Abstract
BACKGROUND: The human respiratory tract harbours diverse microbial communities crucial for health, but their dynamics during environmental perturbations like smoking remain poorly understood. While smoking is a major risk factor for various diseases, its compartment-specific effects on the respiratory microbiome and potential recovery following cessation have not been fully elucidated. Here, we present a longitudinal, multi-site study of respiratory microbiome dynamics in smokers undergoing cessation, benchmarked against healthy never-smokers. METHODS: Using standardized sampling of the anterior nares, oropharynx, and bronchoalveolar lavage (BAL), combined with 16S rRNA gene amplicon sequencing and rigorous contamination controls, we characterized community composition, diversity, inter-individual variability, and microbial interactions across airway compartments. RESULTS: Smokers exhibited pronounced microbiome alterations: nasal richness increased, while lung richness and core taxa were decreased. Smoking-induced changes were compartment-specific and most pronounced in nose and lung. The degree of individual-specific differences in community structure was elevated in smokers and correlated with smoking history. Short-term cessation (6 weeks) led to minor shifts in genus abundance but increased similarity between oropharyngeal and lung communities, whereas long-term cessation (1 year) resulted in partial restoration, particularly in lung and nasal microbiomes. Some genera, including Haemophilus and Prevotella_7, showed persistent alterations, suggesting lasting smoking effects. Network analyses revealed that smoking disrupted microbial co-occurrence and reduced community connectivity, whereas cessation partially restored interaction networks, with central taxa remaining altered and dynamics differing between oropharynx and lung, reflecting different underlying ecological assembly processes. Recovery trajectories were highly individualized, with lung microbiomes influenced by deterministic processes and upper airway microbiomes shaped by stochastic factors, explaining site-specific responses and the persistence of personalized microbial signatures. CONCLUSION: These results provide the first time-resolved, multi-compartment characterization of respiratory microbiome recovery after smoking cessation, revealing that smoking leaves lasting, site-specific imprints on airway microbial communities and interactions. Our findings underscore the need for individual and compartment-specific approaches when designing microbiome-based interventions to support respiratory health. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12931-026-03644-z.