Dysphagia Days as an Assessment of Clinical Treatment Outcome in Eosinophilic Esophagitis

INTRODUCTION: The aim of this study was to evaluate Dysphagia Days as a measure of symptom improvement in patients with eosinophilic esophagitis from the HEROES study. METHODS: Dysphagia Days, defined as a yes answer to the following question: During any meal today, did food go down slowly or get stuck in your throat or chest? was assessed for cendakimab vs placebo. RESULTS: A statistically significant reduction in the mean number of Dysphagia Days experienced was observed with cendakimab 360 mg vs placebo at week 16 (−4.67 vs −1.83; P = 0.0115); an even greater improvement was observed in steroid-refractory patients vs placebo (−4.48 vs −0.04; P = 0.0079). DISCUSSION: Dysphagia Days represents a relevant clinical end point to capture dysphagia-related symptoms.


INTRODUCTION
Finding better treatments of eosinophilic esophagitis (EoE) is an area of active research, but standardization of end points measuring EoE symptoms is lacking. Although the Dysphagia Symptom Questionnaire, EoE Activity Index (EEsAI), and Pediatric EoE Symptom Severity module assess frequency and intensity of dysphagia episodes and have been validated as clinical end point measures for patients with EoE, questions remain about their optimal use (1-3).
Interleukin-13 (IL-13) is implicated in the pathogenesis of EoE (4). Cendakimab is a selective, high-affinity, humanized immunoglobulin G1 monoclonal antibody that recognizes wild-type and variant human IL-13, blocking IL-13 binding to the a1 and a2 subunits of the IL-13 receptors (IL-13Ra1 and IL-13Ra2) with high potency (5). In the randomized, double-blinded, placebocontrolled phase 2 HEROES study (see Supplementary Figure 1, http://links.lww.com/AJG/C794) (6), cendakimab significantly improved histopathologic and endoscopic aspects of disease activity. The mean change in Daily Symptom Diary (DSD) composite score (a different metric than the Dysphagia Symptom Questionnaire), measured over the prior 14 days, resulted in a trend-level improvement with cendakimab 360 mg (P 5 0.0733) at week 16, despite not being powered to demonstrate clinical symptom improvement. A stronger trend was noted at week 16 in the steroid-refractory group treated with cendakimab 360 mg (P 5 0.0547). In addition, a trend toward greater reduction in the EEsAI was also observed with cendakimab 360 mg (P 5 0.1103) in the overall population (6). Cendakimab was well-tolerated; the 360 mg dose was brought forward for evaluation in phase 3 (6).
Given the lack of significant DSD composite score findings, a post hoc analysis was conducted to evaluate a potentially new clinical end point termed Dysphagia Days to assess symptom improvement during EoE treatment. Dysphagia Days was chosen based on interviews conducted in a separate validation study where "trouble swallowing," "food going down slowly," and "food getting stuck in the throat" were the most relevant symptoms associated with EoE.

METHODS
Full details of the HEROES study have been reported (6)(7)(8). For this post hoc analysis, a Dysphagia Day was defined as a yes answer to the following question derived from the DSD: During any meal today, did food go down slowly or get stuck in your throat or chest?
Comparisons of Dysphagia Days between cendakimab and placebo in the intent-to-treat (ITT) population were based on an analysis of covariance model with treatment group, steroidrefractory status, and baseline Dysphagia Days as covariates. Comparisons of cendakimab vs placebo by steroid-refractory status were based on an analysis of covariance model with treatment group, actual steroid-refractory status, and baseline Dysphagia Days as covariates, with treatment group by actual steroid-refractory status as an interaction term. Steroid refractory was defined as failure of an adequate trial of topical steroids to result in meaningful symptom reduction based on investigator assessment (6).

Dysphagia Days
At baseline, mean Dysphagia Days experienced over the previous 14 days ranged from 9.16 to 9.57 days in the ITT, 8.85 to 10.18 days in steroid-refractory, and 8.73 to 9.68 days in nonsteroidrefractory populations. At week 16, cendakimab 360 mg weekly     Figure 2), but no significant differences were observed between cendakimab and placebo in nonsteroid-refractory patients (Table 1). In a separate post hoc analysis, Spearman correlation coefficients indicated a strong correlation between Dysphagia Days at day 1 and week 16 and EEsAI (0.61 and 0.83, respectively, both P , 0.001), with a higher correlation between Dysphagia Days and DSD composite score (0.91 and 0.96, respectively, both P , 0.001) at those time points.

DISCUSSION
In this analysis, Dysphagia Days was chosen as an end point of interest based on patient interviews in which dysphagia was reported as the hallmark symptom of EoE. Correlation between histologic and symptomatic improvement in EoE is imperfect, however, as noted by the lack of significant improvements in Dysphagia Days with cendakimab 180 mg weekly (Table 1), despite the significant decrease in eosinophil counts observed (6). In this analysis, the lack of a significant reduction in clinical symptoms in the nonsteroidrefractory subgroup may be related to the higher placebo effect observed, which may have blunted the improvement observed with cendakimab. Despite this, reduction in Dysphagia Days provides a clear, objective, and meaningful measure of symptom improvement without interpreting a symptom score. In parallel to this analysis, the Dysphagia Days assessment was further modified and is being used as a coprimary end point in a phase 3 study (NCT04753697) investigating cendakimab in adults and adolescents with EoE, complying with United States Food and Drug Administration guidance for investigation of new drugs for EoE (9). The proportion with eosinophilic histologic response (i.e., peak esophageal eosinophil count #6 per high-power field) is the other coprimary end point. Given the simplicity of Dysphagia Days as a patient-reported outcome-targeted to the most common and critical symptom experienced by patients with EoE-and the ease of interpretation from the perspective of the patient and providers, Dysphagia Days has the potential for broader utilization in clinical and community settings for the monitoring of EoE symptoms and could be used as an end point in other diseases where dysphagia is a symptom (e.g., Schatzki rings). This analysis reveals that change from baseline at week 16 in Dysphagia Days was significant for cendakimab 360 mg vs placebo in both the ITT and steroid-refractory populations. Dysphagia Days may provide a more responsive and clinically relevant end point for the assessment of EoE symptom improvement compared with the DSD composite score. As such, Dysphagia Days is being investigated as a coprimary end point in an ongoing phase 3 trial evaluating cendakimab for the treatment of EoE.