Abstract Rationale Asthma and chronic obstructive pulmonary disease (COPD) comprise several phenotypes with distinct characteristics and clinical courses. Most studies have examined these conditions separately, with limited clinical data addressing their overlap. In particular, prospective studies with longitudinal data on clinical phenotypes across a broad spectrum of chronic airway diseases, including asthma and COPD, remain limited. We aimed to identify distinct clinical phenotypes of chronic airway diseases and compare their long-term outcomes. Methods We included patients with asthma, COPD, emphysema, or chronic bronchitis enrolled in the Prospective Integrative Cohort Study of Chronic Airway Diseases, which was a multicenter, prospective cohort study conducted between April 2018 and July 2021 in Hokkaido University Hospital and nine affiliated institutions. We performed a hierarchical cluster analysis using 14 baseline variables: age, sex, body mass index, pack-years, duration of chronic airway disease, post-bronchodilator forced expiratory volume in 1 s (FEV1) as percent predicted, post-bronchodilator FEV1/forced vital capacity, fractional exhaled nitric oxide (FeNO), blood eosinophil count, blood neutrophil count, serum total IgE level, asthma control test, COPD assessment test, and history of exacerbations in the previous year. Participants underwent annual follow-up assessments for 3 years. Outcomes, including exacerbation rates, time to first exacerbation, use of systemic steroids and/or antibiotics, and annual decline in FEV1, were compared across clusters. Results Among 974 patients, six clinical clusters were identified: Cluster 1, predominantly male, smokers, poor lung function, and emphysema; Cluster 2, predominantly male, high type-2 inflammation, and atopy; Cluster 3, mostly obese females, high type-2 inflammation and symptom burden; Cluster 4, mostly females, low smoking exposure, early-onset, and atopy; Cluster 5, low type-2 inflammation with preserved lung function; Cluster 6, smokers with low symptom burden. Among the six clusters, the proportion of patients experiencing exacerbations was the highest for Cluster 3 (34.4%) and the lowest for Cluster 6 (9.2%), and the Cluster 3 exhibited the shortest time to first exacerbation (Kaplan-Meier analysis; log-rank P 0.001). The proportion of exacerbations requiring steroid use was particularly high in Clusters 2 and 3 (90.3% and 93.1%, respectively). Cluster 1 showed the greatest annual decline in FEV1 (P = 0.005). Conclusions We identified distinct clinical phenotypes across various chronic airway diseases, some aligned with conventional diagnostic labels, while others did not. The varied baseline characteristics and clinical trajectory among the clusters indicated the need for a phenotype-based approach to their management. This abstract is funded by: Novartis Pharma K.K.
Wakazono et al. (Fri,) studied this question.