COPD May Follow 2 Distinct Patterns, According to Machine Learning

The disease progression of patients with chronic obstructive pulmonary disease (COPD) follows 2 distinct patterns which likely represent 2 different endotypes, according to study results published in The American Journal of Respiratory Critical Care Medicine.

Computed tomography scans of patients with COPD were subgrouped using a machine learning program called Subtype and Stage Inference (SuStaIn). Disease progression was then compared longitudinally between groups.

Among the 3698 patients with COPD and 3479 control participants, there were 2 distinct subtypes identified: Tissue→Airway (n=2354; 70.4%) and Airway→Tissue (n=988; 29.6%). In the Tissue→Airway subtype, small airway dysfunction and emphysema precede large-airway wall abnormalities whereas, in the Tissue→Airway subtype, large-airway wall abnormalities precede emphysema and small airway dysfunction. Baseline stage in both subtypes correlated with future lung function decline. Furthermore, nearly one-third of healthy smokers showed signs of detectable changes in the imaging, suggesting the potential for a new early biomarker of disease. Patients with early changes were 2.5 times more likely to meet diagnostic criteria for COPD at follow-up.

“[T]he majority of patients develop small airway disease and emphysema before large airway wall changes, but a significant minority (30%) develop large airway wall changes first,” the researchers wrote. “[T]he relationship with lung function in these subtypes is different, with a more rapid initial decline in lung function (greater disease activity) observed in the Tissue→Airway group.

Disclosure: Several study authors declared affiliations with the pharmaceutical industry. Please see the original reference for a full list of authors’ disclosures.

Reference

Young AL, Bragman FJS, Rangelov B, et al; for the COPDGene Investigators. Disease progression modelling in chronic obstructive pulmonary disease (COPD) [published October 28, 2019]. Am J Respir Crit Care Med. doi:10.1164/rccm.201908-1600OC