Eosinophilic esophagitis (EoE): Genetics and immunopathogenesis

INTRODUCTION — Eosinophilic esophagitis (EoE) is a chronic, immune/antigen-mediated esophageal disease characterized clinically by symptoms related to esophageal dysfunction and histologically by eosinophil-predominant inflammation [1]. The pathogenesis of EoE is the result of an interplay among genetic, environmental, and host immune system factors.

This topic reviews the genetics and immunopathogenesis underlying EoE. The clinical features, diagnosis, and management of EoE are discussed in greater detail separately. (See "Clinical manifestations and diagnosis of eosinophilic esophagitis (EoE)" and "Allergy testing in eosinophilic esophagitis" and "Treatment of eosinophilic esophagitis (EoE)" and "Dietary management of eosinophilic esophagitis".)

OVERVIEW OF PATHOGENESIS — The pathogenesis of EoE is incompletely understood but involves genetic, environmental, and host immune system factors. Molecular analysis has elucidated that EoE is caused by a primary defect in esophageal epithelial function rather than an eosinophil defect. This is best illustrated by the finding that genetic susceptibility is mediated by genes expressed by the esophageal epithelia rather than eosinophils. The esophagus of EoE patients has an impairment of epithelial cell differentiation and barrier function.

The esophagus is normally devoid of eosinophils, although they are permanent residents in the rest of the gastrointestinal tract beginning during early embryonic development. Thus, the finding of esophageal eosinophils denotes pathology, typically EoE or gastroesophageal reflux disease (GERD) [2,3]. The diagnosis of EoE originally required the presence of esophageal eosinophilia that was resistant to proton pump inhibitor (PPI) therapy. It is now appreciated that a substantial number of patients with esophageal eosinophilia can respond to PPI therapy, an entity that was referred to as PPI-responsive esophageal eosinophilia (PPI-REE) [4]. It is increasingly appreciated that PPI-REE may be nearly indistinguishable from EoE based upon clinical, endoscopic, histologic, and molecular features [4]. As such, the PPI requirement to diagnosis EoE is no longer a requirement, and PPIs are now considered a therapy for EoE. As such, the diagnosis of EoE is solely based upon the requisite level of eosinophils (at least 15 eosinophils per high-powered field) in patients with clinical manifestations [5]. Diagnosing EoE and differentiating it from GERD are reviewed in greater detail separately. (See "Clinical manifestations and diagnosis of eosinophilic esophagitis (EoE)", section on 'Distinction from GERD'.)

Antigenic proteins, typically derived from food and less commonly from inhaled proteins, trigger an adaptive T helper type 2 (Th2) cell-mediated response that produces cytokines, such as interleukin (IL) 5 and IL-13 [6]. IL-13 subsequently triggers resident cells, such as esophageal epithelial cells, to produce a large set of proteins. The most strongly induced gene in this process is eotaxin 3, which in turn recruits eosinophils from the peripheral blood into the tissue [7]. Antigen-driven Th2 cells also produce IL-5 and IL-13. IL-5 is a chief eosinophil growth and activation factor that primes eosinophils to have enhanced responsiveness to eotaxin 3 and prolongs their cellular survival. IL-13 induces other key mediators of EoE including eotaxin 3 and calpain 14 (CAPN14; the gene product of the primary EoE susceptibility locus) [8]. These type 2 cytokines are largely produced by activated effector memory pathogenic CD4⁺ T cells that accumulate in the esophagus [9]. IL-13 also decreases the expression of genes that encode proteins involved in barrier function such as filaggrin and components of esophageal desmosomes, such as desmoglein 1 [10]. Rare genetic loss-of-function variants in other desmosome genes, including those encoding periplakin and desmoplakin, are also associated with EoE [11].

ROLE OF THE IMMUNE SYSTEM AND ENVIRONMENTAL FACTORS — Host immune system mechanisms in EoE appear to fall somewhere in between pure immunoglobulin E (IgE) mediated and delayed T helper type 2 (Th2) responses [3,12-14]. Eosinophils, T cells, and mast cells are elevated in esophageal mucosal biopsies [15], and mast cell degranulation and eosinophil cytolysis and degranulation are common findings in tissue specimens from patients with eosinophilic gastrointestinal disorders (EGIDs) [16-18]. Mast cell levels are more closely linked with clinical symptoms than eosinophil levels [19]. Studies have identified contributory roles for allergens, cytokines, micro-ribonucleic acid (microRNA or miRNA), chemokines, and polarization of Th2 immunity in the disease pathophysiology. In addition, EoE is uniquely characterized by high levels of systemic and esophageal immunoglobulin G4 (IgG4), which include food-specific responses [20-22]. The high levels of IgG4 may play a role in blocking IgE responses and may explain the dissociation between the absence of positive skin tests and effective elimination diets.

Allergens — The association of EoE with allergies suggests that eosinophil recruitment to the esophagus may be an immune response to environmental antigens in genetically predisposed individuals. Several lines of evidence support an allergic etiology. The role of allergies in EoE is briefly reviewed here and is discussed in greater detail separately. (See "Allergy testing in eosinophilic esophagitis", section on 'Evidence for role of allergies in EoE' and "Dietary management of eosinophilic esophagitis".)

Approximately 75 percent of patients with EoE or other EGIDs are atopic [23-32]. The majority of patients have evidence of food allergen and aeroallergen sensitization, as defined by skin prick and/or allergen-specific IgE tests, although only a minority has a history of food anaphylaxis (an IgE-mediated reaction) [2]. EoE can typically be reversed by institution of an allergen-free diet [23,24,33,34], and relapse is common upon food reintroduction [1]. In addition, priming with epicutaneous exposure to antigens can result in esophageal eosinophil recruitment in a mouse model [35]. Patients with EoE also frequently report seasonal variations in their symptoms, with fewer cases being diagnosed in the winter months (when outdoor aeroallergens are relatively low) [36-38]. In addition, increased eosinophil accumulation is seen in the esophagus of patients with EoE and seasonal allergic rhinitis with hypersensitivity to grass pollen [39]. EoE has also been reported in a few patients treated with sublingual pollen immunotherapy [40,41] and can be triggered by oral immunotherapy [42].

Eosinophils and chemokines — EoE typically occurs in the absence of substantial peripheral blood eosinophilia, indicating the potential significance of esophageal-specific mechanisms for regulating eosinophil levels. When blood eosinophilia occurs in EoE, it is typically mild in magnitude (<500 cells/microL of blood). Eotaxin plays a central role in antigen-mediated eosinophil recruitment [6,7,43,44], and the importance of a local eotaxin pathway in this process has been demonstrated in several studies [43-48]. Collectively, these results strongly implicate eotaxin 3 in the pathoetiology of EoE and offer a molecular connection between Th2 inflammation and the development of EoE.

In murine models of EoE, eosinophil recruitment is attenuated in the absence of eotaxin [47], and mice with a genetic ablation of the eotaxin receptor (C-C motif chemokine receptor 3 [CCR3]) are protected from the development of experimental EoE [45]. The later study also reported on a genome-wide microarray expression profile analysis of esophageal tissue that was used to compare gene transcript expression in the esophageal tissue of patients with EoE or chronic esophagitis (typical of gastroesophageal reflux disease [GERD]) and normal individuals. Eotaxin 3 (also called chemokine [C-C motif] ligand 26, CCL26) was the most overexpressed gene in patients with EoE.

Levels of eotaxin 3 correlated with disease severity in a study in which overexpression of eotaxin 3 alone had a predictive value of 89 percent in diagnosing EoE from a single esophageal biopsy [46]. Another study demonstrated that treatment of EoE with topical glucocorticoids downregulated esophageal eotaxin 3 levels (as well as other cytokines implicated in EoE) [48]. A number of studies in experimental systems in rodents have shown that eosinophils have an effector role in mediating the pathologic features of EoE [49].

Th2 immunity and cytokines — Adaptive T cell immunity, driven by T helper type 2 (Th2) cells and involving interleukin (IL) 13, IL-5, and IL-20 family member expression, appears to play a major role in the pathogenesis of EoE [9,15,50-54].

Studies of cytokines in EoE have shown the following:

●IL-5 is expressed by Th2 cells and eosinophils. It is a key mediator in eosinophil activation and appears to mediate eosinophil-induced esophageal remodeling and collagen deposition [49]. IL-5 expression is increased in tissue samples from children with EoE [55]. In a murine model of EoE, eosinophil recruitment was ablated in the absence of IL-5 [47]. Overexpression of IL-5 in mice leads to increased esophageal eosinophils, and development of EoE is blocked in IL-5-deficient mice [56].

●IL-13 is secreted by Th2 cells in particular. One of its effects is to increase eotaxin 3 expression. Elevated levels of IL-13 messenger RNA (mRNA) are seen in esophageal biopsy specimens from patients with EoE [57]. These findings are ablated with glucocorticoid treatment. In addition, IL-13 stimulation induces an EoE-specific esophageal transcriptome, as well as the gene product from the main EoE susceptibility locus, CAPN14 [58,59]. Repeated delivery of specific allergens or IL-13 to the lung of mice, as well transgenic overexpression of IL-13 in the lung of mice, induces experimental EoE [52,60,61], suggesting that esophageal eosinophilic inflammation is mechanistically linked with pulmonary inflammation. Blockade of IL-13 in vivo with human anti-IL-13 antibodies attenuates esophageal eosinophilia and clinical features and improves endoscopic and molecular markers of EoE [62,63]. The key role of IL-13 and the related cytokine IL-4 is best exemplified by the ability of dupilumab to improve symptomatic, endoscopic, histologic, and molecular aspects of EoE [63], resulting in its US Food and Drug Administration (FDA) approval as the first drug for EoE. (See 'Mast cells' below and "Treatment of eosinophilic esophagitis (EoE)", section on 'Dupilumab'.)

●Thymic stromal lymphopoietin (TSLP) promotes Th2 responses, and elevated levels of TSLP are found in the esophagus of patients with EoE [64]. TSLP and its receptor, in concert with transforming growth factor beta 1 (TGF-beta-1), affect tolerance mechanisms in dendritic cells in vitro [65]. This suggests mechanistically that food tolerance mechanisms are disrupted in patients with EoE. A nonsynonymous polymorphism in the gene for the TSLP receptor was associated with EoE in male patients only [66]. TSLP is overproduced in the absence of the anti-protease serine peptidase inhibitor, kazal type 7 (SPINK7), which is deficient in the esophagus of patients with EoE [67]. (See 'Genetic defects' below.)

●Increased expression of fibroblast growth factor 9 (FGF9) and other profibrogenic cytokine genes, such as IL-5 in the subepithelial layer of the esophagus, is seen in patients with EoE, suggesting that these cytokines may have a role in the fibrogenic response [68,69]. Prolonged treatment with swallowed glucocorticoids is associated with a downregulation of profibrogenic cytokine gene expression [69].

●IL-20 subfamily members, IL-19, IL-20, and IL-24, are increased in the esophagus of EoE patients. Their overexpression induces epithelial barrier defects. Experimental knockout of the IL-20 receptor attenuated experimental EoE associated with esophageal eosinophilia, loss of filaggrin, and type 2 cytokine production [70].

Mast cells — The exact role of mast cells in EoE is unclear. Increased numbers of mast cells are seen in esophageal tissue samples from patients with EoE, and degranulation is common [71]. Results from a murine model of EoE suggest that mast cells may have an important role in esophageal remodeling in EoE by promoting muscle cell hyperplasia and hypertrophy [72]. Elevated TGF-beta, produced by eosinophils and mast cells, contributes to esophageal tissue remodeling and smooth muscle dysfunction in patients with EoE, similar to that seen in the airways of patients with asthma [73], further supporting the link between esophageal and pulmonary inflammation. However, cromolyn sodium, a mast cell stabilizer, has shown little benefit in the treatment of EoE [74]. (See 'Th2 immunity and cytokines' above.)

MicroRNA — MicroRNAs (miRNAs) act as regulators of mRNA expression and translation. In one study, esophageal microRNA expression correlated with eosinophil levels [75]. miR-21 and miR-223 showed the greatest upregulation in patients with EoE compared with normal controls. These two microRNAs may play a role in regulation of eosinophilia and polarization of adaptive immunity. miR-375 was the most downregulated. Levels of these microRNAs mostly reverted to normal with glucocorticoid treatment. Patients in remission showed increased expression of miR-675. In the plasma, the most differentially expressed microRNAs were miR-146a, miR-146b, and miR-223.

GENETICS AND FAMILY HISTORY — A genetic predisposition to EoE is supported by evidence of familial clustering [76] and twin studies, which have revealed a 58 percent concordance in monozygotic twins and a 36 percent concordance in dizygotic twins compared with regular fraternal siblings [77]. The risk of EoE for other siblings is 2 to 3 percent. Collectively, twin studies have indicated a strong heritability of EoE, largely accounted for by shared environment in early life.

Early-life exposures modify the risk of EoE, with maternal fever, pre- and postnatal antibiotics, proton pump inhibitor (PPI) therapy, and neonatal intensive care unit admission associated with increased risk of EoE and pet exposure in the first year of life associated with a decreased risk of EoE [78]. Indeed, the esophagus harbors a unique microbiome, and dysbiosis increases IgE and EoE susceptibility [79].

In addition, several genetic variants that may predispose to EoE have been identified, especially at 5q22 (TSLP gene) and 2p23 (CAPN14 gene) [80]. The identified genetic markers for EoE are primarily common genetic variants, although one appears to be specific for EoE. CAPN14 is induced by interleukin (IL) 13 in esophageal epithelial cells and is involved in regulating barrier function, an impaired process germane to the development of EoE [58,59,81]. Very early onset of EoE (first 18 months of life) is particularly enriched in genetic variants in the CAPN14 gene, as well as associated with cesarean birth compared with late-onset EoE [82].

Family history — Over 30 families with multiple affected individuals were reported by 2008 [83]. In a study of 103 children with EoE, a positive family history was observed in 7 percent [76]. This included three sibling pairs and the mother of one of the pairs of siblings. A report from an eosinophilic gastrointestinal disorder (EGID) registry database showed that 16 percent of patients had an immediate family member with a similar disorder [84]. Almost 10 percent of patients with EoE and evidence of esophageal strictures have parents diagnosed with EoE [3]. There is also a large sibling risk ratio, estimated at approximately 50-fold compared with the general population [3]. This translates into 3 percent of patients with EoE having siblings who develop EoE. A number of rare genetic variants are responsible for familial presentations of EoE including predicted damaging variants in the mitochondrial genes encoding oxidoreductases dehydrogenase E1 and transketolase domain containing 1 (DHTKD1) and oxoglutarate dehydrogenase L (OGDHL) [85,86].

Genetic defects — The earliest work demonstrating that genetic factors have a role revealed dysregulated expression of approximately 1 percent of the entire human genome, constituting an EoE genetic signature [45]. Additional dysregulated transcripts, including several long-noncoding RNAs (lcnRNAs) that act as transcriptional regulators, were identified when RNA sequencing on esophageal biopsies was performed [87].

An EoE susceptibility locus was also identified at 5q22, involving the gene encoding TSLP [64]. The genes for TSLP and its receptor are both associated with EoE [65]. In addition, genetic susceptibility has been found at 5p23, encoding for the CAPN14 gene [58]. Calpain 14 is an esophageal-specific enzyme that regulates IL-13 responses in esophageal epithelial cells and probably explains, at least in part, the tissue-specific nature of the disease [59]. Multiple genetic events collectively contribute to EoE, including an interplay of atopy genes with EoE-specific elements [88]. Meta-analyses of genome-wide association studies have identified at least six other susceptibility loci including EMSY transcriptional repressor, BRCA2 interacting (EMSY)/leucine rich repeat containing 32 (LRRC32; 11q13) and BTB domain and CNC homolog 2 (BACH2; 6q15) [89,90].

EoE involves loss of esophageal barrier function, mediated by depressed expression of desmoglein 1 and SPINK7 [18,67], associated with a loss of epithelial cell differentiation [91]. In addition, genetic variants in the filaggrin gene are linked with EoE, highlighting the potential key role of alterations in barrier function in EoE. Finally, genetic variants in transforming growth factor (TGF) receptors and/or genes associated with inherited connective tissue disorders involving hypermobility syndrome are associated with EoE [65]. These genetic findings have provided insight into molecular etiology and disease susceptibility, but they have not elucidated why patients develop the tissue-specific response characteristic of EoE, since genes such as TSLP have been linked with other atopic disorders.

Relatively large genome-wide association studies and meta-analyses have determined that a major susceptibility linkage exists at 2p23 [58]. This region of the chromosome contains the CAPN14 gene, which encodes for calpain 14. This protein is specifically expressed in the esophagus, is markedly upregulated in patients with EoE compared with controls in a manner dependent upon the CAPN14 genotype, and is dramatically induced by IL-13 in esophageal epithelial cells by an epigenetic mechanism. (See 'Th2 immunity and cytokines' above.)

EoE molecular transcriptome — A 96-gene EoE diagnostic panel (EDP) has been developed based upon analysis of esophageal biopsies and is available commercially. The EDP provides deep information concerning the contribution of individual genes to the pathogenesis of EoE, especially on a patient-to-patient basis [92]. This diagnostic panel, which differentiates EoE from control individuals, including those with gastroesophageal reflux disease (GERD), can also differentiate patients with active and inactive disease and identify glucocorticoid exposure, providing substantial clinical value. Combining analysis of EDP with endoscopic, histologic, and clinical features has led to the identification of three different EoE endotypes that have different disease features and possibly unique course and response to therapy [93]. Molecular profiling has also revealed variable disease groups based upon differential expression of type 2 cytokine-related genes [94]. Genetic testing for the EoE molecular transcriptome may be useful in both diagnosis and management of EoE. (See "Clinical manifestations and diagnosis of eosinophilic esophagitis (EoE)", section on 'Subtypes' and "Clinical manifestations and diagnosis of eosinophilic esophagitis (EoE)", section on 'Laboratory tests'.)

SUMMARY

●Overview – Eosinophilic esophagitis (EoE) is a chronic, immune/antigen-mediated esophageal disease characterized clinically by symptoms related to esophageal dysfunction and histologically by eosinophil-predominant inflammation. (See 'Introduction' above.)

●Pathogenesis – The pathogenesis of EoE is incompletely understood but involves an interplay between genetic, environmental, and host immune system factors. Molecular analysis has elucidated that EoE is caused by a primary defect in esophageal epithelial function rather than an eosinophil defect. The esophagus of EoE patients has an impairment of epithelial cell differentiation and barrier function. (See 'Overview of pathogenesis' above.)

●Role of host immune system – Host immune system mechanisms in EoE appear to fall somewhere in between pure immunoglobulin E (IgE) mediated and delayed T helper type 2 (Th2) responses. Studies have identified contributory roles for allergens, cytokines, microRNAs (miRNAs), chemokines, and polarization of Th2 immunity in the disease pathophysiology. The role of type 2 immunity is underscored by the US Food and Drug Administration (FDA) approval of dupilumab (anti-type 2 biological agent) for the treatment of EoE. (See 'Role of the immune system and environmental factors' above.)

●Genetics – A genetic predisposition to EoE is supported by evidence of familial clustering and twin studies. In addition, several genetic defects that may predispose to EoE have been identified, especially at 2p23, encoding for the esophagus specific gene product, calpain 14. Genetic testing for the EoE molecular transcriptome is available and may be useful in diagnosis, disease subgroup identification (endotyping), and management of EoE. (See 'Genetics and family history' above.)