Background The association between smoking status, cumulative smoking dose, and immunotherapy efficacy in non-small cell lung cancer (NSCLC) remains controversial. We sought to integrate the lifetime pack-years with smoking cessation status to identify optimal immunotherapy beneficiaries. Methods A total of 1,192 immunotherapy-treated NSCLC patients treated between November 2015 and April 2024 were enrolled. Data on demographics, clinical characteristics, pathologic characteristics, treatments, and clinical outcomes were collected. The objective response rate (ORR), disease control rate (DCR), and progression-free survival (PFS) were compared across different smoking statuses (never, current, and former smokers) and cumulative smoking doses (never smokers, non-heavy smokers: 20 pack-years, and heavy smokers: ≥20 pack-years). Multivariate logistic regression and Cox proportional hazards models were used to analyze ORR and PFS, respectively. Results Among the 1,192 patients, 377 were never smokers, 499 were current smokers, and 316 were former smokers. In terms of smoking status, former smokers exhibited the longest median PFS (17.0 months, P 0.001), with the highest ORR (46.8%, P 0.001) and DCR (86.7%, P = 0.008). Regarding cumulative smoking dose, the heavy smoker group demonstrated the longest median PFS (15.9 months, P = 0.001), with the highest ORR (46.6%, P 0.001) and DCR (85.2%, P = 0.012). Notably, further multivariate analysis identified former heavy smokers as independent favorable predictors of ORR (OR = 1.93, 95% CI = 1.25–2.99, P = 0.003) and PFS (HR = 0.75, 95% CI = 0.57–0.99, P = 0.04) in advanced NSCLC patients receiving immunotherapy. Conclusions This real-world cohort analysis establishes a clinical stratification combining smoking cessation status with cumulative smoking dose, identifying former heavy smokers as optimal immunotherapy beneficiaries. These findings advocate integrated smoking history documentation and emphasize clinical prioritization of cessation interventions to enhance treatment efficacy in NSCLC.
Yang et al. (2025) studied this question.