Society: AGA

Background: Novel treatment options for eosinophilic esophagitis (EoE) are needed. Previous studies have assessed mepolizumab (mepo), a monoclonal antibody against IL-5, with mixed results in EoE, and the efficacy of mepo in an adult and adolescent population has yet to be fully examined.

Aim: To determine whether mepo is more effective than placebo for improving symptoms of dysphagia and decreasing esophageal eosinophil counts in EoE.

Methods: We conducted a multicenter, randomized, double-blind, placebo-controlled, investigator-initiated clinical trial comparing mepo to placebo (PBO) (NCT03656380). Patients aged 16-75 with EoE, PPI non-response, ≥15 eos/hpf on biopsy at screening, and active symptoms of dysphagia (EoE Symptom Activity Index [EEsAI] score ≥27) were eligible. Key exclusion criteria were: inability to pass a standard endoscope due to severe stricturing, and esophageal dilation or systemic steroids within 8 weeks, or topical steroids (tCS) within 4 weeks, of baseline endoscopy. Patients were randomized 1:1 to 3 months of either mepolizumab 300mg SQ monthly or matching PBO SQ, after stratification for prior tCS non-response. Primary outcome was change in dysphagia as measured by EEsAI from baseline to month 3 (M3). Secondary outcomes included change in peak eosinophil counts, histologic response thresholds, change in EoE Endoscopic Reference Score (EREFS), EEsAI thresholds, and safety.

Results: Of 66 patients randomized, 64 completed M3 and were analyzed. Baseline characteristics were generally similar between study groups (Table 1). Of note, >80% of subjects had been treated with tCS, 50% had not responded, and >70% had previously had dilation. At M3, EEsAI decreased 15.4±18.1 points with mepo and 8.3±18.0 with PBO (p=0.14; Fig 1A). While 6% in each group had a M3 EEsAI score ≤20 (clinical remission; p=0.95), 35% on mepo had an EEsAI score decrease ≥20 (clinical response) vs 21% with PBO (p=0.20). At M3, the peak eosinophil count decreased with mepo (113±77 to 36±43) and increased (146±94 to 160±133) with PBO (p<0.001 for post-treatment comparison; Fig 1B). With mepo, 42% and 34% achieved histologic responses of <15 and ≤6 eos/hpf compared to 1% and 3% respectively, with PBO (p<0.001 and 0.02; Fig 1C). The change in EREFS at M3 was also larger with mepo than PBO (1.0±1.1 vs 0.4±1.3; p=0.03; Fig 1D). Mepo was generally well-tolerated, with no medication-related SAEs or new safety signals; the most common AEs were injection site reactions.

Conclusions: In this population of previously difficult to treat EoE patients, mepo as stand-alone therapy for 3 months led to a numerical but not statistically significant improvement in dysphagia symptoms, and significant improvements in eosinophil counts and endoscopic severity, as compared to PBO. Longer-term treatment effects and predictors remain to be assessed.

Background: Short-term use of mepolizumab (mepo), a monoclonal antibody against IL-5, has been previously assessed in eosinophilic esophagitis (EoE). However, the optimal dose and length of treatment for mepo in EoE has not been extensively investigated.

Aim: To determine the durability of symptomatic and histologic response to mepo after 6 months of treatment in patients initially randomized to active medication, and to assess response to 3 months of a lower mepo dose in those initially randomized to placebo.

Methods: In the first phase of a multicenter, randomized, double-blind, placebo-controlled, investigator-initiated clinical trial, we compared mepo 300mg SQ monthly (mepo300) to placebo and assessed outcomes after 3 months of treatment (NCT03656380). In the second phase, patients initially randomized to mepo continued 300mg monthly dosing for an additional 3 months, and those initially on placebo started mepo 100mg monthly (mepo100) for 3 months; allocation remained blinded. Eligible patients were aged 16-75 with EoE, had PPI non-response, ≥15 eos/hpf on biopsy at screening, and active symptoms of dysphagia (EoE Symptom Activity Index [EEsAI] score ≥27). The primary outcome for the first phase was change in dysphagia as measured by EEsAI from baseline to month 3 (M3). For the second phase, this same outcome was assessed at month 6 (M6). Other outcomes assessed at M6 included change in peak eosinophil counts, histologic response thresholds, change in EoE Endoscopic Reference Score (EREFS), EEsAI thresholds, and safety.

Results: Of 64 patients who completed M3, 59 entered phase 2, and 56 completed M6 and were analyzed (28 in each group). At M6, EEsAI decreased 18.6±19.2 points from study baseline with mepo100 and 18.3±18.1 with mepo300 (p=0.85), a change of 10.1 and 3.1 points from M3 baseline, respectively (Table). Rates of clinical remission and response at M6 were the same in each group (18% and 46%, respectively). From study baseline to M6, the peak eosinophil count decreased to 50±42 eos/hpf with mepo100 and 26±20 with mepo300 (p=0.008 for post-treatment comparison), but the absolute change was slightly larger with mepo100 (-102±83 vs -88±77; p=0.04). At M6, 21% and 32% had histologic responses of <15 eos/hpf in mepo100 and mepo300, respectively (p=0.37; Table). The change in EREFS at M6 was similar with mepo100 and mepo300 (0.9±1.1 vs 0.5±1.5; p=0.26). Mepo was generally well-tolerated with longer use, with no medication-related SAEs or new safety signals.

Conclusions: Use of 300mg monthly of mepo for 6 months did not lead to additional symptom, endoscopic, or histologic improvement compared to 3 months of use, but responses were generally maintained. Use of 100mg monthly of mepo for 3 months yielded similar improvements to the higher dose. Future studies can investigate how mepo is best positioned in EoE treatment algorithms.

Background:
Eosinophilic esophagitis (EoE) is a predominant cause of food impaction and dysphagia. Patients with undiagnosed or poorly controlled EoE may require emergency department (ED) visits for management of dysphagia or food impactions. The impact of increasing EoE incidence on emergency services remains unknown given ongoing rise in incidence and prevalence of this disorder.
Aims:
We aimed to characterize the burden of EoE on ED visits in the United States (US). We examine trends in ED utilization for adult and pediatric patients with EoE, assess endoscopic, inpatient, and financial resources in EoE patients seeking urgent care, and forecast EoE-associated ED visits to 2030.
Methods:
Data from the US-based Nationwide Emergency Department Sample were used to estimate weighted annual EoE-associated ED visits from 2009-2019. We included patients with either a primary diagnosis of EoE or diagnosis of EoE within the first three diagnostic positions, based on International Classification of Disease (ICD)-9 (530.13) or ICD-10 (K20.0) codes. Data from 2015 and 2016 were excluded due to the transition from ICD-9 to ICD-10 codes and possible misclassifications. Comparisons between covariables were evaluated using the survey-adjusted Pearson χ2 test, adjusted Wald test, or univariable survey-weighted logistic regression. Temporal trends in population-adjusted rates were assessed using joinpoint regression. An autoregressive integrated moving average model was used to forecast ED visits to 2030. We evaluated endoscopic utilization, need for hospitalization, and ED-related charges.
Results:
Baseline demographic and ED visit characteristics are summarized in Table 1. 280,394,748 unweighted visits were sampled, representing approximately 1.2 billion weighted visits. A total of 49,507 weighted ED visits for EoE were included in the analysis, after exclusions for diagnoses associated with secondary esophageal eosinophilia.
The annual volume of EoE-associated ED visits increased from 2,934 [95% CI: 2,437-3,431] in 2009 to 8,765 [95% CI: 7,514-10,015] in 2019 and is forecasted to reach 15,445 [95% prediction interval PI: 14,672-16,218] by 2030 (Figure 1). From 2009 to 2019, the number of EoE-associated ED visits increased by an average of 11.5%/year [95% CI: 10.3%, 12.7%]. The proportion of patients admitted to the hospital decreased from 25.6% in 2009-2011 to 14.0% in 2017-2019. Half of EoE patients presenting to ED required endoscopy, and nearly 40% required esophageal foreign body removal. In 2019, the total mean inflation-adjusted cost for an EoE-associated ED visit was $9,025 (USD).
Conclusions:
The volume of EoE-associated ED visits tripled between 2009 to 2019 and is projected to further double by 2030. This represents a substantial burden of unanticipated healthcare resource utilization and highlights a potential opportunity to optimize outpatient EoE care.

<b>Table 1.</b> Patient demographic and hospital characteristics from emergency department encounters for eosinophilic esophagitis, Nationwide Emergency Department Sample 2009-2014 and 2017-2019

Table 1. Patient demographic and hospital characteristics from emergency department encounters for eosinophilic esophagitis, Nationwide Emergency Department Sample 2009-2014 and 2017-2019

<b>Figure 1.</b> Trends and projections in emergency department visit volume for eosinophilic esophagitis (2009-2030)

Figure 1. Trends and projections in emergency department visit volume for eosinophilic esophagitis (2009-2030)

Introduction: Although eosinophilic esophagitis(EoE) is associated with atopy, EoE has been reported to be associated with immunoglobulin G4(IgG4) antibodies rather than immunoglobulin E(IgE). Food antigens are known to play a key role in driving EoE with 52-81% of paediatric and adult patients responding to empirical elimination diets. IgG4 specific to usual EoE food triggers (dairy, wheat, soy, egg, nuts and seafood) have been demonstrated in EoE patients. We sought to (i) compare serum food-specific IgG4 (FS-IgG4) levels of EoE patients with healthy controls and (ii) assess response to food-specific serum IgG4 led elimination diet in patients with EoE.
Method: Prospective trial at a single tertiary centre with serum FS-IgG4 and IgE from healthy controls and patients with active EoE. All patients underwent endoscopy (8 biopsies in total from proximal and distal oesophagus) with oesophageal eosinophil count and IgG4 staining. Dysphagia symptom questionnaire score and skin prick testing were performed. Patients with elevated FS-IgG4 (measured using Immunocap, using a cut-off of 10mgA/L) were commenced on elimination of these foods for 6 weeks. Follow-up serum FS-IgG4 and endoscopic and histologic examination done at 6 weeks.
Results: Seven controls and 24 patients with EoE were recruited, however 3 patients had inactive EoE (<15 eosinophils per HPF) and therefore not included. Serum FS-IgG4 to milk, wheat, soy, nuts and seafood were significantly higher in EoE patients compared to controls (p=0.008, p=0.003, p=0.012, p=0.004 and p=0.042 respectively). Oesophageal IgG4 staining was negative in 5/5 control patients (not performed in 2) and was positive in 12/12 EoE patients (not performed in 9). Elevated serum FS-IgG4 to one or more food groups (median 2 food groups) were detected in 20/21 patients with EoE. Elevated FS-IgG4 to milk was found in 17, to wheat in 11, 7 to egg white, 1 to soy and 1 to nuts. Sixteen patients have completed 6 weeks of dietary elimination. Median reduction in DSQ score after dietary elimination was 12 (p=0.023). EoE endoscopic reference score fell by a median of 1 (p<0.001). Eight (50%) patients had histological remission (<15 eosinophils per HPF) and a further 2 patients had a reduction in eosinophil count. Oesophageal eosinophil count fell by a median of 33 (p=0.066). Serum FS-IgG4 did not decline significantly in both responders and non-responders.
Conclusion: Food specific IgG4 particularly to milk, wheat, soy and nuts were present at higher levels in adults with EoE than in control adults. FS-IgG4 led targeted elimination diet for 6 weeks in patients with EoE resulted in histological remission in 50%. When successful, FS-IgG4 targeted elimination diet can negate the need for medications and be viewed more favourably by patients due to its smaller endoscopic burden compared to empirical elimination diets.

Introduction: Esophageal dilation may provide rapid symptomatic improvement in patients with eosinophilic esophagitis (EoE) and esophageal strictures but does not modulate the underlying inflammation.Furthermore, dilation may lead to dissociation between histology and symptoms, potentially impacting interpretation of clinical trial data.Results from the 3-part, phase 3 LIBERTY-EoE-TREET study (NCT03633617) demonstrated that weekly (qw) dupilumab 300 mg, an inhibitor of the shared receptor component for IL-4 and IL-13, improved histologic, symptomatic, and endoscopic aspects of EoE at Week 24 and was generally well tolerated in patients ≥12 years. The objective of this pre-specified analysis was to assess the efficacy of dupilumab 300 mg qw vs placebo at Week 24 in patients from Parts A and B with and without prior history of esophageal dilation.
Methods: Patients with and without a history of dilation were stratified for analysis. Of note, patients were excluded from study enrollment if they required dilation during the screening endoscopy. Concomitant dilation during study treatment was prohibited but rescue dilation was permitted. Primary endpoints were the proportion of patients achieving histologic remission (peak esophageal intraepithelial eosinophil [eos] count ≤6/high-power field [hpf]) and absolute change from baseline in Dysphagia Symptom Questionnaire (DSQ) score. Key secondary endpoints included percent change in peak eos count and absolute change in EoE-Endoscopic Reference Score and EoE-Histologic Scoring System grade and stage scores; all at Week 24.
Results: In Parts A and B, respectively, 42.9% and 32.5% of patients in the dupilumab group and 43.6% and 41.8% in the placebo group had a history of dilation. In Parts A and B, respectively, the mean (standard deviation [SD]) number of prior dilations were 1.9 (1.2) and 2.5 (3.1) in the dupilumab group and 2.0 (1.4) and 2.2 (2.8) in the placebo group; days since last dilation were 672.6 (767.2) and 818.8 (962.0) in the dupilumab group and 547.9 (729.9) and 1343.5 (2035.1) in the placebo group. Dupilumab improved outcomes vs placebo for primary and key secondary efficacy endpoints, with comparable results in patients with/without history of dilation (Figures 1 and 2). Of note, symptoms of dysphagia (DSQ score) improved similarly with dupilumab vs placebo, regardless of dilation history (least squares mean change Part A: –18.79 vs –6.53 and –20.79 vs –9.07 in patients with and without history of esophageal dilation, respectively; Part B: –25.24 vs –12.16 and –24.28 vs –16.66 (Figure 1B). Dupilumab was generally well tolerated.
Conclusion: Dupilumab demonstrated substantial improvements in histologic, symptomatic, and endoscopic aspects of EoE in adults and adolescents to 24 weeks, regardless of prior esophageal dilation status.

Background: The Index of Severity for Eosinophilic Esophagitis (I-SEE) has been recently proposed as a metric to capture disease severity in EoE. While it has been preliminarily evaluated in adults, it has yet to be assessed in a pediatric population.

Aims: To assess disease severity of pediatric EoE patients at the time of diagnosis, following initiation of treatment, and over time, as defined by I-SEE.

Methods: We performed a retrospective cohort study of pediatric EoE patients prospectively enrolled into a UCSD/Rady longitudinal database. Patients had a diagnosis of EoE, were enrolled between 2011-2018, maintained follow-up for at least 2 years, and had at least 3 EGDs. Data were extracted from the database and medical record to calculate I-SEE. Features of symptom frequency, complications (i.e. impactions, malnutrition, EoE-related hospitalization, and treatment refractory disease), and inflammatory/fibrostenotic features from both endoscopic and histologic findings were obtained. I-SEE was calculated at three time-points: initial endoscopy, second endoscopy, and last endoscopy. Patients were classified as mild (1-6 points), moderate (7-14), severe (≥15), or inactive (0). We analyzed clinical, endoscopic, and histologic features at each instance/endoscopy by disease severity. We analyzed change in severity between 1^st and 2^nd and 1^st and last EGDs.

Results: Of 67 patients evaluated (mean age 5.2yrs, 22% female, 78% white), the mean I-SEE was 9.7 at baseline with 29 (43%) classified as mild, 24 (36%) moderate, and 14 (21%) severe. Baseline severity was associated with younger age, lower BMI percentile, and difficulty feeding (p<.05 for all, Table 1). At the second EGD, most patients had inactive (9%) or mild (66%) disease (p<0.001 compared to baseline), with accompanying decreased symptom frequency and endoscopic inflammatory features (Table 2). By the last EGD (mean 6.6±2.2 yrs from the first), the mean I-SEE score decreased to 3.9 (p<0.001) with 88% of patients in the inactive or mild disease categories (p<0.001). Malnutrition significantly decreased over this time (15% to 1%; p=0.003), as did symptom frequency (57% asymptomatic; p<0.001), endoscopic inflammatory features (63% with normalization; p<0.001), eosinophil counts (69% with <15 eos/hpf; p<0.001), and histologic lamina propria fibrosis (84% to 49%; p<0.001) (Table 2).

Conclusions: I-SEE reflected longitudinal EoE disease severity features in children. At baseline, severe patients were younger with lower BMI, and difficulty feeding. Chronic therapy improved EoE severity over time in this cohort followed for an average of more than 6 years. I-SEE is responsive to therapy and has clinical utility in children.

<b>Table 1: Baseline characteristics by EoE severity</b><br /> <br /> *ANOVA for continuous variables and chi squared for categorical variables

Table 1: Baseline characteristics by EoE severity

*ANOVA for continuous variables and chi squared for categorical variables

<b>Table 2: Comparison of severity index features and severity classifications (compared 1<sup>st</sup> and 2<sup>nd</sup> instances as well as 1<sup>st</sup> and last instances). </b><br /> <br /> *Paired t-tests for continuous outcomes and cumulative logit model for multi-category ordinal outcomes and with McNemar’s test for dichotomous outcomes comparing 1<sup>st</sup> and 2<sup>nd</sup> instances.<br /> ** Paired t-tests for continuous outcomes and cumulative logit model for multi-category ordinal outcomes and with McNemar’s test for dichotomous outcomes comparing

Table 2: Comparison of severity index features and severity classifications (compared 1^st and 2^nd instances as well as 1^st and last instances).

*Paired t-tests for continuous outcomes and cumulative logit model for multi-category ordinal outcomes and with McNemar’s test for dichotomous outcomes comparing 1^st and 2^nd instances.
** Paired t-tests for continuous outcomes and cumulative logit model for multi-category ordinal outcomes and with McNemar’s test for dichotomous outcomes comparing

Evan Dellon

University of North Carolina School of Medicine

Appearances