Eosinophilic Esophagitis May Persist after Discontinuation of Oral Immunotherapy

Eosinophilic esophagitis (EoE) is a chronic, inflammatory condition perpetuated by allergen exposure.¹ Oral immunotherapy (OIT) is an emerging therapy for IgE-mediated food allergy that desensitizes patients to prevent harm from accidental exposure to a trigger food; however, EoE is a potential complication.² The rate of OIT-induced EoE has been estimated at ~5%³; although a majority of cases remit after stopping OIT.²

We present four patients with persistent EoE despite OIT discontinuation (Table 1). All patients developed symptoms of esophageal dysfunction during OIT. Each met consensus criteria for EoE diagnosis¹ and all had either persistent esophageal eosinophilia (≥15 eos/hpf) despite stopping OIT for ≥6 weeks or required ongoing EoE therapy (e.g. swallowed steroids, high-dose proton pump inhibitor (PPI), diet elimination) to maintain symptom remission. We excluded patients with known EoE, symptoms of esophageal dysfunction, or ongoing treatment for EoE prior to initiating OIT. Endoscopic data were characterized using the EoE Endoscopic Reference score (EREFs) and histologic findings were classified using the EoE Histologic Scoring System (EoEHSS). This study was approved by the Phoenix Children’s Hospital Institutional Review Board (# 19–496).

Table 1:

Summary of Patient Information

Pt#	Age*	Sex	OIT food†	sIgE	Other FAs	Other Atopy	FHx EoE, dilation, impaction	GI sxs	Max OIT dose (mg)	OIT duration at sxs onset (mths)	OIT duration (mths)	Time off OIT at EGD (mths)	AEC during OIT (cells/μL)	EoE EREFs	peak eos/hpf	EoEHSS score (grade)§	EoEHSS score (stage)	Tx at EGD‡	GI sxs resolved
1	3y5mth	F	PN	>100	TN, sesame, unbaked egg	AD	No	chest pain, AP regurgitation	12.5	<1	4	2	940	3	40	0.50	0.58	diagnostic	no
												11		5	35	0.52	0.57	hdPPI	no
												15		4	53	0.42	0.42	diet elimination (milk,wheat)	no
												19		0	0	0.14	0.19	budesonide	yes
												24		3	42	0.46	0.46	diet elimination (milk, wheat, baked egg)	yes
2	4y6mth	F	milk	4.61	PN, egg, TN	AR, AD	Yes	AP, reflux, pharyngitis	7700	6	15	3	460	2	65	0.46	0.21	diagnostic	no
3	6y3mth	M	PN	7.09	TN	AR, As	Yes	dysphagia, AP	8.3	1	4	2	NA	4	75	0.76	0.62	diagnostic	no
												5		2	70	0.58	0.58	hdPPI	no
4	12y8mth	F	PN	72.57	TN	AR, As	No	odynophagia, AP, globus, regurgitation	300	8	10	0	NA	2	60	0.62	0.62	diagnostic	no
												2		0	6	0.19	0.19	hdPPI	no

^*indicates age at OIT initiation

^†PN and milk OIT products were administered by referring allergists and not FDA-approved.

^‡all subjects also avoided IgE-mediated food triggers with exception of patient 4 who continued PN OIT at diagnostic EGD.

^§EoEHSS scores for grade/stage reported from maximally affected biopsy.

Patient #1 was a 3-year-old female with a history of atopic dermatitis during infancy, acute food protein-induced enterocolitis syndrome (FPIES) to avocado and IgE-mediated food allergy to peanut, tree nut, sesame, and unbaked egg. She did not have gastrointestinal symptoms at baseline. Peanut OIT was discontinued after 4 months due to retrosternal chest pain after each OIT dose. She also had dysphagia relieved by excessive lubrication of foods, regurgitation, and throat clearing. Esophagogastroduodenoscopy (EGD) revealed edema, furrows, and microabscesses and pathology was notable for 40 eos/hpf (distal and proximal esophagus). After discontinuation of OIT, the chest pain improved but dysphagia persisted. The patient failed treatment with lansoprazole (15 mg BID, 3 months) and dietary elimination (milk, wheat, +/− egg). Treatment with swallowed budesonide (0.5 mg BID) induced clinical and histologic remission.

Patient #2 was a 4-year-old female with a history of atopic dermatitis, allergic rhinitis, constipation, and IgE-mediated milk allergy. Family history was notable for multiple maternal relatives with prior EoE, esophageal dilation, or food impactions. No gastrointestinal symptoms were present at baseline. The patient developed abdominal pain, reflux, and sore throat after 11 months of milk OIT. Symptoms persisted despite dose adjustment and OIT was discontinued at 21 months. The patient’s symptoms did not resolve and EGD three months later revealed 65 eos/hpf (distal esophagus). She was unable to tolerate lansoprazole. Swallowed fluticasone (440 μg BID) failed to improve symptoms. EoE symptoms resolved after strict diet elimination (milk, egg, peanut, almond, wheat, soy, corn, and peas). She has since successfully reintroduced wheat, soy, and corn without symptom recurrence.

Patient #3 was a 6-year-old male with a history of allergic rhinitis, asthma, and IgE-mediated peanut allergy. The patient’s family history was notable for maternal history of EoE. He did not have gastrointestinal symptoms at baseline. After 1 month of peanut OIT, he developed recurrent emesis temporally related to administration of OIT, abdominal pain and dysphagia. Peanut OIT was discontinued without symptom improvement and EGD 2 months later revealed 75 eos/hpf (mid esophagus). Notwithstanding a trial of omeprazole 20 mg BID for 8 weeks, repeat EGD showed persistent tissue eosinophilia (eos/hpf = 70).

Patient #4 was a 12-year-old female with a history of gastroesophageal reflux, asthma, allergic rhinitis (on subcutaneous immunotherapy), and IgE-mediated peanut allergy. At the time of initiation of OIT, she had not had reflux symptoms for multiple years. After 8 months of peanut OIT, she developed a sore throat and was treated with omeprazole 20mg BID resulting in mild clinical improvement. OIT was continued but she was unable to tolerate PPI weaning. Symptoms of odynophagia, regurgitation, pyrosis, dysphagia, abdominal pain, belching, and nausea ensued and EGD performed on OIT and PPI revealed 60 eos/hpf (mid esophagus). OIT was discontinued after 10 months. The PPI was changed to pantoprazole 40 mg BID and symptoms improved with resolution of esophageal eosinophilia. An impedance probe showed weakly acidic reflux and the patient remains on high-dose PPI.

These four patients developed persistent EoE during OIT. Because 2/4 patients had relatives with known/possible EoE, a thorough family history may identify those at greater risk. Food allergy and EoE are increasing in prevalence, commonly coexist, and are potentially linked.⁴ A recent study examining serial endoscopic biopsies in adults suggests that peanut OIT induces esophageal eosinophilia; however, eosinophilia is usually transient and is not associated with clinical symptoms.⁵ Less commonly, OIT induces EoE.⁵ While this is usually reversible with cessation of exposure to the triggering antigen (i.e. OIT)⁶, our patients likely have ongoing disease driven by other food or environmental allergens. We did not observe seasonal variation in onset or remission of EoE suggesting aeroallergen-triggered disease.⁷ We speculate that OIT triggers local type 2 immune responses to OIT proteins in susceptible individuals that alter mucosal permeability and/or antigen presentation of other food proteins leading to persistent EoE. Clinical responses to PPI, corticosteroids and diet elimination are consistent with this speculation.^{1, 8}

This is a single-center study, therefore, the frequency of this occurrence cannot be generalized to all OIT patients; moreover, we acknowledge certain limitations. First, none of the patients underwent baseline EGD; therefore, some may have had pre-existing asymptomatic esophageal eosinophilia, a phenomenon recently described in patients with IgE-mediated food allergy.^{9, 10} EGDs are not indicated without clinical symptoms; therefore, performing an EGD before OIT is not standard of care. Consequently, we cannot definitively prove that OIT induced esophageal eosinophilia. Notwithstanding, none would have met clinicopathologic criteria for EoE pre-OIT in the absence of clinical symptoms. While all patients were assessed and negative for gastrointestinal symptoms at baseline, these were not evaluated prospectively using a validated patient-reported outcome measure; therefore, subtle symptoms may have gone undetected. Nonetheless, all developed marked symptoms of esophageal dysfunction exacerbated by and temporally correlating with OIT. Finally, longer follow up studies with EoE treatment withdrawal are required to better assess disease permanence. Clinical characterization serves a starting point for larger prospective studies analyzing risk factors and mechanisms underlying the development of EoE during OIT.

Abbreviations used:

AD

atopic dermatitis

AEC

absolute eosinophil count

AP

abdominal pain

AR

allergic rhinitis

As

asthma

BID

twice daily

EGD

esophagogastroduodenoscopy

EoE

eosinophilic esophagitis

EoEHSS

EoE Histologic Scoring System

eos/hpf

eosinophils per high power field

EREFs

EoE Endoscopic Reference score

FA

food allergy

FHx

family history

FPIES

food protein-induced enterocolitis syndrome

GI

gastrointestinal

m

month

mg

milligrams

μg

micrograms

μL

microliter

hdPPI

high dose proton pump inhibitor

NA

not assessed

OIT

oral immunotherapy

PPI

proton pump inhibitor

PN

peanut

pt

patient

sIgE

serum immunoglobuline E

sxs

symptoms

tx

treatment

TN

tree nuts

y

year