ATS 2023 Abstracts Explore Markers of COPD Progression and Morbidity

A symposium at the 2023 American Thoracic Society (ATS) International Conference featured research on predictors of chronic obstructive pulmonary disease (COPD) occurrence and disease progression.

COPD is characterized by progressive damage to the lungs and breathing difficulty over time among affected individuals, with a host of clinical signs manifesting as the disease progresses.

In a symposium at the 2023 American Thoracic Society International Conference, taking place May 20-24 in Washington, DC, researchers presented abstracts exploring the prediction of COPD and markers of progression.

Predicting Future Lung Health With Baseline Lung Function

Overall lung health is typically determined by measuring forced expiratory volume in 1 second (FEV1), with a decline in FEV1 increasing the risk of COPD over time. However, baseline FEV1 as a predictor of long-term respiratory disease and overall health outcomes has not been well studied, according to the first abstract presented at the session.1

The researchers hypothesized that low or high baseline FEV1 (LLF or HLF) could be an indicator of future lung health for younger patients and tested the theory in a cohort of 532 Lovelace Smokers Cohort (LSC) participants between 40 and 60 years of age.

All patients had a post bronchodilator FEV1/forced vital capacity (FVC) ≥ 0.7 and were stratified into a HLF group (mean FEV1 of 105% predicted) and a LLF group (mean FEV1 of 73% predicted). Eleven years later, 56 patients from the HLF and 24 participants from the LLF groups were assessed, with a mean follow-up of 5.5 years.

Those in the HLF group saw an FEV1 decline of 30 ml/year vs 20 ml/year in the LLF group. Individuals in the LLF group had a higher COPD incidence compared with the HLF group (9% vs 3%) and a higher risk of death (18% vs 6%). The LLF group also had higher rates of diabetes (13% vs 4%) and chronic bronchitis (36% vs 25%).

After adjustment, those with LLF at baseline were at higher risks of COPD, diabetes, hypertension, cardiovascular disease, and mortality.

“What we see is that in current and ever smokers that are heavy smokers, low lung function as determined by spirometry is an important predictor for death, as well as incidence of cardiovascular disease, diabetes, and other disease as well. Therefore, this study supports that one should consider having spirometry at an early age,” said presenter Yohannes Tesfaigzi, PhD, professor of medicine at Harvard University.

The Impact of Respiratory Exacerbations on Lung Function in Those With Normal Spirometry

Among individuals with normal spirometry, the effect of respiratory disease exacerbations is not as clear, according to the authors of another abstract.2 The study explored the association of exacerbations with lung function decline in patients in the COPDGene study with at least 10 pack-years of smoking history who had normal lung function (FEV1/FVC <0.7 and/or FEV1 predicted < 80%) at enrollment.

“It is well-established that in COPD, respiratory exacerbations are associated with lung function decline and mortality, but the effect of respiratory exacerbations on lung function and mortality on those who have normal spirometry is understudied,” said presenter Spyridon Fortis, MD, a pulmonologist and clinical associate professor of internal medicine at the University of Iowa.

In the study population (n = 2628), all of whom had normal lung function at enrollment, 1 exacerbation per year was associated with COPD at a 5-year follow-up visit, with an odds ratio of 1.32 (95% CI, 1-1.74, P = .045). Severe exacerbations per year did not show a statistically significant association with COPD at visit 2, however.

“Exacerbations between visit 1 and visit 2 were associated with increased mortality, but the relationship between respiratory exacerbations and mortality is not significantly affected by having COPD at visit 2,” Fortis and co-authors wrote.

Overall, they concluded that respiratory exacerbations in individuals with normal spirometry resulted in lung function decline and COPD, but also predicted mortality without progression to COPD first.

Factors Impacting COPD-Specific Mortality in a Novel Risk Model

While prognostic risk scores for COPD exist, most model mortality overall and not COPD-specific mortality.3 This is why researchers at the University of Pittsburgh School of Medicine developed CausalCoxMGM, a novel method “that learns causal probabilistic graphical models of heterogeneous clinical datasets containing continuous, categorical, and censored variables.”

They applied the model to individuals in phase 1 and 2 of the COPDGene study to learn the clinical, imaging, and blood-derived gene expression features that are linked to overall and adjudicated COPD-related mortality.

Four features were associated with overall and COPD-specific mortality: age, 6-minute walk distance (6MWD), resting oxygen saturation, and the standard deviation of lung attenuation from CT imaging.

Features linked to COPD-specific mortality in the phase 1 group were FEV1/FVC and ATS/ERS classification of COPD severity. Factors contributing to overall mortality were body mass index, heart rate, duration of smoking, and overall health status. In the phase 2 group, 6MWD was linked to COPD-specific mortality and overall mortality, while age and airway wall thickness were linked COPD-specific mortality. The gene expression markers NELL2 and GRB10 were directly linked to overall mortality in the phase 2 cohort, and NRG1 was directly linked to COPD-specific mortality.

“If we’ve done a good job of identifying the features that are directly linked to COPD-specific mortality, a predictor constructed on these features should do a good job of generalizing to new patient populations,” explained Tyler Lovelace, PhD student and graduate research assistant, who presented the study.

Additional markers, including gene expressions and microbiome profiles, were explored in abstracts presented at the session, emphasizing a need for more research in this area and the exploration of additional markers for COPD and other lung conditions.


1. Tesfaigzi Y, Brown MN, Liu C, et al. The FEV1 as a predictor of future health in young non-obstructed smokers. Abstract presented at: American Thoracic Society International Conference; May 20-24, 2023; Washington, DC. Accessed May 23, 2023.

2. Fortis S, Strand M, Bhatt SP, et al. Respiratory exacerbations and lung function decline in people with normal spirometry and smoking exposure. Abstract presented at: American Thoracic Society International Conference; May 20-24, 2023; Washington, DC. Accessed May 23, 2023.

3. Lovelace TC, Benos PV. Disentangling predictors of overall and COPD-specific mortality with probabilistic graphical models. Abstract presented at: American Thoracic Society International Conference; May 20-24, 2023; Washington, DC. Accessed May 23, 2023.

Related Videos
Sheila Garland, PhD, MSc, Memorial University
Andrew McHill, PhD, Oregon Health and Science University
Dayna Johnson, PhD, MPH, MSW, MS, Rollins School of Public Health at Emory University
Related Content
© 2023 MJH Life Sciences
All rights reserved.