Microbiome-driven precision medicine in asthma: Roles of targeted therapies and artificial intelligence

To evaluate current evidence on airway and gut microbiome alterations in asthma pathogenesis, inflammatory heterogeneity, and therapeutic response, and to examine microbiome-targeted interventions and artificial intelligence (AI)-based approaches for precision asthma medicine.

A structured narrative review was conducted using PubMed, Scopus, and Google Scholar to identify key literature published from 2015 to 2026, supplemented by landmark studies where appropriate. The review focused on microbiome dysbiosis, host-microbe interactions, inflammatory endotypes, microbiome-targeted interventions, biologic therapies, and AI- and machine-learning-based approaches relevant to precision asthma care.

Asthma-associated microbiome alterations were linked to disease heterogeneity, inflammatory endotypes, exacerbation risk, and treatment response, although evidence remained largely associative. Airway dysbiosis, particularly enrichment of Haemophilus, Moraxella, and Neisseria, was associated with neutrophilic inflammation and exacerbation-prone disease. Gut microbiota and short-chain fatty acid-producing bacteria were implicated in immune regulation through the gut-lung axis. Probiotics, prebiotics, synbiotics, postbiotics, engineered microbial consortia, and bacteriophages showed promise but remain investigational. Biologics targeting IgE, IL-5/IL-5R, IL-4Rα, and TSLP improved outcomes in selected patients, but microbiome-based response predictors remain unvalidated. AI/ML integrating microbiome, multi-omic, clinical, and environmental data showed potential for endotype classification, exacerbation prediction, biomarker discovery, and treatment-response prediction, but require robust validation.

The airway and gut microbiomes offer promising biomarkers and therapeutic targets for precision asthma care. Integration of multi-omic, clinical, and environmental data with AI may improve patient stratification and therapeutic decision-making, but standardized methodologies and large prospective, multicenter, externally validated studies are needed before routine clinical implementation.