The Potential of Long-Term Study Cohorts

Lung development and function is notoriously difficult to study. Both are affected by a range of environmental factors, such as first- and second-hand smoke, air pollution, and infections and genetic factors can also play a role. Retrospective studies can be helpful, but it’s difficult for individuals to be sure of their environmental exposures, and such studies carry a strong risk for bias that can confound the results.

Those limitations put a premium on prospective cohort groups, who are recruited at a young age with the intent of following lung function over a lifespan and deliberately tracking environmental exposures. The oldest such cohort in the United States, and one of the oldest in the world, is the Tucson Children’s Respiratory Study (TCRS). It began between 1980 and 1984 with the recruitment of 1246 healthy newborns in the Tucson area.

After the first survey at birth, the study included exams, surveys, and blood draws every 4-6 years, beginning at age 6 years.

“We have a full picture of the natural history of asthma for the children who did develop disease, and a full profile of factors that could be examined for their role in making some of these participants more likely or less likely to develop a disease over time,” said Stefano Guerra, MD, PhD, who is a professor of medicine and director of the Population Science Unit at Asthma and Airway Disease Research Center, The University of Arizona, Tucson, Arizona.

“They really wanted a snapshot of the general population of newborns in Tucson at the time. The value of that is that the conclusions that you can reach in your study apply to the general population,” said Guerra, who noted that the area’s large Hispanic population ensured good representation of that group, with 25%, but there are few African Americans (4%).

Early on, in a study published in 1995, the cohort led to key insights into phenotypes of early childhood wheezing, where TCRS researchers found that most infants who wheeze have transient conditions linked to diminished airway function at birth and are not at increased risk for asthma later in life, though in a substantial minority the condition is associated with a heightened risk for asthma. TCRS has also provided insight into early life risk factors for chronic obstructive pulmonary disease (COPD) and the identification of a potential COPD biomarker that could assist with identifying at-risk individuals.

Prospective Value

Prospective studies like TCRS, which is the longest running respiratory birth cohort in the United States, are valuable because they link clinical information to disease development and outcomes in a way that is very difficult to capture in retrospective analyses, according to Guerra. Retrospective data can have high variability and inconsistent capture of key data, and it also suffers from recall bias, wherein patients with an illness are more likely to recall events of their childhood than those without an illness because they naturally search their memories for a potential cause, according to Guerra.

There are other respiratory cohorts including: COPDGene, which is searching for genetic clues to explain why some smokers develop COPD and others don’t; the NIH CARDIA study, which since 1983 has tracked development of respiratory, cardiovascular, and other conditions during aging; and the American Lung Association BELung cohort, which aims to follow lung health and predictors of lung disease in 4000 individuals between the ages of 25 years and 35 years.

TCRS is also one of the few such cohorts that begin at birth. Others that start tracking participants at school age “are missing that critical part of first 6 years of life that is targeted very well in TCRS,” said Guerra.

Although COPD has long been thought to be a disease that develops during adult life and a result of smoking, there is increasing evidence that the roots of the disease extend to early childhood, according to Guerra. The TCRS study opens a window into that period. “We can now see the subgroup of these participants that develop COPD and look at their data very early in life, even in utero during pregnancy, and connect those risk factors to the onset of the disease,” said Guerra.

As many as half of COPD patients may have low lung function even at a young age. These “low flyers,” as Guerra calls them, experience natural lung function decline with aging, but their low baseline lung function causes them to experience COPD even if they don’t smoke. “Something happened in childhood and their lungs didn’t grow the way they were supposed to,” said Guerra.

Evidence from adult TCRS participants with low spirometry values, referred to as spirometric restriction, lends some insight into the problem. Some individuals with the condition have small lungs, and spirometric restriction is linked to cardiovascular and metabolic problems as well as heightened risk for mortality, said Guerra. His group examined the background of participants in TCRS with spirometric restriction and found strong associations with nutritional problems of the individual’s mother during pregnancy, being born small for gestational age, and being underweight in early life. “All factors, if you think about it, that point towards poor growth and malnutrition,” said Guerra.

To confirm the finding, Guerra’s group partnered with the Swedish BAMSE and UK MAAS cohorts and found the same pattern.

Risk Factors Revealed

TCRS participants also provide blood samples at every visit, which gives researchers the opportunity to look for biomarkers that might predict onset or progression of disease. For example, the protein CC16 is a possible biomarker, and a TCRS study supported that. “People that have low levels of this protein not only experience a steeper decline of their lung function when they are adults but they have an impaired lung function growth when they are children, implying that this protein might play a role in [the trajectory of COPD],” said Guerra.

That work was performed in collaboration with Erik Melén, MD, PhD, who is the principal investigator of the BAMSE cohort and a professor of pediatrics at Karolinska Institutet, Stockholm, Sweden. The BAMSE study is examining thousands of biomarkers, as well as exposure to air pollution and early life infections, along with genetic studies. The group has demonstrated that a reduction in air pollution can lead to improvement in lung function and lower risk for asthma 10 years later. Melén credits TCRS as a key inspiration for creating the BAMSE cohort.

An August 12 publication in eClinicalMedicine found that genetic risk factors for COPD were strongly associated with low lung function across age groups in BAMSE and other European cohorts, and this finding was independent of smoking habits, sex, or asthma diagnosis. The results suggest that preventive measures for COPD should extend into early childhood, according to the authors.

The finding helps illustrate the value of such cohorts. “We can tie the ends together. We can rewrite our medical textbooks to say that COPD is not just a self-inflicted smoker’s disease. The other side is poor lung development, and there are numerous factors that lead to impaired lung function and growth, and that itself puts you at risk. [It helps provide an] understanding of disease origins, and also the possibility of monitoring, treating, and preventing disease,” said Melén.

Analyses across cohorts can be particularly powerful. “The beauty of this collaborative network is that having multiple cohorts to validate your results provides robustness to your findings because we can see spurious associations sometimes in our single cohort. We can validate those findings by replicating them in additional cohorts, which gives us confidence that this is a robust association that we’re looking at,” said Guerra.

That’s because, like retrospective studies, cohorts can be vulnerable to bias. “Sometimes it’s difficult to differentiate an association from something that’s causal because sometimes things sort of travel together like cigarette smoking is also more common in people with binge drinking,” said David Mannino, MD, who is a professor at the University of Kentucky, Lexington, Kentucky. He wrote a review of COPD epidemiology studies.

Nevertheless, cohorts have changed medicine, according to Mannino. “When you look at the modern world of medical science, a lot of what we have learned has come from these longitudinal studies,” said Mannino. For example, the British Doctors Study, started in the 1950s, was instrumental in demonstrating a causal link between cigarette smoking and lung cancer and all-cause mortality. “It also showed the benefits of stopping smoking,” he said.

“The benefit of longitudinal studies is that it shows you what happens to a population over time, whereas in cross-sectional studies, it’s tough to separate cause from effect,” said Mannino, who spent much of his career working at Centers for Disease Control on asthma-related issues.

He continues to champion such studies when serving on funding review boards, emphasizing their long-term value. He participated in the review of the Austrian LEAD study, which follows a cohort of individuals aged 6 years through 80 years, with an emphasis on normal and pathological lung growth and development; genetic, environmental, and socioeconomic risk factors tied to abnormal lung function; and nonpulmonary consequences of abnormal lung function like cardiovascular disease, metabolic disorders, depression, and anxiety. “I lobbied hard as a reviewer that this is something that absolutely has to be studied. The challenge with these studies is that they never bear fruit for years. It’s something that you’re investing in,” said Mannino.

Guerra is a co-founder of Aspiro Therapeutics, which is developing novel CC16-based therapeutics.Melén had no financial disclosures. Mannino consults for GlaxoSmithKline, AstraZeneca, Regeneron, Genentech, Up-to-Date, and the COPD Foundation. He is also an expert witness on behalf of people suing the Tobacco and vaping industries.