Imaging-based clusters in former smokers of the COPD cohort associate with clinical characteristics: the SubPopulations and intermediate outcome measures in COPD study (SPIROMICS)

Abstract

Background Quantitative computed tomographic (QCT) imaging-based metrics enable to quantify smoking induced disease alterations and to identify imaging-based clusters for current smokers. We aimed to derive clinically meaningful sub-groups of former smokers using dimensional reduction and clustering methods to develop a new way of COPD phenotyping. Methods An imaging-based cluster analysis was performed for 406 former smokers with a comprehensive set of imaging metrics including 75 imaging-based metrics. They consisted of structural and functional variables at 10 segmental and 5 lobar locations. The structural variables included lung shape, branching angle, airway-circularity, airway-wall-thickness, airway diameter; the functional variables included regional ventilation, emphysema percentage, functional small airway disease percentage, Jacobian (volume change), anisotropic deformation index (directional preference in volume change), and tissue fractions at inspiration and expiration. Results We derived four distinct imaging-based clusters as possible phenotypes with the sizes of 100, 80, 141, and 85, respectively. Cluster 1 subjects were asymptomatic and showed relatively normal airway structure and lung function except airway wall thickening and moderate emphysema. Cluster 2 subjects populated with obese females showed an increase of tissue fraction at inspiration, minimal emphysema, and the lowest progression rate of emphysema. Cluster 3 subjects populated with older males showed small airway narrowing and a decreased tissue fraction at expiration, both indicating air-trapping. Cluster 4 subjects populated with lean males were likely to be severe COPD subjects showing the highest progression rate of emphysema. Conclusions QCT imaging-based metrics for former smokers allow for the derivation of statistically stable clusters associated with unique clinical characteristics. This approach helps better categorization of COPD sub-populations; suggesting possible quantitative structural and functional phenotypes. Electronic supplementary material The online version of this article (10.1186/s12931-019-1121-z) contains supplementary material, which is available to authorized users.

Longitudinal Imaging-Based Clusters in Former Smokers of the COPD Cohort Associate with Clinical Characteristics: the SubPopulations and Intermediate Outcome Measures in COPD Study (SPIROMICS)

“…These subjects were recruited in the NIHfunded SPIROMICS multicenter research study. A subset of these subjects were utilized to derive cross-sectional cluster membership in the previous study (14), allowing for the comparison of longitudinal and cross-sectional clusters. All subjects had a baseline visit and a one-year follow-up visit.…”

“…The separate quantification of emphysema and small airway disease can dissociate the emphysematous region(s) from the air-trapped region(s) (3). fSAD% and Emph% with the respective air-fraction threshold of 90% and 98.5% were employed to eliminate the effects of scanners (13)(14)(15). Tissue fractions at TLC (β tissue TLC ) and at RV (β tissue RV ) (14) were introduced to measure the proportion of tissue volume in each voxel for detection of tissue destruction and inflammation.…”

“…We compared the cross-sectional clustering (14) with our longitudinal clustering by studying the 269 participants who were in both studies. The confusion matrix between crosssectional clustering and longitudinal clustering (Table 10) Table 7).…”

“…Cluster analysis of clinical variables by Burgel et al (10) concluded that COPD patients with similar airflow obstruction may belong to different phenotypes and may differ in symptoms, outcomes, age and comorbidities. Besides clinical variables, qCT imaging-based machine learning of diseased lungs has advanced recently, and multiscale imaging-based cluster analyses have provided clinically meaningful clusters in both asthma and COPD cohorts (11)(12)(13)(14). By considering inter-site and inter-subject variations through normalization schemes (15,16), their approach enabled the analysis of datasets acquired by multiple centers utilizing multiple make and model CT scanners.…”

“…In this study, we performed a longitudinal cluster analysis of former smokers based on the imaging-derived variables at two visits with a one-year interval. We focused on former smokers because smoking status may introduce confounding factors that interfere with qCT quantification (13,14). The feature vector for each subject contains the baseline imagingderived variables and their percentages of change over one year.…”