Randomised Clinical Trial

The Safety and Tolerability of Fluticasone Propionate Orally Disintegrating Tablets Versus Placebo for Eosinophilic Oesophagitis

Ikuo Hirano; Ekaterina Safroneeva; Marie C. Roumet; Gail M. Comer; Gina Eagle; Alain Schoepfer; Gary W. Falk


Aliment Pharmacol Ther. 2020;51(8):750-759. 

In This Article


A randomised, double-blind, placebo-controlled, multicentre, phase 1b/2a study was conducted at seven medical centres in the US to evaluate the safety and tolerability of APT-1011 (Adare Pharmaceuticals, Inc [formerly Aptalis Pharma US, Inc.]) over 8 weeks in adults and adolescents with EoE (ClinicalTrials.gov registration #NCT01386112). The study was conducted in accordance with the principles set forth in the Declaration of Helsinki and in compliance with International Conference on Harmonization—Good Clinical Practice standards. The study protocol was approved by Copernicus Group Independent Review Board (protocol #PR-021; 01 September 2011).

Eligibility Criteria

Eligible patients between 12 and 55 years age had histologically confirmed EoE (oesophageal mucosal peak eosinophil count ≥24 per high-power field [HPF] [HPF; radius = 0.275 mm; 400×] in at least one biopsied site, within 30 days prior to and 21 days after the screening visit), histologically confirmed prior treatment failure of a high-dose proton pump inhibitor (PPI), defined as peak eosinophil counts ≥24 per HPF after 8 weeks of 2× standard PPI dose per investigator, and at least one of the following symptoms: chest pain or discomfort, dysphagia or food impaction continuously or intermittently present within 30 days prior to the screening visit. Specific symptom severity was not required for this study. Females of child-bearing potential must have agreed to use adequate contraception during the study and could not be pregnant or lactating at time of enrolment. Written informed consent was obtained from each participant or caregiver/parent/guardian at screening.

Patients were excluded from participating in the study if they met any of the following criteria: presence of any condition, other than EoE, that affected the oesophageal mucosa or motility; any contraindication to completing oesophagogastroduodenoscopy, including stricture that blocked the passage of a standard endoscope; history or presence of Crohn's disease, celiac disease or other gastrointestinal inflammatory disease; use of systemic, inhaled, intranasal or high-potency dermal topical corticosteroids during the 30 days prior to enrolment; morning serum cortisol level ≤5 μg/dL or use of anti-inflammatory or immunosuppressant drugs.


Participants were randomised 1:1:1 to APT-1011 at 1.5 mg BID, APT-1011 at 3.0 mg QD or placebo by a blinded study coordinator, according to a computer-generated list of randomisation numbers, with a block size of 3. A unique randomisation number was assigned to each participant in sequential order of enrolment. All study site personnel, including the endoscopist, the central pathologist, the participants/caregivers and the sponsor, were blinded to the treatment groups.


Participants were given two bottles—one for the morning dose and one for the evening dose—and were instructed to take their morning dose at least 30 minutes before food intake and their evening dose at least 30 minutes after a meal. Participants were instructed not to drink or eat for 30 minutes after each dose. The placebo group had a placebo tablet in both morning and evening bottles; the APT-1011 1.5 mg BID group had a 1.5-mg fluticasone propionate tablet in both morning and evening bottles; and the APT-1011 3.0 mg QD group had a placebo tablet in the morning bottle and a 3.0-mg fluticasone propionate tablet in the evening bottle. Participants were instructed to place the tablet on the tongue until it dissolved completely and to swallow until it was completely ingested (ie, no visible drug), then to thoroughly rinse mouth with water (ie, rinse and spit) to remove any residual fluticasone propionate from the oral cavity. Participants were allowed to continue PPI therapy if they met the histology inclusion criteria above (oesophageal mucosal peak eosinophil count ≥24 per HPF) and had a stable PPI dose for at least 30 days prior to study enrolment. Antihistamines were permitted if the participant required them for a medical condition.


The primary endpoint was safety and tolerability of APT-1011, as assessed by treatment-emergent adverse events (TEAEs), morning serum cortisol and salivary cortisol levels, laboratory tests (haematology, serum chemistry, liver function tests, urinalysis and urine chemistry), physical examination and vital signs at Weeks 2, 4, 6, 8 and at follow-up.

Exploratory efficacy end points were also measured. At screening and at Week 8, participants underwent oesophagogastroduodenoscopy during which two to four biopsies from the distal, middle and/or the proximal oesophagus were taken. Oesophageal eosinophil counts per HPF in all parts of the oesophagus were assessed. Endoscopic appearance of the oesophagus was evaluated by a treatment-blinded endoscopist. The endoscopist graded the presence and severity of oedema (decreased vascular markings), rings, exudates, furrows, felinisation, linear shearing and stricture(s) in accordance with the endoscopic reference score (EREFS) grading and classification system.[17] Total EREFS was calculated by summing the severity scores of the individual components (oedema 0–1, rings 0–3, exudates 0–2, furrows 0–2 and strictures 0–1) assessed separately for both the proximal and distal oesophagus (ranges 0–18, with higher scores indicating more severe endoscopic findings).

Questionnaires developed for the Eosinophilic Esophagitis Activity Index (EEsAI) study were adapted for use in this clinical trial. The following were completed prospectively by the study participants and the physicians:[18] EREFS grading and classification system completed by the endoscopist at screening and at Week 8; Physician Global Assessment completed by the investigator at randomisation, Weeks 4 and 8; EEsAI patient-reported outcomes (PRO) instrument[18] with 7-day recall period, and Patient Global Assessment with 7-day recall period completed at randomisation, Weeks 4 and 8 by the patient; and Pathologist Questionnaire (querying presence of various histologic features in biopsies obtained in distal, proximal and/or middle part of the oesophagus) completed at screening and at Week 8.

The Physician Global Assessment of the participant's overall EoE activity (Likert scale 0–10, where 0 signifies no EoE activity and 10 represents the most severe EoE activity) takes into account participant-reported symptoms, endoscopic findings and histologic activity. Study participants provided the Patient Global Assessment of symptom severity and EEsAI PRO (score range 0–100, score increases with increasing symptom severity)[18] at randomisation, Weeks 4 and 8. Participants also completed the Mayo Dysphagia Questionnaire-30[19,20] and the Gastrointestinal Symptom Rating Scale Questionnaire[21] at randomisation, Weeks 4 and 8.

Statistical Analysis

All statistical analyses were performed using the statistical program R (version R3.3.1).

The sample size of 24 participants (8 per group) was expected to provide sufficient observation to assess safety and tolerability. As formal sample size calculations were not performed for the study, the analyses presented here can only be considered directional. All efficacy analyses are based on the intent-to-treat (ITT) population. All randomised participants received at least one dose of a study drug, had at least one efficacy measure and were included in all safety and efficacy analyses.

Treatment-emergent adverse events were summarised by treatment group, according to preferred term from the Medical Dictionary for Regulatory Activities (MedDRA, version 14.0).

As per protocol, the mean, median, min and max values of distal and proximal eosinophil counts per HPF (for subjects in whom these values were available and stratified by participant treatment group) are reported at baseline and at Week 8. Change over time in the mean and median values of these counts also is presented. A participant was considered to show a response to treatment if the peak eosinophil count decreased from >24 to <15 eosinophils/HPF and a complete response if the peak eosinophil count decreased from >24 to 0 eosinophils/HPF at Week 8.

Additional exploratory post hoc analyses were carried out in consistent with current reporting standards. Study participants were first categorised into three treatment groups: placebo, APT-1011 at 1.5 mg BID and APT-1011 at 3.0 mg QD. We also categorised study participants into placebo and APT-1011–treated groups (the APT-1011 at 1.5 mg BID and the APT-1011 at 3.0 mg QD groups combined). For each patient at each visit we calculated the mean and median value of the EREFS, the EEsAI PRO score (scoring system was published after protocol development and study completion) and the overall inflammatory burden by taking peak eosinophils/HPF value from proximal, middle and distal oesophagus. Changes in EREFS, EEsAI PRO score, Patient Global Assessment and maximum peak oesophageal eosinophils/HPF from baseline to Week 8 were calculated as the difference between the values observed at Week 8 and baseline. The variation in these changes was analysed using linear regression models that included the effect of treatment with APT-1011 (relative to placebo) and adjustment for the baseline value. We first investigated the effect of treatment on the change in these values by comparing the patients from the three treatment groups: placebo, APT-1011 at 1.5 mg BID and APT-1011 at 3.0 mg QD. In addition, we combined the patients from both APT-1011–treated groups. We compared the change in values in APT-1011–treated patients with those in placebo group. A P < 0.05 was considered to be significant.