Cell-free DNA levels associate with COPD exacerbations and mortality

Abstract

The question addressed by the study: Good biological indicators capable of predicting chronic obstructive pulmonary disease (COPD) phenotypes and clinical trajectories are lacking. Because nuclear and mitochondrial genomes are damaged and released by cigarette smoke exposure, plasma cell-free mitochondrial and nuclear DNA (cf-mtDNA and cf-nDNA) levels could potentially integrate disease physiology and clinical phenotypes in COPD. This study aimed to determine whether plasma cf-mtDNA and cf-nDNA levels are associated with COPD disease severity, exacerbations, and mortality risk.

Materials and methods: We quantified mtDNA and nDNA copy numbers in plasma from participants enrolled in the Evaluation of COPD Longitudinally to Identify Predictive Surrogate Endpoints (ECLIPSE, n = 2,702) study and determined associations with relevant clinical parameters.

Results: Of the 2,128 participants with COPD, 65% were male and the median age was 64 (interquartile range, 59-69) years. During the baseline visit, cf-mtDNA levels positively correlated with future exacerbation rates in subjects with mild/moderate and severe disease (Global Initiative for Obstructive Lung Disease [GOLD] I/II and III, respectively) or with high eosinophil count (≥ 300). cf-nDNA positively associated with an increased mortality risk (hazard ratio, 1.33 [95% confidence interval, 1.01-1.74] per each natural log of cf-nDNA copy number). Additional analysis revealed that individuals with low cf-mtDNA and high cf-nDNA abundance further increased the mortality risk (hazard ratio, 1.62 [95% confidence interval, 1.16-2.25] per each natural log of cf-nDNA copy number).

Answer to the question: Plasma cf-mtDNA and cf-nDNA, when integrated into quantitative clinical measurements, may aid in improving COPD severity and progression assessment.

Fig. 1

cf-mtDNA levels are higher in participants with COPD. Baseline (A) cf-mtDNA and (B) cf-nDNA levels were assessed in participants with COPD (n = 2,128), smokers (n = 331), and non-smokers (n = 243). The results are presented as median, with the box indicating Q1 and Q3 quartiles and the whiskers indicating the range. To format the y-axis, we used the natural logarithmic scale of data and then transformed it back to nominal data for ease of interpretation. Statistical analysis was performed using ANOVA with Bonferroni post hoc pairwise comparison. ***P value < 0.001. cf-mtDNA = cell-free mitochondrial DNA; cf-nDNA = cell-free nuclear DNA; COPD = chronic obstructive pulmonary disease

Fig. 2

cf-mtDNA levels are higher in participants classified as GOLD stage I/II while cf-nDNA levels are higher in participants classified as GOLD stage IV. Baseline (A) cf-mtDNA and (B) cf-nDNA levels were assessed in participants with COPD who were classified as GOLD stage I/II (n = 944), III (n = 893), or IV (n = 290). The results are presented as median, with the box indicating Q1 and Q3 quartiles and the whiskers indicating range. To format the y-axis, we used the natural logarithmic scale of data and then transformed it back to nominal data for ease of interpretation. Statistical analysis was performed using ANOVA with Bonferroni post hoc pairwise comparison. *P value < 0.05, ***P value < 0.001. cf-mtDNA = cell-free mitochondrial DNA; cf-nDNA = cell-free nuclear DNA; COPD = chronic obstructive pulmonary disease; GOLD = Global Initiative for Obstructive Lung Disease

Fig. 3

Higher cf-mtDNA levels at baseline are associated with an increased rate of yearly exacerbations while cf-nDNA levels are associated with an increased risk of mortality in participants with COPD. Baseline (A) cf-mtDNA and (B) cf-nDNA levels were evaluated for their association with the average number of exacerbations per year during the 3-year study period in all participants with COPD (n = 2,106). The results are presented as adjusted mean with 95% confidence intervals. Statistical analysis was performed using gamma regression after adjusting for age, sex, and body mass index. In (A) and (B), we used the natural logarithmic scale of data and then transformed back the x-axis to nominal data for ease of interpretation. Baseline (C) cf-mtDNA and (D) cf-nDNA levels were evaluated for their association with the risk of mortality in all participants with COPD (n = 2,126). Statistical analysis was performed using Cox proportional hazard regression. cf-mtDNA = cell-free mitochondrial DNA; cf-nDNA = cell-free nuclear DNA; COPD = chronic obstructive pulmonary disease; HR = hazard ratio

Fig. 4

Combined measures of low cf-mtDNA and high cf-nDNA levels are associated with increased mortality risk. (A) Samples across the COPD population were binned into quartiles. The frequency of the top quartile of cf-mtDNA and cf-nDNA was determined across GOLD stages. Statistical analysis was performed using test for linear trend. (B) Low and high quartiles of cf-mtDNA and cf-nDNA were combined, demonstrating an increase in the strength of the association with mortality. cf-mtDNA = cell-free mitochondrial DNA; cf-nDNA = cell-free nuclear DNA; GOLD = Global Initiative for Obstructive Lung Disease; HR = hazard ratio