysis revealed that high VNIC (>31.2%), high Zeff (>8.0), and high NLR (≥2.7) significantly shortened postoperative DFS (all P<0.001).

CONCLUSIONS

The nomogram model based on VNIC, Zeff, and NLR combined with clinical characteristics showed high predictive accuracy for postoperative DFS in resectable NSCLC patients in this study, potentially offering guidance for clinical risk stratification and personalized treatment.