Asthma Risk Among Individuals With Androgen Receptor Deficiency

In studying why men have increased SARS-CoV-2 pneumonia risk, we discovered that androgen receptors (ARs) are expressed in human airways.¹ Androgens may benefit asthma, and asthma prevalence is lower in men than in women.^2,3 Indeed, AR-deficient mice have increased severity of antigen-induced airway inflammation and bronchoconstriction,⁴ and androgens are protective against allergic airway inflammation. We therefore hypothesized that in humans, as in the mice, AR deficiency would be associated with increased asthma risk.

Children and adults with complete androgen insensitivity syndrome (AIS; International Statistical Classification of Diseases, Tenth Revision, Clinical Modification code E34.51) or incomplete AIS (E34.52) have an inherited loss of AR.⁵ To our knowledge, the risk of asthma in human AR deficiency has not been studied.

We analyzed the prevalence of asthma in AIS among 76 200 860 unique patients in IBM Explorys using International Statistical Classification of Diseases, Tenth Revision, Clinical Modification codes for asthma (J45*). Individuals with AIS were compared with the rest of the cohort, and risk was adjusted for age category, obesity, and nicotine addiction using a logistic regression model. The analysis was repeated after excluding individuals with a history of nicotine addiction and those who simultaneously carried the diagnosis of chronic obstructive pulmonary disease (COPD) (J44*). In a validation study, we compared all patients with AIS with randomly selected age-matched control individuals (2 control individuals per patient with AIS) among approximately 18 million unique individuals in the Indiana Network for Patient Care Oracle database at the Regenstrief Institute. Institutional review board approval for deidentified data analysis was provided by the Cleveland Clinic (Explorys analysis) and Indiana University (Regenstrief analysis). Informed consent was not required owing to the use of deidentified data. Additional details can be found in the eMethods of the Supplement.

In Explorys, 310 patients had complete AIS and 670 had incomplete AIS. Asthma prevalence was higher in patients with AIS (both complete and incomplete) than in unaffected individuals (18.4% vs 6.8%; P < .001). The odds ratio (OR) for having asthma was higher in complete than incomplete AIS vs control individuals (OR, 4.01; 95% CI, 3.07-5.24 vs 2.65; 95% CI, 2.16-3.26; Figure). Results were similar after excluding COPD and nicotine dependence for both complete AIS (3.89; 95% CI, 2.76-5.48; n = 220) and incomplete AIS (2.28; 95% CI, 1.76-2.99; n = 530). Asthma risk was also increased in all AIS after adjusting for age, obesity, and nicotine dependence (adjusted OR, 2.01; 95% CI, 1.77-2.29). Younger patients (<24 years; n = 130) also had a high OR for having asthma (3.35; 95% CI, 2.23-5.04; P < .001). However, the OR for those taking testosterone was 1.72 (95% CI, 1.14-2.58; P = .009; n = 130), lower than those not taking testosterone (3.35; 95% CI, 2.97-3.97; n = 1630; P < .001), further supporting a benefit of androgens in human asthma.^3,4 In the validation cohort, there were 65 patients with complete and incomplete AIS and more than 130 age-matched control individuals. The adjusted OR for having asthma was 2.26 (95% CI, 1.2-4.28; Figure).

Androgen insensitivity syndrome is strongly associated with increased asthma risk. This risk is higher in complete than in partial AIS. Neither increased age nor COPD appeared to increase the risk, arguing somewhat against an effect of lifetime environmental confounders. Although there are sex-based differences in asthma risk between children and adults,² both younger and older patients with loss of AR function were at comparable asthma risk in our analysis.

While large population studies are limited because they cannot identify granular details of clinical presentation and of mechanistic considerations, our data suggest that large, electronic medical record–based data sets may be useful for studying genetic deletions in human diseases and in sexually dimorphic diseases in particular. We demonstrate that patients, both young and old, with phenotypic evidence for AR deficiency have increased asthma risk. This information will be important to a variety of pediatric, medical, and surgical practitioners. Further, it suggests the possibility that androgen signaling could be exploited to therapeutic benefit in asthma. However, prospective human mechanistic studies remain to be performed.

Corresponding Author: Benjamin Gaston, MD, Herman B. Wells Center for Pediatric Research, 1044 W Walnut St, R476 Indianapolis, IN 46202 (begaston@iu.edu).

Accepted for Publication: February 2, 2021.

Published Online: April 12, 2021. doi:10.1001/jamapediatrics.2021.0281

Author Contributions: Dr Gaston had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Dr Zein from the Cleveland Clinic was primarily responsible for analysis of the Explorys data, and Dr Marozkina from Indiana University was primarily responsible for analyzing the data from Regenstrief Institute.

Concept and design: Gaston, Newcomb, Sharifi, Zein.

Acquisition, analysis, or interpretation of data: Gaston, Marozkina, Zein.

Drafting of the manuscript: Gaston, Marozkina, Zein.

Critical revision of the manuscript for important intellectual content: Gaston, Newcomb, Sharifi, Zein.

Statistical analysis: Gaston, Marozkina, Zein.

Obtained funding: Gaston.

Administrative, technical, or material support: Gaston, Marozkina, Newcomb.

Supervision: Gaston, Newcomb, Sharifi.

Conflict of Interest Disclosures: Dr Gaston reported grants from the National Heart, Lung, and Blood Institute during the conduct of the study. Dr Newcomb reported grants from the National Insitutes of Health during the conduct of the study. Dr Sharifi reported being a paid consultant for Pfizer and Celgene. Dr Zein reported grants from National Heart, Lung, and Blood Institute during the conduct of the study. No other disclosures were reported.

Funding/Support: This work was funded by grants 1 P01 HL128192 (Drs Gaston and Marozkina) and K08HL133381 (Dr Zein).

Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.