Toward More Efficient Dietary Elimination Therapy for Eosinophilic Esophagitis: The Fantastic 4?

First-line treatment for eosinophilic esophagitis (EoE) after proton pump inhibitor nonresponse consists primarily of pharmacologic or dietary therapy.^1,2 Dietary therapy is centered on eliminating food triggers that drive the pathogenesis of EoE,³ and currently consists of 3 main modalities: elemental diet, allergy testing-directed food elimination, and empiric 6-food elimination diet (SFED). Although elemental diet is the most effective of the 3 methods in achieving histologic remission,⁴ it has limitations that make adherence difficult, such as the need to avoid regular foods, high cost, and, in some cases, unpalatable taste. Allergy testing-directed elimination diets are appealing conceptually, but have the lowest remission rates of the available strategies.⁴ The SFED has good response rates in both adults and children,^5-8 is more palatable, and is easier to implement. Despite eliminating only 6 food groups, adherence still may be an issue, and the need for repeated endoscopies after sequential food re-introduction to identify triggers poses a significant burden for patients. Because most patients on the SFED typically have between 1 and 3 food triggers,^6,7,9 diets eliminating fewer food groups would be appealing if they were shown to be equally effective.

In adults, a 4-food elimination diet was found to induce remission in 54% of EoE patients.¹⁰ Based on these data and other existing evidence on less-restrictive strategies than SFED,^11,12 the prospective multicenter study by Kagalwalla et al¹³ published in this issue of Clinical Gastroenterology and Hepatology investigated the clinical, histologic, and endoscopic outcomes of a 4-food elimination diet excluding dairy, egg, wheat, and soy in a pediatric EoE cohort. The study comes from an expert group considered to be pioneers in dietary elimination therapy for EoE. The investigators enrolled pediatric EoE patients who were treatment-naive or failed topical steroid therapy from 4 study sites. After 8 weeks of treatment with the 4-food elimination diet, subjects underwent repeat upper endoscopy with biopsies. Those who achieved the primary outcome of histologic remission (defined as <15 eosinophils/high-powered field) underwent sequential re-introduction of single food groups in the order of soy, eggs, wheat, and, finally, dairy. Upper endoscopy was performed at least 8 weeks after each food group was added back. In addition to identification of specific trigger foods, secondary outcomes included endoscopic and symptomatic improvement. Symptoms were assessed using a nonvalidated 17-item questionnaire with each symptom dichotomized as absent or present.

A total of 64% of subjects achieved histologic remission after 8 weeks of treatment with the 4-food elimination diet. Although just 36% had complete symptom resolution, 91% had a decrease in symptom score compared with baseline. Responders also had a decrease in mean endoscopic scores, but not a complete resolution of findings. Cow’s milk was the most common food trigger, followed by eggs, wheat, and soy. Serum-specific IgE and skin prick testing did not predict food triggers accurately. However, a family history of food allergies and food sensitization to the 4 foods based on serum-specific IgE testing predicted poor treatment response. Although initiation of the diet led to a decrease in body mass index at onset, participants regained this lost weight and subsequently gained further weight after the food re-introduction protocol was completed.

The histologic response rate of 64% with 4-food elimination in this well-designed study was close to the efficacy reported in previous studies of the SFED in both adults and children. A prospective study by Lucendo et al⁹ showed a 73% histologic response with SFED in adult EoE patients. However, this study used a more restrictive strategy by excluding rice and corn in addition to wheat, legumes, and peanuts. Histologic response rates with SFED have been as high as 70% in adults and 74% in children,^6-8 and meta-analyses have shown pooled response estimates ranging from 69% to 72%.^4,5 In all of these studies, cow’s milk and wheat were the most common food triggers, similar to what the investigators of the present study found,¹³ but nuts and seafood were always the least frequent triggers. The studies differed in the number of foods that triggered EoE. EoE was triggered by a single food in only 36% of adult patients on the SFED in 1 Spanish study,⁹ but rates of EoE being triggered by a single food group were much higher in American studies, with reports of 85% in adults⁶ and 72% in children.⁷ This difference in the number of food triggers may be owing to differences in regional dietary practices. Based on this theory, it is not surprising that the frequency and types of food triggers in this study were identical to that of SFED data published by the same group previously.⁷

The results of this study also were similar to existing data on 4-food elimination diets. In 1 multicenter prospective Spanish study¹⁰ of adult EoE patients, clinicopathologic remission was achieved in 54% of cases, which was lower than the current study, but response rates increased to 72% after rescue SFED therapy in those who failed 4-food elimination. In addition to age, a major difference was that in the Spanish study nearly a quarter of the patients were refractory to steroid therapy, and may have had a more severe EoE phenotype. Response rates were comparable in another prospective study¹¹ of adults and children treated with a 4-food elimination diet: 85% of adults and 87% of children had a greater than 50% decrease of baseline eosinophil counts with concurrent symptomatic and endoscopic improvement.

There also have been studies focusing solely on dairy elimination because dairy typically is the most common EoE food trigger in both adults and children. In a prospective pediatric EoE cohort,¹⁴ cow’s milk elimination resulted in histologic response in 64% of patients, along with a significant improvement in quality of life and symptom scores. Prior retrospective data also showed similar results with a 65% remission rate by eliminating only cow’s milk.¹² A more recent prospective study showed lower, but relatively good, histologic response rates of 43% and symptomatic improvement in 58% of responders in a pediatric cohort.¹⁵

Synthesizing these data, it is clear that dietary elimination is an effective treatment for EoE in both adults and children, and histologic response rates, although higher for SFED, are on the same order of magnitude among the different elimination strategies. Dairy elimination alone has response rates ranging from 43% to 65%, which approach the rate found in the current study by Kagalwalla et al.¹³ Results from this current study also align with existing data on 4-food elimination, acknowledging that different studies use different histologic response thresholds. In contrast to prior data on 4-food elimination, this study had a large sample of pediatric patients recruited from multiple sites. Although symptom data were not collected with a validated symptom score, symptom and endoscopic data do not always align with histologic response and the latter is used most commonly to define remission. However, an assessment of symptom response using a validated instrument would be welcome. The results of this study in the context of existing literature suggest that the 4-food elimination diet likely should be considered over SFED in pediatric patients with EoE who choose dietary therapy, given that the added marginal benefit with SFED likely is offset by increased patient burden of endoscopy, complexity of the diet, and associated costs.

Balancing the number of foods eliminated, dietary therapy efficacy, and efficiency of the food re-introduction process would be ideal, and a group from Spain recently reported the results of a multicenter prospective study of adult and pediatric EoE patients doing just that.¹⁶ In this study, patients initially were treated with 2-food elimination of animal milk and gluten-containing cereals. Failure to improve both symptoms and histology resulted in subsequent expansion to 4-food elimination (egg/legumes eliminated), and then to 6-food elimination (seafood and nuts) in nonresponders. Histologic response rates were 42%, 58%, and 71% with the sequential 2-, 4-, and 6-food elimination diets, respectively, and the investigators estimated this approach would save up to 30% of endoscopies for food trigger identification. This is an intriguing approach that could become standard practice because of the efficiency of the process. In addition, there is currently a study underway by the Consortium of Eosinophilic Gastrointestinal Disease Researchers investigating the efficacy of single-food vs 4-food dietary elimination in children and single-food vs SFED in adults with EoE.

While we await these upcoming data, what approach should providers recommend right now? Are dairy, wheat, eggs, and soy the fantastic 4? It is clear that empiric elimination diets are effective for EoE, and are a good choice in motivated patients and when adequate nutritionist/dietician support is available. The specific regimen should be individualized to the patient, potentially taking into account individual atopic history including food sensitization as well as family history of allergy (as suggested in the present study), and selecting a strategy that is manageable for the patient, both for the initial elimination and for subsequent food group re-introduction and trigger identification. Therefore, based on the present study as well as prior data showing that nuts and seafood are infrequent EoE triggers, selecting an empiric 4-food elimination diet, especially in children, is a reasonable initial therapeutic approach.