In September 2007, the largest study of COPD in the United States was funded by the NHLBI to broadly address unanswered questions about the causes of COPD. The study was funded as paired projects at National Jewish Health in Denver, Colorado, and Brigham and Women’s Hospital in Boston, Massachusetts, under the direction of Co-Principal Investigators Dr. James D. Crapo, FCCP; and Dr. Edwin K. Silverman, PhD. The ambitious goals of the project were to enroll 10,000 smokers with and without COPD from two ethnic groups, precisely characterize the subjects using spirometry, respiratory symptoms, medical history, 6-minute walk test, and high-resolution CT scans, followed by genome-wide association testing for susceptibility to COPD and COPD-related traits (Regan et al. COPD. 2010;7:32). The extensive clinical, radiographic, and genetic phenotyping would be analyzed to detect novel subtypes of COPD. In addition, the study investigators planned to enhance enrollment of African Americans to address apparent disparities in COPD susceptibility and to potentially discover unique COPD susceptibility genes in this population.
Initially, some skepticism was expressed about the ability to complete enrollment for such a complex project over the 5-year period of funding. Twenty-one clinical centers across the country commenced enrolling subjects in March 2008. Enrollment was completed in just over 3 years due largely to a committed group of investigators and clinical coordinators at each of the sites. CT scans are being analyzed by the Quantitative Imaging Lab at National Jewish, using the VIDA software, for automated measures of emphysema, gas trapping, and airway disease and by the Brigham and Women’s Hospital imaging team using an alternative software package. A separate project to maintain longitudinal contacts was established, and 80% of the cohort has been reporting ongoing health data approximately every 6 months.
More than 30 scientific papers are either published or in preparation with early results from the study. These include a number of genetics papers in which COPDGene GWAS data from an early group of 1,000 subjects was used in combination with other large cohorts to identify or confirm genetic findings, such as associations to FAM13A for COPD affection status (Cho et al. Nat Genet. 2010;42:200), FTO variants to BMI in COPD subjects (Wan et al. Am J Res Cell Mol Biol. 2011;45:304), and smoking behaviors to regions of CYP2A6 and a locus at chromosome 15q25 (Siedlinski et al. Thorax. 2011;66:894).
Interstitial lung disease in the context of smoking was the topic of another study from COPDGene (Washko et al. N Engl J Med. 2011; 364:897). Using visual reads of the CT scans from 2,416 subjects enrolled in COPDGene, the group found 194 subjects (8%) had evidence of interstitial lung abnormalities (ILA). These subjects with CT evidence of ILA had greater smoking exposure, were more current smokers, had lower total lung capacity, and were more likely to fall into the unclassified category of subjects. However, there was a definite overlap of ILA with COPD— 63 (32%) of the ILA subjects fell in the GOLD 2-4 group, and these subjects showed less emphysema and lower total lung capacity than subjects without ILA.
A unique group of smokers has emerged from the COPDGene project that is not classified by GOLD criteria but shows significant health effects. These subjects have a normal (>0.7) FEV1/FVC ratio but reduced (<80% predicted) FEV1. They have been termed GOLD unclassified subjects. Early in the recruitment process for COPDGene, it was decided that the study would encompass all smokers with the specified smoking exposure and not exclude subjects who failed to meet the current classification criteria for COPD. The rationale for this was that these subjects had smoking exposure and either represented an alternative control group or potentially a group “at risk” of smoking-related disease.
For purposes of studying genetic associations to COPD, this group represents an important comparison group. The GOLD unclassified subjects constituted 9% of the subjects in COPDGene study (Wan et al. Am J Respir Crit Care Med. 2011;184:57). They showed greater impairment of the 6-min walk, fewer pack-years of smoking, higher BMI, reduced total lung capacity, more comorbid cardiovascular disease, and lower oxygen saturation. The unclassified group is heterogeneous, with about half of the group having a reduced total lung capacity, slightly more than half of the group is obese with BMI>30, and the group overall has significantly more comorbid disease and worse physical function than the GOLD 1 group. Further study will be done on this group of smokers as the full cohort is analyzed and genetic data become available.
Several groups of investigators have used the early COPDGene data to look at the unique aspects of race and COPD. Using the large population of African American subjects enrolled in the study, investigators found significant differences in quality of life and function (Han et al. Chest. 2011; 140:1169). Although African American subjects had similar mean percent predicted FEV1 values compared with non-Hispanic white subjects, they had fewer pack years of smoking exposure and worse 6-minute walk distance (381 meters ±135 vs 298 meters ± 119, P <.001).
Within COPD subjects who reported exacerbations, African Americans had worse quality of life and more dyspnea.
Another study that addressed differential impacts of race looked at Early Onset COPD (Foreman et al. Am J Respir Crit Care Med. 2011; 184:414). In this study, the authors found that severe early onset COPD (age < 55 years and FEV1< 50% predicted) was identified more often in African American subjects (present in 42% of African Americans compared with 14% of the non-Hispanic white group, P <.0001). Women were also overrepresented in the early onset group, and both maternal smoking and maternal history of COPD were significant predictors of early disease.
Chronic bronchitis as a distinct phenotype in COPD was identified by another group (Kim et al. Chest.2011;140:626). Chronic bronchitis was defined as cough and phlegm production for greater than 3 months, and 27% of the subjects with COPD studied fell into that category. These subjects were younger, had greater pack-years of exposure, were more likely to be current smokers, had more exacerbations, and were more likely to have severe exacerbations than subjects without chronic bronchitis, although their mean FEV1 was the same, and there was no difference in percent emphysema or gas trapping. The subjects with chronic bronchitis did have evidence of greater airway disease with mean segmental wall area percent significantly greater. This group is posited to need directed therapy for reducing smoking and airway mucin production.
The COPDGene study is poised to provide important information about both genetic causes of COPD and the natural history of the disease in the coming years. The investigators hope to obtain additional funding to perform a second evaluation of these subjects in order to determine disease progression over time. The genetic data from the whole cohort are in preparation for analysis. With data on 10,000 well-characterized smokers, we anticipate detecting additional important genetic findings over the next few years that, hopefully, will offer opportunities to improve treatment or reduce the incidence of this disabling disease.
Dr. Elizabeth A. Regan, PhD
Assistant Professor, Department of Medicine
Associate Director, COPDGene
National Jewish Medical and Research Center
Guest Editor’s Note
The COPDGene project offers an incredible degree of promise in our efforts to better understand the disease we call “COPD.” The project is already starting to nicely demonstrate that COPD is not merely one homogeneous disease but rather multiple different diseases with some commonality in their presentation. The early data from COPDGene are pointing us in that direction, with evidence supporting some specific phenotypes, such as the restrictive or “unclassified” group, the early-onset group, and the chronic bronchitis group. Additional work will, in all likelihood, identify additional important subgroups, along with genetic or other markers of disease activity and severity. The ultimate goal of this work and other related projects around the world is to prevent, treat, and even cure COPD.
We have come a long way over the last 5 years, but still have a long way to go toward our goal of decreasing the morbidity and mortality of this group of diseases.
Dr. David M. Mannino, FCCP
Director of Graduate Studies, Masters of
Science in Clinical Research Design
Director, Pulmonary Epidemiology
Department of Preventive Medicine and Environmental Health
University of Kentucky College of Public Health