Eosinophilic esophagitis (EoE) is an emerging chronic local immune-mediated disease of the esophagus. Beside proton pump inhibitors and food-restriction-diets swallowed topical corticosteroids (STC) can be offered as a first line therapy according to current guidelines. This review describes the background and practical management of STCs in EoE. So far, mainly asthma inhalers containing either budesonide or fluticasone have been administered to the esophagus by swallowing these medications “off label”. Recently esophagus-targeted formulations of topical steroids have been developed showing clinicopathological response rates up to 85% - an orodispersible tablet of budesonide has been approved as the first “in label” medication for EoE in Europe in June 2018. Whereas it was shown that disease remission induction of EoE by STCs is highly effective, there is still a lack of data regarding long-term and maintenance therapy. However, current studies on STC maintenance therapy add some movement into the game.

Eosinophilic esophagitis (EoE) is defined as a chronic local immune-mediated esophageal disease, characterized clinically by esophageal dysfunction and histopathologically by eosinophil-predominant inflammation[1-3]. Since EoE was first described as a new disease entity in the early 1990s[4,5], its incidence and prevalence have remarkably increased with a pooled prevalence of 34.4 cases per 100000 inhabitants (adults 42.2/100000; children 34/100000) and a pooled incidence of 7.7/100000 patient-years for adults and 6.6/100000 patient-years for children[6]. Although, EoE right now still has to be considered a rare disease, the increasing incidence and prevalence make it more likely to encounter a patient with EoE in everyday practice.

Adult EoE-patients typically suffer from dysphagia, bolus impaction and retrosternal pain whereas in children predominant symptoms are abdominal pain, vomiting and failure to thrive. Upper endoscopy with esophageal biopsies is always required to diagnose EoE. Typical endoscopic findings are mucosal edema, white exudates and furrows as signs of acute inflammation as well as rings and strictures as signs of fibrostenotic remodelling of the esophagus. However, approximately 10% of adult patients and up to 32% of children will have a normal appearing esophagus[7] - therefore at least 6 biopsies from different levels of the esophagus (typically proximal and distal) have to be taken and the proof of a minimum of 15 eosinophils per high power field (hpf) in at least one biopsy is mandatory for EoE diagnosis. Additional biopsies of the stomach and duodenum should be taken when EoE is newly diagnosed to exclude other causes of esophageal eosinophilia, such as eosinophilic gas-troenteritis[2,3].

EoE patients predominantly are white young males with concomitant atopic conditions[8] and IgE related food allergies. In these genetically susceptible patients, the esophageal exposure to patient-specific food-antigens, probably promoted by an increased mucosal permeability trigger a T-helper type 2 (Th2) cell response leading to the production of key-cytokines including IL-4, IL-5, and IL-13. These cytokines promote further eosinophilic leukocyte activation, recruitment and inflammation of the esophagus mainly driven by eotaxin-3. Esophageal eosinophilia also promotes development of tissue fibrosis by transforming growth factor beta and periostin which enhances remodelling, fibrosis and stricture formations[9,10]. Hence, natural history of EoE has to be understood as a progression from inflammation to fibrosis[11-14]. Patients with a fibrostenotic phenotype more often report dysphagia and food impaction, have a longer duration of dysphagia symptoms, and thus, a longer disease duration and are older than patients with the inflammatory phenotype[13]. EoE has a substantial negative impact on deglutition and on the patients’ health-related quality of life (HRQoL)[15], especially if not treated sufficiently. Moreover, once advanced fibrosis and stricture formation are existent, medical or dietary treatment are not able to fully improve dysphagia and the quality of life – in these cases endoscopic dilatation-therapy with the possible risk of complications has to be added[16].

Current guidelines lack explicit disease management goals, considering the deficient available data. The optimal treatment objectives have yet to be defined. So far, the major goals of (short-term, remission-induction) EoE-therapy are improvement of histologic, symptomatic, and endoscopic outcomes. Other outcome measures include quality of life, occurrence of complications like food impaction, biomarkers and especially in children normal growth/thrive[13]. The main histopathological measure normally looked at is the (peak) eosinophil count (eos/hpf) and in studies the therapeutic objective is normally defined as decrease of this peak eosinophil count to at least < 15 eos/hpf (variation between studies). Also other features such as basal zone hyperplasia, eosinophil abscesses, eosinophil surface layering, dilated intercellular spaces, surface epithelial alteration, dyskeratotic epithelial cells and lamina propria fibrosis can be observed on which a standardized histological measure has been proposed by Collins et al[17] Concerning the symptoms of EoE, the therapeutic goal is an improvement, or stricter a cessation of symptoms after therapy. Different validated symptom measures are available, for example the Eosinophilic Esophagitis Activity Index or the Dysphagia Symptoms Questionnaire (DSQ) for adults and the Pediatric Quality of Life Inventory EoE for children. Endoscopic-macroscopic signs of EoE can be best objectively assessed with the Endoscopic Reference Score (EREFS - acronym for exudates, rings, edema, furrows, and strictures), which allows an endoscopic assessment of EoE disease activity – therapeutic goal is at least a decreased score to less than baseline or better a normalization of esophageal changes[18-23].

As already explained, EoE progresses from inflammation to fibrosis, so long-term treatment goal has to be maintenance of disease remission. Current guidelines do not include recommendations in terms of long-term management strategies (e.g., intermittent or continuous therapy) and the exact extent of disease remission with regard to prognosis or disease progression still needs to be defined[24]. While there is quite good correlation between endoscopic signs (EREFS-score) and histologic improvement[25], a problems remains only modest correlation between symptoms and endoscopic/macroscopic as well as histologic disease activity[26,27]. Therefore, the efficacy of any first line anti-inflammatory therapy should be checked by a follow-up endoscopy after a 6- to 12-wk initial course[3].

From patients’ perspective, symptoms and quality of life improvement are the most important short- and long-term therapy goals in EoE[28]. In addition, through patient information about the connection of preventing symptom recurrence by inflammation-control, patients can probably be motivated to better long-term therapy adherence[28].

Proton pump inhibitors (PPIs), food-restriction-diets and swallowed topical corticosteroids (STCs) can be offered as first line anti-inflammatory therapy in EoE-patients[3].

The mechanism of action of glucocorticoids in EoE is mainly based on the reduction of inflammatory cells and fibrosis: in short term treatment with STC results in a significant decrease of esophageal eosinophils, mast cells, T-cells and proinflammatory cytokines as well as restoration of epithelial barrier function and reduction of tissue remodeling[29-31]. Moreover, it has been shown that the EoE IL-13–induced tran-scriptome (IL-13 induced pathways and genes) is largely reversible with glucocorticoid treatment in vivo[32].

In 1998, Liacouras et al[33] was the first to describe successful treatment of EoE with STCs in children. A subsequent randomized controlled trial (RCT) in 80 children showed equal effectiveness of systemic prednisone and topical fluticasone in achieving histological and clinical improvement but systemic prednisolone was associated with more adverse events[34]. Because of these findings STCs should generally be preferred. Systemic steroids may only be considered as an exception in patients who require rapid improvement of symptoms (e.g., children with severe symptoms, malnutrition or feeding intolerance)[2].

The effectiveness of STC-therapy in EoE has been shown in several studies in both children and adults. The most thoroughly evaluated topical corticosteroids are budesonide and fluticasone. Until today, seventeen randomized trials[34-50] comprising nine placebo-controlled studies have been published (Table 1). Parts of these data have been summarized in different meta-analyses[51-56]. These studies considerably vary regarding inclusion criteria, agents, daily dosages, length of treatment and end point definitions, showing histologic remission rates from 39%-100% and mainly sufficient clinical response. However, a lack of correlation between symptoms and histology has been observed in some studies. This might be explained by possible insufficient durations of therapies, by fixed fibrosis or most likely by the usage of non-validated and/or insufficient patient reported outcome metrics[57].

EoE is considered a chronic progressing disease and most patients suffer from rapid relapse when therapy is stopped: Greuter et al[63] analyzed 351 EoE patients retrospectively, treated with a topical steroid maintenance therapy (0.25 mg twice daily). Disease activity was assessed on an annual basis. In patients, who achieved long-lasting (≥ 6 mo) clinical, endoscopic and histological remission, treatment was stopped, which was only possible in 9.4% of patients after a median of 89 wk. In 82% of the remitters a clinical relapse occurred after a median of 22.4 wk and only 1.7% of all included patients were able to discontinue therapy in the long term[63]. Another retrospective interval follow-up trial over three years including 51 adult EoE patients, who were initially treated successfully with a swallowed fluticasone inhaler, found a relapse in 91% of these patients after an average of nine months after therapy was stopped. 69% of patients repeated treatment with the steroid inhaler at least once[68]. Dellon et al[69] reported recurrence of symptoms after initial successful remission-induction therapy in 33 of 58 patients (57%) before one year. At symptom recurrence, 78% of patients were found to have histologic relapse. Additionally, in the very recently conducted double-blind, placebo-controlled maintenance trial (Eos-2 trial) with budesonide orodispersible tablets (BOT), EoE-patients in remission (n = 204) received either 1 mg BOT BID, 0.5 mg BOT BID or Placebo BID. After 48 wk of treatment, only 20.6% of patients in the placebo group were still in clinical remission, 89.7% were proven to have a histological relapse (intention-to-treat-analysis)[70]. These data show, that long-term-treatment of EoE to maintain remission would make much sense.

STCs have been mainly successfully evaluated for short-term treatment of EoE but there is only limited data for long-term-treatment available. A retrospective cohort study on 206 patients with a median follow-up time of five years showed that an increased frequency of use of swallowed topical steroids was associated with a lower risk for bolus impactions[71]. Another retrospective study which evaluated the long-term efficacy of STC in 229 adult EoE patients over four years showed significantly higher proportions of patients on STCs in clinical remission (31.0%) compared to patients off STCs (31% vs 4.5%, P < 0.001). The same was described for endoscopic remission (48.8% vs 17.8%, P < 0.001), histologic remission (44.8% vs 10.1%, P < 0.001), and complete remission (16.1% vs 1.3%, P < 0.001). Higher cumulative doses of STCs and longer durations of treatment were associated with higher proportions of clinical and complete remission[72]. One small prospective study in 10 patients who received 1 year of treatment with fluticasone propionate showed a significant reduction of esophageal subepithelial fibrosis compared to controls[73]. Recently, results of an open-label extension study with the aforementioned BOS have been published. 82 EoE-patients who completed 12 wk of either budesonide (2 mg twice daily) or placebo therapy received another 24 wk budesonide (2 mg once daily for 12 wk, with optional dose increase (1.5-2 mg twice daily) for 12 wk thereafter). 42% of the therapy responders maintained a histologic remission during the open-label extension and 4% of non-responders gained response[46].

Only a few placebo-controlled EoE-maintenance trials exist: Straumann et al[74] reported 28 adult patients who, after having responded to swallowed budesonide, were randomized to continue low-dose budesonide at 0.25 mg twice daily or to receive placebo for one year. Histological remission (< 5 eosinophils/hpf) after one year was only achieved in 36% of the budesonide patients and in 0% of the placebo patients, but this difference was not significant. The comparably low number of long lasting clinical remission was most likely due to underdosing of the used steroid treatment (only 0.5 mg/d), moreover no specific esophagus-targeted formulation was used[74]. High-quality data are available from the aforementioned Eos-2 trial, comparing maintenance therapy with 1mg BOT BID, 0.5 mg BOT BID or Placebo BID over a period of 48 wk in EoE patients in clinical-histopathological remission. As previously alluded, EoE relapse was high in the placebo-group. Placebo treatment was significantly inferior to both BOT-treatment groups, where maintenance of clinic-histological remission (defined as absence of clinical relapse, lack of histological relapse, absence of food impaction, no need for endoscopic dilation therapy and no premature withdrawal for any reason) was described in 75% (1 mg BID) and 73.5% (0.5 mg BID) of patients vs 4.4% (placebo BID)[70]. Additionally, the BOT maintenance treatment was shown to maintain and even to further improve patients’ quality of life (evaluated with the EoE-QoL-A) compared to the placebo group[75]. The 0.5 mg BOT is expected to be officially approved for EoE maintenance therapy in 2020.

Generally, STC-treatment in EoE has been found to be well tolerated, with a good safety profile. Most common side effect is candida esophagitis which was estimated to occur in 8.7% of patients[54]. For the newest, esophagus-targeted formulations candida esophagitis in short-term treatment was described for the orally disintegrating fluticasone tablet in a total of 9.4% of patient (up to 20% in the highest dose group – 3 mg BID)[59], for the mucoadherent BOS (2 mg BID) in 2% of patients[44] and for the BOT (1 mg BID) in 10% of patients[50]. In the long-term therapy, esophageal candidiasis was described in 10% of patient with BOS-treatment (for a total of 24 wk, at least first 12 wk 2 mg/d)[46]. For treatment with BOT clinically suspected symptomatic candidiasis was described in 11.8% of patients in the 1 mg BID group, with histological confirmation only in 2.9%. In the 0.5 mg BID group, there was suspected esophageal candidiasis in 16% of patients. Histological confirmed candidiasis was present in 7.4% of patients[70]. However, most esophageal candidiasis-cases are asymptomatic and usually only incidentally detected at follow-up endoscopy. If candida esophagitis is suspected, histologically confirmation should be carried out; particularly eosinophilic microabscesses can mimic white candida plaques. Antifungal treatment with either fluconazole or nystatin over 7-14 d should be performed[3].

As shown in asthma-patients, long-term use of high-dose topical steroid therapy may cause systemic side effects such as impaired growth in children, decreased bone mineral density, skin thinning and bruising, and cataracts[76]. Hypothalamic-pituitary-adrenal-axis suppression, measured by reduced serum or urine cortisol levels seems to correlates with the occurrence of systemic side effects. All aforementioned clinical trials on topical steroids for EoE-treatment (short- and longer-term) have not found relevant changes in cortisol levels or even systemic cortisol side effects. However, one pediatric study reported adrenal suppression, which was measured by a low-dose adrenocorticotropin stimulation test, in 10% of children treated with swallowed glucocorticoids for > 6 mo. However, adrenal suppression was only found in those children treated with fluticasone > 440 μg/d compared to lower doses of fluticasone. After discontinuation of fluticasone the hypothalamic-pituitary-adrenal axis recovered in all children[77]. Another small series reported in 6/14 children (43%) treated with OVB suboptimal cortisol stimulation measured by ACTH stimulation test independently of treatment duration[78]. Thus, measurement of morning cortisol levels as well as close observation for potential systemic steroid side effects should be done on a regular basis in all patients receiving long-tem topical steroid treatment of EoE. Additional ACTH-stimulation test might be helpful in unclear cases.

STCs can be regarded as the medical treatment of choice for most EoE patients. Recommended standard regimes include either the use of budesonide 2 mg/daily or fluticasone 880-1760 μg/daily in adults and budesonide 1mg/daily or fluticasone 88-440 μg/daily in children, respectively[2]. Most effective are esophagus-targeted formulations of topical steroids, but so far, only in the EU, the BOT (1 mg BID 6-12 wk) is officially approved for EoE remission induction therapy. In the rest of the world esophagus-targeted formulations may only be available within studies. So, in daily practice in most of the world non-esophagus-targeted formulations of STCs will still have to be used. Regarding this scenario, Dellon et al[47] recently compared the efficacy of budesonide vs. fluticasone for initial treatment of EoE in a randomized controlled trial. For a period of 8 wk 129 EoE-patient were randomized to treatment either with an OVB (aqueous budesonide mixed into a slurry with a sugar substitute, 1 mg BID) plus a placebo inhaler or with a fluticasone inhaler (MDI, swallowed, 880 g BID) plus a placebo slurry. Both STCs resulted in a significant decrease in esophageal eosinophil counts (histologic response with budensonide: 71%; with fluticasone: 64%) and improved dysphagia and endoscopic features. The rates of esophageal candidiasis were comparable (OVB: 12%, MDI: 16%, all asymptomatic). There was one event of food-impaction in the MDI-group, however this was recorded in a patient who had stopped taking medications during the 8-wk treatment phase. Overall, OVB was not superior to MDI, so either one can be regarded as an acceptable treatment for EoE. However, the OVB involves increased patient or pharmacy effort to mix, as well as added costs[47].

For inducing remission of the disease, treatment duration of 6 to 12 wk is initially recommended. As symptoms alone cannot safely predict therapy success, endoscopy with biopsies for histopathological evaluation is still mandatory. To avoid repeated endoscopic procedures much of an effort has been put in the evaluation of potential non-invasive biomarkers but until today no biomarker has been found reliable enough to replace endoscopy[79]. Best results most likely exist for the easy and cheap to measure but rather non-specific serum absolute eosinophilic count, which thus might be helpful for therapy monitoring in a least some EoE patients[80,81].

For the reasons already stated above, namely the chronic inflammatory and fibrosing character of EoE, long-term treatment seems highly advisable for most patients. Given the results of the long term maintenance study with oral orodispersible budesonide, the dosage of STCs in this scenario can be chosen lower than dosage used for inducing remission (in this case 0.5 mg BID)[70]. Further studies are needed to define the optimal lowest dosing for maintenance of remission. Furthermore, there is no data about when sustained clinical remission can be expected and long-term therapy can possibly be stopped. If either an interval treatment with regular therapy breaks might be superior to a continuous application of topical steroids, is under current investigation. Nevertheless, long-term treatment of EoE with STCs can be considered safe and particularly systemic corticosteroid side effects rarely occur. However, only a few safety data are available yet. Thus, careful surveillance of possible local side effects (particularly candidiasis) and unlikely systemic side effects (particularly adrenal suppression) have to be performed. This includes a regular clinical and endoscopic assessment as well as monitoring of morning cortisol levels and physical and ophthalmologic examinations.

In summary, response rates of remission induction therapy with STCs are very high, especially for esophageal targeted formulations and thus can be regarded as therapy of choice for the majority of EoE patients. But what are the alternatives, particularly if this treatment fails? Beside the other first-line options such as PPIs and elimination diets emerging drugs including antibodies that specifically target key players in EoE inflammation pathways like IL-4, IL-5 or IL-13 signaling are currently under development and investigation.