Effect of randomized treatment withdrawal of budesonide oral suspension on clinically relevant efficacy outcomes in patients with eosinophilic esophagitis: a post hoc analysis.

BACKGROUND

Relapse after corticosteroid withdrawal in eosinophilic esophagitis is not well understood.

OBJECTIVES

Budesonide oral suspension (BOS) 2.0 mg twice daily (b.i.d.) was evaluated in two consecutive phase III studies (12 and 36 weeks, respectively). For clinicopathologic responders after 12 weeks of BOS treatment, we assessed randomized treatment withdrawal for up to 36 weeks of therapy.

DESIGN

Post hoc analysis of a phase III, double-blind, randomized withdrawal study.

METHODS

Clinicopathologic responders (⩽6 eosinophils per high-power field (eos/hpf) and ⩾30% reduction in Dysphagia Symptom Questionnaire (DSQ) score from baseline) after 12 weeks of BOS were randomized to continue BOS 2.0 mg b.i.d. (BOS-BOS) or withdraw to placebo (PBO; BOS-PBO) for up to 36 weeks. Relapsers (⩾15 eos/hpf (⩾2 esophageal regions) and ⩾4 days of dysphagia (DSQ)) could reinitiate BOS 2.0 mg b.i.d. This post hoc analysis assessed a more clinically relevant relapse definition (⩾15 eos/hpf (⩾1 esophageal region) and ⩾4 days of dysphagia (DSQ)) for BOS-BOS versus BOS-PBO patients over 36 weeks. To account for BOS-PBO patients who reinitiated BOS before week 36, patients' last observations before reinitiating BOS were carried forward (last observation carried forward (LOCF)) for histologic, symptom, and endoscopic efficacy endpoints (at weeks 12 and 36).

RESULTS

Of 48 patients included (BOS-BOS, n = 25; BOS-PBO, n = 23), significantly more BOS-PBO than BOS-BOS patients relapsed over 36 weeks using this post hoc relapse definition (60.9% vs 28.0%; p = 0.022). More BOS-BOS than BOS-PBO patients maintained histologic responses (all thresholds) and showed improvements in symptom and endoscopic efficacy endpoints.

CONCLUSION

More BOS-PBO than BOS-BOS patients relapsed, determined by a more clinically relevant post hoc relapse definition. Using LOCF, more BOS-BOS than BOS-PBO patients also maintained or had improvements in efficacy endpoints.

TRIAL REGISTRATION: ClinicalTrials.gov identifiers (https://clinicaltrials.gov/): NCT02605837, NCT02736409.