• Unsupervised identification of...

*Result*: Unsupervised identification of asthma symptom subtypes supports treatable traits approach.

Title:
Unsupervised identification of asthma symptom subtypes supports treatable traits approach.
Authors:
Hamada K; Department of Systems Bioinformatics, Graduate School of Medicine, Yamaguchi University, Yamaguchi, Japan; Department of Respiratory Medicine and Infectious Disease, Graduate School of Medicine, Yamaguchi University, Yamaguchi, Japan., Abe T; AI Systems Medicine Research and Training Center, Graduate School of Medicine and University Hospital, Yamaguchi University, Yamaguchi, Japan; The Division of Systems Medicine and Informatics, Research Institute for Cell Design Medical Science, Yamaguchi University, Yamaguchi, Japan., Oishi K; Department of Respiratory Medicine and Infectious Disease, Graduate School of Medicine, Yamaguchi University, Yamaguchi, Japan., Murata Y; Department of Respiratory Medicine and Infectious Disease, Graduate School of Medicine, Yamaguchi University, Yamaguchi, Japan., Hirano T; Department of Respiratory Medicine and Infectious Disease, Graduate School of Medicine, Yamaguchi University, Yamaguchi, Japan., Hayano T; Department of Systems Bioinformatics, Graduate School of Medicine, Yamaguchi University, Yamaguchi, Japan; The Division of Systems Medicine and Informatics, Research Institute for Cell Design Medical Science, Yamaguchi University, Yamaguchi, Japan., Nakatsui M; AI Systems Medicine Research and Training Center, Graduate School of Medicine and University Hospital, Yamaguchi University, Yamaguchi, Japan; The Division of Systems Medicine and Informatics, Research Institute for Cell Design Medical Science, Yamaguchi University, Yamaguchi, Japan., Asai Y; Department of Systems Bioinformatics, Graduate School of Medicine, Yamaguchi University, Yamaguchi, Japan; AI Systems Medicine Research and Training Center, Graduate School of Medicine and University Hospital, Yamaguchi University, Yamaguchi, Japan; The Division of Systems Medicine and Informatics, Research Institute for Cell Design Medical Science, Yamaguchi University, Yamaguchi, Japan. Electronic address: asai@yamaguchi-u.ac.jp., Matsunaga K; Department of Respiratory Medicine and Infectious Disease, Graduate School of Medicine, Yamaguchi University, Yamaguchi, Japan.
Source:
Allergology international : official journal of the Japanese Society of Allergology [Allergol Int] 2026 Jan; Vol. 75 (1), pp. 92-102. Date of Electronic Publication: 2025 Jul 28.
Publication Type:
Journal Article
Language:
English
Journal Info:
Publisher: Elsevier Country of Publication: England NLM ID: 9616296 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1440-1592 (Electronic) Linking ISSN: 13238930 NLM ISO Abbreviation: Allergol Int Subsets: MEDLINE
Imprint Name(s):
Publication: 2015- : London : Elsevier
Original Publication: Carlton, Vic., Australia : Blackwell Science, c1996-
MeSH Terms:
Asthma*/diagnosis , Asthma*/classification , Asthma*/therapy , Unsupervised Machine Learning*, Humans ; Male ; Female ; Middle Aged ; Adult ; Cluster Analysis ; Precision Medicine ; Surveys and Questionnaires ; Algorithms ; Patient Reported Outcome Measures ; Aged
Contributed Indexing:
Keywords: Asthma symptom subtypes; Healthcare disparities; Patient-reported outcome; Treatable traits; Unsupervised machine learning
Entry Date(s):
Date Created: 20250729 Date Completed: 20260104 Latest Revision: 20260105
Update Code:
20260130
DOI:
10.1016/j.alit.2025.06.004
PMID:
40730774
Database:
MEDLINE

*Further Information*

*Background: Heterogeneity of asthma requires a personalized therapeutic approach. However, objective measurements, such as spirometry and fraction of exhaled nitric oxide (FeNO) for implementing treatable traits approach, are limited in low- and middle-income countries and non-specialist settings. To implement precision medicine even with minimal resources, we developed an algorithm using unsupervised machine learning techniques that estimates key treatable traits (airflow limitation, type 2 [T2] inflammation, and frequent exacerbations) based on an asthma patient-reported outcome (PRO).
Methods: We applied hierarchical clustering and Uniform Manifold Approximation and Projection (UMAP) to Asthma Control Questionnaire (ACQ)-5 including five residual symptoms from two asthma cohorts (the discovery cohort with 1697 patients and validation cohort with 157 patients).
Results: We identified five symptom clusters, characterized by key treatable traits: Cluster 1, minimal asthma symptoms; Cluster 2, a little symptom, mild airflow limitation; Cluster 3, predominant shortness of breath and wheezes, airflow limitation; Cluster 4, predominant morning symptoms and nocturnal awakening, T2 inflammation; and Cluster 5, all symptoms severe, airflow limitation, T2 inflammation and frequent exacerbations. The UMAP projections of ACQ-5 (five-dimensional) to two-dimensions allowed to visualize datapoints and clusters, which visually revealed that patients with poorly-controlled asthma were divided into Clusters 3, 4 and 5. These results were externally validated in an independent cohort.
Conclusions: Based on asthma PRO data, the developed algorithm categorized asthma patients into five symptom-based subtypes that provide insights into key treatable traits. Our data-driven digital health approach will extend precision medicine of asthma to medical facilities even in resource-constrained settings.
(Copyright © 2025 Japanese Society of Allergology. Published by Elsevier B.V. All rights reserved.)*

*Conflict of interest Our machine learning technique has been applied for a patent in Japan (application number: JP2022-200914) and PCT applications (application number: PCT/JP2023/38330 (KH, KM, and YA). KH received speaker fees from AstraZeneca, Kyorin and Sanofi. KO has received speaker fees from AstraZeneca, Boehringer Ingelheim and Sanofi. THi has received speaker fees from AstraZeneca, Novartis Pharma, and Sanofi. KM received speaker fees from AstraZeneca, Boehringer Ingelheim, GlaxoSmithKline, Kyorin, Novartis Pharma, and Sanofi. The rest of the authors have no conflict of interest.*