MALT lymphoma and eosinophilic oesophagitis: incidental finding— review of possible factors influencing the aetiopathogenesis of eosinophilic oesophagitis

Abstract

An 11-year-old boy with a history of atopy and allergic rhinitis under treatment with sublingual immunotherapy was referred following several episodes of food impaction. Diagnosis of eosinophilic oesophagitis, chronic gastritis and gastric mucosa-associated lymphoid tissue (MALT) lymphoma associated to Helicobacter pylori were confirmed. Results of the extension study were negative and the lymphoma was resolved with eradicating treatment for H. pylori. No improvement was observed in the oesophagitis after the withdrawal of immunotherapy and treatment with high proton pump inhibitor doses or following the omission of several foods to which subclinical sensitisation was confirmed. Oesophagitis was finally resolved by removing cow's milk protein. After 10 years, neither eosinophilic oesophagitis nor MALT lymphoma was observed.

Gastric MALT lymphoma associated to H. pylori is a rare disorder in children. Although coexisting H. pylori infection is common in patients with eosinophilic oesophagitis, the association of gastric MALT lymphoma with eosinophilic oesophagitis has not been reported before.

Background

The incidence of eosinophilic oesophagitis in children is on the rise and is today a frequent indication for upper gastrointestinal endoscopy. Paediatric mucosa-associated lymphoid tissue (MALT) lymphoma is rare, even though our case illustrates how at least its gastric variant can be an incidental finding during an endoscopic study. The relationship between MALT lymphoma and gastric infection by Helicobacter pylori is a known fact in adult population; however, it has been barely described in paediatric ages. Our case is also interesting as it prompted us to consider the correlation between H. pylori infection and eosinophilic oesophagitis, and between the later and immunotherapy treatment, so frequently used nowadays.

Case presentation

An 11-year-old boy presented with several episodes of food impaction in the proximal oesophagus in the past year. There was no associated abdominal pain or dysphagia. His medical history included atopic dermatitis in early childhood, bronchial hyperreactivity and seasonal allergic rhinitis. He was allergic to olive, grasses and banana and had been receiving sublingual immunotherapy for grasses for 1.5 years. Findings in a physical examination were unremarkable: the abdomen was soft, depressible and nonpainful to the touch, there were no palpated masses or megaly, heart and lung sounds were normal, and there were no skin lesions.

Given the symptoms and medical history of the patient, our main suspicion was eosinophilic oesophagitis, and an upper gastrointestinal endoscopy was scheduled.

Investigations

In the endoscopy, we detected oedema (vascular pattern), mild longitudinal linear furrows, whitish severe exudates (more than 10% of the surface) and crepe paper-like appearance at the oesophagus but no rings or strictures erosions, along with antral nodularity as the only significant finding in the stomach (figures 1–3). Rapid urease test at the antrum was positive. These findings were all suggestive of eosinophilic oesophagitis and H. pylori infection.

Figure 1.

Oedematised antral and gastric mucosal nodularity.

Figure 2.

Congestive mucosa and hematine observed at the gastric body.

Figure 3.

Oedema, mild longitudinal linear furrows throughout the oesophagus, whitish exudates, crepe paper-like appearance of the mucosa.

Histological findings revealed lymphoplasmacytic proliferation at the antral gastric mucosa, with lymphoepithelial lesions and monotypic lambda chain expression suggestive of MALT lymphoma. We also observed the presence of H. pylori bacilli. In the distal and proximal oesophageal mucosa, inflammatory changes were noted associated with eosinophilia (>15 eosinophils/field).

The patient neither report weight loss nor a family history of gastric neoplasms. There were no relevant findings in blood tests (including no iron deficiency anaemia), chest X-ray and thoracic abdominal CT scan. Furthermore, all requested serological tests, hepatitis B virus, hepatitis C virus and HIV, were negative. The patient was referred to the Paediatric Oncology Department of a reference tertiary care centre where a bone marrow biopsy and cervical and thoracic MRI were performed, with no pathological findings. Another endoscopy at that hospital confirmed the presence of gastric MALT lymphoma.

Treatment

Given the macroscopic findings observed in the first endoscopy, sublingual immunotherapy was discontinued and sequential H. pylori eradication therapy with high-dose proton pump inhibitors was prescribed in order to eradicate H. pylori infection and after that lansoprazole 60 mg per day was maintained until histopathological findings were available (1 month later). After completing this treatment, the patient was asymptomatic. Diagnosis of MALT lymphoma was confirmed in both endoscopies and H. pylori was not observed in the second one. Sublingual immunotherapy was discontinued.

Outcome and follow-up

The boy returned a negative result in a H. pylori breath test. Over the subsequent year, no evidence of MALT lymphoma was found in several endoscopies and analytical controls at the paediatric oncology unit, so the follow-up was continued at our centre.

During that year, treatment with esomeprazole was maintained at 1 mg per kilo and day, so the patient remained completely asymptomatic although oesophageal eosinophilia was persisted. This prompted an allergological study and foods to which the boy had shown sensitisation not eliciting an endoscopy or histological response were excluded from his diet. Subsequently and following new guidelines of eosinophilic oesophagitis, treatment with higher dose proton pump inhibitors (omeprazole at 2 mg per kilo and day) was initiated without resolution of oesophageal eosinophilia. After that a dietary exclusion of two foods (gluten and cow’s milk proteins) was prescribed. This resulted in endoscopic and histological healing along with relapse when reintroducing cow’s milk proteins but not when reintroducing gluten.

After 10 years of follow-up, the patient continues cow’s milk proteins exclusion on his diet and remains asymptomatic. Neither further signs of MALT lymphoma nor eosinophilia in oesophagus in follow-up endoscopies has been founded.

Discussion

MALT lymphoma is a form of non-Hodgkin's lymphoma of extranodal B cells, which develops in the lymphoid tissue associated with mucous membranes. It typically occurs in the stomach, although it can arise at other sites such as conjunctiva, duodenum, salivary glands, thyroid or larynx. It accounts for 5% of non-Hodgkin lymphomas, appears more frequently in persons older than 50 years and mostly affects women regardless of age group.¹ In children, this lymphoma is rare, high-grade lymphomas (Burkitt lymphoma and anaplastic B and T cell lymphomas) being more frequent in this population group.²

In 10%–20% of cases, H. pylori infection leads to comorbidities such as peptic ulcer, iron deficiency anaemia, atrophy of the gastric mucosa, metaplasia, dysplasia, gastric adenocarcinoma and MALT lymphoma.³ H. pylori associated with chronic gastritis is more frequent in developing countries and populations with poor hygiene conditions and is one of the most important factors for the development of these lymphomas. This is thought to be the consequence of clonal lymphocyte stimulation due to inflammation and permanent lymphocyte infiltration of the gastric mucosa.^{1 4} Extragastric MALT lymphomas have been also associated with autoimmune diseases involving an inflammatory substrate. These lymphomas also appear more frequently in immunosuppressed patients and in children, especially in those with HIV infection.²

Most paediatric MALT lymphomas are localised and indolent. The first choice treatment when the gastrointestinal tract is the affected site is eradication treatment against H. pylori. Cure rates are high and the prognosis is good, as in our patient. However, the literature lacks reports of this disease except those describing isolated cases in children. The resolution of some extragastric MALT lymphomas by eradication of H. pylori has also been observed. If progression to high-grade lymphoma occurs, additional systemic treatment would need to be added.²

The main interest of this clinical case is that, as far as we know, no relationship between MALT gastric lymphoma and eosinophilic oesophagitis has been reported in the literature neither in children nor in adult patients.

Although coexisting H. pylori infection is common in patients with eosinophilic oesophagitis as in our patient, numerous recent studies suggest an inverse relationship between both conditions. In effect, this was confirmed in a recent meta-analysis,⁵ and this inverse relationship has also been seen between H. pylori infection and asthma and other allergies.⁶ The way H. pylori infection protects against the development of eosinophilic oesophagitis remains unclear although it is thought to be related to immunomodulation events associated with the bacterium. Some authors propose that chronic H. pylori infection provokes constant Th1 response activation and consequently inhibition of the Th2 response, reducing the overall likelihood of an allergic condition, in particular, eosinophilic oesophagitis.⁷

Other authors addressing the mechanisms underlying this association^{5 8} have proposed a role of other factors. Among others, we should mention an inverse epidemiological relationship, as eosinophilic oesophagitis appears more frequently in persons of higher socioeconomic level (hygiene hypothesis) and in those with allergies, while H. pylori infection is associated with poor hygiene conditions and a lower socioeconomic status. Further studies are needed to clarify the mechanisms underpinning this relationship.

In contrast, allergic diseases are frequently associated with eosinophilic oesophagitis. Between 50% and 81% of patients with eosinophilic oesophagitis have an associated allergic disease such as atopic dermatitis, rhinitis, asthma or a food allergy mediated by IgE class antibodies.⁹

Several studies have also shown that oral immunotherapy used in patients with IgE-mediated food allergies (especially those to egg or cow’s milk proteins) can trigger eosinophilic oesophagitis.^{10 11} To date, however, the relationship between eosinophilic oesophagitis and sublingual immunotherapy has not been explored.

In our case, the patient’s mother suspected a possible relationship between onset of symptoms and this treatment. So, once diagnosis of eosinophilic oesophagitis was confirmed, sublingual immunotherapy was discontinued. In animal models, a causal relationship has been effectively confirmed between the two conditions.¹¹ In human clinical practice, while this vaccine is very safe, there are several cases in the literature suggesting that eosinophilic oesophagitis may be related to sublingual immunotherapy. These cases include the use of therapy with different antigens such as cedar, hazelnut, birch and alder, among others.

In etiopathogenical terms, both oral and sublingual immunotherapy could exert a constant antigenic stimulus in the vicinity of the oesophagus for a period of time, which would elicit the production of cytokines, such as IL5, causing local activation of the eosinophilic response.^{12 13} There could be two reasons for not having considered this possible association until a few years ago: (1) studies examining the safety of these vaccines have only been carried out in adults in whom the symptoms of eosinophilic oesophagitis can be nonspecific making it difficult to diagnose and (2) it could be several years after the antigen stimulus that the disease arises.¹³

Although more studies are needed to confirm this association, some research groups propose eosinophilic oesophagitis as a contraindication for sublingual immunotherapy¹⁴ and warn about the need to monitor oesophagitis symptoms in patients who undergo such treatments.^{11 12}

Furthermore, while subcutaneous immunotherapy does not appear to be as strongly linked to the development of eosinophilic oesophagitis, its use is limited in patients with this pathology, although it could be used in patients with eosinophilic oesophagitis and allergic rhinoconjunctivitis of poor control. In fact, this is presently considered a therapeutic option pending further studies.¹⁴

Finally, we want to highlight the important role of endoscopic procedures in the diagnosis of severe gastrointestinal diseases in patients with gastrointestinal symptoms.

Learning points.

• While the incidence of mucosa-associated lymphoid tissue (MALT)-type gastric lymphoma in children is very low, it can be an incidental finding during an endoscopic study and can be related to Helicobacter pylori infection.

• Despite recent studies suggesting an inverse relationship between H. Pylori and eosinophilic oesophagitis, this relationship warrants further investigation.

• A relationship between MALT-type gastric lymphoma and eosinophilic oesophagitis has not been previously reported.

• Further studies are needed to establish whether the onset of eosinophilic oesophagitis could be related to sublingual immunotherapy, as confirmed for oral immunotherapy.

• Withdrawal of sublingual immunotherapy should be considered in atopic patients at the time of eosinophilic oesophagitis diagnosis.