Cluster analysis in the COPDGene study identifies subtypes of smokers with distinct patterns of airway disease and emphysema

Abstract

Background There is notable heterogeneity in the clinical presentation of patients with COPD. To characterize this heterogeneity, we sought to identify subgroups of smokers by applying cluster analysis to data from the COPDGene Study. Methods We applied a clustering method, k-means, to data from 10,192 smokers in the COPDGene Study. After splitting the sample into a training and validation set, we evaluated three sets of input features across a range of k (user-specified number of clusters). Stable solutions were tested for association with four COPD-related measures and five genetic variants previously associated with COPD at genome-wide significance. The results were confirmed in the validation set. Findings We identified four clusters that can be characterized as 1) relatively resistant smokers (i.e. no/mild obstruction and minimal emphysema despite heavy smoking), 2) mild upper zone emphysema predominant, 3) airway disease predominant, and 4) severe emphysema. All clusters are strongly associated with COPD-related clinical characteristics, including exacerbations and dyspnea (p<0.001). We found strong genetic associations between the mild upper zone emphysema group and rs1980057 near HHIP, and between the severe emphysema group and rs8034191 in the chromosome 15q region (p<0.001). All significant associations were replicated at p<0.05 in the validation sample (12/12 associations with clinical measures and 2/2 genetic associations). Interpretation Cluster analysis identifies four subgroups of smokers that show robust associations with clinical characteristics of COPD and known COPD-associated genetic variants.

Can metabotyping help deliver the promise of personalised nutrition?

“…In the field of personalised medicine, metabotyping may play a role by stratifying patients for the development of tailored healthcare solutions (45) . Examples of this concept are also available from the medical literature where cluster analysis has been used to identify different phenotypes for a range of diseases including chronic obstructive pulmonary disease (46) and Parkinsons disease (47) . This concept has been particularly useful for multisymtomatic and multifactorial diseases where presentation and severity of the disease can differ from one individual to the next.…”

From systems biology to P4 medicine: applications in respiratory medicine

“…They found that B-cell-related genes were significantly enriched in emphysema (compared with COPD patients without emphysema), paving the way for differential therapeutic research on inflammatory pathways of the adaptive immune response. 5) Two COPD studies demonstrated the utility of unsupervised k-means clustering by identifying robust cluster associations with clinical characteristics and known COPD genetic variants [95,96]. 6) Very recently, ROSS et al [97] introduced a new Bayesian method for COPD subtyping.…”

Clustering with flexible constraints and application to disease subtyping

“…Although being called one disease, clinicians believe that COPD is a heterogeneous disease and the discovery of different disease subtypes (clusters) can lead to tailored medical care for each patient [1]. The dataset we use consists of a wide range of features extracted from patients, including their demographics, clinical information, lung function measurements, and variables extracted from computed tomography (CT) chest images.…”

“…In a recent seminal work [1], instead of manually defining COPD subtypes, some COPD investigators apply K-means clustering using a feature set they consider as important and generate a clustering solution that has significant association with some key genetic variables. This work demonstrates that it is possible to define meaningful COPD subtypes using data-driven approaches, such as clustering algorithms.…”

“…For example, clinicians would like to cluster patients according to their disease subtypes so that patients belonging to different clusters can receive personalized treatments [1]. News editors would like to cluster documents according to topics to efficiently organize a a large set of news stories [2].…”