Genetics of asthma

INTRODUCTION — Asthma is a condition that arises from complex interactions between multiple genetic and environmental influences. Studies of twins and of families of asthmatic individuals (predating current genomics technology) demonstrate a range of heritability of asthma from 25 to 80 percent [1]. There are clearly components of the asthma phenotype that appear strongly heritable, although these inherited components do not follow the simple Mendelian pattern that is seen in monogenic (or single gene) disorders such as cystic fibrosis. (See "Basic genetics concepts: DNA regulation and gene expression", section on 'Modes of inheritance'.)

Genetic testing in patients with asthma should be used only to exclude monogenic obstructive lung diseases that can be misdiagnosed as asthma, including:

●Cystic fibrosis and primary ciliary dyskinesis, in patients with recurrent bronchitis or chronic productive cough from bronchiectasis

●Alpha-1 antitrypsin deficiency, in patients with a diagnosis of severe asthma, irreversible airflow obstruction, and reduced serum alpha-1 antitrypsin levels

In contrast, testing for asthma susceptibility variants is not recommended because asthma pathogenesis is determined by many interacting genomic and environmental factors, most of which have not yet been identified. Nonetheless, studies of asthma genetics are helping to improve our understanding of the pathogenesis of asthma and identify new therapies.

The genetics of asthma risk and response to asthma drugs (ie, pharmacogenetics) will be reviewed here. The principles of complex trait genetics and the epidemiology and risk factors for asthma are discussed separately.

●(See "Principles of complex trait genetics".)

●(See "Epidemiology of asthma".)

●(See "Increasing prevalence of asthma and allergic rhinitis and the role of environmental factors".)

●(See "Risk factors for asthma".)

CHALLENGES IN STUDYING ASTHMA GENETICS — Human and animal data indicate that the syndrome of asthma is likely transmitted by multiple genes. Multiple genes act together in any given individual to give rise to the asthma phenotype (oligogenic/polygenic inheritance), but different genes result in the same phenotype in different individuals (locus heterogeneity) [2].

Some genes may influence the development of asthma, while others modify asthma severity or the patient's response to therapy. Multiple variants in genes associated with asthma are strongly associated with autoimmune disease and also overlap with inflammatory diseases including cardiovascular diseases and cancer (called pleiotropy) [3-5]. Finally, interactions between genetic factors and environmental influences provide another layer of complexity.

Exploration of the genetics of asthma is hampered by the absence of a "gold standard" diagnostic test for asthma, few if any agreed upon biomarkers, the inconsistent application of a clinical diagnosis, and the heterogeneity of asthma disease expression. To circumvent these issues, investigators have studied the distribution of asthma-related traits, including [6]:

●Bronchial hyperresponsiveness (BHR)

●Measures of atopy (eg, total serum immunoglobulin E ([IgE]) levels or skin test reactivity)

●Patient self-report of asthma diagnosis

●Physician-report or medical record indicating asthma diagnosis

●Objective evaluation (symptoms plus positive bronchodilator response or BHR)

Additional information on the evaluation and diagnosis of asthma may be found elsewhere. (See "Asthma in adolescents and adults: Evaluation and diagnosis".)

GENETIC TECHNIQUES — Four main strategies are being employed to identify the genetic factors that predispose to the development of asthma:

●Case-control or family-based candidate or genome-wide association studies

●Next-generation sequencing (targeted, whole genome) studies

●Multi-omics studies

●Animal models of asthma traits

Association studies — Association studies determine the relationship between certain disease characteristics and the presence of specific forms (referred to as alleles) of a particular DNA sequence [7]. With the publication of initial efforts in sequencing the human genome [8,9], the opportunity to genotype markers directly in genes of interest was greatly expanded, facilitating genetic association studies.

Evolution of genomewide association studies (GWAS) for asthma — Early association studies focused on candidate genes, some of which (eg, chitinase 3-like 1 [CHI3L1], human leukocyte antigen [HLA-D]) had been identified by prior genomewide linkage studies [6,10-13]. The development in the early 2000s of dense DNA microarrays consisting of hundreds of thousands of single nucleotide polymorphisms (SNPs) enabling large-scale genome-wide association studies (GWAS), marked a turning point in asthma genetics [14,15]. (See "Genetic association and GWAS studies: Principles and applications".)

The first asthma GWAS was published in 2007, followed soon after by dozens more [16-25]. These early studies yielded only a handful of loci as they were individually insufficiently powered to identify variants of modest effect sizes. To address this, researchers formed consortia and performed "meta-analyses," a method of pooling GWAS results to improve the power to detect true associations and test the generalizability of findings across study populations [26]. One of the largest GWAS meta-analysis for asthma confirmed prior discoveries, identified novel associations, and suggested that variants discovered so far only explain a small proportion of the genetic risk for asthma [5]. Additional (larger) GWAS meta-analyses identified variants uniquely associated with childhood onset asthma versus adult onset asthma and asthma in women versus men [27-29].

Major findings of GWAS studies for asthma — Accumulated GWAS studies over the past two decades have found associations between asthma and many genetic loci. In the sections below, we summarize the strongest and some of the most replicated genetic signals from GWAS for asthma susceptibility.

●ORMDL3/GSDMB (17q12-q21) – The ORM1-like protein 3 and gasdermin protein locus (ORMDL3/GSDMB) on chromosome 17q12-q21 has reproducibly demonstrated one of the strongest associations for asthma in those with European ancestry [5,30-32]. The locus was identified in the first asthma GWAS in a European sample [30], with increasing evidence of the strength of the association in later studies [5,21,31]. An important observation has been that the SNPs significantly associated with asthma in the discovery population of early GWAS were not associated with subjects of African ancestry and African American subjects in several independent studies [33-35], which supports the argument that European-associated (and European-selected) SNPs do not adequately tag common causal variants relevant for non-European individuals. (See 'Ancestral diversity' below.)

Another diverse GWAS focused exclusively on multiple populations with varying degrees of African ancestry showed evidence of ancestry heterogeneity, demonstrating that the strength of genetic association (measured by effect size) increased as the average proportion of European ancestry increased; conversely, the higher the average proportion of African ancestry, the smaller the effect size [36]. In separate confirmatory studies, GWAS of African American individuals alone, the chromosome 17q12-q21 locus was not significantly associated with asthma while in a GWAS of individuals of Puerto Rican heritage (who have relatively higher global European and lower African ancestry), this locus provided the strongest association with asthma [37-39]. Detailed mapping of this locus in African descent populations has demonstrated the importance of several alternative subhaplotypes that confer a higher risk for asthma in African-admixed populations and a variant that regulates GSDMB splicing, which confers an increased risk independent of ancestral background [36,40].

●Thymic stromal lymphopoietin – GWAS meta-analyses have also confirmed an association with the susceptibility locus for thymic stromal lymphopoietin (TSLP), which is an epithelial-cell-derived cytokine that is important in initiating allergic inflammation [5,31]. In one study, a single TSLP SNP (rs1837253) was identified as protective against risk for allergy, asthma, and airway hyperresponsiveness. Airway epithelial cells derived from individuals heterozygous for this SNP demonstrate reduced inducible TSLP secretion [41].

GWAS data re-emphasized the importance of prior genetic and experimental studies identifying TSLP as a gene of interest [42-44], further motivating the development of a biologic agent. In clinical studies, tezepelumab, a human anti-TSLP monoclonal immunoglobulin G2-lambda antibody, reduces allergen-induced bronchoconstriction and exacerbation rates and improves lung function, as described separately. (See "Treatment of severe asthma in adolescents and adults", section on 'Anti-thymic stromal lymphopoietin (tezepelumab)' and "Treatment of severe asthma in adolescents and adults", section on 'Persistently uncontrolled asthma'.)

●IL13-RAD50-IL4 – Additional GWAS studies and metanalysis identified the 5q31 IL13-RAD50-IL4 as a susceptibility locus for the development of asthma [5,45]. Interleukin (IL)-13 promotes immunoglobulin E (IgE) production by B cells, generation of eosinophil chemoattractants, and contractility of airway smooth muscle cells, and is associated with asthma symptom severity [46]. IL-4 helps in the differentiation of uncommitted T cells into Th2 cells, switch of B-lymphocyte immunoglobulin synthesis to IgE production, and selective endothelial cell expression of vascular cell adhesion molecule-1 (VCAM-1) that mediates eosinophil, basophil, and T cell-specific recruitment. Based in part on the encouraging genetic data, monoclonal antibodies targeting these cytokines have been developed. While the anti-IL-13 monoclonal antibodies (ie, lebrikizumab and tralokinumab) have not demonstrated consistent benefit in clinical trials to date, dupilumab, an anti-IL-4 receptor antibody that impacts both IL-4 and IL-13 signaling, has been very effective in the treatment of asthma and other allergic diseases. (See "Investigational agents for asthma", section on 'Anti-IL-13 antibodies' and "Treatment of severe asthma in adolescents and adults", section on 'Anti-lL-4 receptor alpha subunit antibody (dupilumab)'.)

●IL-33 and IL-33 receptor – Several large GWAS metanalyses have identified and replicated associations between asthma and the IL33 and ST2 (IL1R1) genes [5,31,47].

IL-33 is a member of the IL-1 family that induces Th2 cytokines, and the IL-33 receptor ST2 (known as interleukin 1 receptor-like 1 or IL1RL1) exists as a membrane and circulating soluble protein. IL-33, initially described for its role in expulsion of extracellular parasites from the gut [48], is produced by mast cells following IgE-mediated activation [49]. Greater IL33 expression has also been observed in airway smooth muscle cells [50] and airway epithelium of patients with asthma compared with healthy individuals [51].

In an independent GWAS based on eosinophil counts in more than 9000 Icelanders, one group observed that the IL33 and IL1RL1 SNPs associated with asthma in prior studies were linked to atopic asthma, but not nonatopic asthma [47]. This group also identified a rare IL33 loss-of-function variant associated with lower eosinophil counts, less production of the IL-33 protein, reduced binding to the ST2 receptor, and a reduced risk of asthma in those with European ancestry [52]. Biologic agents targeting IL-33 for the treatment of atopic asthma are under development. (See "Investigational agents for asthma", section on 'Anti-IL-33 and anti-IL-33/ST2 monoclonal antibodies'.)

●Additional loci – The Trans-National Asthma Genetic Consortium (TAGC) published a worldwide asthma GWAS meta-analysis that included more than 142,000 individuals from 75 independent GWAS comprising diverse racial/ethnic groups, a doubling of the largest GWAS previously performed [5]. TAGC confirmed nine known loci (IL1RL1, TSLP, IL13-RAD50-IL4, HLA-DRB1/HLA-DQA1, IL33, LRRC32, RORA, SMAD3-SMAD6-AAGAB, 17q12-q21), identified two new associations at known asthma loci, and identified five new loci (NDFIP1-GNDPA1-SPRY4, GPX5-TRIM27, BACH2-GJA10-MAP3K7, STAT6-NAB2-LRP1, and ZNF652-PHB). In the TAGC GWAS, the chromosome 17q12-q21 locus had the strongest association for asthma but had an effect size (odds ratio) of only 1.16. However, when studies limited analyses to more homogeneous subgroups of cases and account for age of onset of disease and exposure, the effect size was increased [53].

Based on these loci, one group has postulated that the following genes are potentially the strongest candidates for asthma: IL1RL1 (ST2), HLA-DQA1, HLA-DQB1, TLR1, IL6R, ZPBP2, GSDMA, and GSDMB [54].

GWAS of asthma-associated traits — GWAS of asthma-associated traits, such as lung function, have also been performed. For example, a meta-analysis on percent predicted forced expiratory volume in one second (ppFEV₁) values as a measure of lung function in asthma was performed on GWAS data from four European American asthma populations: (1) Severe Asthma Research Program (SARP); (2) The Epidemiology and Natural History of Asthma: Outcomes and Treatment Regimens (TENOR) multicenter study; (3) a subset of the Collaborative Studies on the Genetics of Asthma (CSGA); and (4) two subsets from the Asthma Clinical Research Network (ACRN) clinical trials (the Tiotropium Bromide as an Alternative to Increased Inhaled Corticosteroid in Patients Inadequately Controlled on a Lower Dose of Inhaled Corticosteroids [TALC] and Best Adjustment Strategy for Asthma in Long Term [BASALT] trials). Four of the 32 loci associated with ppFEV₁ values were Th1 pathway genes (IL12A, IL12RB1, STAT4, and IRF2), and together they explained 2.9 to 7.8 percent of the variance in this lung function outcome [55].

GWAS for the identification of shared and distinct risk genes for asthma and other allergic traits — An increasingly consistent message from GWAS is that a large number of asthma loci overlap with loci associated with allergy-related phenotypes and immunologically-related diseases such as autoimmune or infection-related diseases [5,56,57]. This phenomenon of variants in one gene affecting more than one disease or phenotype is called “pleiotropy” [58]. Disentangling the genetic basis for eczema, hay fever, and asthma has been difficult because these traits often present sequentially in the life of the patient (ie, the atopic march), occur together, and the development of one allergic phenotype can predict the development of another [59]. Large meta-analyses have been performed to determine to what extent genetic risk factors are shared between asthma and other allergic phenotypes including hay fever, eczema, and food allergy, and which ones are distinct. A GWAS of asthma and collateral allergic diseases (asthma, allergic rhinitis, or eczema) in approximately 360,000 individuals identified 136 independent variants, of which nearly all were shared risk loci across allergic diseases [57].

Successes and limitations of GWAS — Despite the success of asthma GWAS, several factors appear to have limited discovery, including: (1) the clinical heterogeneity of asthma disease expression (most frequently categorized by allergic T2 or non-T2 inflammation), (2) the age of onset of disease, and (3) environmental exposures that interact with genes to drive asthma risk and progression. Integration with other methods may be the key to determining the causal variants and the underlying role they play in asthma pathogenesis.

●Missing heritability – GWAS to date have mostly identified loci with individual small effect sizes, all of which together only account for a small proportion of asthma heritability and burden, a large so-called “missing heritability.” This missing heritability has not grown appreciably smaller with increasingly large GWAS meta-analyses, suggesting possibly diminishing returns to this strategy. For example, The UK Biobank Study has approximately 500,000 phenotyped and genotyped individuals from the United Kingdom with existing genome-wide data to test for replication of GWAS findings. Although limited by a self-reported asthma diagnosis, interrogation of this large resource replicated at least 28 of 31 published SNPs associated with asthma in European individuals from the UK Biobank. However, these 31 SNPs only explain 2.5 percent of the risk of disease [54].

●Biologic variation – Several large GWAS meta-analyses have focused on genetic architectures of asthma specific to age of onset, demonstrating 23 variants uniquely associated with childhood onset asthma versus one variant uniquely associated with adult onset asthma, and 37 shared loci with overall stronger effect sizes in children [27,28]. Sex-specific shared and unique loci for asthma susceptibility were reported in a recent GWAS of the UK BioBank highlighting genetic variation in pathways known to be differentially expressed between the sexes [29]. These GWAS highlight the varying genetic architectures across biologically different groups that must be considered as larger GWAS are performed to enable discovery.

●Determining causal variants – Homing in on the causal variant(s) in asthma risk loci has been challenging. For example, while initial studies implicated the ORMDL3 gene, this locus is within a 300,000 base region that includes a dozen genes of wide-ranging functions, and three of these genes, GSDMB, IKZF3, and ZPBP2, are in tight linkage disequilibrium with ORMDL3, which has complicated identification of the true causal variant(s) driving association in this region [60,61]. (See 'Major findings of GWAS studies for asthma' above.)

One of the most significant findings in the TAGC meta-analysis is confirmation of an observation from other GWAS of complex traits: most significantly associated SNPs are located in noncoding regions of the genome [5]. In TAGC, all top SNPs in the 18 loci were in noncoding sequences except for a single missense variant in IL13. These top TAGC associations were enriched in enhancer marks, especially in immune cells, and are therefore most likely to be involved in gene regulation. Enhancers play an important role in driving cell-type-specific gene expression, but they activate transcription of their target genes at great distances. Integration of GWAS variant data with other omics analyses, especially RNA expression data, may be the key to understanding how these variants regulate gene expression.

•In an "expression quantitative trait locus" (eQTL) study (see 'Multi-omics studies' below), SNPs associated with asthma in a subset of the GABRIEL sample were consistently and strongly associated with transcript levels of ORMDL3 [30]. A separate study identified polymorphisms in a putative promoter region of ORMDL3 that altered transcriptional regulation of the gene and correlated with changes in Th2 cytokine levels [62], shedding light on its potential role in asthma. (See "The adaptive cellular immune response: T cells and cytokines", section on 'Th2'.)

•In a separate review incorporating gene expression analysis, researchers suggested that unlike the "core" 17q locus, which is associated with childhood asthma, the associations at the proximal and distal loci are not specific to early-onset asthma and are mediated by ORMDL3, PGAP3, and GSDMA expression [32]. (See 'Major findings of GWAS studies for asthma' above.)

•An additional eQTL study of cells from the lower respiratory tract demonstrated SNPs associated with asthma susceptibility and altered GSDMB (not ORMDL3) transcript expression, as well as expression of TSLP and IL33 [63].

Next generation sequencing — Next-generation sequencing (NGS), also referred to as massively parallel or high-throughput sequencing, has significantly advanced genomic research [64,65]. With NGS, nearly all of the 3 billion bases of the human genome are sequenced multiple times and much faster – as much as 50,000-fold faster – than conventional sequencing platforms (eg, Sanger sequencing). NGS can be used to sequence the entire genome, or specific parts of the genome, such as the exome (coding genes) [66] or targeted individual or groups of genes [67]. (See "Next-generation DNA sequencing (NGS): Principles and clinical applications".)

Analyzing genetic variation outside of protein-coding regions and variants beyond SNPs is a major step forward in dissecting common, complex diseases, but is a daunting, computationally exhaustive task. Some advantages of sequencing include being able to identify and analyze novel or ancestry-specific rare variants and additional forms of variants that are not SNPs, including variants resulting from the insertion or deletion of one of more nucleotides. Some notable examples of success in identifying underlying genetic mechanisms for asthma using NGS technology are described in the following sections.

Exome sequencing — Exome sequencing, while useful when applied to Mendelian traits in which most causal alleles disrupt protein-coding sequences, has proved of limited value in complex traits. Exceptions include those diseases for which highly penetrant variants (ie, copy number variants) play a key role in disease risk (ie, autism) and in families where multiple individuals are affected by a complex disease [68].

Using exome sequencing technology, investigators have identified a functional asthma variant in the GSDMB gene [69]. However, in a comprehensive exome study of asthma, the EVE Consortium investigated the role of rare (<1 percent frequency) and low-frequency (1 to 5 percent) variants using Illumina's HumanExome BeadChip array and identified several ethnic-specific associations, but concluded that rare coding variation is not likely to explain a significant proportion of asthma heritability [70]. These observations do not exclude the role of rare individual or ancestry-specific variants with strong effects on asthma risk or as a modifier of disease severity [71-75].

Targeted deep resequencing — The development of NGS technology allowed for targeted deep resequencing of candidate gene loci, typically in coding exons and flanking noncoding regions, to identify rare variants potentially contributing to disease. The plausibility of multiple rare variants within known asthma genes as a risk factor for asthma was first demonstrated by a sequencing study performed by the EVE Consortium of nine candidate asthma-associated genes in a subset of European American, African American, and Hispanic asthma cases and controls. As this was prior to NGS, the coding exons and flanking noncoding regions were resequenced using a prior method (Sanger sequencing) [76]. Although the team expected to discover associations between rare coding variants and risk of asthma, the majority of significant associations were for rare noncoding variants flanking the coding exons. Most importantly, multiple rare variants at single loci contributed to disease, with four genes being more strongly associated in African American subjects and one in non-Hispanic White subjects [76].

By refining the phenotype to asthma following severe respiratory syncytial virus (RSV) bronchiolitis in infancy, four novel, potentially functional, nonsynonymous (coding) variants were identified by targeted sequencing in the following genes: ADRB2 (beta-2 adrenergic receptor), FLG (filaggrin), NCAM1 (neural cell adhesion molecule 1), and NOS1 (nitric oxide synthase 1) [77].

Whole genome sequencing — Because the vast majority of asthma risk variants identified by GWAS lie outside of coding regions, the optimal tool is whole genome sequencing (WGS) in large numbers of well-characterized cases and controls. WGS allows for an unbiased analysis of all types of variation (SNPs, structural variants [SVs] including duplications or deletions of bases, inversions, and translocations), variants of common to rare frequencies, and both coding and noncoding changes. Additional challenges and complexities surrounding WGS are covered elsewhere. (See "Next-generation DNA sequencing (NGS): Principles and clinical applications", section on 'Technical considerations' and "Next-generation DNA sequencing (NGS): Principles and clinical applications", section on 'Limitations'.)

In one of the first examples of applying WGS technology to asthma, multiple copy number variants (CNVs), SVs, and rare coding variants were identified with limited validation in a founder population [78]. Subsequently, the Consortium of Asthma in African Populations of the Americas (CAAPA) [79,80] and consortia who are part of the National Heart, Lung, and Blood Institute (NHLBI)-supported Trans-Omics for Precision Medicine (TOPMed) program have begun leveraging WGS technology to identify novel associations with asthma [81]. With the availability of these large research consortia, replication and discovery of asthma-associated genes using WGS is underway (ie, IL33 [52], CRISPLD2 [82]). WGS from TOPMed has identified a higher mitochondrial DNA copy number as a genetic risk factor for asthma, while another TOPMed WGS study identified a novel loci for lung function and bronchodilator response in children with asthma [82-84].

Next-generation genotyping arrays — A major application of NGS has been improved imputation panels and the design of next-generation genotyping arrays [79,85]. GWAS is a powerful tool for potentially identifying associations between a trait and variants in the genome; however, GWAS arrays are designed to capture relatively common variation SNPs (with a minor-allele frequency >0.05) and the majority of commercially available GWAS arrays were designed from European populations (therefore missing much of the common variation in non-European populations). Even with the addition of imputation to fill in the gaps, it is not possible to account for low-frequency (minor-allele frequency between 0.005 and 0.05) or rare variants (<0.005), although this category of variation accounts for the vast majority of human variation [86]. Sequence-based reference datasets cataloging population variation (Thousand Genomes Project [86], CAAPA [79,80], the National Institutes of Health (NIH) NHLBI-sponsored TOPMed program, and the Michigan Imputation Server) have facilitated genotype imputation accounting for increasingly diverse background genetic ancestries [87].

Multi-omics studies — "Omics" refers to a global assessment of a set of molecules, and omics research has been driven by the revolution of technological advances that include array and NGS technology, methodological approaches that support the interrogation of "big data," and the modeling of biologic networks [88].

A molecular paradigm, which takes into account that DNA is transcribed into RNA, RNA is translated to proteins subsequently controlling metabolism, and DNA methylation controls this process at the transcriptional level, has a greater potential for a more complete characterization of the mechanisms of asthma onset and progression than any single "omics" approach.

Other omics technologies, such as transcriptomics (gene expression), proteomics and metabolomics (measures of protein and metabolites, respectively), and epigenomics (measuring the collection of epigenetic marks throughout the genome) have been incorporated into asthma genetics research, with increasing success [89]. Examples of the application of each of these omics approaches in asthma are described below.

Epigenetics — Epigenetics (chemical modifications of DNA that switch parts of the genome on and off without changing the DNA sequence code) is thought to be one of the mechanisms by which the environment interacts with the genome to cause changes in gene expression (ie, heritable changes in gene expression due to noncoding changes to the DNA) [90-92]. As an example, DNA methylation, a reversible covalent modification of DNA in which a methyl group is transferred from S-adenosylmethionine to cytosine residues at cytosine-guanine (CG) dinucleotides by DNA-methyltransferases, is associated with reduced gene expression [93]. (See "Genetics: Glossary of terms" and "Principles of epigenetics".)

Several observations suggest epigenetics may be involved in pathogenesis of asthma [94].

●First, the concordance rate in monozygotic twins of only approximately 50 percent argues in favor of multiple nongenetic factors [95,96].

●Second, complex gene-environment interactions are involved in asthma, which could reflect epigenetic genome modifications. Supportive evidence for this hypothesis comes from studies of in utero exposures:

•Interactions have been observed between maternal smoking during pregnancy and markers, such as interleukin-1 receptor antagonist (IL-1RN), with a resulting significant increase in risk of asthma in offspring [97].

•In utero exposure to rural environments (ie, endotoxin exposure) seems to have protective effects against the development of asthma [98,99].

•Neonatal immune cells were found to harbor nearly 600 differentially methylated regions that could distinguish children who would go on to develop asthma by the age of nine years [100]. The same study found that methylation of SMAD3 was significantly increased in asthmatic children of asthmatic mothers and associated with childhood risk of asthma.

●Third, sex-specific differences in disease timing and severity, such as the higher prevalence of asthma pre-pubertal males and post-pubertal females, suggest potential epigenetic effects [101].

●Fourth, a parent-of-origin effect exists in susceptibility to asthma and elevated total serum IgE [102], with a clear difference in risk to offspring of asthmatic mothers compared with asthmatic fathers; this effect strongly suggests that imprinted genes may be involved [103-107].

●Finally, a number of genetic studies of asthma have reported linkage and associations in clusters of related genes [58], which suggests a higher-order, epigenetic level of gene regulation.

Multiple studies have identified candidate epigenetic loci for asthma and related phenotypes using DNA from peripheral blood [108-110], buccal [111], and upper airway sampling of nasal cells as a proxy for the lower airway [112,113]. Some notable examples of candidate gene and genome-wide methylation studies include:

●Studies examining methylation near the 17q21 susceptibility locus – A study combining GWAS data with methylation and transcriptome data from whole blood and isolated CD4⁺ T cell samples from asthmatics and nonasthmatics showed that local CG methylation mediates some of the functional effects of cis-acting asthma risk variants on ORMDL3 and GSDMB gene expression [114]. Another study of a Swedish birth cohort demonstrated that methylation at this locus was associated with ORMDL3 expression and risk for childhood asthma [115].

●Novel associations from EWAS of peripheral blood cells – One epigenome-wide association study (EWAS) of children from urban neighborhoods based on DNA from peripheral blood mononuclear cells demonstrated lower methylation associated with childhood asthma in 73 CG sites, including sites next to inflammatory pathway genes relevant to asthma (IL13, RUNX3) [116]. Another EWAS of peripheral blood cells identified the gene encoding the IL5 receptor, IL5RA, a highly relevant T2 inflammatory pathway gene [117]. A large European meta-analysis EWAS identified 14 replicated CG sites associated with asthma in peripheral blood cells and more strongly associated in DNA from purified eosinophils, demonstrating the importance of: (1) the epigenetic regulation of T2 inflammation as a driver of asthma risk and (2) considering the appropriate cell type for epigenetic studies [118].

●EWAS from nasal epithelium samples – EWAS of asthma based on DNA from nasal epithelial cells are few, but performed based on evidence that nasal epithelial cells can serve as a proxy for the lower airways [112,113]. An EWAS in Black children from urban neighborhoods (36 asthmatic children and 36 controls) reported CG sites adjacent to genes related to the T2 inflammatory pathway genes, (arachidonate 15-lipoxygenase [ALOX15] and periostin [POSTN]), the extracellular matrix, epigenetic regulation, cell adhesion, and immunity [114]. A larger EWAS based on nasal epithelium from 312 Puerto Rican children with and 171 without atopy resulted in the identification of 8664 atopy-associated methylation loci [119]. The cumulative effects of the top 30 CG sites were shown to be highly predictive of asthma in individuals of Puerto Rican ancestry, but also in children of European and African American genetic backgrounds [119]. This EWAS demonstrates that while DNA variation across genomes differs due to differences in background ancestry, there are strong similarities across all people in how epigenomic modifications and gene regulation occurs in response to environmental factors.

Expression quantitative trait (eQTL) mapping — DNA is transcribed into messenger RNA (mRNA) which plays a role in making proteins. Transcriptomics refers to the study of the collection of all these gene readouts, or transcripts, present in a cell. Gene expression is heritable [120,121]. Expression quantitative trait loci (eQTL) mapping is a process connecting RNA expression levels to genetic variation, and its overlay with GWAS is a powerful tool for identifying networks of genes involved in disease [122-124].

Several studies have integrated findings from asthma GWAS with cataloged genome-wide gene expression data, such as the Genotype-Tissue Expression (GTEx) public resource [125-134]. Many of the asthma eQTL studies to date have focused on modestly relevant target tissue such as immortalized lymphoblastoid cell lines (LCLs) [135-137] and peripheral blood mononuclear cells (PBMCs) [62,138,139]. Notable studies include:

●In one of the first asthma eQTL studies, SNPs associated with asthma in a subset of the GABRIEL consortium sample were consistently and strongly associated with transcript levels of ORMDL3 from immortalized lymphoblastoid cell lines [30].

●A separate eQTL analysis used lung samples from transplant patients to identify variants affecting gene expression in human lung tissue, then integrated their lung eQTLs with GWAS data from GABRIEL to determine one of their strongest eQTLs was a SNP in the chromosome 17q21 region [126].

●Resting PBMCs and upper airway epithelial cells from children participating in the Children’s Respiratory and Environmental Workgroup (CREW) were used to perform an eQTL focused on SNPs associated with childhood-onset asthma in the 17q12-21 genes; SNPs regulating GSDMB expression in airway epithelial cells appeared to contribute to childhood-onset asthma, but SNPs regulating expression of 17q12-21 genes in blood did not [140].

●In transcriptome-wide analyses of respiratory epithelium conducted by two research groups, genotype-dependent GSDMB, but not ORMDL3, expression in the airway mucosa was associated with a pro-inflammatory cell-lytic type 1 immune transcriptome signature involving interferon signaling in three independent asthma cohorts (BRIDGE, ALLIANCE, GALA) [141,142]; an NK signature was also reported in one cohort [141]. These gene signatures were enhanced after respiratory viral infection of ex vivo cultured upper airway epithelial cells with higher GSDMB expression [141,142], as well as in virus-exposed mice expressing human GSDMB in airway epithelium [142]. Mechanistically, GSDMB was shown to bind to (virally produced) double-stranded RNA and stimulate interferon pathways via activation of MAVS (mitochondrial antiviral-signaling protein)-TBK1 (TANK binding kinase 1) transcription factor signaling [142].

Only a small number of cohorts have access to lower respiratory tract samples for genomic studies, as this requires an invasive collection of samples through bronchoscopy. The NHLBI-sponsored SARP has collected sufficient samples of bronchial epithelial cells (BEC) and alveolar lining cells (via investigative bronchoscopy, including bronchoalveolar lavage [BAL]) in patients spanning all severities of asthma. Using these samples, an eQTL study of 34 asthma genes identified by GWAS demonstrated the following: (1) that the TSLP locus was associated with TSLP expression in BAL and BEC samples; (2) that SNPs in the ORMDL3/GSDMB locus associated with GSDMB, but not ORMDL3, transcript expression in BAL and BEC; and (3) that the disease-associated SNP in IL33 associated with IL-33 expression in BEC, but not BAL cells [63].

A major advance in the study of genome-wide transcriptomics has been the development of RNA sequencing, called "RNA-Seq," which uses NGS technology to determine both the presence and quantity of RNA. This has facilitated transcriptome-wide association studies (TWAS), where transcript expression across the genome can be compared. Notable TWAS studies in asthma include:

●A small preliminary TWAS compared transcript expression in nasal epithelial cells and lower airway bronchial epithelium between atopic asthma patients and nonatopic controls [143]. This study found a strong correlation between the transcriptomes of the nasal airway epithelium and the lower airway bronchial epithelium, indicating nasal airway epithelium might be a reasonable proxy for future TWAS. The TWAS also identified several asthma GWAS genes (ORMDL3/GSBMA/IKZF3, IL33, PYHIN1, IL1RL1, IL13) and T2 pathway genes (IL13, POSTN, SERPINB2) that were differentially expressed in the nasal airway epithelium of atopic asthma patients compared to controls [143].

●A larger TWAS for asthma based on nasal epithelial cell RNA-Seq verified the findings in the previously identified asthma GWAS genes while also discovering novel associations relevant to airway mucus pathobiology, including at genes encoding the mucin 5AC (MUC5AC) and FOXA3, a transcriptional driver of mucus metaplasia [144]. A separate GWAS from two United Kingdom cohorts independently demonstrated an SNP in MUC5AC associated with moderate-to-severe asthma; the accompanying eQTL found that the SNP led to increased MUC5AC expression in BEC [145].

Animal models — Several animal species, including Drosophila fruit flies, rats, guinea pigs, dogs, and swine, have been used to study asthma, but the most common animal model is the mouse [146]. Certain strains of mice either naturally manifest or can be manipulated to express certain aspects of the asthma phenotype, including the allergic response and innate airway hyperresponsiveness. Both linkage studies of cross-bred strains of phenotypic extremes and broad cross-strain association studies have been employed in an attempt to map the underlying genetic determinants of these traits. Animal models also serve as a powerful tool for understanding asthma disease mechanisms and the evaluation of both safety and efficacy of novel therapeutics.

Murine models have been especially useful for validation of genetic variants identified in human association studies, including ORMDL3 and other genes in the 17q12-q21 locus [147-152], IL33 [153,154], ST2 [154,155], IL4R [156], IL13 [157], and TSLP [158]. In an early application of integrating a mouse model for asthma (eg, airway hyperreactivity) and asthma GWAS data (eg, the EVE Consortium), the Kv channel interacting protein 4 (KCNIP4) was identified as a novel candidate gene for asthma and subsequently replicated in the GABRIEL consortium [159].

ANCESTRAL DIVERSITY — Studies of asthma genetics have not adequately sampled the diverse human populations afflicted with asthma, although representation of non-European populations in meta-analyses is increasing [5,31]. One review revealed that genetic studies in White populations (ie, of northern and western European ancestry) represented 60 percent of the association studies reported from 1987 to 2005 [10]; conversely, there were only 25 studies (3 percent) of African or African American populations and 41 studies of Hispanic populations. These populations are clearly under-represented despite the fact that they suffer disproportionately from asthma morbidity and mortality [111,160,161].

●Allelic variants associated with asthma in non-European ancestries prior to GWAS – Certain allelic variants in allergic response candidate genes are more common in people of non-European descent. In other cases, the "wild type" allele, rather than the variant, confers risk of the trait. One example is the functional variant T allele at position -260 in the CD14 gene, which has been associated with lower total serum immunoglobulin E (IgE) levels and less severe asthma [162-168]. In contrast, a functional polymorphism leading to lack of expression of Duffy antigen/receptor for chemokines (DARC) appears to increase susceptibility to asthma and atopy among certain populations of African descent [169].

●GWAS studies in non-European ancestries – Replication of genome-wide association studies (GWAS) findings in non-European groups have not always confirmed associations observed on populations of European ancestry, supporting the notion that certain genes (or polymorphisms in those genes) may be unique to different ethnic groups.

For example, as described above, associations in single nucleotide polymorphisms (SNPs) near the ORMDL3 gene reported in the first asthma GWAS have been widely replicated in several ethnically diverse populations. However, the SNPs significantly associated in the discovery population (European) were not significantly associated with asthma in several independent African American populations [33,34,170] until the Trans-National Asthma Genetic Consortium (TAGC) meta-analysis [5]. In general, the chromosome 17q12-q21 locus has not been associated with asthma in GWAS of African American individuals alone, while this locus is strongly associated with asthma in GWAS of those with Puerto Rican ancestry because of the relatively higher global European and lower African ancestry in the Puerto Rican diaspora [37-39]. The weak and inconsistent associations at the 17q12-q21 locus among African ancestry individuals may be attributable to different early-life asthma endotypes, the breakdown of linkage disequilibrium, and difference in 17q allele frequencies in African-derived genomes [32].

In the largest GWAS focused exclusively on populations of African ancestry (Consortium on asthma among African Ancestry Populations, or CAAPA) [36], investigators both verified associations initially found in European populations and identified novel associations that may be specific to asthma risk in those with African ancestry.

Collectively, these studies of the 17q21 locus, along with GWAS that have discovered novel asthma loci in minoritized populations, demonstrate the importance of diversity in genetic studies to enable discoveries applicable to all people. These studies are also evidence that polymorphisms with varying frequencies across different ancestries could contribute to disparities in asthma risk and severity in under-represented minority populations [31,37-39,171].

●Pharmacogenetic variants – As with many asthma GWAS loci, pharmacogenetic variants associated with response to therapy vary considerably in frequency according to ancestry. The most notable example is a common coding variant substituting a glycine for an arginine (Gly¹⁶Arg) in the gene encoding the beta-2 adrenergic receptor (ADRB2) [171]. Asthma patients of African and Asian ancestry are more likely to carry an Arg¹⁶ homozygote genotype (Arg¹⁶Arg), which is associated with a decreased likelihood of beneficial short-acting beta-agonist response than seen in homozygotes for the common allele, Gly¹⁶ (Gly¹⁶Gly) [172]. Sequencing studies of ADRB2 in a multi-ethnic asthma cohort identified low frequency, ancestry-specific coding variants nearly exclusive to the non-Hispanic White population (Thr¹⁶⁴Ile) and minority groups with African ancestry (25 base-pair promoter insertion only in individuals with African or Puerto Rican ancestry), both associated with adverse responsiveness to long-acting beta agonists [71,173].

Lower frequency and rare variants, all of which are derived from a common ancestry, are commonly limited to individuals from a specific racial or ethnic group and can therefore be missed if these groups are excluded from genetic studies [171].

GENE-ENVIRONMENT INTERACTIONS — Concordance rates for atopy among monozygotic twins raised together are only 50 to 60 percent, suggesting that differences in exposure to certain environmental triggers may account for some of the disparity of disease expression [95]. In a study of asthma and allergic rhinitis among 3808 pairs of Australian twins, about 40 percent of the genetic liability for both diseases was due to environmental factors [1]. Elsewhere, it has been demonstrated that genes and the environment contribute approximately equally to asthma and associated traits such as total serum immunoglobulin E (IgE) [174]. The marked increase in asthma prevalence over the past century likewise supports the role of unknown environmental factors.

Added to this complexity is the observation that associations with alleles at candidate genes and interactions between these genes might only be observed among certain subpopulations with nearly identical environmental and genetic backgrounds. As an example, an association between the glycoprotein CD14(-260)C>T variant (chromosome locus 5q23-31) and low total serum IgE was noted in school children living in urban/suburban Tucson, AZ; however, the opposite association was reported in a farming community [162,175]. In other studies, the association of asthma with CD14(-260)C>T was influenced by pet exposure, the dose of endotoxin in the domestic environment, or country living [163,176,177], suggesting that the unusual role that endotoxin exposure plays in asthma may be due to a unique combination of susceptibility genes and the degree of exposure to endotoxin.

Additional evidence supports divergent effects of certain alleles depending on environmental exposures:

●In a study of 983 children, single nucleotide polymorphisms (SNPs) adjacent to ORMDL3 and GSDMB at chromosome locus 17q21 were associated with decreased risk of asthma by age six among children exposed to farm animals, but increased risk of asthma among those with an older sibling [178]. (See 'Association studies' above.)

●Environmental tobacco exposure has been shown to affect the extent to which significant associations are observed between candidate gene variants for both asthma and atopy [176,179-181]. For example, in one study of 372 families with asthma, certain SNPs in the 17q21 region increased the risk of early-onset asthma; this risk was further increased by early life tobacco smoke exposure [182]. Variants in the NAT1 gene are likewise associated with risk of asthma among children exposed to secondhand smoke [183].

●Certain SNPs in the transforming growth factor-beta (TGF-beta) gene have been associated with increased airway responsiveness and increased asthma exacerbation rates in cohorts of Costa Rican and non-Hispanic White children [184]. In this study, high dust mite exposure further increased the likelihood of asthma exacerbations among patients with alleles of two SNPs (rs2241712 and rs1800471).

●In a separate study, dust mite exposure modified the effect of an SNP (rs117902240) in the interleukin (IL)-2 receptor subunit beta (IL2RB) gene on forced expiratory volume in one second (FEV₁) among children with asthma [185]. The SNP was positively associated with FEV₁ in conditions of low dust mite exposure, but negatively associated with FEV₁ in conditions of high dust mite exposure.

PHARMACOGENETICS — Some asthmatic patients do not respond to the most commonly prescribed therapeutics [186], and it appears likely that genetic factors influence an individual patient’s response to a specific medication. Pharmacogenetics is a field of precision medicine research focused on how genetic variation determines interindividual differences in drug response with the goal to optimize drug efficacy and preventing adverse effects.

●Candidate gene studies – Initial pharmacogenetic candidate gene studies identified variants associated with differential response to inhaled beta-2 adrenergic agonists (beta-agonists), including those in the beta-2 adrenergic receptor and nitric oxide biosynthetic pathways (ADRB2, corticotropin releasing hormone receptor-2, arginase-1, S-nitroso-glutathione reductase [187-190]). Although most pharmacogenetic studies focus on common variants due to the issue of small sample sizes with drug-response phenotype data, lower frequency and rare variants, including those in ADRB2, have also been investigated as determinants of therapeutic responsiveness to beta-agonist therapy.

Pharmacogenetic candidate gene loci for leukotriene modifiers include loci from the leukotriene biosynthetic pathway (arachidonate 5-lipoxygenase, leukotriene A4 hydrolase, leukotriene C4 synthase, cysteinyl leukotriene receptor-2) [189]. Candidate loci associated with response to inhaled glucocorticoids have included the corticotropin hormone releasing hormone receptor (CRHR1), the corticosteroid chaperone complex (STIP1), the T bet transcription factor implicated in mice studies (TBX21), the low affinity immunoglobulin E (IgE) receptor (FCER2), and the chromosome 17q21 locus [189,191-193]. An adrenal restrictive HSD3B1 coding variant allele (Asn³⁷⁶ of Thr³⁶⁷Asn) was associated with oral glucocorticoid resistance in the Severe Asthma Research Program (SARP) cohort [194].

●Whole genome pharmacogenetic studies in asthma – Genome-wide association studies (GWAS) have identified novel pharmacogenetic variation associated with responsiveness to inhaled glucocorticoids and beta agonists in patients with asthma [192,195,196]:

•The first pharmacogenetic GWAS in asthma was performed in the Childhood Asthma Management Program (CAMP) trial and reported SNPs in the gene encoding the glucocorticoid-induced transcript 1 (GLCCI1) associated with lung function response to inhaled glucocorticoids [195]. The increase in forced expiratory volume in one second (FEV₁) in the treated subjects who were homozygous for the minor allele was only about one-third of that seen in similarly treated subjects who were homozygous for the common allele (3.2 versus 9.4 percent predicted). A subsequent study of the GLCCI1 locus in adults did not replicate associations with inhaled glucocorticoid response, which may suggest that this locus only significantly associates with inhaled glucocorticoid response in children [197].

•A subsequent GWAS evaluating the adverse effect of adrenal suppression in response to inhaled glucocorticoids identified a common variant within the platelet-derived growth factor gene (minor allele frequency approximately 0.35) that demonstrated a strong effect on adrenal suppression (odds ratio [OR] 5.9; 95% CI 3-11 in two independent pediatric asthma cohorts). [198].

•Peripheral blood DNA-based epigenome-wide association studies (EWAS) have identified novel differentially methylated cytosine-guanine (CG) loci for therapeutic responsiveness to inhaled glucocorticoids, including within the BOLA2 gene (upregulated in the lung during respiratory syncytial virus infection), the IL12B gene (which regulates T2-associated bronchial hyperresponsiveness), and the CORT gene (which regulates inflammatory cytokine production) [199,200].

•The first GWAS for FEV₁ bronchodilator response to inhaled short-acting beta agonists was performed in a cohort of non-Hispanic White individuals with asthma and identified a gene locus shown to regulate beta-2 adrenergic receptor downregulation, SPATS2L [201]. Subsequent GWAS for albuterol bronchodilator response in asthma cohorts identified a locus on a nitric oxide signaling pathway gene (PRKG1) in African American asthmatics, an intergenic region of chromosome 9q21 with a substantially higher allele frequency in regions and people of higher continental African ancestry, and a gene locus previously associated with lung function (DNAH5) in a multi-ethnic asthma cohort of primarily Hispanic/Latino subjects [84,202]. Both GWAS were focused on underrepresented minorities and were unable to replicate or confirm their findings in another independent cohort, illustrating the challenges of genetic studies in populations that are vastly underrepresented in genetic research.

•Admixture mapping is an alternative approach to GWAS which tests for associations between phenotype and local ancestry for an SNP allele (ie, African versus European) rather than an SNP allele itself. This technique is used to increase power and facilitate discovery of novel associations in a recently admixed minoritized population. Admixture mapping in the National Heart, Lung, and Blood Institute (NHLBI) AsthmaNet-sponsored Best African American Response to Drug (BARD) trial resulted in the first genome-wide study to discover and independently replicate pharmacogenetic loci for response to inhaled glucocorticoids and long-acting beta-agonist controller medications in minorities of recent African descent [203,204].

●Limitations of pharmacogenetic studies – Despite the expansion of pharmacogenetic studies across asthma drug classes, there remains limited clinical applicability for several reasons: (1) the small numbers of patients in clinical trials limits the power of genome-wide studies, especially in minority populations who experience the most morbidity; (2) the increasing understanding that numerous genes likely influence drug response; (3) the heterogeneity of asthma likely modifies associations for a particular pharmacogenetic locus; and (4) the polygenic nature of asthma susceptibility, severity, and drug response [196].

ON THE HORIZON — The most precise prediction of asthma susceptibility, progression, severity, and drug responsiveness will require the development of multi-omic profiles that consider the cumulative effects of many genes, ancestral diversity, and dynamic environmental and drug exposures. Genome-wide association studies (GWAS) findings combined with whole-transcriptome and other "omics" will be used to inform drug repurposing, the development of novel therapies, and the identification of appropriate responder subgroups for specific asthma therapies. As one example of this approach, an integrative quantitative trait locus (QTL) mapping and pathway enrichment study was used to examine data from a GWAS of zileuton response plus mRNA expression profiles plus leukotriene production data to identify genes and pathways associated with differential responsiveness to zileuton [205]. Similarly, another study evaluated six to eight week longitudinal changes in the whole-transcriptomes of sputum airway cells in response to a systemic corticosteroid bursts in order to identify pharmacogenomic endotypes predictive of lung function responsiveness [206].

Polygenic risk scores (PRSs) are a summation of tens of thousands to hundreds of thousands of genetic variants in an individual weighted by their disease-specific effect sizes (typically derived from previously completed GWAS but evaluated or confirmed independently). The premise for PRS is that there is greater predictive power when larger number of variants with smaller effects spanning the genome are combined and modelled accounting for cumulative effects. PRSs have demonstrated success in predicting risk of disease for coronary artery disease and breast cancer [207]. Early asthma PRS attempts have demonstrated poor predictive utility even when based on the best asthma GWAS data available [208-211]. As with GWAS, the development of future PRSs that account for more of the "missing heritability" of asthma should be addressed by increasing ancestral diversity, using objectively confirmed asthma whenever possible, decreasing heterogeneity by developing PRS specific to particular subtypes, and integrating different omics, including epigenomics and transcriptomics (resulting in a "multi-omic" PRS).

SUMMARY

●Heritability of asthma – Asthma is a syndrome that is passed down through families in complex patterns. Some components of the asthma phenotype are clearly heritable, although the specific genes and their complex interactions remain under investigation. Genetic testing for risk of asthma is not yet clinically useful. (See 'Introduction' above.)

●Challenges in studying asthma genetics – Genetic research in asthma is complicated by a number of factors, such as different genes in different individuals leading to the same phenotype, multiple genes acting in one individual to produce the given phenotype, and complex interactions between environmental factors, in addition to the problematic lack of a "gold standard" diagnostic test for asthma. Further complexity arises from the influence of asthma pathogenesis genes, asthma severity modifying genes, and genes that modify the response to asthma treatments. (See 'Challenges in studying asthma genetics' above.)

●Genetic techniques – The four main strategies used to identify the genetic factors that predispose to the development of asthma are case-control or family-based candidate gene or genomewide association studies (GWAS), next-generation sequencing (NGS; targeted, whole genome sequencing [WGS]) studies, multi-omics studies, and animal models of asthma traits. (See 'Genetic techniques' above.)

•Association studies – Association studies determine the relationship between certain disease characteristics and the presence of specific alleles of a particular DNA marker. Using this technique, a number of genes, such as ORMDL3/GSDMB, thymic stromal lymphopoietin (TSLP), and interleukin-33 (IL33), have been identified as possibly important. (See 'Association studies' above.)

A further observation from GWAS of asthma is that most of the significantly associated single nucleotide polymorphisms (SNPs) are located in noncoding regions of the genome. (See 'Major findings of GWAS studies for asthma' above.)

•Next generation sequencing – Some advantages of sequencing include being able to identify and analyze novel or ancestry-specific rare variants and additional forms of variants that are not SNPs, including variants resulting from the insertion or deletion of one of more nucleotides.

Exome sequencing, while useful when applied to Mendelian traits in which most causal alleles disrupt protein-coding sequences, appears to be of limited value in complex traits like asthma. (See 'Exome sequencing' above.)

WGS allows for an unbiased analysis of all types of variation (SNPs, structural variants [SVs], variants of common to rare frequencies, and coding and noncoding changes) and is a promising approach to characterizing asthma risk variants identified by GWAS that lie outside of coding regions. (See 'Whole genome sequencing' above.)

•Multi-omics studies – Other omics technologies, such as transcriptomics (gene expression), proteomics and metabolomics (measures of protein and metabolites, respectively), and epigenomics (measuring the collection of epigenetic marks throughout the genome) have been incorporated into asthma genetics research, with increasing success.

The role of epigenetic mechanisms in the pathogenesis of asthma remains poorly understood, but several observations suggest that epigenetics is likely to be involved. (See 'Epigenetics' above.)

There appears to be a strong correlation between the transcriptomes of the nasal airway epithelium and the bronchial epithelium, and some of the top published asthma GWAS genes are differentially expressed between asthmatics and nonasthmatics in the nasal airway epithelium. (See 'Expression quantitative trait (eQTL) mapping' above.)

●Ancestral diversity – Certain groups of people, such as those of African and/or Hispanic ancestry, have been under-represented in asthma genetics research despite the fact that they suffer disproportionately from asthma morbidity and mortality. (See 'Ancestral diversity' above.)

●Gene-environment interactions – Evidence supports divergent effects of certain alleles depending on environmental exposures, complicating genetic association studies. (See 'Gene-environment interactions' above.)

●Pharmacogenetics – Pharmacogenetics is a field of precision medicine research focused on how genetic variation determines inter-individual differences in drug response with the goal to optimize drug efficacy and prevent adverse effects. Pharmacogenetic studies have expanded across multiple asthma drug classes, but the heterogeneity of the disease and polygenic nature of susceptibility responses have so far limited the clinical applicability. (See 'Pharmacogenetics' above.)

●The future of genetic research in asthma – The most precise prediction of asthma susceptibility, progression, severity, and drug responsiveness will require the development of multi-omic profiles that consider the cumulative effects of many genes, ancestral diversity, and dynamic environmental and drug exposures. (See 'On the horizon' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Kathleen C Barnes, PhD, who contributed to earlier versions of this topic review.