Emerging insights into the epidemiology, pathophysiology, diagnostic and therapeutic aspects of eosinophilic oesophagitis and other eosinophilic gastrointestinal diseases

Summary

Background

Eosinophilic gastrointestinal diseases (EGIDs) are chronic, immune-mediated disorders characterized clinically by gastrointestinal symptoms and histologically by a pathologic increase in eosinophil-predominant tissue inflammation in the gastrointestinal tract, in the absence of secondary causes of eosinophilia.

Aims

To highlight emerging insights and research efforts into the epidemiology, pathophysiology, diagnostic and therapeutic aspects of eosinophilic oesophagitis (EoE) and non-EoE EGIDs, and discuss key remaining knowledge gaps

Methods

We selected and reviewed original research, retrospective studies, case series, and randomized controlled trials, and meta-analyses.

Results

Standardized nomenclature classifies EGIDs as EoE, eosinophilic gastritis (EoG), eosinophilic enteritis (EoN), and eosinophilic colitis (EoC). Incidence and prevalence of EoE is rising, emphasizing the need to better understand how environmental risk factors and genetic features interact. Advances in understanding EoE pathophysiology have led to clinical trials of targeted therapy and the recent approval (in the United States) of dupilumab for EoE. Several therapies are under current investigation, which hope to satisfy both histologic and clinical targets. For non-EoE EGIDs, efforts are focused on better defining clinical and histopathologic disease determinants and natural history, as well as establishing new therapeutics.

Conclusions

Unmet needs for research are dramatically different for EoE and non-EoE EGIDs. In EoE, non-invasive diagnostic tests, clinicopathologic models that determine risk of disease progression and therapeutic failure, and novel biologic therapies are emerging. In contrast, in non-EoE EGIDs, epidemiologic trends, diagnostic histopathologic thresholds, and natural history models are still developing for these more rare disorders.

Introduction

Eosinophilic gastrointestinal diseases (EGIDs) are chronic, immune-mediated disorders characterized clinically by gastrointestinal (GI) symptoms – ranging from dysphagia to abdominal pain and diarrhea – and histologically by a pathologic increase in eosinophil-predominant tissue inflammation in specific regions of the GI tract, in the absence of secondary causes of eosinophilia.¹ The most rigorously studied EGID is eosinophilic oesophagitis (EoE).^1,2,3 EoE is a T helper type-2 (Th2) cell mediated disorder of the oesophagus diagnosed with symptoms of oesophageal dysfunction and eosinophil-predominant oesophageal inflammation (≥15 eosinophils per high-powered field [eos/hpf]). EoE was first described as a distinct clinical entity by Attwood and Straumann in the early 1990s.^2–6 Since then, there has been a rapid expansion of literature pertaining to EoE, investigation of EoE pathophysiology, and conduct of clinical trials.^7–9 The pipeline for investigative therapeutics for EoE and other EGID disorders continues to expand.¹⁰

Clinical awareness of non-EoE EGIDs has increased over the past decade, leading to expansive efforts to understand the clinical presentation, epidemiology, pathophysiology and natural history, and effective therapeutics for these disorders.^11–16 In 2022, an international consensus was achieved to standardize EGID nomenclature.¹ The umbrella term “EGID” was preferred to represent all GI tract eosinophilic inflammation in the absence of secondary causes. Non-EoE EGID nomenclature uses an “Eo” abbreviation convention and includes: eosinophilic gastritis (abbreviated EoG), eosinophilic enteritis (EoN), and eosinophilic colitis (EoC). Overall, however, compared to EoE, non-EoE EGIDs are far rarer, which has limited data collection and rapid expansion of knowledge for these diseases.

This review will highlight emerging insights into the epidemiology, pathophysiology, diagnostic and therapeutic aspects of EoE and non-EoE EGIDs. In each section, we will first review EoE, then discuss non-EoE EGIDs, and finally will highlight areas of unmet need in the field.

Epidemiology

Eosinophilic Oesophagitis

Incidence and prevalence of EoE has consistently increased globally in both children and adults, beyond rates which would be explained by increased disease awareness or endoscopic procedures. Two systematic reviews and meta-analyses have characterized the global incidence and prevalence of EoE. The first was published in 2019 and included 27 population-based studies from 2004 to 2019, spanning both children and adults, and demonstrated a pooled EoE incidence of 6.6 cases per 100,000 person-years (95% confidence interval (CI), 3 – 11.7) in children and 7.7 cases per 100,000 person-years (95% CI, 1.8–17.8) in adults over the study period.¹⁷ Pooled EoE prevalence was 34 cases per 100,000 persons (95% CI, 22.3–49.2) among children and 42.2 cases per 100,000 persons (95% CI, 31.1–55) among adults.¹⁷ From the earliest studies, published before the 2007 consensus guidelines,¹⁸ to the most recent guidelines in 2018,¹⁹ the incidence increased over 2-fold and prevalence over 4-fold.¹⁷ The most recent meta-analysis, published in 2023, included 40 population-based studies from 15 countries across 5 continents between 1976 to 2022, including both children and adults.²⁰ The global pooled incidence of EoE was 5.3 cases per 100,000 person-years (95% CI, 4.0-6.6) and prevalence was 40.0 cases per 100,000 persons (95% CI, 31.1-49.0).²⁰ Prevalence increased >800% from the earliest studies between 1976-2001, to the most current studies between 2017-2022.²⁰ A recent study in the United States (US) Veterans Health Administration also demonstrated a >300% increase in annual incident cases from 2009 to 2018.²¹ This substantial rise in incidence rates is outpacing endoscopy and biopsy rates, suggesting that changes in environmental factors are contributing to EoE pathogenesis.²²

Multiple risk factors for the development of EoE have been postulated. Several studies have suggested early-life exposures such as pre-term birth,²³ neonatal intensive care unit (NICU) admission,²³ acid suppression,^24,25 antibiotic use,^25–27 Cesarian delivery,^26–27 and material smoking²⁵ all pose an increased risk for developing EoE. In contrast, having furred pets²⁵ or experiencing Helicobacter pylori infection may decrease the risk of developing EoE. One hypothesis is that these early-life exposures may influence microbiota colonization or constitution, such as in the case of Helicobacter pylori exposure, and affect EoE development.^28–31 Other environmental factors associated with EoE include low population density, arid or cold climates, seasonal variation, certain housing materials, and poor water quality.^32–36

Non-EoE EGIDs

Non-EoE EGIDs are much rarer than EoE and there are limited data on the epidemiology of non-EoE EGIDs, particularly since the standardized nomenclature was established in 2022¹ and diagnostic and histopathologic criteria are still being developed (see the Diagnosis section below).³⁷ Based on claims code data in the US, prevalence of EoG, eosinophilic gastroenteritis (as reflected by the billing codes rather than the new nomenclature), and EoC was 6.3 cases per 100,000 persons, 8.4 cases per 100,000 persons, and 3.3 cases per 100,000 persons, respectively.³⁸ Prevalence was similar in another US population-based study utilizing claims codes: 5.1 cases per 100,000 persons for eosinophilic gastroenteritis and 2.1 per 100,000 persons for EoC.³⁹ Although seemingly rare based on these claims-based studies, a systematic review and meta-analysis of 13,377 patients showed that the prevalence of non-EoE EGIDs was 1.9% (95% CI, 0.58-3.89) among patients with gastrointestinal symptoms.⁴⁰ Further, a prospective study found a high prevalence of gastric and duodenal eosinophilia in highly symptomatic patients undergoing upper endoscopy, but these results have yet to be replicated.⁴¹

Regarding risk factors, non-EoE EGIDs are often associated with atopy (ranging from 40-73%),³⁷ and have demonstrated a slight female predominance,³⁸ which is in distinct contrast to EoE, though the reasons for this are not known. Non-EoE EGIDs present at any age; however eosinophilic gastroenteritis has been demonstrated more commonly in children less than 5 years whereas EoG was more prevalent among older age groups.^38,42 Like EoE, there is evidence that early life exposures including pregnancy complications, NICU admission, and antibiotics in infancy are associated with an increased risk of developing non-EoE EGIDs.⁴³ Multisite inflammation has been seen more commonly in children compared to adults (68% vs 37%; P < 0.001), with the oesophagus being the most common additional site.³⁶ If there is multisite inflammation, then standardized nomenclature includes all sites involved; for instance, eosinophilic oesophagitis and eosinophilic gastritis: EoE and EoG.³⁷

Knowledge Gaps

Specific environmental etiologic factors in EoE are still not well elucidated. There are many potential exposures that have been hypothesized, such as the effects farming practices (e.g. exposure to barn environments in early-life),⁴⁴ genetic modification of foods, mass production and monoculture, packaging, and air and water pollutants which have been understudied. The role of detergents and factors that disrupt epithelial barriers have also been queried.^45,46 For example, a recent study demonstrated that mice exposed to sodium dodecyl sulfate, a common ingredient in household products such as dish soap and toothpaste, increased IL-33 protein expression, basal zone hyperplasia, CD4+ cell infiltration, and oesophageal eosinophilia.⁴⁷ There may be several “pathways” to EoE development which include, but not limited to, early-life exposures that influence the gut microbiome and immunogenicity and chronic environmental exposures that trigger a Th2 mediated immune response. To understand the epidemiology and pathogenesis will require in vitro studies looking at epithelial barrier dysfunction and immune mediated pathways as well as large epidemiologic studies that have access to changing environmental exposure patterns such as air and water quality indices.

Additionally, there is a need to understand how various exposures interact not only with EoE susceptibility genes,⁴⁸ but also with disease progression from an inflammatory to fibrostenotic phenotype. Large, validated cohort studies with access to environmental exposure data and modifiable behaviors are needed advance our understanding of EoE epidemiology.²¹ Another important gap in knowledge pertains to the architecture of the gut microbiota, and how dysbiosis of the microbiome, as well as effects of environmental exposures such as diet, can influence EoE pathogenesis. A broad range of similar studies are needed in non-EoE EGIDs, as epidemiologic and etiologic data are lacking.

Pathophysiology

Eosinophilic Oesophagitis

Epithelial Barrier Dysfunction, Immunopathogenesis, and Tissue Remodeling

Epithelial barrier dysfunction, Th2 inflammation, and tissue remodeling are hallmarks of EoE pathophysiology.^49–52 In susceptible individuals, exposures to exogenous allergens such as foods and/or aeroallergens lead to the production of epithelial-derived cytokine thymic stromal lymphopoietin (TSLP) and interleukin (IL) -33, also known as alarmins.^3,50 Epithelial barrier dysfunction, caused by increased endogenous protease activity and abnormal epithelial junctions, allow translocation of exogenous allergens to dendritic cells. Dendritic cells are influenced by TSLP and IL-33 to upregulate Th2-biased effector T cells and ILC2 cells, a type-2 innate lymphoid cell.^3,51 These cell types both secrete cytokines, in particular IL-4, IL-5 and IL-13 which recruit and activate eosinophils, mast cells, and basophils. Eosinophils and mast cells propagate allergic inflammation through inflammatory mediators (eg. leukotrienes, prostaglandin D2 [PGD2]) and cytokine production, leading to epithelial changes and immune cell activation which further lead to epithelial barrier dysfunction.³ A recent article by Kaymak et al, identified that IL-20 cytokines, which are overexpressed in active EoE, lower the expression of genes involved in oesophageal barrier integrity, including filaggrins. Furthermore, targeting and inhibiting IL-20 subfamily signaling pathways may restore epithelial integrity.⁵³ Tissue remodeling and fibrosis is influenced through the cytokine transforming growth factor beta (TGF-β), fibroblast activation, pro- and anti-fibrotic mediator (eg. TSPAN-12, TSP1) modulation, and epithelial-mesenchymal transition.^3,52 The food-allergic basis of EoE is confirmed by the near universal response to elemental formula treatment.⁵⁴

Genetic Contributions to Pathogenesis

Genetic factors play a role in the development of EoE, but EoE is not a primarily genetic disease. For example, while some genetic predisposition for EoE is supported by familial clustering and twin studies,⁵⁵ there is only a 2.4% concordance in non-twin siblings, suggesting there is a substantial influence of a shared twin environment in the underpinnings of EoE pathogenesis.⁵⁵

There is a complex interplay between genetic and environmental factors that lead to EoE development. Genome-wide association studies (GWAS) have led to the identification of specific gene variants associated with EoE susceptibility. Genetic variants at four loci (5q22 [TSLP], 2p23 [CAPN14], 11q13 [LRRC32/EMSY], and 16p13 [CLEC16A/DEX1]) have been reproduced across multiple studies.^56–61 As described above, TSLP is an epithelium-derived alarmin that induces a Th2-mediated immune response and directly affects allergenic effector cells.⁶² CAPN14 is an intracellular calcium-activated protease specifically expressed in the oesophagus that plays a role in the epithelial barrier dysfunction.⁶³ LRRC32 and EMSY are expressed in oesophageal epithelial cells; LRRC32 is involved in regulator T-cell effects and EMSY is involved in transcriptional regulation.⁶⁴ CLEC16A and DEX1 are also expressed in oesophageal epithelial cells.⁵⁹ Their roles in EoE pathogenesis are not well characterized.⁶¹ Most recently, a GWAS meta-analysis was performed which described variants at these four loci and additionally identified 11 novel genome-wide significant risk loci,⁶¹ including 15q23 (SMAD3) which has been associated with other allergic diseases.⁶⁵ Table 1 highlights genes with purported pathogenic mechanisms in EoE.

Table 1.

Genes with purported pathogenic mechanisms in EoE

Genetic Risk Loci	Gene	Purported Pathologic Mechanism
2p23.1	CAPN14	Involved in repairing IL-13–induced epithelial changes and is implicated in downregulation of DSG1, a protein that is important for epithelial barrier function
	CCL26/ eotaxin-3	Potent eosinophil chemoattractant and activating factor induced by IL-13
	CRLF2	TSLP receptor gene
	FLG	Regulates oesophageal epithelial barrier function
5q22.1	TSLP	Regulates Th2-mediated immune response and directly affects allergenic effector cells
11q13.5	LRRC32	TGF-β binding protein
	EMSY	Involved in transcriptional regulation
12q13	STAT6	Involved in IL-4/IL-13 signaling pathway
	TGFB1	Regulates oesophageal epithelial barrier function and influences Th2 fibrosis
18q12.1	DSG1	Regulates oesophageal epithelial barrier function
	SPINK5	Regulates oesophageal epithelial barrier function

Candidate gene–phenotype association studies have identified additional candidate genes that influence EoE pathogenesis. Genes coding CCL26,^66,67 filaggrin (FLG),⁶⁸ desmoplakin (DSP),⁶⁹ desmoglein 1 (DSG1),⁷⁰ and serine peptidase inhibitor kazal type 5 (SPINK5)⁷¹ influence epithelial barrier dysfunction, whereas STAT3⁷² regulates Th2 response and TGFβ1/SMAD⁷³ contribute to oesophageal fibrosis and remodeling.

Non-EoE EGIDs

The pathogenesis of non-EoE EGIDs is a focus of current investigative efforts, but is less well described than for EoE. Two multi-site studies have evaluated the molecular profiles of EoG and EoC compared to controls. In former study, which evaluated 74 subjects with EoG and 111 controls, a subset of 18 genes, referred to as the EGDP₁₈ score, were identified and found to be 88–95% sensitive and 100% specific in the discovery and validation cohort to identify EoG diagnosis relative to controls. Expression of CCL26, CLC, IL13RA2, BMP3, IL5, CDH26, CCL18, NPY, HPGDS, and SST tracked with tissue eosinophilia and therefore may be important molecular markers in the EoG pathogenesis.⁷⁴ In latter study, evaluating 27 subjects with EoC and 20 controls, 987 differentially expressed genes (constituting the EoC transcriptome) were found between EoC and controls. Colonic eosinophil count correlated with 31% of EoC transcriptome, most notably with CCL11 and CLC (r = 0.78 and 0.77, P < .0001) which may be significant molecular markers in EoC pathogenesis. The EoC transcriptome was notably distinct from inflammatory bowel disease and other EGIDs, distinguishing EoC as a distinct disease entity.⁷⁵ Most recently, gene expression data were reported for eosinophilic duodenitis (EoD),⁷⁶ which suggested that there is a Th2 profile, similar to EoG and EoD, but different from EoC. Overall, these genomic studies demonstrate some overarching similarities (namely a Th2 profile) between EGIDs but also genetic differences, particularly for EoC, and call for further studies focused on unraveling the pathogenesis of each EGID. For example, similar to EoE, a study demonstrated that elemental-exclusive formula leads to improvement in nearly all subjects with EoG +/− EoD, suggesting a food-driven process in most.⁷⁷

Knowledge Gaps

In EoE, GWAS have mostly been limited to pediatric subjects and those of European ancestry.⁷⁸ GWAS performed in adult subjects as well as studies focusing on gene-environment interactions are needed to better characterize EoE pathogenesis and understand the complex interactions between genetic and environmental factors. GWAS in non-EoE EGIDs have yet to be performed and are challenged by the rarity of the diseases which limits power needed to find genome-wide significance. Identifying patients for GWAS in non-EoE EGIDs and comparing results from these studies to known genetic variants for EoE will shed light on the unique pathophysiology of different EGIDs and how therapeutic targets may differ.

Another knowledge gap pertains to the fundamental question of whether eosinophils are inherently good vs bad or important vs non-important in EGID pathogenesis. This is a critical topic as EGID diagnostic criteria relies on absolute eosinophil counts. Moreover, absolute eosinophil counts are pivotal metrics for clinical trials. Yet with recent randomized control trials, such as with the IL-5RA antibody benralizumab, there has been a discordance between histologic resolution of eosinophils and symptomatic resolution or improvement. More research in these areas is needed to determine if other inflammatory cells, such as histological infiltration of lymphocytes or neutrophils, or molecular processes are contributing to EGID pathogenesis. This is further supported by the recently published VOYAGE (NCT04682639) phase 2 randomized, double-blind, placebo-controlled trial. In this 24-week trial, Etrasimod, a selective sphingosine 1-phosphate (S1P) which regulates lymphocyte egress from secondary lymphatic organs, significantly improved peak eosinophil counts, endoscopic features of EoE and overall symptom severity.⁷⁹ Potentially there are several variants or phenotypes of EGID disorders and the landscape of EGID pathophysiology is more complex than we currently understand.⁸⁰

Diagnosis

Eosinophilic Oesophagitis

The diagnosis of EoE is based upon symptoms, endoscopic appearance, and histological findings. Retrospective studies have shown a median delay in diagnosis of 4-10 years for adults which has remained stable in some locations,^81–83 but has improved in others.^84,85 Timing to diagnosis is critical, as longer duration of symptoms translates to increased risk of fibrostenotic disease. A cohort study evaluating 721 patients from the Netherlands with EoE estimated that for every year of symptoms of EoE that were untreated, the risk of stricture increased by 9% (95% CI, 1.05–1.13),⁸⁶ and similar data have been reported in other locations as well.^87,88

International consensus for diagnostic criteria for EoE diagnosis¹⁹ require: (1) symptoms of oesophageal dysfunction (e.g. dysphagia, food impaction, food refusal, heartburn, regurgitation, vomiting, or non-cardiac chest pain); (2) ≥15 eos/hpf (~60 eos/mm²) on oesophageal biopsy; and (3) exclusion of non-EoE disorders that cause or potentially contribute to oesophageal eosinophilia.¹⁹ Conditions associated with oesophageal eosinophilia are noted in Table 2. When performing endoscopy, the validated EoE Endoscopic Reference Score (EREFS) classification should be utilized,^89,90 and it has been shown to correlate well with histopathologic scoring and treatment response.^91–93 A detailed characterization of EREFS is shown in Figure 1, and care should be taken during upper endoscopy to optimize the exam with full insufflation and complete visualization.⁹⁴ Biopsies should be collected from the oesophagus whenever EoE is a clinical possibility, even when normal mucosa is visualized.¹⁹ Because EoE can be patchy, at least 2-4 biopsies from two or more levels in the oesophagus (e.g. the lower, middle or upper oesophagus) should be collected. Notably, studies have demonstrated taking 6 to 9 oesophageal mucosal biopsies increases the sensitivity for EoE to 100%.^95–98 It is also important to target areas where there are endoscopic findings of EoE, if they are present.⁹⁹

Table 2.

Differential diagnosis of gastrointestinal eosinophilia

Disorders

Drug hypersensitivity

Parasitic or fungal infection

Vasculitis (eg. Churg-Strauss syndrome, Polyarteritis nodosa)

Inflammatory bowel disease

Hypereosinophilic syndrome

Hyper-IgE syndrome

Graft-vs-host disease

IgG4-related disease

Pemphigus

Leukemia

Hodgkin disease

Celiac disease

Systemic mastocytosis

Connective tissue disease

Gastrooesophageal reflux disease (oesophagus-specific)

Achalasia (oesophagus-specific)

Figure 1.

EoE Endoscopic Reference Score (EREFS)

EREFS severity assessment features are displayed. Endoscopic examples with the EREFS scores are also provided for reference. The endoscopic EREFS example constitutes a total score of 6.

Recently, a multidisciplinary team of experts iteratively created a clinically usable EoE severity scoring system, denominated “I-SEE,” to guide practitioners in EoE management by standardizing disease components reflecting disease severity beyond eosinophil counts.¹⁰⁰ Initial data show that this provides an accurate severity characterization that shows improvement in overall severity with successful treatment.¹⁰¹

Although endoscopy with biopsy is currently required to complete the EoE diagnostic algorithm, there is a movement towards more non-invasive testing to decrease procedure burden to patients and healthcare costs. One approach is a 1-hour oesophageal string test (EST), where a string attached to a capsule is swallowed and then subsequently removed out of the mouth and analyzed for inflammatory mediators absorbed by the string. The 1-hour EST samples were able to achieve nearly equivalent diagnostic accuracy as tissue biopsy with endoscopy,¹⁰² and have shown promise in treatment monitoring.¹⁰³ Another approach is cytosponge, a swallowed gelatin capsule containing a compressed mesh sponge attached to a string. Using a threshold of ≥15 eosinophils per high-power field to define active disease, cytosponge had a sensitivity of 75%, specificity of 86%, positive-predictive value of 88%, and negative-predictive value of 76% when compared to histology as the gold standard.¹⁰⁴ Unsedated transnasal endoscopy is another diagnostic modality which is effective, results in lower costs, and does not carry the risk of sedation.¹⁰⁵ A number of other non-invasive biomarkers have been assessed, but none are validated or available for clinical use.¹⁰⁶

Non-EoE EGIDs

Prior to 7/2023 no guidelines existed for the evaluation of non-EoE EGIDs. Recently, the European Society for Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) and the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition (NASPGHAN) formed a task force to provide consensus guidelines for childhood non-EoE EGIDs.³⁷ The approach to diagnosis for non-EoE EGIDs parallels that of EoE. Diagnosis for non-EoE EGIDs requires the presence of (1) gastrointestinal symptoms, (2) histologic evidence of eosinophilic infiltration of the gastrointestinal tract, and (3) exclusion of other causes of intestinal eosinophilia.^38,107 Historically, incident cases of non-EoE EGIDs are rare. The low incidence may be reflective of several diagnostic challenges. First, symptoms can vary widely, including early satiety, abdominal pain, bloating, weight loss, nausea, vomiting, diarrhea, among others, and lead to diagnostic delay, with one population-based study reporting a mean delay of 3.6 years.¹⁰⁸ When compared to patients with EoE, those with non-EoE EGIDs reported more frequent symptoms as well as more extra-luminal symptoms such as fatigue or muscle or joint pain.⁴⁶ Aside from widespread symptomatology, other diagnostic challenges include a lack of clinical awareness by clinicians and under recognition of eosinophils on histologic evaluation.^109–111

Potential symptoms of a non-EoE EGID should lead to an endoscopic examination. Endoscopic findings described in non-EoE EGIDs have been mostly non-specific, and none are pathognomonic. A recently published scoring system for disease severity of EoG, termed the eosinophilic gastritis reference system (EG-REFS), includes seven endoscopic features: (1) erosions/ulcers, (2) granularity, (3) raised lesions (nodularity), (4) erythema, (5) friability/bleeding, (6) folds, (7) pyloric stenosis (Table 3).¹¹² Among 98 patients with EoG, the most common EG-REFS features included erythema (72.3%), raised lesions (49.2%), erosions/ulcers (46.2%), and granularity (35.4%), and were most pronounced in the gastric antrum. In patients with active histopathology, 8% had no endoscopic findings.¹¹² An endoscopic reference system has not been established to-date for EoN or EoC. Common endoscopic findings have been described in EoN and include villous dropout or flattening, erythema, friability and ulceration,¹¹³ whereas erythema, friability, congestion and ulceration are commonly seen in EoC.^113-115

Table 3.

Eosinophilic Gastritis Reference System (EG-REFS) Features and Severity Score

Feature	Severity Assessment
Erosions/Ulcerations	Grade 0: None Grade 1: Less than 5 erosions Grade 2: Five or more erosions Grade 3: Shallow/superficial ulceration(s) Grade 4: Deep/excavated ulceration < 25 % of the surface area of specified location Grade 5: Deep/excavated ulceration 25–50 % of the surface area of specified location Grade 6: Deep/excavated ulceration > 50 % of the surface area of specified location
Granularity	Grade 0: None Grade 1: Fine Grade 2: Coarse
Raised Lesions (Nodularity)	Grade 0: None Grade 1: Mild (raised focal nodules with width > height) Grade 2: Severe (raised nodules with height > width)
Erythema	Grade 0: None Grade 1: Mild (pink) Grade 2: Severe (red/hemorrhagic)
Friability/Bleeding	Grade 0: None Grade 1: Mild (contact bleeding) Grade 2: Severe (spontaneous bleeding)
Folds	Grade 0: None Grade 1: Thickened folds
Pyloric Stenosis	Grade 0: None Grade 1: Present (inability to pass diagnostic 8–10 mm upper endoscope)

^*Adapted from the American Journal of Gastroenterology¹⁰³

Eosinophilic-predominant tissue inflammation is a key diagnostic requirement for all EGIDs; however, eosinophils are normal tissue resident cells of the GI tract distal to the oesophagus and can complicate the diagnosis of non-EoE EGIDs. Estimates of normal eosinophil counts in various portions of the GI tract are as follows: stomach (5-10 eos/hpf), duodenum (10-25 eos/hpf), and terminal ileum and cecum (likely >50 eos/hpf), with a decreasing number in the distal colon.^116–119 Histologic thresholds of eosinophil values are not standardized for non-EoE EGIDs in adults. Some studies have suggested EoG should be defined by the presence of ≥30 eos/hpf in ≥5 hpfs,^120,121 and EoD requires ≥30 eos/hpf in ≥3 hpfs.^113,122 EoC suggested thresholds include ≥100 eos/hpf in cecum and ascending colon, ≥85 eos/hpf in transverse and descending colon, ≥65 eos/hpf in rectosigmoid colon.¹²³ In the new pediatric guidelines,³⁷ the recommend thresholds in children are ≥30 eos/hpf in the stomach, ≥50 eos/hpf in the duodenum, ≥60 eos/hpf in the terminal ileum, ≥100 eos/hpf in the cecum and ascending colon, ≥80 eos/hpf in the transverse and descending colon, and ≥60 eos/hpf in the rectum (Table 4).³⁷ Additionally, eosinophil-predominant tissue inflammation can be patchy and involve deeper mucosal layers including the muscularis and/or serosa,¹²⁴ which may lead to misdiagnosis on histology in some cases. It is therefore recommended that multiple biopsies be collected; however, the optimal number of biopsies is not standardized. A recent study evaluating the specificity of biopsy quantity demonstrated that an evaluation of multiple nonoverlapping hpfs in 8 gastric and 4 duodenal biopsies was required to capture 100% of EoG/EoD cases in the study.¹²⁵ It is important to note that neither endoscopic findings nor symptom severity correlated with eosinophil counts, which reinforces the need to collect multiple biopsies from normal tissue if your clinical suspicion is high for an non-EoE EGID. However, in another study, EG-REFS severity was significantly correlated with active histology (P=0.0002).¹¹² It is a reasonable approach to target abnormal tissue for all EGIDs. Another recent study demonstrated a higher diagnostic yield of biopsies for EoG when patients had peripheral eosinophilia and hypoalbuminemia.¹²⁶ As a general practice, biopsies for non-EoE EGIDs should be collected when there is a clinical suspicion for these disease pathologies based on symptomatology or if there are endoscopic abnormalities in the stomach. For example, if an index esophagogastroduodenoscopy is performed for the initial evaluation of dyspepsia, it would be clinically appropriate to collect gastric biopsies to assess for EoG. Often a discussion with the pathology department is required to ensure appropriate assessment of non-EoE EGIDs in collected tissue samples as not all pathology departments may routinely examine eosinophil counts beyond the oesophagus.

Table 4.

Non-EoE EGID Pediatric Consensus Peak Eosinophil Thresholds

Gastrointestinal Site	Peak Eosinophil Threshold (eos/hpf)
Stomach	≥30
Duodenum	≥50
Terminal ileum	≥60
Cecum and ascending colon	≥100
Transverse and descending colon	≥80
Rectum and sigmoid colon	≥60

^*Adapted from the Journal of Pediatric Gastroenterology & Nutrition³⁶

Knowledge Gaps

There are several diagnostic facets which could improve the yield for identifying EGIDs. A critical facet for EoE is optimizing diagnosis at the time of endoscopy, as fibrostenotic features can manifest if diagnosis is delayed, and strictures or narrowing are frequently missed by endoscopists.^127,128 Identifying diagnostic predictors of disease phenotype are also of interest. Functional luminal imaging of the oesophagus has shown promise of determining oesophageal compliance in EoE patients to distinguish inflammatory vs fibrostenoic physiology. Another goal is to be able to use blood-based biomarkers to diagnose EoE. No biomarkers have yet been prospectively validated and none are in routine clinical use at this time. For non-EoE EGIDs, though there are now proposed guidelines in children, a consensus diagnostic approach is needed for adults, and efforts are also needed for less invasive diagnosis, as well as potentially methods of diagnosis that are not solely focused on eosinophil counts. Lastly, more efforts are needed to understand the sub-epithelial inflammatory processes in EGIDs. Sub-epithelial tissue is not always obtained using standard forceps, and therefore has limited the evaluation of sub-epithelial histopathology. A study by Schoepfer et al.¹²⁴ demonstrated eosinophilic infiltration, eosinophil degranulation, fibrosis, and lymphoid follicles represent the key histologic features of EoE in the sub-epithelial layer, and that approximately one-third of patients had higher peak eosinophil counts in the sub-epithelial layer compared to the epithelium. Routine assessment of sub-epithelial eosinophil counts may increase the diagnostic yield for EGIDs, as well as lead to further phenotypic characterization of EGIDs, though the ideal approach to assessing for submucosal and serosal involvement in non-EoE EGIDs remains to be determined.

Therapeutics

Eosinophilic Oesophagitis

The mainstay therapies for EoE have included proton pump inhibitors (PPIs), swallowed/topical corticosteroids, and dietary elimination therapy. Due to a lack of head-to-head trials, all three options are recommended as first-line therapies for the treatment of EoE. However, with the approval of dupilumab and with other biologic medications in the pipeline, there is a shift in focus on positioning therapies in the treatment algorithm.

PPI Therapy

PPI therapy is well tolerated and easily accessible, which is often why this therapy is used upfront.¹⁹ Several mechanisms are purported for PPI therapy in EoE. PPI have an anti-secretory mechanism, increasing the pH refluxate and restoring integrity to the oesophageal mucosa.¹²⁹ PPI also are thought to reduce oesophageal inflammation by inhibiting eotaxin-3-medidated eosinophil migration to the oesophagus,¹³⁰ activating the aryl hydrocarbon receptor which plays a role in Th2 mediated responses,¹³¹ reducing STAT6 and exotoxin-3 expression,¹³² and inhibiting ATP12A,¹³³ the non-gastric P2-type H+, K+-ATPase. A systematic review and meta-analysis of 23 observational studies demonstrated that 41.7% of subjects on PPI achieved histopathologic remission (<15 eos/hpf) compared to 13.3% of a placebo comparison group.¹³⁴ This resulted in a 34% reduction in the risk of failing to achieve histopathologic remission (RR=0.66, 95% confidence interval (CI), 0.61 – 0.72) in those taking PPI compared to placebo, and further resulted in a guideline recommendation of using PPI over no treatment (conditional recommendation with very low quality evidence).¹³⁴ Retrospective data supports the use of once or twice daily PPI use, with many studies using double-dose (equivalent to omeprazole 40mg daily or 20mg twice per day) in the evaluation of PPI-responsive oesophageal eosinophilia.^19,135,136 Current recommendations are to initially use a double-dose similar to erosive oesophagitis for a duration of 8 weeks.¹⁹ For children this is equivalent to 1-2mg/kg of omeprazole daily or equivalent.¹³⁷ Dose reduction of PPI to the lowest effective dose can then be attempted following histologic and clinical remission.¹⁹ Long-term histologic and clinical response rates to PPI therapy (mean follow up 3.6 ± 2.9 years) have been described at 60% and 64%, respectively.^138,139

Topical Corticosteroids

TCs inhibit inflammatory pathways induced by IL-13,¹⁴⁰ reduce eosinophil¹⁴¹ and mast cell responses¹⁴² in the oesophageal epithelium, reduce epithelial apoptosis,¹⁴¹ and decrease oesophageal molecular remodelling.¹⁴¹ Analysis of 8 double-blind placebo-controlled randomized control trials demonstrated that 64.9% of subjects on tCs therapy achieved histologic remission compared to 13.3% of a placebo comparison group.¹⁴⁴ This resulted in a 61% reduction in the risk of failing to achieve histopathologic remission (RR=0.39, 95% CI, 0.26 – 0.58) in those taking tCs compared to placebo, and further resulted in a guideline recommendation of using tCs over no treatment (strong recommendation with moderate quality evidence).¹⁴⁴ Adverse effects did not significantly differ between those taking tCs or placebo. Additionally, tCs should be continued as maintenance therapy as opposed to discontinued after short-term use.¹⁴⁴

Budesonide and fluticasone propionate are presently recommended for topical use in EoE. Several different formulations of budesonide and fluticasone have been studied including oral viscous budesonide (OVB), budesonide oral suspension (BOS), nebulized budesonide, compounded budesonide suspension, orally disintegrating budesonide tablets, fluticasone from a multi-dose inhaler (MDI) or a disk device, compounded fluticasone suspension, and orally disintegrating fluticasone tablets.^143–145 Dosages and instructions for preparation/administration are detailed in Table 5. A recent double-blind, double-dummy trial of patients with a new diagnosis of EoE evaluated the efficacy of OVB compared to fluticasone MDI. Patients were randomized to OVB slurry plus placebo inhaler or swallowed fluticasone from a MDI plus a placebo slurry. Post-treatment eosinophil counts, dysphagia symptom questionnaire, and post-treatment total EREFS were not significantly different between the two treatment groups, suggesting that budesonide and fluticasone may be equally efficacious.¹⁴⁶ Additional formulations are being developed and studied in the drug pipeline (Table 6).

Table 5.

Topical Corticosteroid Dosing and Formulations

Topical Corticosteroid	Adult Dosing	Children Dosing	Formulations and Instruction*
Budesonide	Induction: 2-4mg/kg Maintenance: 2mg/day	Induction: 1-2mg/kg Maintenance: 1mg/day	1. Oral Viscous Budesonide (OVB) – Mix aqueous budesonide with sucralose (dosages and preparation vehicle may vary, but typically the mixture is a ratio of 0.5mg/2mL of aqueous budesonide with 5g of sucralose). Swallow oral budesonide viscous liquid/suspension slowly over 5 to 10 minutes. Available dosing: Pulmicort respules 0.5 to 1mg/2mL; can substitute honey or other thickening agents for sucralose; a similar formulation can be made at a compounding pharmacy without the sucralose 2. Orodispersible Tablets – Budesonide tablet is placed on the tip of the tongue and pressed gently against the hard palate until it completely disintegrates by contact with saliva. The dissolved medication is then continuously swallowed with the saliva over the several minutes. Available dosing: Jorveza 0.5 to 1mg tablet 3. Budesonide Oral Suspension (BOS) – Studied Dosing: Budesonide suspension 2mg/10mL was used twice per day in a phase 3 trial, but is not currently available.
Fluticasone propionate	Induction: 1760 mcg/day Maintenance: 880–1760 mcg/day	Induction: 880–1760 mcg/day Maintenance: 440–880 mcg/day	1. Fluticasone Metered-Dose Inhaler (MDI) – Spray 1-4 puffs of a 220 mcg inhaler into the mouth and swallow. Available dosing: Flovent HFA 220 mcg inhaler; the Diskus can also be used (250 mcg device) and opened to obtain the fluticasone powder inside, which can then be put on the tongue. A compounded fluticasone formulation can also be made at a compounding pharmacy 2. APT-1101 Orally Disintegrating Tablets – Fluticasone propionate tablet is placed on the tip of the tongue and pressed gently against the hard palate until it completely disintegrates by contact with saliva. The dissolved medication is then continuously swallowed with the saliva over the several minutes. Studied dosing: The 3 mg at bedtime (qHS) dose is being studied in a phase 3 trial

^*Eating and drinking should be avoided for 30-60 minutes after ingesting the topical corticosteroid therapy for all formulations.

Table 6.

Therapeutic Development Pipeline for EoE and non-EoE EGIDs

Pharmacotherapy	Mechanism	Route of Administration	Development Status	Clinical Trials Identifier
Eosinophilic Oesophagitis
CALY-002	Interleukin-15 inhibitor	Intravenous	Phase 1	NCT04593251
NDX-3315	Radioactive Imaging Agent	Oral	Phase 1	NCT05757856
ESO-101 (Mometasone furoate)	Glucocorticoid receptor agonist	Oral	Phase 2	NCT04849390
Zemaira	Alpha-1 trypsin inhibitor	Intravenous	Phase 2	NCT05485155
IRL201104	Immunomodulatory Peptide	Intravenous	Phase 2	NCT05084963
Barzolvolimab	Receptor tyrosine kinase KIT inhibitor	Subcutaneous	Phase 2	NCT05774184
Etrasimod	Sphingosine 1-phosphate receptor modulator	Oral	Phase 2	NCT04682639
Mepolizumab	Interleukin-5 inhibitor	Subcutaneous	Phase 2	NCT03656380
AK002 (Lirentelimab)	Siglec-8 inhibitor	Subcutaneous	Phase 3	NCT04322708
CC-93538 (Cendakimab)	Interleukin-13 inhibitor	Subcutaneous	Phase 3	NCT04991935
APT-1011 (fluticasone propionate disintegrating tablet)	Glucocorticoid receptor agonist	Oral	Phase 3	NCT05634746
Tezepelumab	Thymic stromal lymphopoietin inhibitor	Subcutaneous	Phase 3	NCT05583227
Eosinophilic Gastritis
Dupilumab	Interleukin-4 and Interleukin-13 inhibitor	Subcutaneous	Phase 2	NCT03678545

Biologics

Despite the efficacy of PPI and tCS, there are substantial non-response and loss-of-response rates, and these treatments are not targeted to EoE pathogenesis. With a greater understanding of potential pathways to target in EoE, a number of biologics have been tested. The first examined tested in EoE were antibodies against circulating IL-5, such as mepolizumab and reslizumab. In randomized studies, both biologics led to a significant decrease in eosinophil counts compared to placebo, but did not show significant improvements in clinical endpoints compared to placebo.^147–149 Because these studies used non-validated clinical outcome assessments, a more recent multicenter randomized trial compared a 300mg monthly dose of mepolizumab to placebo.¹⁵⁰ This study showed a more robust histologic response rate and a modest improvement in endoscopic severity compared to placebo, but again did not show improvement in symptoms of dysphagia, which was the primary outcome.¹⁵⁰ The next biologic trial tested cendakimab, an antibody against circulating IL-13. The phase 2 dose-finding trial showed significant improvement in endoscopic and histologic endpoints, and a strong trend towards symptom improvment.¹⁵¹ Treatment effects were sustained in a 52-week open labeled trial.¹⁵² On the basis of these results, cendakimab is being studied in an ongoing phase 3 trial (NCT04753697).

The biologic that has completed development is dupilumab, a human monoclonal IgG4 antibody, which inhibits both IL-13 and IL-4 signaling by binding to the IL-4Rα subunit. Based on promising data in a phase 2 study,¹⁵³ a phase 3 trial was conducted.⁸ In Part A of the trial, 81 subjects received either 300 mg of dupilumab weekly (without a loading dose) or placebo for 24 weeks of treatment. Part B (n=240) randomized patients to either placebo, 300 mg dupilumab weekly, or 300 mg every other week. Weekly dosing achieved 59% vs 6% (placebo) for histologic response (P<0.001) as well as greater dysphagia symptom improvement (P<0.001). Every other week dosing met the histologic endpoint, however, showed no difference in symptoms compared to placebo.⁸ Based on these findings, the FDA approved the 300 mg weekly dosing of dupilumab for the indication of EoE in May of 2022, and this medication is now approved in Europe, Canada, and other parts of the world. Dosing is 300mg weekly, given with a subcutaneous injection, and approvals for use are in patients 12 years and older, weighing at least 40 kg. The medication is well tolerated, and the safety profile is known given long-term use for other approved indications including asthma and atopic dermatitis. The most common side effects are injection site reactions and discomfort.

Another biologic tested in EoE was an antibody against the siglec-8 receptor, which is primarily found on eosinophils and mast cells, as when activated by the medication called lirentelimab, leads to eosinophil depletion and mast cell inactivation. In phase 2/3 study of EoE, lirentelimab let to marked histologic improvement with decreased eosinophil counts compared to placebo, but symptoms of dysphagia did not improve compared to placebo.¹⁵⁴ Another phase 3 trial of an eosinophil depleting agent, the IL-5RA antibody benralizumab, had similar results. It led to very high histologic response rates compared to placebo, but did not lead to symptomatic or endoscopic improvement compared to placebo, and non-eosinophil count histologic findings were also unchanged.¹⁵⁵ These two studies, taken together with consistent results that depleting eosinophils also did not also result in clinical improvement, raise the question of whether the eosinophil count is the best biomarker of treatment response (especially with a targeted therapy) and what other cells or mediators are causing persistent disease activity in the absence of the cell for which the disease is named.

Other biologics tested in EoE include an anti-immunoglobulin (Ig)E antibody (omalizumab) and a monoclonal anti-tumor necrosis factor alpha (TNFα) (infliximab). Omalizumab was studied in a randomized study of 30 adolescents and adults and failed to improve histologic or clinical endpoints compared to placebo.¹⁵⁶ Infliximab was studied in refractory EoE in the capacity of a case series and did not show benefit.¹⁵⁷ It has not been studied further in EoE. Neither of these are recommended for use in EoE.

Other Small Molecules

Other pharmacologic therapies have been tested for the treatment of EoE including montelukast (a leukotriene receptor antagonist), cromolyn (a mast cell stabilizer), and losartan (an angiotensin II receptor blocker). Montelukast has not been shown to maintain histopathologic or clinical response in adults following treatment with tCs.¹⁵⁸ In a randomized, double-blinded, placebo-controlled trial of viscous oral cromolyn for EoE in 16 pediatric patients, cromolyn did not significantly decrease oesophageal eosinophilia or symptoms scores compared to placebo.¹⁵⁹ In an 16-week open-label trial of losartan, only 17% of patients (denominator 6 subjects) met the primary outcome of a peak eosinophil count <15 eos/hpf in distal and proximal oesophagus (NCT01808196). While these therapies have not made it into the guidelines for EoE treatment, there are several medications in the pipeline for EoE, including etrasimod (a sphingosine-1-phosphate receptor agonist) and IRL201104 (an immunomodulatory peptide), among others. Therapies in the pipeline and their development status are noted in Table 6.

Dietary Therapy

Dietary therapy offers a nonpharmacologic treatment option that may be more cost-effective compared to pharmacotherapy.¹⁶⁰ There are currently three approaches to dietary avoidance of potential food allergen triggers in individuals with EoE: amino-based (elemental) diet, empiric elimination diet, and allergy test-directed elimination diet. The initial choice of dietary therapy is influenced by more than efficacy alone and requires a patient-centered approach as each dietary therapy has unique advantages and disadvantages. However, the targeted diet has fallen out of favor because current allergy test results (ie. skin prick testing or IgE blood testing) do not correlate with food triggers in EoE.^161–163 Additionally, elemental diet, which removes all food allergens from the diet and has the highest documented efficacy, is not commonly used as an initial approach to treatment since the diet is very limiting. In a systematic review and meta-analysis of 6 observational studies reported that 93.6% of subjects on an elemental diet achieved histologic remission compared to 13.3% of a placebo comparison group,¹⁶⁴ resulting in a 93% reduction in the risk of failing to achieve histopathologic remission (RR=0.07, 95% CI, 0.05 – 0.12) in those on an elemental diet compared to placebo.¹⁶⁴

Food elimination approaches are most used in clinical practice. Six food groups are identified as the most common allergen triggers in EoE: (1) cow’s milk, (2) eggs, (3) wheat, (4) soybean, (5) peanuts/tree nuts, and (6) fish/shellfish. Systematic reviews and meta-analyses of observational studies demonstrated higher histologic remission rates with the more food eliminated from the diet than with fewer, ranging from 67.9% with a six-food elimination diet (6FED), to 56.9% with a four-food elimination diet (4FED), and 42.1% with a two-food elimination diet (2FED) and 54.1% with a one-food elimination diet (1FED; ie. cow’s milk).^134,164

Since this time, a randomized trial in adults with EoE randomly assigned to 1FED or 6FED failed to detect a difference in rates of histologic remission between the two groups at six weeks (34% vs 40%, respectively; absolute risk difference 6%, 95% CI -11 to 23%).^165,166 Another randomized trial involving children with EoE showed that 1FED compared with 4FED resulted in similar rates of histologic remission and slightly lower improvement in symptoms but greater improvement in quality of life.¹⁶⁷ This implies that from a clinical standpoint, it would be reasonable to start with a less restrictive diet, for examine a 1FED or 2FED, assess for response, and then if there is non-response, move to more restrictive diets if a patient is interested. Two main approaches are used in clinical practice which include a top-down approach – beginning with a 6FED and adding a single food/food group back at a time – and a step-up approach – beginning with 1FED or 2FED and adding back foods/food groups in intervals. Details of each approach are displayed in Figure 2.

Figure 2.

Dietary Elimination Approaches – Step-up vs Top-down

FED: food elimination diet

Endoscopic Therapy

A sub-set of patients with EoE will develop a fibrostenotic disease phenotype and form narrowing of the oesophagus. Currently, in adults with dysphagia from a stricture associated with EoE, endoscopic dilation is recommended over no dilation.¹³⁴ This recommendation is based on a technical review demonstrating symptomatic improvement in 87% of patients who underwent oesophageal dilation with no significant increase in safety risks (pooled perforation rate was 0.4%, significant GI hemorrhage was 0.1% and hospitalization after dilation was 1.2%).¹³⁴ Dilation may be performed with bougie dilators or with through-the-scope (TTS) balloon. The pull-through technique using the TTS balloon approach can be used if there is diffuse narrowing.¹⁶⁸ At this time there is no evidence indicating a preferred dilation technique for EoE. 16 mm or greater is the optimal post-procedural oesophageal caliber in adolescent and adults which relieves dysphagia and avoids food impaction.^169,170 This target diameter can be achieved over multiple endoscopy sessions if needed. Dilation should not be used as an isolated chronic management strategy.

Non-EoE EGIDs

Treating non-EoE EGIDs is challenging as there are no approved medications for the indication of non-EoE EGIDs, and only one published phase 2 clinical trial for a treatment that is not currently available.¹²²

In non-EoE EGIDs, the most commonly used class of pharmacotherapy is glucocorticoids. Evidence to support the use of glucocorticoids is limited to a small series of patients. Improvement in symptoms typically occurs within two weeks, especially with systemic corticosteroids.^171,172 Dose, duration, and formulations are not standardized, but a goal of glucocorticoid therapy is to use the minimum dose needed to ameliorate symptoms and reduce tissue eosinophilia, as long-term and higher dose systemic glucocorticoid therapy has multiple side effects and is not a viable long-term treatment strategy.¹⁷³ Symptoms may reoccur after initial treatment and taper from glucocorticoid therapy, and in some cases low-dose maintenance therapy with prednisone (eg. 5 or 10 mg per day) may be needed.¹⁷¹ However, given the desire to avoid chronic systemic steroid exposure, use of topical budesonide has been reported.^174–177 Generally, enteral-release budesonide formulations can be strategically modified to target the area of the GI tract that is involved. For example, intact capsules with timed release in the colon should be used for EoC; capsules can be opened, and the granules swallowed without modification for EoN; and capsules must be opened and the granules crushed for EoG.^174–177

Other pharmacotherapies for the treatment of non-EoE EGIDs include biologic and non-biologic therapy. Three biologics have been evaluated in EoG/EoN. The first was omalizumab (anti-IgE) in 2007 which by current standards for active disease in EoG and EoN showed equivocal results.¹⁷⁸ Benralizumab (anti–IL-5Rα) was studied in a phase 2, randomized, double-blind trial for PDGFRA-negative hypereosinophilic syndrome (HES) with a 12-week blinded phase followed by a 36-week open-label extension.¹⁷⁹ There were 7 of 20 patients with EGID overlap (5 of 10 patients in the active arm and 2 of 10 in the placebo arm). For the EGID end point, after 24 weeks of benralizumab therapy, all EGID patients achieved histologic depletion of tissue eosinophils (counts dropped to the 0-1 range). In additional follow-up in 7 individuals with concomitant EGID treated with benralizumab, eosinophil count reduction was persistent at 1 year.¹⁸⁰ Clinical responses however varied. A subsequent randomized trial of benralizumab in EoG assessed 13 patients on active therapy and 13 patients on placebo and demonstrated that the primary outcome of less than 30 eos/hpf was achieved in 77% of those on benralizumab compared with 8% on placebo (P<0.001).¹⁵⁵ Clinical symptoms however did not differ between the groups. Lirentelimab (anti–Siglec-8) was studied in a phase 2 randomized control trial in which a total of 63 adult patients were randomized 1:1:1 to low- or high-dose active treatment or placebo over the 85-day blinded part of the trial.¹⁸¹ There was an average of an 86% decrease in GI eosinophil counts compared with a 9% increase in the placebo arm (P<0.001). Clinically, symptom scores were reduced 48% in the active arm compared with 22% in the placebo arm (P<0.004). There were higher rates of infusion-related reactions in the active arm. A subsequent phase 3 study (NCT04322604) has demonstrated a greater histologic improvement in the active arm (85%) compared with placebo (5%) (P<0.001).¹⁸² However, symptoms improved in both arms without a difference between active and placebo. Dupilumab is currently under investigation in a phase 2 trial for EoG in subjects 12-70 years of age (NCT03678545).

Non-biologic medications which have been evaluated in non-EoE EGIDs include mast cell agents (cromolyn and ketotifen),^183,184 PPIs,¹⁸⁵ humanized anti-IL-5 antibody,¹⁸⁶ vedolizumab,¹⁸⁷ and the leukotriene antagonist montelukast;¹⁸⁸ however, none of these agents can be recommended for routine use based on limited available data.

Dietary therapy has been studied in case series of EoG and EoN, with favorable outcomes for a 6FED¹⁸⁹ or elemental diet.¹⁹⁰ The first prospective study of elemental diet used for 6 weeks in adults with EoG and eosinophilic gastroenteritis demonstrated that 100% of patients who completed the study achieved histologic remission (<30eos/hpf) by the end of the study period.⁷⁷ Eosinophils decreased in the stomach (50 to 11 eos/hpf, p<0.001) and duodenum (49 to 16 eos/hpf, p=0.001) over the study period. Endoscopic and symptomatic endpoints also significantly improved.⁷⁷ These data suggest a significant role for food antigens in the EoG and eosinophilic gastroenteritis pathogenesis.

What is Missing

With emerging biologic therapies and novel steroid formulations, critical areas of focus will be: (1) how to position therapies in the treatment algorithm for EoE, (2) to understand the risk/benefits for combination therapy, and (3) to evaluate if treatment cycles may impact disease outcomes. Currently, initial monotherapy with PPI, topical corticosteroids, or dietary therapy, with or without oesophageal dilation in the setting of fibrostenotic disease features, is the mainstay of therapy for EoE. There is a paucity of research regarding use of combination therapy in EoE refractory disease or as initial therapy for active EoE, but these few studies suggest a significant, positive role for combination therapy.^191–193 Two retrospective studies have demonstrated symptomatic and histologic improvement with combination therapy in monotherapy refractory disease in both adults and children.^191,192 The first study (n=12) showed improvement with the use of combined PPI and food elimination diet (FED) compared to those with PPI monotherapy or FED monotherapy; comparative analysis of peak eosinophil counts showed that patients achieved a median of 4.5 eos/hpf (interquartile range [IQR], 2–6.5) on combination therapy, which was significantly decreased compared to baseline (median, 45; IQR, 35.5–50; P < .001), PPI monotherapy (median, 41; IQR, 26–50; P < .001), and FED monotherapy (median, 45; IQR, 17–67.5; P < .001).¹⁹¹ The second study (n=23) showed improvement with combined topical corticosteroids and FED compared to PPI monotherapy; comparative analysis showed significant improvement in dysphagia and endoscopic furrows (84% vs 55%; P = .02).¹⁹² A recent randomized control trial (n=63) in children with EoE found that initial combination therapy with 4FED and PPI (omeprazole 1 mg/kg twice daily, max dose 20 mg twice daily) was significantly more effective than PPI monotherapy at inducing histologic remission. On per-protocol analysis, those who achieved partial remission (<10 eosinophils/HPF) were more likely to be on combination therapy vs monotherapy (88% vs 45%; p=0.002).¹⁹³ With this evidence in mind, it is reasonable to consider combination therapy for monotherapy refractory disease in an attempt to achieve histologic remission, with the goal of attempting to deescalate to monotherapy for long-term maintenance therapy if feasible. There is currently a dearth of evidence regarding the use of treatment cycles or therapy holidays.¹⁹⁴

As more therapeutic targets are identified through genetic and pathophysiologic models, therapies will need to be positioned not only to achieve a histologic improvement but also to improve symptoms. From clinical trials such as benralizumab, we see that histologic response does not always translate to improvement in symptoms. Understanding the association between histology and clinical response will help with understanding the pathogenesis of EGIDs as well as with selecting therapeutic outcomes for future drug development trials. EGIDs are complex disorders which are influenced by demographic, clinical, environmental, and genetic factors. As such, the treatment paradigm must also shift towards individualized management strategies. As in other inflammatory disorders such as inflammatory bowel disease or asthma, an important direction for EGID treatment will be to better characterize unique EoE phenotypes and endotypes which may help personalize treatment regimens and disease surveillance. Examples of potential phenotypes include early-onset vs late-onset disease, high atopic overlap vs no atopic comorbid overlap, and PPI-responsive vs non-PPI responsive. Research efforts are needed to understand how these phenotypic groupings may impact treatment response, disease outcomes, and disease natural history.

Summarizing Research Focus for the field

Unmet needs for research are dramatically different for EoE and non-EoE EGIDs. In EoE, non-invasive diagnostic tests, clinicopathologic models that determine risk of disease progression and therapeutic failure, and novel biologic therapies are emerging and pushing the envelope for the field. In contrast, in non-EoE EGIDs, epidemiologic trends, diagnostic histopathologic thresholds, and natural history models are moving the needle in the field for these rare disorders.