Eosinophilic Oesophagitis Assessment With String Testing

Eosinophilic oesophagitis (EoE) is a chronic progressive inflammatory condition of the oesophagus first described in 1978 but not recognised as distinct diagnosis until 1993, affecting both children and adults. It is characterised by eosinophilic infiltration of the oesophageal epithelium and can lead to difficulty swallowing, oesophageal strictures, food bolus obstruction, and symptoms mimicking gastro-oesophageal reflux disease (GORD). 23% adults with dysphagia and 50% with food bolus obstruction are diagnosed with EoE, meaning it is a common condition encountered by Gastroenterologists. Recent systematic reviews have estimated that EoE has a pooled prevalence of 28-34 per 100,000 population, but the incidence has risen remarkably over the past four decades.

EoE can occur at any age, with a rising incidence in children and adults aged 30-50 years and studies have shown that it can significantly impact on health related quality of life by impairing social and psychological functioning. It is more common in males and in those with allergic conditions such as asthma, eczema and rhinitis, suggesting a food or aeroallergen trigger. Diagnosis is obtained by gastroscopy with a minimum of six oesophageal biopsies recommended by international guidelines, due to its potentially patchy distribution. Hallmark endoscopic appearances of the oesophagus have been described and include oesophageal rings, longitudinal furrows, oedema, exudate, and narrowing. Oesophageal biopsies must show > 15 eosinophils (eos) per high powered field (HPF) infiltrating the oesophageal mucosa, as to not be mistaken with other conditions such as GORD which may also exhibit eosinophilic infiltration but at lower numbers. Other histological features include eosinophil micro-abscesses, basal zone hyperplasia, dilated intercellular spaces, eosinophil surface layering, papillary elongation and lamina propria fibrosis.

Once a diagnosis has been made, treatment of EoE is with food elimination diets, proton pump inhibitors (PPI), or topical glucocorticoids, such as budesonide orodispersible tablet (BOT). In a small subgroup of patients, immunomodulatory drugs such as azathioprine or 6-mercaptopurine might be required, and research is ongoing into monoclonal antibody therapy. The aim of treatment, as described in recent European and American Gastroenterological Association (AGA) and Joint Task Force on Allergy Immunology Practice Parameters (JTF) guidelines, is to alleviate symptoms, prevent stricture formation and achieve histological remission with < 15 eos/HPF infiltration.

Previous studies have shown that PPI may induce remission in 42% patients. Budesonide orodispersible table (BOT) may induce remission in 85% patients, but can be associated with the side effect of candidiasis in up to 10% patients. An empiric 6-food (dairy, wheat, soy, eggs, nuts, and seafood/shellfish) elimination diet induces histological remission in around 75% of patients, whereas a 4-food elimination diet (dairy, wheat, soy, and egg) induced histological remission in around 50% patients, and a 2-food elimination diet (dairy and wheat) may be effective for 40% patients. Elemental diet (amino acids) induces histological remission in 90% of patients but is impractical and only recommended for consideration after failure of medical therapy or elimination diets. Prolonged avoidance of dietary triggers may lead to drug-free sustained clinical and histological remission. Once a stricture has developed then endoscopic balloon dilatation may be required and can improve dysphagia in up to 75% of adults with oesophageal strictures, and is relatively safe with a perforation rate of only 1%, but has no effect on the underlying inflammatory process.

Systematic review with meta-analysis found no statistical difference between PPI and BOT, however there are few head to head studies comparing the treatments and long term evidence for BOT is still lacking; indicating the need for further prospective studies. European guidelines therefore currently do not favour one treatment over another and state that the choice of therapy should be individually discussed with the patient and may be interchangeable over time. Treatment effectiveness should be assessed with repeat gastroscopy with oesophageal biopsies after 6-12 weeks.

Gastroscopy involves placing a thin flexible camera into a patient's mouth and advancing it to the oesophagus, stomach and duodenum. The procedure takes approximately 5 minutes, during which biopsy forceps are placed through a 2.8mm channel on the standard gastroscope to obtain oesophageal mucosal biopsies for histological analysis. Although topical anaesthetic may be applied to the back of the throat and intravenous sedation given, for many patients, gastroscopy remains an unpleasant and daunting experience. Transnasal endoscopy (TNE) has more recently been developed and potentially offers patients a more comfortable procedure due to the reduced diameter of the gastroscope, but due to the need for lifelong surveillance, there has been a move for even less invasive and less costly methods.

The GI string test (EnteroTestTM) was first developed in 1970 for sampling microorganisms from the small intestine. The original test involved swallowing a small, weighted gelatin capsule to which a string was attached and the other end taped to the patients cheek. The capsule reaches the duodenum where it detaches and is excreted naturally and the string left in situ overnight. The string is then pulled out via the mouth and can be analysed. Over the years it has been used to analyse biliary composition, culture intestinal tuberculosis, and profile the distinct microbiome in the oesophagus using 16S rRNA gene sequencing, as accurately as with mucosal biopsy.

The potential for oesophageal string testing (EST) to monitor EoE activity has been suggested by a previous study using an overnight swallowed string test in 41 children correlated with EoE activity on biopsy, which could distinguish between active EoE, treated EoE, GORD and normal oesophagus, using measured ESGP (eosinophil secondary granule proteins) - including major basic protein 1 (MBP1), esosinophil peroxidase (EPX), eosinophil derived neurotoxin (EDN) and eosinophil cationic protein (ECP). Also measured was the Chacot-Leyden crystal protein/galectin-10 (CLC/Gal-10), which is a primary cytosolic protein but is also present in the eosinophil's primary (coreless) granule. Each granule-associated protein carries different biochemical and functional properties relevant to proposed pathogenic mechanisms for oesophageal dysfunction and remodelling in EoE. These biomarkers were found to be elevated uniquely in active EoE, compared to treated EoE, GORD and normal controls.

Another previous study performed a one hour string test on 134 people with EoE or healthy control, comparing levels of eosinophil associated protein (EAP) including eotaxin-2 (Eot2), eotaxin-3 (Eot3), and MBP on EST to mucosal biopsy. Ackerman showed that these EST-captured eosinophil-associated biomarkers were proportionate on EST and mucosal biopsy levels, and also correlated with EoE activity as defined by mucosal eosinophil counts. These proteins on EST therefore had excellent sensitivity and specificity for identifying active EoE. They also surveyed patients, and 92% adults preferred the EST over gastroscopy to monitor their EoE.

Whilst these studies have shown the potential of intraluminal oesophageal biomarkers to reflect EoE activity, additional research is required to further characterise their relation to disease activity, correlation with symptoms and response to different treatments, which has not yet been described in the literature. This knowledge may improve our understanding of the pathogenesis of EoE and lead to new therapeutic targets in future, as well as reduce the need for surveillance endoscopy in EoE, which would be safer and more convenient for patients, and offer cost savings for healthcare resources.

The role of the microbiome is increasingly been appreciated in the development of immune mediated disease such as EoE, yet to date there are only a handful of studies on the oesophageal microbiome, finding distinct changes in EoE compared to GORD, adenocarcinoma and healthy patients. Further research in this area would be interesting to establish if the microbiome is different in active EoE vs remission, and how this may affect disease severity and symptoms, which again may lead to new therapeutic targets in future.

In addition, it is not known whether people with EoE have altered sensitivity in their oesophagus as part of the remodelling from the chronic inflammatory process. If sensation is increased this could lead to increased symptoms and if it is reduced then this coud lead to undiagnosed conditions including strictures and GORD. The calcitonin gene related peptide has recently been shown to be a marker of oesophageal sensory nerves, and has been shown to be closer to the oesophageal lumen in oesophageal hypersensitivity, however its relation to EoE has not yet been examined.

In this study the investigators will sample oesophageal biomarkers and the microbiome from adults with EoE, using a 30 minute string test, which will be performed on the same day as their routine gastroscopy an addition to validated symptom questionnaires. The investigators will repeat the string test after treatment courses with PPI, budesonide or dietary elimination, alongside their routine gastroscopy. During their gastroscopy the investigators will also take biopsies to check for the density and positioning of CGRP, to see if this is different in active and inactive EoE. The aim is to correlate biomarkers, CGRP and microbiome with endoscopic EoE activity and symptoms, to assess for changes in response to different treatments. and again may provide further insight into disease pathogenesis and lead to new therapeutic targets in future.