Machine learning-based MPO and Lp-PLA2 profiling reveals asthma-predominant inflammatory signature in asthma-COPD
Mingtao Liu1,2,3, Jiaxi Chen4, Jiani Yao4
1Department of Clinical Laboratory, State Key Laboratory of Respiratory Disease, National Center for Respiratory Medicine, National Clinical Research Center for Respiratory Disease, Guangzhou Institute of Respiratory Health, The First Affiliated Hospital of Guangzhou Medical University, Guangzhou 510120, China.
The World Allergy Organization Journal
|July 12, 2026
Summary
Machine learning analysis reveals asthma-COPD overlap (ACO) shares inflammatory traits with asthma, not COPD. Biomarkers like the MPO/Lp-PLA2 ratio and ACORN score aid in precision phenotyping for improved treatment strategies.
Area of Science:
- Pulmonary Medicine
- Biomarkers
- Machine Learning
Background:
- Asthma-COPD overlap (ACO) is poorly understood at the molecular level, causing diagnostic and treatment challenges.
- Current understanding limits precise identification and management of ACO patients.
Purpose of the Study:
- To investigate if machine learning analysis of myeloperoxidase (MPO) and lipoprotein-associated phospholipase A2 (Lp-PLA2) can define ACO as a distinct inflammatory endotype.
- To explore the potential of these biomarkers in differentiating ACO from asthma and COPD.
Main Methods:
- Prospective enrollment of 609 patients with obstructive airway diseases (asthma, COPD, AECOPD, ACO).
- Quantification of serum MPO and Lp-PLA2 using chemiluminescence immunoassay.
- Application of supervised (XGBoost, LightGBM, CatBoost, Neural Network, Random Forest) and unsupervised (hierarchical, k-means, DBSCAN) machine learning algorithms for classification and phenotyping.
- Model interpretability using SHAP analysis and dimensionality reduction (PCA, UMAP, t-SNE).
Main Results:
- The MPO/Lp-PLA2 ratio was lower in ACO patients, resembling asthma more than COPD.
- Unsupervised clustering showed 72.2% of clinically diagnosed ACO patients grouped with asthma.
- An ensemble machine learning model achieved high performance (AUC 0.954), outperforming individual biomarkers and clinical criteria.
- The ACORN score demonstrated a 91.2% positive predictive value for ACO recognition.
Conclusions:
- Machine learning analysis indicates ACO exhibits predominantly asthma-like inflammatory characteristics.
- The MPO/Lp-PLA2 ratio and ACORN score facilitate precision phenotyping of obstructive airway diseases.
- Biomarker-guided treatment strategies are advocated for improved patient outcomes in ACO.
Related Concept Videos
01:28Asthma I: Introduction
Asthma is a chronic inflammatory disorder of the airways characterized by variable airflow obstruction and heightened bronchial responsiveness to a wide range of triggers. The underlying inflammation leads to airway swelling, mucus hypersecretion, and smooth muscle constriction, all of which narrow the airway lumen and impede airflow. Clinically, asthma presents with recurrent episodes of wheezing, shortness of breath, chest tightness, and coughing, symptoms that typically vary in intensity and...
01:24Chronic Obstructive Pulmonary Disease III: Chronic Bronchitis Features
Chronic bronchitis is a key phenotype of chronic obstructive pulmonary disease (COPD), characterized by airway-centered inflammation and mucus overproduction. It develops from long-term exposure to harmful particles or gases, most commonly cigarette smoke, which triggers a persistent inflammatory response.Cellular and Structural ChangesInflammation initially affects the large bronchi and later the smaller airways, with infiltration by immune cells, including neutrophils, macrophages, and...
01:26Asthma-II: Pathophysiology and Classification
Asthma is a prevalent chronic respiratory condition marked by inflammation and hyperresponsiveness of the airways. Its pathophysiology involves complex interactions among inflammatory pathways, immune responses, and neural mechanisms.
Additionally, environmental and genetic factors play crucial roles in determining an individual's susceptibility to asthma and the severity of their condition.
Critical processes in asthma pathophysiology include:
Additionally, environmental and genetic factors play crucial roles in determining an individual's susceptibility to asthma and the severity of their condition.
Critical processes in asthma pathophysiology include:
01:29Asthma-I: Introduction
Asthma is a chronic respiratory ailment that requires careful management due to its varying symptoms and influencing factors. It is characterized by airway inflammation, bronchial hyperresponsiveness, and reversible airflow obstruction, leading to symptoms like wheezing, shortness of breath, chest tightness, and coughing. The symptom frequency and intensity may vary considerably over time. It is also linked to immune system responses to allergens and irritants, highlighting the complex...
01:20Chronic Obstructive Pulmonary Disease-II: Pathophysiology
Chronic Obstructive Pulmonary Disease (COPD) pathophysiology is intricate and multifaceted, involving a complex interplay of physiological processes. Understanding these mechanisms is crucial for effectively managing and treating COPD. Here is an in-depth look at the critical elements in the pathophysiology of COPD:
Chronic Inflammation
Chronic Inflammation
01:20Asthma: Pathogenesis and Management
Asthma is a chronic pulmonary condition involving inflammation of the airways, hyper-reactivity, and reversible obstruction of the airways. This condition can significantly impact a person's quality of life, making breathing difficult and leading to distressing symptoms.
Asthma is classified as allergic and non-allergic. Allergens such as dust mites, pollen, and pet dander trigger allergic asthma, while factors like cold air, intense emotions, or exercise can induce non-allergic asthma.
Asthma is classified as allergic and non-allergic. Allergens such as dust mites, pollen, and pet dander trigger allergic asthma, while factors like cold air, intense emotions, or exercise can induce non-allergic asthma.