Abstract: Background　In advanced non-small cell lung cancer, the efficacy of PD-1/PD-L1 (Programmed Cell Death Protein 1/Programmed Death-Ligand 1) inhibitors varies considerably, and early prognostic predictors remain unclear. Objective　To evaluate the prognostic significance of early changes in the neutrophil-to-lymphocyte ratio (ΔNLR), prognostic nutritional index (PNI), and their combination as the inflammation-nutrition score (INS) in advanced non-small cell lung cancer (NSCLC) patients receiving PD-1 inhibitor therapy. Methods　A retrospective analysis was conducted on patients with advanced NSCLC who received anti-PD-L1 monoclonal antibody monotherapy or combination therapy at Shandong Second Medical University Affiliated Hospital from January 2020 to December 2024. Peripheral blood NLR and serum albumin levels were measured before treatment and after completing two treatment cycles. ΔNLR and PNI were calculated and used to group the patients: ΔNLR ≥ 20% as the elevated group, PNI ＜ 45 as the low-nutrition group, and ΔNLR ≥ 20% combined with PNI ＜ 45 as the high INS risk group. Kaplan-Meier method was used to compare progression-free survival (PFS) and overall survival (OS) among the groups, and Cox regression was applied to identify independent prognostic factors. A prognostic prediction model was then constructed, and the prediction performance was evaluated using ROC curve analysis. Results　A total of 163 patients with advanced NSCLC were included in this analysis, comprising 100 males (61.35%) and 63 females (38.65%). The age range was 54 to 68 years, with a median age of 61 years. The median OS was 16.4 months. Survival analysis showed that the PFS of elevated NLR group was significantly shorter than that of the non-elevated group (6.5 months vs 8.6 months, HR=1.55, 95% CI: 1.07 - 2.26, P=0.041). Low PNI predicted poorer PFS (6.9 months vs 8.9 months, HR=1.52, 95% CI: 1.04 - 2.21, P=0.048). The INS high-risk group also showed reduced PFS compared with non-high-risk patients (6.0 months vs 8.4 months, P=0.047). Multivariate Cox regression analysis revealed that an increase in ΔNLR ≥ 20% and PNI ＜ 45 were both adverse prognostic factors for PFS, while PD-L1 expression (≥50%) was a favorable prognostic factor (P＜0.05). ROC analysis indicated that Models 1 and 2, constructed based on regression risk probabilities, demonstrated high predictive performance for the prognosis of advanced NSCLC patients, with AUC values of 0.852 (95% CI: 0.753 - 0.937) and 0.843 (95% CI: 0.698 - 0.978), respectively. Conclusion　Early elevation of ΔNLR and reduction of PNI are strongly linked to poor prognosis in advanced NSCLC patients treated with PD-1 inhibitors. INS, integrating inflammatory and nutritional status, may serve as a practical tool for risk stratification and prognosis prediction in immunotherapy.