Pretreatment Neutrophil-to-Lymphocyte Ratio as a Risk Marker for Brain Metastasis in Patients with Stage I–III Non-Small Cell Lung Cancer Receiving Definitive Thoracic Radiotherapy

Abstract Background Brain metastasis (BM) represents a major pattern of distant failure in non-small cell lung cancer (NSCLC) following definitive thoracic radiotherapy (RT). Systemic inflammation plays a critical role in tumor progression and the metastatic cascade. This study aimed to evaluate the prognostic value of the baseline neutrophil-to-lymphocyte ratio (NLR) for predicting the occurrence, cumulative incidence, and temporal dynamics of brain metastasis in patients with Stage I–III NSCLC receiving definitive thoracic RT. Methods A total of 138 patients with pathologically confirmed Stage I–III NSCLC who received definitive-intent thoracic RT (60–66 Gy) between 2014 and 2024 were retrospectively analyzed. Pre-treatment hematological parameters were obtained within two weeks prior to any oncological intervention. Time-dependent Receiver Operating Characteristic (ROC) curve analysis evaluated the dynamic predictive performance of baseline NLR. Independent predictors of brain failure were identified using cause-specific Cox proportional hazards regression, accounting for death as a competing risk. A pre-specified sensitivity analysis was restricted to Stage III patients (n = 108) with mandatory baseline and routine follow-up brain MRI. Results Over a median follow-up of 28.0 months, 40 patients (29.0%) developed BM. Median baseline NLR was significantly higher in patients who developed BM compared to those who did not (3.21 vs. 2.22; p < 0.001). Time-dependent ROC analysis at 28 months yielded an Area Under the Curve (AUC) of 0.79, with an optimal NLR threshold of 2.26 (sensitivity: 91.7%, specificity: 54.6%). Multivariable cause-specific Cox regression demonstrated that higher continuous baseline NLR was independently associated with an increased hazard of BM (HR: 1.77, 95% CI: 1.48–2.12; p < 0.001), alongside adenocarcinoma histology (HR: 2.96, 95% CI: 1.41–6.24; p = 0.004). In the Stage III sub-cohort sensitivity analysis, continuous baseline NLR maintained its independent prognostic significance (HR: 1.74, 95% CI: 1.44–2.09; p < 0.001). High baseline NLR (> 2.26) was significantly associated with a higher cumulative incidence of BM (Gray’s test p < 0.001), inferior overall survival (median 34.0 vs. 65.0 months; p = 0.015), and worse progression-free survival (median 18.0 vs. 42.0 months; p = 0.006). Baseline PLR was not significantly associated with BM (p = 0.136). Conclusions Continuous baseline NLR serves as a readily accessible, independent biomarker for predicting the risk and accelerated onset of brain metastasis in patients with NSCLC undergoing definitive chemoradiotherapy. Baseline NLR may aid in risk-stratified neuro-radiological surveillance protocols.