Cracking the Code: New ISAR Study Reveals How T2- and Non-T2 Inflammatory Pathways Predict Severe Asthma Exacerbations

A novel Bayesian Network analysis on the International Severe Asthma Registry (ISAR) has identified key clinical and biological pathways that contribute to the risk of future severe exacerbations in patients with severe asthma. Leveraging real-world data from over 6,800 biologic-naïve adults across 17 countries, the study—recently published in CHEST under the title “Prediction Pathway for Severe Asthma Exacerbations: A Bayesian Network Analysis”—provides significant insights into how complex clinical factors interact to influence exacerbation risk.

Using the Bayesian Network approach—a robust probabilistic model that integrates expert knowledge with machine learning—the researchers uncovered the relationships between key predictors and how they led to severe asthma exacerbations. Blood eosinophil count (BEC), fractional exhaled nitric oxide (FeNO), and percentage predicted FEV₁ were identified as both directly contributing as well as mediating the transition from past to future severe exacerbations. Chronic rhinosinusitis (CRS) influenced this transition indirectly by altering these biological markers. In a separate pathway, prior macrolide use also mediated the transition from past to future exacerbations, highlighting a non-T2 inflammatory pathway.

“This study moves beyond identifying isolated predictors by elucidating how they interact within a broader clinical framework,” said PI Prof. Wenjia Chen. “It further provides an influence diagram and a counterfactual prediction framework for more informed risk stratification and actionable risk modification in severe asthma management.”

By integrating clinical, functional, and inflammatory domains into a unified predictive framework, this study marks a significant advance toward precision medicine in severe asthma. The findings provide actionable insights that could inform risk stratification, guide targeted therapies, and ultimately improve patient outcomes worldwide.

To learn more about the study, please read the full publication in CHEST, as well as the accompanying slide deck.

Acknowledgement: This work was supported by the Singapore National Medical Research Council – Open Fund – Young Individual Research Grant (MOH-001337-00), the International Severe Asthma Registry (ISAR) Expert Panel and the Observational and Pragmatic Research Institute (OPRI). ISAR is operated by Optimum Patient Care Global (OPCG) and co-funded by OPCG and AstraZeneca Ltd.  We thank all investigators, collaborators, and patients who contributed to this research.

About OPRI   

The Observational and Pragmatic Research Institute (OPRI) is an internationally recognized independent research organization dedicated to providing real-world evidence that supports best practices in chronic disease management in primary care. Learn more at https://www.opri.org.uk/. For media inquiries and additional information, please contact https://www.opri.org.uk/contact.  

About ISAR 

The International Severe Asthma Registry (ISAR) is the first global adult severe asthma registry, providing a rich, standardized dataset to advance research, clinical practice, and policy in severe asthma care. It fosters international collaboration to improve outcomes for patients worldwide. ISAR is operated by Optimum Patient Care Global Ltd. (OPCG) and co-funded by OPCG and AstraZeneca. Learn more at https://www.isar.opcglobal.org.