Metabolic Markers and Chronic Obstructive Pulmonary Disease

COPD, lung x ray
COPD, lung x ray
Certain metabolic signatures are found in the blood of people with lung disease. Researchers sought to better identify plasma metabolites associated with chronic obstructive pulmonary disease.

A network analysis found the dysregulation of systemic metabolism has a modest genetic component related with chronic obstructive pulmonary disease (COPD) phenotypes. These findings were published in Network and Systems Medicine.

The National Institutes of Health (NIH)-sponsored multicenter Genetic Epidemiology of COPD (COPDGene) trial was conducted in 2 phases. In phase 1, 10,198 White and Black individuals with ³10 pack-year smoking history and 465 age- and gender-matched healthy nonsmokers enrolled as controls were recruited between 2008 and 2011. In phase 2 (between 2013 and 2017) 5697 of participants had an in-person, 5-year follow-up. Ultimately, 995 metabolites were assessed in frozen plasma samples from 957 of the participants and related using a network analysis to 5 COPD phenotypes (chronic bronchitis, exacerbation frequency, percent emphysema, postbronchodilator forced expiratory volume at one second [FEV1]/forced vital capacity (FVC), and FEV1 percent predicted [FEV1pp]). Results were compared with two independent cohorts from the Subpopulations and Intermediate Outcome Measures in COPD study (SPIROMICS; Metabolon: n=445; SPIROMICS; UC: n=76).

The 957 study participants were aged mean 68.3 (standard deviation [SD], 8.4) years, 51.2% were men, BMI was 29.1 (SD, 6.2) kg/m2, 21.3% were current smokers with 46.0 (SD, 24.9) smoking pack-years. Individuals with preserved ratio impaired spirometry (8.9%), COPD (48.9%), and controls (40.8%) differed by age, gender, BMI, smoking pack-years, and percent emphysema (all P £.001).

Compared with the discovery cohort, the replication cohorts were younger (P <.001), they had higher FEV1pp (P =.0015), and lower percent emphysema (P =.038).

Of the 995 metabolites assessed, 40.2% were associated with age (androgenic steroids, acylcarnitines, and dicarboxylates), 32.1% with gender (sphingomyelin, androgenic steroids, and phosphatidylcholines), 37.2% with ethnicity (xanthines and dicarboxylates), 25.1% with BMI (diacylglycerols and branched chain amino acids), and 12.9% with smoking status. Additionally, 14.8% of metabolites were associated with ³1 of the 5 COPD phenotype(s). For phenotypes characterized by airflow obstruction, 145 metabolites from 55 subpathways were identified.

The genetic contributions to metabolites found 4281 metabolite quantitative trait loci (met-QTL) single-nucleotide polymorphisms associated with 10.95% of the metabolites tested. The strongest predictor was a missense variant on chromosome 10 which explained 50.48% of the variance in N6-methyllysine, however, the clinical variables only accounted for 0.64% of the variance in this metabolite.

With the replication cohort, a total of 72 (66%) of the met-QTL associations were validated. These findings indicated that genetic and metabolite abundance were associated–but not necessarily specific–to COPD phenotypes.

To integrate these data, the investigators created a network of co-abundant metabolites in order to identify associations with disease phenotypes. They identified 26 significant modules associated with COPD, 6 of which were associated with airflow obstruction phenotypes. Between 5%-18.9% of the variance of the individual metabolites was accounted for by genetic components.

This study may have been biased by assessing metabolites using serum blood samples rather than bronchial lavage fluid, which the investigators acknowledged may have better represented COPD phenotypes.

“While we found nongenetic factors to explain more variance in COPD-associated metabolites than genetic, further work is needed, potentially integrating the metabolome with other omics data types (eg, epigenomics and proteomics), to elucidate and characterize dysregulated pathways in COPD pathogenesis.

Reference

Gillenwater LA, Pratte KA, Hobbs BD, et al. Plasma metabolomic signatures of chronic obstructive pulmonary disease and the impact of genetic von phenotype-driven modules. Net Sys Med. 2020;3.1:159-181. doi:10.1089/nsm.2020.0009