Effect of Biologics against Interleukin-5 Administered to Patients with Eosinophilic Gastroenteritis

Abstract

Biologics against interleukin-5 were administered to five patients with eosinophilic gastroenteritis (EGE) and bronchial asthma (BA). BA and abdominal symptoms as well as changes in steroid dose and the blood eosinophil count were examined. The man-to-woman ratio was 1:4. The average age of onset was 63 years old. The duration of the disease was three to nine years. Four patients showed improved BA symptoms with mepolizumab or benralizumab, and three successfully discontinued steroids. Regarding abdominal symptoms, mepolizumab was effective in one of three cases, while benralizumab improved symptoms in three of four cases. Biologics targeting interleukin-5 are effective in some patients with EGE accompanied by BA.

Introduction

A diagnosis of eosinophilic gastrointestinal disease (EGID) is made when a patient presents with symptoms originating in the digestive tract, other diseases are ruled out, and pathological eosinophilic infiltration of the digestive tract is confirmed. Although EGID is occasionally accompanied by an increase in the eosinophil (Eo) count in the peripheral blood, it is thought to be an allergic disease that causes localized damage to the digestive tract with no damage to other organs (1).

EGID can be classified into two types: eosinophilic esophagitis (EoE) and eosinophilic gastroenteritis (EGE), both of which are characterized by an abnormally high number of Eo infiltrates into the digestive tract from the stomach to the rectum. EGE is also known as an EGID (2).

EGE differs from EoE in that it does not have a specific age of onset and there is no marked difference in the incidence between men and women or among different age groups (3). However, the cause of EGE remains unclear, although it is thought to be an allergic disease in which individuals who are prone to Th2 reactions react to food antigens, causing an increase in the production of cytokines, such as interleukin (IL)-5, IL-13, IL-15, and eotaxin in the digestive tract and activation of Eos and mast cells, resulting in damage to the digestive tract epithelium. IL-5 plays an important role in the activation of Eos and is involved in their mobilization and activation in tissues. A Th2-type immune response may cause both EGE and EoE; however, it has been suggested that there may be differences in the pathology of esophagitis and gastroenteritis, as IgE antibodies, which are ineffective in EoE, are effective in EGE (4-7).

Given the above, it is currently difficult to perform curative therapy for EGE, and symptomatic treatment is administered to manage inflammation by suppressing the allergic and immune reactions. Most treatment reports are case reports or case series, and no randomized controlled studies with a high level of evidence have been conducted.

Previously, the main treatment options for EGE were steroids, and while systemic steroid administration was effective in many cases, steroid-resistant cases have been reported. In addition, many patients who respond to steroids experience worsening symptoms when the dose is tapered. Although food elimination therapy is effective, it is often difficult to maintain. In recent years, there has been a growing body of evidence supporting the use of biological agents for EoE, and the development of biologics (e.g. benralizumab, mepolizumab, and dupilumab) for EGE is currently underway.

Biological drugs targeting IL-5 have been approved in Japan as therapeutic agents for bronchial asthma (BA) and other allergic diseases. Although biologics against IL-5 have been reported to be effective in some patients with EGE, they have yet to be approved for patients with EGE in Japan. In the present study, we investigated the clinical course of treatment with biologics for IL-5 in patients with EGE and BA.

Case Reports

This study included five patients [1 man, 4 women; mean (range) age: 63 (42-76) years old] with EGE who received biologics against IL-5 for BA. None of the patients had any food allergies. Benralizumab was administered as an anti-IL-5 receptor α monoclonal antibody and mepolizumab as an anti-IL-5 monoclonal antibody.

We evaluated allergic disease, abdominal symptoms, blood Eo count, esophagogastroduodenoscopy (EGD) findings, colonoscopy findings, clinical course, and steroid dosage reduction (Table 1).

Table 1.

Summary of the Five Cases.

Case	Age	Sex	Asthma onset age	Time intervals between the onset of BA and EGE	Allergic disease	EoE findings	Symptoms	Eo (/μL)	Eo after biologics(/μL)	1st biologics	2nd biologics	Clinical course
1	60s	F	40s	15	Bronchial asthma (BA)	・Longitudinal farrow ・No eosinophilic infiltration	Abdominal pain, diarrhea, nausea	8,589 (43.6)	0	Mepolizumab	Benralizumab	Benralizumab is effective for BA and EGE
2	70s	F	50s	11	BA, allergic rhinitis	None	Abdominal pain, diarrhea	1,797 (25.6)	0	Mepolizumab	Benralizumab	Benralizumab is effective for BA and EGE
3	50s	F	40s	10	BA, eosinophilic pneumonia	None	Abdominal pain, diarrhea, vomiting	1,502 (35.0)	0	Benralizumab	-	Benralizumab is effective for BA and EGE
4	40s	M	30s	9	BA	None	Abdominal pain, diarrhea, vomiting	668 (12.0)	90	Mepolizumab	-	Mepolizumab is effective for BA and EGE
5	40s	F	30s	6	BA	None	Abdominal pain, diarrhea, vomiting	742	104	Benralizumab	Dupilumab	Dupilumab is effective for BA and EGE

Case 1

A woman in her 60s who had been receiving treatment for BA with high-dose inhaled corticosteroids (ICS)/long-acting beta2-agonist (LABA) starting in her 40s reported that she had been experiencing abdominal pain, diarrhea, and nausea for 5 years. The time interval between BA and EGE was approximately 15 years. Blood tests showed a white blood cell count of 19,700/μL and Eo count of 8,589/μL (43.6%). The first biologic administered was mepolizumab, but the abdominal symptoms did not improve, and BA symptoms did not stabilize. Therefore, the patient was switched to benralizumab, which was effective for BA and EGE. EGD revealed longitudinal furrows in the esophagus, gastric atrophy, fixed rings, flattening of villi, erosion, and edema in the duodenum. Colonoscopy revealed no abnormalities in the terminal ileum or colon (Fig. 1). However, eosinophilic infiltration was not observed in the esophagus, and minimal eosinophilic infiltration was observed in the stomach. Conversely, eosinophilic infiltration was observed in many cases in the duodenum and ileum (Fig. 2).

Figure 1.

Upper gastrointestinal endoscopy showing longitudinal furrows in the esophagus, gastric atrophy, fixed rings, and villi flattening. Colonoscopy showed no abnormalities in the terminal ileum or colon. a: esophagus, b: stomach, c: indigocarmine sprayed on the stomach, d: duodenum, e: jejunum, f: terminal ileum

Figure 2.

A biopsy of the duodenum showing eosinophilic infiltration. Eosinophil infiltration was minimal in the stomach. The chorion is flat and depressed, and there are many eosinophils among the chronic inflammatory cells that infiltrate the stroma. a: stomach (Hansel stain ×400), b: duodenum (Hansel stain ×40), c: jejunum (Hansel stain ×400), d: terminal ileum (Hematoxylin and Eosin staining ×400)

Benralizumab was effective for the treatment of BA and EGE. Although 5 mg prednisolone (PSL) was administered to treat EGE, PSL was discontinued with the introduction of biologics. The Eo count was maintained at 0%. The patient's abdominal symptoms improved. The histological eosinophilic infiltration also disappeared.

Case 2

A 70-year-old woman with a history of BA lasting approximately 20 years, also presenting with allergic rhinitis, reported that her symptoms included abdominal pain and diarrhea that had started 9 years prior. The interval between BA and EGE was approximately 11 years. Her blood test results showed a white blood cell count of 7,020/μL and Eo count of 1,797/μL (25.6%). The esophagus and duodenum showed no specific findings, although the stomach showed atrophy on EGD. Eosinophilic infiltration was observed in all the upper gastrointestinal tracts.

Colonoscopy revealed flattening of the villi in the terminal ileum and erosion of the colon; diffuse eosinophilic infiltration was observed in the terminal ileum and colon. Mepolizumab was used as the initial biologic agent; however, as the abdominal symptoms did not improve and the BA symptoms did not stabilize, it was changed to benralizumab, which was effective for BA and EGE. PSL (20 mg/day) was administered to treat EGE, although it was gradually reduced and discontinued after the introduction of the biologics. The Eo count was maintained at 0%. The histological eosinophilic infiltration also disappeared, and the patient's abdominal symptoms improved.

Case 3

A 50-year-old woman was receiving treatment with high-dose ICS/LABA for BA and eosinophilic pneumonia that had started in her 30s. She had experienced abdominal pain, diarrhea, and vomiting in her 40s, and was diagnosed with eosinophilic gastroenteritis. The patient had started taking PSL (10 mg) 10 years prior to presentation. Therefore, the interval between BA and EGE was 10 years.

Her white blood cell count was 4,290/μL, and her Eo count was 1,502/μL (35.0%). Benralizumab was effective in the treatment of BA and EGE. EGD revealed no abnormalities in the esophagus, stomach atrophy, or flattening of villi in the duodenum. Eosinophilic infiltration was observed in all organs. Colonoscopy revealed flattening of villi in the terminal ileum and erosion of the colon, although no eosinophilic infiltration was observed. After initiating benralizumab, PSL was started at 10 mg to treat EGE and reduced to 3 mg, although discontinuation was not possible. The PSL dose was gradually reduced to 10 mg, and the patient became steroid-free. The Eo count was maintained at 0%. The histological eosinophilic infiltration disappeared.

Case 4

A 40-year-old man had undergone treatment with high-dose ICS/LABA for BA approximately 10 years prior. He had been experiencing abdominal pain, diarrhea, and vomiting for one year. The time interval between BA and EGE was approximately nine years. The results of a blood test performed before treatment showed a white blood cell count of 5,570/μL and an Eo count of 668/μL (12.0%). EGD revealed no abnormalities in the esophagus or duodenum, but redness was observed in the stomach. Eosinophilic infiltration was observed in the esophagus. Colonoscopy revealed no abnormalities in the terminal ileum, but eosinophilic infiltration was observed. Mepolizumab was initiated and was effective in treating BA and EGE. PSL was not initially administered. The Eo count was 90/μL. The histological eosinophilic infiltration disappeared, and the patient's condition was good.

Case 5

A woman in her 40s who had been receiving treatment with high-dose ICS/LABA for BA for approximately 10 years had been experiencing abdominal pain, diarrhea, and vomiting for the previous 4 years. The patient was treated with 5 mg PSL for EGE. The time interval between BA and EGE was approximately six years. Blood tests showed a white blood cell count of 6,290/μL and Eo count of 742/μL (11.8%). EGD revealed no abnormalities in the esophagus, but redness was observed in the stomach and duodenum. Eosinophilic infiltration was not observed in any of the organs. Colonoscopy revealed erosions and ulcers in the cecum, ascending colon, and rectum, with eosinophilic infiltration. Benralizumab was the first biologic drug used; however, as the abdominal symptoms did not improve and the BA symptoms did not stabilize, benralizumab was changed to dupilumab, which was effective in improving her abdominal symptoms. PSL (5 mg) was continued, and the Eo count was 104/μL. The histological eosinophilic infiltration disappeared.

Discussion

We herein report our experience with five cases of EGE treated with biologics. Most patients with EGE and BA complain of abdominal pain, diarrhea, and vomiting. The average peripheral Eo count before treatment was 2,660/μL, and eosinophilia was observed in all cases. After the administration of biologics, the peripheral Eo count decreased, and abdominal symptoms improved in all cases. PSL was discontinued or decreased in three patients. Benralizumab was administered to four patients, and abdominal symptoms improved in all cases. In one of the three cases treated with mepolizumab, there was an improvement in abdominal symptoms.

Concerning the endoscopic findings, the authors previously reported that, in four out of six cases, flattening of the villi of the small intestine was observed at close inspection, even in cases without any abnormalities noted during routine observation, except for some slight edema of the mucosa (8). In the present study, flattening of the villi was observed in two out of five cases. Villi flattening is not specific to an EGE. For example, villous flattening and atrophy in the small intestine are characteristic findings in CD, but similar phenotypes can also be observed in inflammatory bowel diseases, such as Crohn's disease and intestinal tuberculosis. In such cases, a tissue biopsy may be recommended. In EGID, many cases present without any distinct endoscopic findings, although villous flattening in the small intestine may prompt tissue biopsy to histopathologically confirm eosinophilic infiltration.

Currently, EGID is not a disease entity covered by health insurance in Japan, and the cost of biologics is only covered for diseases such as BA and atopic dermatitis (9). In the present study, all five patients had BA and were able to use biologics. Although steroids are effective for EGID, approximately half of the patients relapse during steroid tapering, and there are concerns regarding long-term steroid use. Therefore, it is desirable to establish effective biological treatment as soon as possible (10).

Clinical research on drugs targeting the various mechanisms of EGID is currently underway. Based on a small number of studies, promising evidence has emerged regarding the effectiveness of monoclonal antibodies targeting IL-13, IL-4/13Rα, and Siglec-8 (11). Antibody drugs targeting IL-5 are expected to be effective in the treatment of EGE. The Eo count in the blood is a good biomarker for predicting the efficacy of anti-IL-5 therapy; the higher the Eo count in the blood, the greater the suppression effect on exacerbations.

Mepolizumab and reslizumab, which bind to soluble IL-5, and benralizumab, which binds to IL-5 receptor alpha subunit (IL-5Rα), are used to treat EoE. In a pilot study involving four patients with EGE treated with reslizumab, there was a decrease in the Eo count in the peripheral blood and tissues, and symptoms improved in three of the four patients (12). However, a reactive increase in the Eo count was reportedly observed after administration of the drug (13). Mepolizumab is also effective in patients with asthma-complicated EGE (14,15) (Table 2).

Table 2.

Previous Reports on the Biologics Used for Eosinophilic Gastroenteritis Treatment.

Biologics	Country	Study type	Comments		Reference
Mepolizumab
	Japan	Case	68-year-old woman, mepolizumab that was maintained in steroid-free remission.		(14)
	Japan	Case	46-year-old man, after mepolizumab (anti-interleukin 5 antibody) was administered, the patient’s disease was controlled.		(15)
Benlalizumab				Clinical trial number
	U.S.	Phase 2 clinical trial	n=7, Benralizumab treatment completely depleted blood and GI tissue eosinophilia in patients with eosinophilic GI disorders.	NCT02130882	(18)
	U.S.	Phase 2 clinical trial	n=26, Benralizumab treatment induced histological remission, as defined by absence of tissue eosinophilia, in most patients with eosinophilic gastritis.	NCT03473977	(20)

Concerning its mode of action, benralizumab differs from mepolizumab in that it exerts antibody-dependent cellular cytotoxicity and inhibits the effects of IL-5. Benralizumab demonstrates rapid and strong peripheral blood Eo-removing power as well as high efficacy in removing Eos from tissues, including lung tissue (16).

In most patients with EGE who experience symptom recurrence, long-term low-dose maintenance therapy with PSL is mandatory. Mepolizumab may be effective for the treatment of eosinophilic disorders other than asthma, such as EGE and aspirin-exacerbated respiratory disease (17). Benralizumab treatment completely depletes blood and gastrointestinal tissue eosinophilia in patients with EGIDs; however, the clinical response is heterogeneous (18) (Table 2).

A phase-3 study on the administration of benralizumab to patients with EoE found that a greater proportion of patients in the treatment arm achieved histological remission at 24 weeks than those in the placebo arm. However, there was no significant difference between the groups in the change in the dysphagia symptom questionnaire score from baseline (19), which is why this treatment is not currently reimbursed by health insurance in Japan. A phase-2 study of eosinophilic gastritis showed that histological remission was induced in most patients treated with benralizumab. However, the persistence of histological, endoscopic, and other features of the disease suggests the involvement of an Eo-independent pathogenic mechanism (20).

Finally, in a study in which benralizumab was administered to patients with EGE syndrome, Eo counts in the tissue decreased, and all patients reported improvement in gastrointestinal symptoms, although some reported symptom relapse over time (18). Symptoms improved in three of the four cases treated with benralizumab in our study, suggesting that this antibody treatment may be beneficial to patients, warranting its approval for this indication.

Recently, a systematic review of biological agent treatments for non-EoE-EGIDs was conducted in Japan (21), and only five agents were used in eight reports, including the present study. In addition to the effects of omalizumab and mepolizumab, the effects of the Siglec-8-targeting drug rilentlimab have also been demonstrated. However, the differences between the five biological agents used for non-EoE-EGID in the reviewed literature remain unclear, and the most effective is not yet known. Nevertheless, the evidence supporting their effectiveness is compelling.

Conclusions

In conclusion, biologics against IL-5 are useful for not only BA but also for EGE in patients with EGE accompanied by BA. However, further evidence is required to establish the effectiveness of IL-5-targeting biologics, specifically in patients with EGE.

The authors state that they have no Conflict of Interest (COI).