See “Are We Making Progress in the Hunt for Biomarkers that Could Differentiate Proton Pump Inhibitor-Responsive Eosinophilic Esophagitis?” by Sayej on page 173.
What Is Known/What Is New
What Is Known
- Proton-pump inhibitor (PPI) is a proven treatment method for eosinophilic esophagitis (EoE).
- Prevalence of PPI-responsive EoE in children ranges greatly from 30% to 71%.
- There are no predictive tests to determine the best treatment strategies for EoE.
- Eosinophil-derived neurotoxin (EDN) has been shown to be a viable measure of disease activity in EoE.
What Is New
- EDN levels are lower in patients with EoE who are responsive to PPI treatment than in EoE patients who do not respond to PPI.
- EDN is more predictive of PPI-responsiveness in EoE than either endoscopic reference score or peak eosinophilic count.
- EDN measurement is a viable tool to help guide treatment strategies for patients with EoE.
Eosinophilic esophagitis (EoE) is a chronic immune antigen-mediated disease causing esophageal dysfunction due to an eosinophil predominant inflammatory response in the mucosa of the esophagus (1). Since its first identification in the early 1990s, the EoE incidence has increased, with recent pediatric studies showing a prevalence of up to 34.4 cases per 100,000 (1–3). Diagnostic criteria has evolved over time and there are no tests or patient factors to predict the best treatment approach. The endoscopic reference score (EREFS), peak eosinophil count (PEC), and peripheral eosinophilia have all been studied, but have not been shown to accurately predict treatment response or disease progression (4–6). Treatment with proton-pump inhibitors (PPI), swallowed steroids, and diet elimination therapy have all shown efficacy, but require adjustment periods as well as response monitoring which includes the need for repeat endoscopies (1,7). Endoscopy is invasive and confers increased risk, discomfort, and costs for our pediatric patients and their caregivers (8). In many cases, PPI therapy is trialed first as it is safe and simple in comparison to swallowed steroids or diet elimination; however, many do not respond and require escalation of treatment. It would be beneficial to find a test to help guide clinicians to determine who may benefit from a PPI trial versus who may need to go straight to other therapies.
Eosinophil-derived neurotoxin (EDN) from esophageal brushing samples is a viable measure of disease activity in EoE (9). An EDN level of 10 mcg/mL or higher is highly sensitive (97%) and specific (89%) for diagnosis of active EoE (9). EDN levels have not been previously studied as a predictive tool in EoE. It is possible that certain levels of EDN, at baseline endoscopy, could indicate whether a patient may or may not respond to PPI therapy. Therefore, our aim is to determine if EDN via esophageal brushing can be used to predict proton-pump inhibitor (PPI)-responsiveness in pediatric patients with EoE.
This was a prospective cross-sectional pilot study in children and young adults (<21 years old) who underwent routine endoscopy with biopsies and esophageal brushing at the EoE center in Arnold Palmer Hospital for Children between 2018 and 2020. Patients were enrolled if they underwent upper endoscopy for suspicion of EoE. and met criteria for EoE based on consensus recommendations of histological evidence of eosinophils on esophageal biopsy. Enrolled patients with active EoE were then placed on high dose PPI (1 mg/kg per dose twice per day) as their sole therapy and underwent repeat endoscopy to determine PPI-responsiveness. After the second endoscopy, subjects were divided into PPI-responsive (PPI-R) and nonresponsive (PPI-NR) EoE groups based on histological findings. EDN was measured at baseline endoscopy, before any treatment, and at follow up endoscopy, after PPI therapy.
Patients were excluded if they did not meet diagnostic criteria of EoE or were found to have generalized eosinophilic gastrointestinal disease. Exclusion criteria also included any history of PPI use, steroids, or diet elimination within 6 months before endoscopy. The study was reviewed and approved by the Institutional Review Board at Orlando Health Arnold Palmer Hospital for Children, Orlando, FL, USA.
Active EoE was defined based on symptoms of esophageal dysfunction and increased esophageal intraepithelial eosinophilia (>15 eos/hpf) in at least one biopsy of the esophagus during esophagogastroduodenoscopy (EGD) (1). PPI-responsive EoE was defined by improvement of esophageal eosinophilia to less than 15 eso/hpf in all biopsies upon repeat endoscopy done at least 6 weeks after baseline endoscopy. High dose PPI was defined as 1 mg/kg per dose twice per day.
Data collected included clinical symptoms, diet, medications, comorbidities, EREFS score at time of endoscopy, PEC, and EDN. Both esophageal brushing and esophageal biopsies were obtained during time of baseline and follow up endoscopy. EDN was measured in esophageal epithelial brushing samples by enzyme-linked immunosorbent assay (ELISA), as described previous by Smadi et al (9).
Esophageal brushing was performed before obtaining biopsy samples. A standard cytology brush (Kimberly-Clark #60314) was passed through the endoscopy channel and the entire distal, mid, and proximal esophagus was brushed under direct visualization. To improve the contact of the brush with the epithelial lining of the esophagus, the esophagus was moderately deflated before brushing. The brush was then removed, and the head of the brush was clipped off (approximately 3–4 cm from the brush end), placed in an empty tube, frozen immediately, and transferred to our laboratory for analysis later in batches.
A minimum of four biopsies were obtained from various esophageal levels during EGD. Additional biopsies were obtained at the discretion of the performing physician. Biopsy samples were collected and sent to pathology for standard processing and microscopical observation. Pathology reports were obtained from chart review.
Eosinophil-Derived Neurotoxin Analysis
We followed the same protocol as described by Smadi et al (9) for processing, storage, and analysis of brushing samples. The frozen samples were processed in batches at our Pediatric Gastroenterology Translational Laboratory. Samples were thawed on ice for 15 minutes, then centrifuged for 1 minute at 3500 RPM (swing bucket rotor), and the volume of the extracted fluid was measured. The brush was removed from the plastic sheath and the sheath was flushed twice with a measured volume (135 μL × 2 = 270 μL) of extraction buffer (0.5% bovine serum albumin, 10 mM ethylenediaminetetraacetic acid, 10 mM ethylene glycol tetraacetic acid, 1% protamine sulfate, 0.5 M NaPO4, 1% Triton X-100). The sheath was then discarded. A protease inhibitor cocktail (Sigma-Aldrich I3911-1BO) of 30 μL was added for a final concentration of 1:10. The brush with the extraction buffer was then mixed by vortex at high speed for 1 minute and centrifuged at 3500 rpm (swing bucket rotor) for 5 minutes. The extracted brushing sample was then transferred to a 1.5 mL sterile microcentrifuge tube either for immediate EDN analysis or for storage in a freezer at −20°C for less than 7 days or at −80°C for long-term storage. The EDN concentration in the serially diluted suspension was measured using ELISA as per the manufacturer's instruction (MBL International, Woburn, MA, USA; EDN Kit #7630) using wells coated with anti-human EDN monoclonal primary antibody. The absorbance in each microwell was measured at 450 nm using a microplate reader. The EDN concentration was calculated using the standard curve.
Patient age is reported as mean with standard deviation (mean ± SD), whereas sex and medications are reported as counts with percentages (n; %). Data were non-normal (Shapiro–Wilk test values were <0.05). Accordingly, all statistical testing was conducted with a non-parametric test and reported as mean with standard deviation. Differences in levels of EDN, PEC, and EREFS were assessed with the Mann–Whitney U test. Changes in EDN levels before and after treatment were conducted with a Wilcoxon signed-rank test, separately for PPI responsive and nonresponsive patients. Statistical testing was conducted in SPSS 27.0 (two-tailed; P < .05) was used for statistical significance.
Thirty-six subjects were enrolled with EoE (mean age 11.6 years, range 2–19 years). Fifteen subjects (42%) were found to have PPI-R EoE and 21 (58%) were PPI-NR. There were no statistically significant age differences between groups (Table 1). Twenty-one patients (58%) were found to have an existing atopic comorbidity, with some patients having more than one.
TABLE 1 -
Demographics of study population
||PPI-responsive (n = 15)
||PPI-nonresponsive (n = 21)
||All subjects (n = 36)
|Mean age (y)
|Atopic comorbidities (n)
| Allergic rhinitis
| Milk protein Intolerance
| Allergies (other)∗
Demographics of patients enrolled in study.PPI = proton-pump inhibitor.
∗Allergies (other) included unspecified allergies on chart review.
EDN levels before and after treatment showed that among PPI-R patients there was a statistically significant change (75.7 ± 60 mcg/mL vs 3.84 ± 4.9 mcg/mL, P = 0.001), but not among PPI-NR patients (219.1 ± 229 mcg/mL vs 212.5 ± 228 mcg/mL, P = 0.877; Fig. 1).
At baseline endoscopy, EDN in the PPI-NR group was significantly higher than in the PPI-R group (219.1 ± 229 mcg/mL vs 75.7 ± 60 mcg/mL, respectively, P = 0.036; Fig. 2).
There was no statistically significant difference in PEC between PPI-NR and PPI-R groups at baseline endoscopy (62 ± 31 vs 48 ± 22, respectively, P = 0.15; Fig. 3). Similarly, there were no statistically significant differences in EREFS between PPI-NR and PPI-R groups (3.1 ± 1 vs 2.9 ± 1, respectively, P = 0.55; Fig. 3).
Most patients in the PPI-R group were prescribed omeprazole (73.3%) followed by lansoprazole (13.3%), pantoprazole (6.67%), and esomeprazole (6.7%). Prescribed treatment was similar in the PPI-NR group: omeprazole (71.4%), followed by lansoprazole (14.3%), esomeprazole (9.5%), and pantoprazole (4.8%). Dosing was similar among both groups with 1 mg/kg per dose twice per day.
This is the first study to evaluate EDN as a marker for PPI responsiveness in EoE. In the 36 patients enrolled, we found EDN levels correlated with patient response to PPI therapy. At initial endoscopy, lower EDN levels indicate a higher likelihood of PPI-R EoE. Previous studies have shown similar baseline clinical, histological, and endoscopic characteristics in PPI-responders and nonresponders (4–6,10,11). EREFS and PEC measurements were similar in our study between the two groups, adding to the evidence that they are not useful predictors of PPI responsiveness. In fact, the mean PEC was higher in the PPI-R group versus the PPI-NR group which further illustrates how this is not a reliable tool to determine treatment response. There were 58% of patients with coexisting atopic disease, but our sample size was too small to find any correlation between atopic disease and PPI-responsiveness. EDN is a much more useful marker based on these initial studies.
Response to PPI treatment has been difficult to predict and currently ranges from 30% to 71% (5,12–14). This variability is likely due to differences in diagnostic criteria among centers as well as difficulty in distinguishing between PPI-R EoE and reflux disease. Intra-epithelial eosinophilia in the distal esophagus along with clinical symptoms of dysphagia and heartburn can be indicative of both diseases. Additionally, esophageal pH monitoring has not been shown to predict PPI-responsiveness (11,15). EDN levels, however, would only be elevated in EoE. Therefore, EDN is helpful in differentiating between reflux and PPI-R EoE (9).
EDN is an easy test to obtain during endoscopy. We use a standard cytology brush that passes through the endoscopy channel and under direct visualization the distal, mid, and proximal esophagus is brushed. Our center uses a commercially available ELISA that is run in-house. We also have previously published data on the use and outcomes of EDN in esophageal brushing for EoE (9). EDN values can have a large range, and while a level of 10 mcg/mL or higher is diagnostic for EoE, levels as high as 7600 mcg/mL have been reported (9). We have found EDN to be a useful tool in general and a way to now potentially assess PPI responsiveness. There is an EoE genetic diagnostic panel offered by Cincinnati which may be able to differentiate between PPI responsive and resistant EoE (16); however, that panel is currently only available through research. It is useful for both families and clinicians to have a feasible tool to guide them in treatment strategies and to decrease the number of endoscopies.
Study limitations include a small population size from a single clinical center, and variability among the prescribed PPIs due to provider choice or insurance requirements. Despite the size, this pilot study has significant results proving the hypothesis that EDN can be used to predict responsiveness to PPI therapy. Additional studies will be needed to determine a possible cut-off value or range for EDN concentrations as a larger sample size is necessary. We are actively enrolling patients to continue this research and expand on our initial findings. Different PPIs are metabolized via different pathways, which could affect treatment response and potentially be a confounding factor altering study results (17,18). Future studies may prove that certain individuals will respond better to one PPI over another due to their genetic makeup and metabolism. PPI dosing in this study was standardized at 1 mg/kg per dose twice per day. Of the PPI prescribed, omeprazole was used 77% of the time among the two groups, making it more homogeneous and a good study of the use of omeprazole in these patients.
EDN levels measured from the esophageal epithelium obtained by brushing during endoscopy are lower in pediatric patients whose EoE responds to PPI therapy, showing EDN may be a good predictive marker for PPI responsiveness. Further studies in larger populations across different treatment centers will be useful. EDN measurement is an easy and widely available tool, can be used to individualize therapy, and may potentially reduce costs and sedation risk for pediatric patients.
1. Lucendo AJ, Molina-Infante J, Arias Á, et al. Guidelines on eosinophilic esophagitis: evidence-based statements and recommendations for diagnosis and management in children and adults. United European Gastroenterol J
2. McGowan EC, Keller JP, Dellon ES, et al. Prevalence and geographic distribution of pediatric eosinophilic esophagitis in the 2012 US Medicaid population. J Allergy Clin Immunol Pract
3. Navarro P, Arias Á, Arias-González L, et al. Systematic review with meta-analysis: the growing incidence and prevalence of eosinophilic oesophagitis in children and adults in population-based studies. Aliment Pharmacol Ther
4. Lucendo AJ, Arias A, Molina-Infante J. Efficacy of proton pump inhibitor drugs for inducing clinical and histologic remission in patients with symptomatic esophageal eosinophilia: a systematic review and meta-analysis. Clin Gastroenterol Hepatol
5. Dranove JE, Horn DS, Davis MA, et al. Predictors of response to proton pump inhibitor therapy among children with significant esophageal eosinophilia. J Pediatr
6. Ngo P, Furuta GT, Antonioli DA, et al. Eosinophils in the esophagus-peptic or allergic eosinophilic esophagitis? Case series of three patients with esophageal eosinophilia. Am J Gastroenterol
7. Hirano I, Chan ES, Rank MA, et al. AGA institute and the joint task force on allergy-immunology practice parameters clinical guidelines for the management of eosinophilic esophagitis. Ann Allergy Asthma Immunol
8. Paterson N, Waterhouse P. Risk in pediatric anesthesia. Paediatr Anaesth
9. Smadi Y, Deb C, Bornstein J, et al. Blind esophageal brushing offers a safe and accurate method to monitor inflammation in children and young adults with eosinophilic esophagitis. Dis Esophagus
2018; 31: doi: 10.1093/dote/doy056.
10. Dellon ES, Speck O, Woodward K, et al. Clinical and endoscopic characteristics do not reliably differentiate PPI-responsive esophageal eosinophilia and eosinophilic esophagitis in patients undergoing upper endoscopy: a prospective cohort study. Am J Gastroenterol
11. Molina-Infante J, Ferrando-Lamana L, Ripoll C, et al. Esophageal eosinophilic infiltration responds to proton pump inhibition in most adults. Clin Gastroenterol Hepatol
12. Shaheen NJ, Mukkada V, Eichinger CS, et al. Natural history of eosinophilic esophagitis: a systematic review of epidemiology and disease course. Dis Esophagus
2018; 31:doy015doi: 10.1093/dote/doy015.
13. Dellon ES, Hirano I. Epidemiology and natural history of eosinophilic esophagitis. Gastroenterology
14. Dellon ES. Red between the lines: evolution of eosinophilic esophagitis as a distinct clinicopathologic syndrome. Dig Dis Sci
15. Francis DL, Foxx-Orenstein A, Arora AS, et al. Results of ambulatory pH monitoring do not reliably predict response to therapy in patients with eosinophilic oesophagitis. Aliment Pharmacol Ther
16. Wen T, Stucke EM, Grotjan TM, et al. Molecular diagnosis of eosinophilic esophagitis by gene expression profiling. Gastroenterology
17. Ward RM, Kearns GL. Proton pump inhibitors in pediatrics: mechanism of action, pharmacokinetics, pharmacogenetics, and pharmacodynamics. Paediatr Drugs
18. El Rouby N, Lima JJ, Johnson JA. Proton pump inhibitors: from CYP2C19 pharmacogenetics to precision medicine. Expert Opin Drug Metab Toxicol