Eosinophils account for a small fraction of peripheral blood leukocytes and tissue resident cells under healthy conditions. However, they can accumulate in
large numbers in the blood and/or tissues in a variety of diseases including hypereosinophilic syndrome (HES), Churg-Strauss syndrome, helminth infections,
allergy (including asthma), nasal polyposis, immunodeficiencies, and EGIDs. EGIDs comprise distinct diseases, including eosinophilic esophagitis (EoE),
eosinophilic gastritis (EG), and eosinophilic colitis (EC), defined by the anatomical location of eosinophil-rich inflammation along the GI tract in the
absence of known causes of eosinophilia (e.g. drug reactions, parasitic infections, malignancy).
EoE, EG, and EC are rare, chronic, primary disorders, and the diagnosis and management of each disease is complex and requires adult and pediatric
multidisciplinary input inclusive of gastroenterologists, allergists, immunologists, and pathologists. Of the eosinophil-associated diseases, there has
been growing attention on EoE, EG, and EC over the past decade, as these diseases are now recognized to be increasing in incidence and prevalence,
associated with substantial morbidity and to lack satisfactory treatment.
The importance of these issues is underscored by a recent NIH report calling for more research on these diseases, for which there are no USA Food and Drug
Administration (FDA)-approved drugs. The NIH Office of Rare Diseases Research (ORDR) lists EoE, EG, and EC under the broad term of eosinophilic
enteropathy. These distinct diseases have been grouped together for several reasons.
First, the diagnoses of these disorders have the common characteristic of eosinophilic inflammation infiltrating the GI wall as evidenced by microscopic
analysis of tissue biopsies.
Second, the defining clinical constellation and molecular pathogenesis of these disorders are poorly understood and are in desperate need of scientific
and clinical advances. In fact, although we strive to distinguish these diseases, they can sometimes overlap, highlighting their relatedness and the need
to develop diagnostic and prognostic markers.
Third, these diseases appear to be largely mediated by eosinophils rather than by other cells, although definitive clinical research is needed to prove
Fourth, although it is now appreciated that most forms of EoE are food antigen–driven, these diseases are largely idiopathic, and the pathogenesis of EG
and EC is unknown.
Fifth, glucocorticosteroids (GCs) are typically the first line treatment for these disorders, yet GC dependence, toxicity, and/or resistance are major
concerns limiting both short and long-term use.
Sixth, these diseases are likely to benefit from recently emerging eosinophil-directed therapy such as anti–interleukin (IL)-5 (mepolizumab, reslizumab),
the eosinophil-depleting anti–IL-5 receptor antibody (benralizumab), and anti–eotaxin-1 (bertilimumab).
Seventh, by understanding these rare diseases in which eosinophils are likely to be the primary cause of disease, our findings may contribute to our
understanding of eosinophilia in more common diseases (e.g. asthma and allergic rhinitis).
Eighth, the FDA has begun to focus on how to best meet the needs of these disorders and has engaged our consortium members in a partnership, as outlined
in recent workshops and articles.
The diagnoses of EG and EC are generally reserved for cases with markedly increased numbers of eosinophils [in the case of EG, we will use ≥30
eosinophils/HPF in 5 or more HPF, and for EC we will use ≥65 eosinophils/HPF] in which eosinophils are the predominant inflammatory cells in the tissue and
are associated with the presence of architectural abnormalities, including excessively branched and/or coiled glands and eosinophilic cryptitis. Yet, there
remains a diversity of opinions as to the exact quantification of some of these features, and to date, consensus guidelines have not been published. These
diagnostic features of EG and EC have been highlighted by our consortium members, including our expert GI pathologist Dr. Margaret Collins, Professor of
Pediatrics (CCHMC), and members of the CEGIR Pathology Core, including Dr. Kelley Capocelli and Dr. Guang-Yu Yang.
These conditions also present in diverse ways reflecting the sites of tissue eosinophilia, as EoE has symptoms of feeding difficulties and vomiting in
young children and food impactions and dysphagia in teenagers and adults, EG evidences upper abdominal pain, nausea, and vomiting, and EC manifests with
diarrhea, hematochezia, and lower abdominal and rectal pain. Despite suggestions of an EGID mini-epidemic over the last decade, EoE, EG, and EC remain rare
diseases. With the introduction of ICD-9-CM codes for these conditions, recent investigations by CEGIR Investigators were able to accurately estimate the
prevalence of EGIDs. Using a highly specific case definition, the prevalence of EoE has been estimated to be between 39.0-56.7/100,000 in the USA,
consistent with earlier work, with an extrapolated national prevalence of approximately 105,000-150,000 cases. In a separate study, the prevalence of EG
was found to be 6.4/100,000, extrapolated to approximately 17,000 cases in the USA, and the prevalence of EC was estimated at 3.5/100,000, extrapolated to
approximately 9,000 cases. Due to the rarity of these conditions, single centers are unable to compile sufficient patient numbers to study these diseases
comprehensively. Although the etiologies of EoE, EG, and EC remain unknown, several lines of evidence support allergic, environmental, and genetic
components in EoE. The high response rate to food elimination diets, especially amino acid–based elemental diets, and the frequent recurrence of disease
with food reintroduction imply that EoE is mediated by immune sensitization to foods. Indeed, experimental EoE, EG, or EC in murine models can be induced
by exposure to dietary or inhaled allergens or to the Th2-associated cytokine interleukin (IL-13) via a variety of entry points including the skin,
respiratory tract, and GI tract. Consistent with an allergic etiology, SGCs are effective for the treatment of EoE and elicit local changes in esophageal
expression. However, better treatments are needed due to the substantial rates of GC resistance, high rates of disease relapse upon GC cessation, and/or
relapse while on therapy. Unlike food anaphylaxis, which is typically caused by sensitization to a single food in a given individual, most patients with
EoE have sensitization to a large panel of foods, requiring removal of several major food groups from their diet in order to achieve disease remission.
Even if remission is achieved, dietary restriction is typically a very challenging long-term therapeutic option.
The limitations of dietary and GC therapy, compounded with the lack of an FDA-approved drug for the treatment of EoE, EG, or EC, highlight the pressing
unmet medical needs of patients with these diseases. The impact of EGIDs on patients and their families is underscored by data collected by APFED. Over 30%
of APFED members who responded to a survey in 2011 reported traveling more than 100 miles from their home for treatment. Over 30% reported that EGIDs had
an adverse impact on their finances, 30% reported an effect on their marriage, over 55% reported an effect on school, and over 60% reported an effect on
their job CURED has also emphasized the desperate need to provide medical formula for patient with EGID. Most patients using elemental formula do not have
insurance coverage or have only partial coverage for medical foods, even though more than half of these patients use elemental formula as a supplement for
a severely restricted diet (six or fewer foods). The annual expense for elemental formula typically exceeds $15,000, and the annual out-of-pocket cost for
most families with EGIDs is at least $2,000 not including formula. Administration of elemental formula often requires placement of feeding tubes.
Collectively, there is substantial evidence for the growing unmet medical needs of patients and families suffering from EoE, EG, and EC.
Areas of uncertainty:
Little is known regarding the natural history and pathophysiological mechanisms, especially for EG and EC, and FDA-approved drugs do not exist for any of
these diseases. Although elevated eosinophil levels are required for diagnosis of these diseases and generally accepted to correlate with disease activity,
few studies have directly addressed whether eosinophils are the histological component that best correlates with clinical symptoms, and these studies have
been primarily limited to EoE. Evaluating the fundamental assumption that tissue eosinophil counts correlate with disease activity is a major focus of the
CEGIR consortium. A number of issues are likely contributing to the uncertainty about the best tissue features to monitor for correlation with clinical
symptoms. These include 1) lack of validated metrics to assess whether symptoms or PROs align with tissue histology measures; 2) lack of prospective,
randomized, placebo controlled trials that have utilized the same primary outcome variables; 3) lack of studies that have cohort sizes that are truly
powered to find correlations between symptoms and tissue histology elements; 4) the intermittent nature of symptoms and the ability of patients to
institute behavioral and lifestyle changes to compensate for their symptoms; and 5) inconsistent histological measurement methodologies of tissue
eosinophil counts in various studies. We will overcome many of the limitations of prior studies by 1) utilizing a series of diverse clinical outcome
measures (COMs); 2) conducting a multisite prospective trial with well-defined entry criteria; 3) increasing cohort sizes; and 4) standardizing our
histological methodologies. The histological assessment of EoE is relatively straightforward as the normal esophagus is void of mucosal eosinophils. In
contrast, the diagnosis of EG and EC are particularly complex as eosinophils are normal resident cells in the GI mucosa. Importantly, increased mucosal
eosinophils are not pathognomonic for EoE, EG, or EC, as mucosal eosinophilia can be seen in other diseases including reflux esophagitis, food allergy,
IBD, celiac, and malignancies. Therefore, definitions of key clinical and histological features, especially for patients with EG and EC, are urgently
needed. Currently, on the basis of limited data and clinical experiences, it has been proposed that the diagnostic criteria for EG and EC includes the
histological finding of at least twice the normal peak eosinophils/high-power field (HPF) reported in different regions of the non-diseased stomach and/or
colon. Through this project, we seek to advance the field by determining the relationships between COMs and mucosal eosinophilia, which will ultimately
assist in developing diagnostic criteria, understanding mechanisms, and identifying endpoints for treatment efficacy of each of the three rare diseases
(EoE, EG, and EC). Pathophysiological mechanisms defining EoE have made great strides in the last decade. For instance, a number of studies identified
dysregulation of the allergic arm of the immune system in EoE pathogenesis. This etiology is supported by the reversibility of EoE following dietary
avoidance of specific foods, the reoccurrence of EoE upon re-introduction of the removed foods, the induction of the EoE-like disease in mice by exposure
to both food allergens and aeroallergens, and genome-wide transcriptome analysis of esophageal tissue implicating interplay between the innate and adaptive
immune responses. EoE has a strong hereditary component with a large sibling risk ratio (λs~80). Early genetic analyses have identified susceptibility loci
in the regions that contain candidate genes expressed in epithelial cells and strongly implicated in antigen recognition (TSLP, thymic stromal
lymphopoietin), inflammatory cell recruitment/activation (CCL26, eotaxin-3), and calpain, an esophageal-specific enzyme. The Th2 cytokine IL-13
programs transcription of key EoE-related genes and pathways, and transforming growth factor (TGF)-β1 has been proposed to be a regulator of EoE
pathogenesis. In contrast, our understanding of EG and EC is in its infancy, with current murine model studies suggesting roles for the
eosinophil-associated cytokines eotaxin-1 and IL-5. Thus, while early evidence suggests that EoE, EG, and EC occur as a result of immune dysfunction,
specific targets and biomarkers remain to be fully recognized.