Capsule summary
High circulating IL-6 may define a phenotype of asthma associated with obesity and systemic inflammation. We demonstrate that circulating IL-6 is higher in African-American patients with asthma, and that race-specific thresholds should be considered.
Keywords: Asthma, Interleukin-6, phenotypes, endotypes, biomarker, eosinophils, obesity, race, ethnicity
To the editor:
Asthma is increasingly recognized as a heterogeneous clinical syndrome with diverse underlying molecular mechanisms. While type 2 (T2) inflammation is important in some patients with asthma, many patients with severe asthma lack T2 airway inflammation, do not respond to current T2-high therapies, and have few therapeutic options (1). Obesity in adulthood co-occurring with asthma has been suggested as a distinct phenotype that is associated with type-1 inflammation, is more common in women, and is more likely to be non-atopic (2, 3). Systemic inflammation related to obesity and low-grade metabolic syndrome, due to production of pro- inflammatory mediators such as interleukin-6 (IL-6) by adipocytes and macrophages, is associated with diabetes, atherosclerosis, and dyslipidemia (4). These circulating mediators may also influence airway inflammation. A recent study of two cohorts of patients with mild to severe asthma suggests that a subset of these patients may belong to a new “IL-6 high phenotype” characterized by high circulating concentrations of IL-6 (5). The “IL-6 high” patients in both cohorts, one a single-center cohort and the other derived from the Severe Asthma Research Program (SARP), had a higher body mass index (BMI) and a higher prevalence of diabetes and hypertension, suggesting an association of this phenotype with metabolic disorders. These patients also had worse lung function and a greater number of asthma exacerbations. Importantly, the cohorts had a racial balance that was majority white (62 and 66%) with relatively few African-Americans (11 and 23%). As African-Americans have a greater prevalence of diseases associated with low-grade systemic inflammation (6, 7), the definition of a “high-IL-6” phenotype may differ based on race.
To examine the role of race on the IL-6 phenotype in asthma, we conducted a prospective, observational clinical study that was approved by the Institutional Review Board at the University of Chicago. A description of the clinical protocol and methods is provided in the Online Repository, and data from a subset of these subjects has previously been reported (8, 9). A total of 68 control subjects with no history of pulmonary disease and 124 subjects with asthma defined by the EPR 3 Guidelines on Asthma were studied. The clinical characteristics of the subjects are shown in Table 1. Assays of IL-6 were done using aliquots of ethylenediaminetetraacetic acid-treated plasma and high sensitivity human (2018) IL-6 Quantikine ELISA kits (R&D Systems, Minneapolis, MN, USA; lower and upper limit of detection, 0.16 and 10.0 pg/mL, respectively). When higher concentrations were detected, diluted plasma kept at 4°C overnight was assayed again the following day. To calculate a reference value for plasma IL-6 concentrations in control subjects we log-transformed values to normalize the distribution of values (confirmed by kernel density estimate and quantile-quantile plot) and applied the following equation to determine the upper 95th centile value:
where t(0.95, n-1) is the 95% quantile of a Student’s t-distribution with n-1 degrees of freedom, and μlog and σlog are the mean and standard deviation of the log distribution of values, respectively. Incorporation of t adjusts for the size of the population. A single-tailed test was used as values defined as below the lower 95th centile threshold were not of clinical interest. The reference threshold value for the entire control population was 4.13 pg/ml; the reference value for subjects of self-reported European ancestry (EA) was 3.69 pg/ml and for subjects of self-reported African ancestry (AA) was 4.89 pg/ml. Eleven subjects (9 control, 2 asthmatic) were of ancestry other than EA or AA; these individuals were not included in the ancestry-specific analyses.
Table 1.
Baseline characteristics of the control and asthmatic subjects in this study.
| Parameter | Control (N = 68) | Asthma (N = 124) |
|---|---|---|
| Age, median (IQR) | 31 (21, 48) | 33.5 (25, 48) |
| Women, N (%) | 43 (63%) | 81 (65%) |
| Ancestry (EA : AA : Other), N | 27 : 32 : 9 | 44 : 77 : 2 |
| BMI, median (IQR) | 26.1 (23.5, 31.4) | 28.7 (25.1, 34.6) |
| BMI > 30 kg/m2, N (%) | 19 (28%) | 57 (46%) ¶ |
| History of sinusitis, N (%) | 1 (1%) | 22 (18%) |
| History of GERD, N (%) | 2 (3%) | 22 (18%) |
| Serum IgE, median (IQR) | 39 (17, 157) | 134 (39, 383) * |
| Inhaled corticosteroid use, N (%) | - | 72 (58%) |
| Oral corticosteroid use, N (%) | - | 21 (17%) |
| Blood eosinophils, N / μL, (mean ± SD) | 123 ± 82 | 297 ± 77 * |
| Blood eosinophils > 300 / μL, N (%) | 1 (1%) | 35 (28%) * |
| Blood neutrophils, N / μL, (mean ± SD) | 3718 ± 1649 | 3818 ± 1984 |
| FeNO, ppb, median (IQR) | 15 (12, 22) | 24 (14, 42) * |
| FEV1, % predicted (mean ± SD) | 96.2 ± 12.2 | 79.2 ± 21.0 * |
| FEV1 ≥ 80% predicted N (%) | - | 65 (52%) |
| FEV1 60 – 80% predicted (N (%) | - | 35 (28%) |
| FEV1 ≤ 60% predicted N (%) | - | 24 (19%) |
| Plasma IL-6 in all subjects, pg/ml (mean ± SD) | 1.77 ± 1.26 | 2.69 ± 3.19 † |
| Plasma IL-6 in EA subjects, pg/ml (mean ± SD) | 1.58 ± 1.26 | 2.67 ± 4.20 † |
| Plasma IL-6 in AA subjects, pg/ml (mean ± SD) | 2.15 ± 1.30 § | 2.75 ± 2.52 |
, P < 0.05 for asthma versus control subjects by Fisher’s exact test;
, P < 0.05 for asthma versus control subjects by Mann-Whitney rank sum test;
, P < 0.05 for asthma versus control subjects by Student t test;
, P < 0.05 for EA versus AA subjects by Student t test. Corrections made for multiple tests as required. Abbreviations as in text.
Consistent with the results of the Peters et al. study (5), IL-6 concentrations were higher in patients with asthma (Table 1), and approximately 15% of asthma patients overall had a concentration greater than the total group reference threshold (Figure S1). Patients with asthma and IL-6 concentrations above the threshold had a higher BMI but did not differ with respect to blood eosinophil or neutrophil counts, serum IgE concentration, or FEV1 (Table 2). While the age difference in the high IL-6 group was significant (Table 2), there was no significant relationship between age and plasma IL-6 in patients with asthma (R2 = 0.03, p = 0.07). As expected, a higher BMI correlated with a higher plasma IL-6 concentration in both control and asthmatic subjects (Figure S2). As in the study of Peters, et al. (5), there were no significant correlations between plasma IL-6 concentrations and the number of blood eosinophils, fraction of exhaled nitric oxide (FeNO), or serum IgE in asthmatic subjects (Figure S3).
Table 2.
Characteristics of patients with asthma in the low and high IL-6 groups†
| Parameter (asthmatic subjects only) | Low IL-6 group (N = 105) | High IL-6 group (N = 16) |
|---|---|---|
| Age, median (IQR) | 32 (24,47) | 44 (34.8, 55.8) * |
| Women, N (%) | 64 (61%) | 15 (94%) ¶ |
| Ancestry (EA : AA : Other), N | 36 : 69 : 0 | 8 : 8 : 0 |
| AA subjects in phenotype, N (%) | 69 (90%) | 8 (10%) |
| EA subjects in phenotype, N (%) | 36 (82%) | 8 (18%) |
| BMI, median (IQR) | 28.1 (24.9, 34.0) | 42.7 (27.5, 50.3) * |
| BMI > 30 kg/m2, N (%) | 46 (44%) | 11 (69%) |
| Inhaled corticosteroid use, N (%) | 59 (56%) | 12 (75%) |
| Oral corticosteroid use, N (%) | 16 (15%) | 5 (31%) |
| Blood eosinophils, N / μL, median (IQR) | 200 (110, 320) | 155 (82, 290) |
| Blood neutrophils, N / μL, median (IQR) | 3200 (2490, 4540) | 3690 (2898, 4920) |
| FeNO, ppb, median (IQR) | 25 (14, 44) | 16 (7, 33) |
| Serum IgE, median (IQR) | 141 (41, 505) | 101 (25, 231) |
| FEV1, % predicted (mean ± SD) | 79.4 ± 21.9 | 75.1 ± 15.7 |
Race-specific thresholds as defined in the text are used to group EA and AA subjects into the low and high IL-6 groups.
, P < 0.05 for asthma versus control subjects by Fisher’s exact test;
, P < 0.05 for asthma versus control subjects by Mann-Whitney rank sum test. Abbreviations as in text.
IL-6 concentrations were higher in AA subjects compared to EA subjects in both control and asthmatic groups (Table 1). IL-6 concentrations were significantly higher in asthmatic subjects of EA ancestry versus EA control subjects, a difference not seen in subjects of AA ancestry. Although the proportion of asthma patients with IL-6 values above the threshold was higher in subjects of EA versus AA ancestry, this difference was not significant (Table 2). However, while 8 of 77 AA asthmatic subjects had IL-6 values above the threshold determined for AA subjects, applying the threshold for EA subjects to the AA subjects would have moved 8 AA subjects from the low to the high IL-6 group, doubling the proportion of AA subjects in the IL-6 high phenotype. Likewise, applying the total group threshold would have moved 6 AA subjects to the IL-6 high phenotype.
We demonstrate that control AA subjects without asthma have higher IL-6 concentrations compared to control EA subjects without asthma, suggesting that a higher threshold value is required in AA subjects to define the IL-6-high phenotype. Using race-specific thresholds, the proportion of patients with the IL-6-high phenotype in each ancestral group with a high IL-6 concentration is appreciable, and concentrations are similar in both groups. We specifically note that a IL-6 concentration above the reference value threshold does not necessarily predict a threshold for drug responsiveness or of clinical significance; such thresholds remain to be determined. Our data suggest that future studies examining the IL-6 high phenotype in asthma should account for racial differences.
Respectfully,
Supplementary Material
Acknowledgments
We thank Stephany Contrella, Rebecca Anderson, Jerrica Hill, Leidy Gutierrez, and Cynthia Warnes in the Asthma Clinical Research team for their assistance in patient recruitment and study. We thank Jyotsna Sudi for her technical assistance.
This work was supported by U19 AI095230 from the National Institute for Allergy and Infectious Diseases, by T32 HL007605 from the National Heart, Lung and Blood Institute, by UL1 TR000430 from the National Center for Advancing Translational Sciences to the Institute for Translational Medicine of the University of Chicago.
This work was supported by U19-AI-095230 co-funded by the National Institute of Allergy and Infectious Disease and the Office of Research on Women’s Health, by UL1-TR000430 from the National Center for Advancing Translational Sciences, and by the Institute for Translational Medicine of the University of Chicago.
Conflict of Interest
Dr. White is a principal investigator the PrecISE Asthma Network (NHLBI UG1-HL139125) that is considering asthma therapies based on the work cited in reference #5 of the manuscript. Dr. White does not have a direct interest in the outcomes of that consideration. None of the other authors have a conflict of interest with regard to financial support, consultant arrangements, speakers’ bureau participation, stock or other equity ownership, patent licensing arrangements, support such as financial or materials grants for research, employment, or expert witness testi-mony.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Fahy JV. Type 2 inflammation in asthma--present in most, absent in many. Nat Rev Immunol. 2015;15(1):57–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Moore WC, Meyers DA, Wenzel SE, Teague WG, Li H, Li X, et al. Identification of asthma phenotypes using cluster analysis in the Severe Asthma Research Program. Am J Respir Crit Care Med. 2010;181(4):315–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Muc M, Mota-Pinto A, Padez C. Association between obesity and asthma - epidemiology, pathophysiology and clinical profile. Nutrition research reviews. 2016;29(2):194–201. [DOI] [PubMed] [Google Scholar]
- 4.Lumeng CN, Saltiel AR. Inflammatory links between obesity and metabolic disease. J Clin Invest. 2011;121(6):2111–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Peters MC, McGrath KW, Hawkins GA, Hastie AT, Levy BD, Israel E, et al. Plasma interleukin-6 concentrations, metabolic dysfunction, and asthma severity: a cross-sectional analysis of two cohorts. Lancet Respir Med. 2016;4(7):574–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Stepanikova I, Bateman LB, Oates GR. Systemic Inflammation in Midlife: Race, Socioeconomic Status, and Perceived Discrimination. American journal of preventive medicine. 2017;52(1s1):S63–s76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Moore JX, Chaudhary N, Akinyemiju T. Metabolic Syndrome Prevalence by Race/Ethnicity and Sex in the United States, National Health and Nutrition Examination Survey, 1988–2012. Prev Chronic Dis. 2017;14:E24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.White SR, Nicodemus-Johnson J, Laxman B, Denner DR, Naureckas ET, Hogarth DK, et al. Elevated levels of soluble humanleukocyte antigen-G in the airways are a marker for a low-inflammatory endotype of asthma. J Allergy Clin Immunol. 2017;140(3):857–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Nicodemus-Johnson J, Myers RA, Sakabe NJ, Sobreira DR, Hogarth DK, Naureckas ET, et al. DNA methylation in lung cells is associated with asthma endotypes and genetic risk. JCI insight. 2016;1(20):e90151. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.