Abstract
BACKGROUND: Immunotherapy with PD-1/PD-L1 inhibitors has transformed advanced non-small cell lung cancer (NSCLC) treatment, yet optimal strategies for elderly patients remain uncertain. Elderly patients (≥ 65 years) exhibit immune senescence (e.g., T-cell dysfunction, chronic inflammation), which may compromise immunotherapy efficacy and amplify toxicity risks, yet direct comparisons of monotherapy versus combination regimens in this population are lacking. Real-world comparisons of monotherapy versus combination therapy in this vulnerable group are lacking, hindering personalized clinical decisions. OBJECTIVE: This real-world study aimed to compare the efficacy and safety of PD-1/PD-L1 inhibitor monotherapy versus combination therapy with chemotherapy in elderly patients (≥ 65 years) with advanced NSCLC and develop a prognostic nomogram to guide personalized treatment decisions. METHODS: In this multicenter retrospective study, 641 patients (149 monotherapy, 492 combination therapy) with stage IIIB/IV NSCLC were analyzed. Primary endpoints included overall survival (OS), progression-free survival (PFS), and adverse events (AEs). A nomogram incorporating clinical variables was constructed using LASSO-Cox regression. RESULTS: In a retrospective analysis of 641 elderly patients (≥ 65 years) with advanced NSCLC, combination therapy (n = 492) demonstrated superior median OS compared to monotherapy (n = 149) (35.37 vs. 20.53 months; HR = 0.62, 95% CI 0.48-0.80, P < 0.001), though PFS did not differ significantly (11.87 vs. 10.67 months; HR = 0.94, P = 0.535). Age-stratified analysis revealed marked OS benefits for patients < 75 years receiving combination therapy (36.10 vs. 18.67 months, P < 0.001), whereas no advantage was observed in those ≥ 75 years (29.23 vs. 34.93 months, P = 0.645). Cox regression identified combination therapy as a protective factor (HR = 0.54, P < 0.001), while ECOG PS ≥ 2 (HR = 1.87, P = 0.002), liver metastasis (HR = 1.62, P = 0.013), bone metastasis (HR = 1.84, P < 0.001), and malignant pleural effusion (HR = 1.64, P < 0.001) independently worsened prognosis. The incidence of AEs of any-grade (P < 0.001) and grade 3-4 (P = 0.003) in the immunotherapy combination group was significantly higher than that in the immunotherapy monotherapy group. A prognostic nomogram integrating treatment type, ECOG PS score, and other six variables had an AUC value of 0.70-0.71 for predicting 1-2 year OS. CONCLUSIONS: For elderly patients with advanced NSCLC, immune combination therapy improved median OS over monotherapy. It was safe and effective, suggesting a viable treatment option, though further evaluation is needed for those aged 75 and older. A prognostic nomogram for OS following immunotherapy was developed, showing superior accuracy.