Abstract
Background:
Eosinophilic esophagitis (EoE) disease severity is not typically reported in clinical practice.
Objectives:
To assess real-world EoE disease severity (assessed by physicians and using an adapted Index of Severity for EoE (I-SEE)), physician/patient characteristics, healthcare resource utilization (HCRU), and treatment patterns among adolescents and adults with EoE in the USA.
Design:
A noninterventional, retrospective, physician-reported medical chart review.
Methods:
Gastroenterologists and allergists/immunologists from a US nationwide panel completed a chart review of medical records (December 2021–January 2022) of patients aged ⩾11 years with a histologically confirmed diagnosis of EoE. Baseline data were collected ⩾6 months before diagnosis of EoE (index date), and study outcomes were collected ⩾6 months after the index date. The study outcomes assessed the EoE diagnostic process, disease severity, treatment patterns, and EoE-related HCRU. All data were stratified by physician-assessed EoE disease severity (mild, moderate, or severe (clinical severity and markers of severity)); index date data were also mapped to an adapted I-SEE scoring system post hoc.
Results:
Overall, 74 adolescents (11–17 years old) and 325 adults with EoE (⩾18 years old) were included; patient demographics were generally similar across severity levels. The presence of some symptoms (e.g., food impaction), allergic comorbidities (e.g., allergic rhinitis and food allergy (adults only)), duration between first symptom and index date, and the number of treatment lines received by patients increased with increasing physician-assessed EoE disease severity. However, fewer than 10% of adolescents and adults with EoE were assessed as having severe disease, and physician-assessed EoE disease severity was generally higher than the adapted I-SEE-assessed severity.
Conclusion:
Real-world US data indicated that most patients with EoE had mild or moderate EoE disease severity. The duration between first symptom and index date and the number of treatment lines received by patients with EoE increased with worsening EoE disease severity.
Introduction
Eosinophilic esophagitis (EoE) is a chronic, immune-mediated esophageal disease characterized by symptoms of esophageal dysfunction and eosinophilic inflammation in the esophageal mucosa (⩾15 eosinophils per high power field (eos/hpf)). 1 Early diagnosis of EoE is crucial; if left untreated, the disease can progress from an inflammatory to a fibrostenotic phenotype that is associated with stricture formation and esophageal remodeling.1–3 The clinical symptoms associated with EoE can vary by age4,5; difficulty swallowing (dysphagia) and food impaction are the most frequently reported symptoms in adolescents and adults.4,5
Clinical practice guidelines for EoE strongly recommend treatment with topical corticosteroids over no treatment. 6 Dietary restriction and proton-pump inhibitors (PPIs), which are used off-label, are both conditionally recommended.6,7 Since the publication of the American Gastroenterological Association and the Joint Task Force on Allergy-Immunology Practice Parameters (AGA/JTF) 2020 guidelines, Dupixent (dupilumab), a biologic therapy, has become the first US Food and Drug Administration (FDA)-approved treatment for patients aged 1 year and older with EoE.8,9 In 2024, Eohilia (budesonide oral suspension (BOS)) became the first US FDA-approved swallowed corticosteroid for the 12-week treatment of patients aged 11 years and older with EoE. 10
Despite the identification of standardized metrics and validated outcomes to monitor EoE and patient response to treatment, disease severity is not consistently measured in clinical practice. 11 Some metrics that could be used to assess EoE disease severity include: patient-reported outcomes,12,13 endoscopic assessments, 14 and histologic indicators of disease activity.15,16 A clinical severity index, known as the Index of Severity for EoE (I-SEE), was developed in 2022 by a multidisciplinary team of experts during an AGA-sponsored consensus conference; the AGA has developed an application to provide physicians with easy access to the I-SEE.11,17 This standardized EoE disease severity scoring system considers three domains: symptoms and complications; inflammatory features; and fibrostenotic features. Initial studies have shown that the I-SEE is suitable for capturing patient response to topical corticosteroid treatment in EoE, and results correlate with aspects of molecular disease activity.18–20 However, the distribution of clinical severity across patients with EoE in real-world clinical practice is not well established. This information could improve understanding of EoE disease features and consequently inform management strategies in EoE.
Here, we aim to assess the distribution of physician-assessed and an adapted I-SEE-assessed EoE disease severity among patients with EoE in the USA and identify any correlation between EoE disease severity and physician/patient characteristics or treatment patterns.
Methods
Study design and population
This retrospective cohort study recruited gastroenterologists and allergists/immunologists from a US nationwide panel of physicians from both academic and community practices. Recruited physicians completed an online, custom-designed electronic case report form (eCRF) for their patients with a histologically confirmed diagnosis of EoE (as reported in their patients’ medical charts) between December 18, 2021 and January 28, 2022. The targeted sample size of 400 medical charts from approximately 200 physicians was based on the maximum sample availability. The reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement.
The study design is shown in Figure 1. The baseline period was defined as the period at least 6 months before the first EoE diagnosis, the index date was defined as the date of the first EoE diagnosis, and the follow-up period was defined as the time between the date of the first EoE diagnosis and the date of chart abstraction (at least 6 months). Patients’ baseline clinical characteristics were collected for a period of at least 6 months before the index date, and study outcomes were evaluated over at least 6 months during the follow-up period after the index date. These data were collected from preexisting medical records of patients with EoE. Physicians were selected at random by Sago (previously Schlesinger Associates) from an online panel of physicians and were eligible for enrollment in this study if they were board-certified, licensed gastroenterologists, or allergists/immunologists in the USA; had been actively treating patients with EoE for at least the past 3 years; had treated at least five patients with EoE in the past 2 years; had direct access to patients’ medical charts containing information for at least 6 months before the first documented EoE diagnosis; and routinely assessed and recorded data related to patients’ EoE disease severity (as defined below).
Study design. Data were stratified by physician-assessed EoE disease severity; data were mapped to an adapted I-SEE scoring system at the index date post hoc.
Each participating physician could contribute a maximum of five charts. Charts were selected for data extraction via a randomization scheme: the eCRFs use a built-in algorithm to generate a random sequence of letters to determine the selection of the medical chart for review. The medical charts included in the study were for patients who had a histologically confirmed first diagnosis of EoE in the past 10 years; were at least 11 years old at the index date; had at least 6 months of medical history and treatment history data before the index date (baseline period); and had medical records available for at least 6 months and for at least one EoE-related physician visit following the index date. Patients with gastric or esophageal neoplasms, human immunodeficiency virus, or other life-threatening conditions during the baseline period were excluded from this study.
All study materials were reviewed by the Western Institutional Review Board-Copernicus Group Independent Review Board, and the study was granted exemption status per 45 code of federal regulations 46.104(d)(2).
Study outcomes
Physician characteristics and patient demographics and clinical characteristics at baseline, as well as the study outcomes, were captured and stratified by physician-assessed EoE disease severity (mild, moderate, or severe; detailed in the EoE disease severity classification section).
EoE diagnostic process
For the EoE diagnostic process, the outcomes that were assessed included: duration from the first EoE symptom to the index date, symptoms present when EoE was first suspected, symptoms present at the index date, and histopathologic and endoscopic support for diagnosis.
EoE disease severity classification
Physician-assessed EoE disease severity (the physician’s impression of whether the overall clinical picture for a patient was mild, moderate, or severe) was characterized using the most comprehensive clinical data available in each of their patients’ medical charts. The assessment methods included clinical severity (symptoms, histopathology, endoscopy, and EoE-related healthcare resource utilization (HCRU)) and markers of severity (treatment failure and presence of adaptive or coping behaviors).
The severity assessment data elements used by physicians were defined during a feasibility assessment that assessed how EoE disease severity was determined in a real-world clinical setting. The severity assessment data elements used were conclusions from histology or pathology reports, high-power microscopy peak eosinophil count, history of treatment failure with PPIs and topical corticosteroids, non-EoE-related comorbidities, number and frequency of esophageal dilations, number of food impaction removals and related emergency room visits, patient-reported adaptive behavior, treatment failure with dietary modification, and use of systemic steroids.
In addition to the physician assessment, patient data collected on symptoms and complications, inflammatory features, and fibrostenotic features were mapped to an adapted version of the I-SEE scoring system 11 at the index date post hoc. An adapted I-SEE system was used because data relating to a history of esophageal perforation and some fibrostenotic features were not available during this study (Supplemental Table 1) and because the I-SEE was developed after the initial study was conducted. 11
EoE phenotypes were defined using data extracted from patients’ medical charts and were based upon clinical features specific to EoE phenotypes, as previously described in the literature.21–24 The mild phenotype was characterized by a normal-appearing esophagus with relatively mild histologic and endoscopic changes. The refractory phenotype was characterized by an inflammatory and corticosteroid-refractory phenotype and high expression of cytokines and steroid-responding genes. The fibrostenotic phenotype was characterized by a narrow-caliber esophagus and a high degree of histologic and endoscopic severity, and low expression of epithelial differentiation genes.
Severity assessment
The average duration from the index date to the first severity assessment was captured. The medical chart review also identified the assessment methods most used by physicians to determine EoE disease severity in real-world clinical practice.
Patient treatment patterns
The number of lines of pharmacological treatment received for EoE (a line of treatment refers to the order in which patients received treatments (single or multiple concomitantly); the treatment lines included medications, dietary modifications, procedures, surgeries, or a combination of these) and the pharmacological treatments used after the index date were assessed.
EoE-related HCRU
HCRU was assessed during three time periods: from the index date to the first severity assessment, from the index date to the end of follow-up, and from the first severity assessment to the end of follow-up.
Data analysis
Analyses were conducted separately for adolescents (11–17 years old) and adults (at least 18 years old). Descriptive analyses were used to summarize physician and patient characteristics, severity assessments, HCRU, and treatment patterns; statistics were reported as numbers and percentages (categorical variables) or as means and standard deviations (SDs; continuous variables). HCRU measures were calculated as annualized rates, that is the number of occurrences of each HCRU outcome per year. Data were stratified by the physician-assessed EoE disease severity (mild, moderate, or severe). Appropriate tests (e.g., Chi-squared test, Fisher’s exact test (categorical variables), and analysis of variance tests (continuous variables)) were used for comparisons across the different severity groups; statistical significance (α = 0.05) was assessed.
Results
Physician characteristics
Overall, 126 gastroenterologists and 60 allergists/immunologists contributed 411 patient medical charts in total; 399 charts met the study eligibility criteria and were included. Each physician contributed an average (interquartile range (IQR)) of 2.2 (1.0, 3.0) charts. Physicians had spent a mean (SD) duration of 13.2 (7.3) years treating adolescent and/or adult patients with EoE and had treated a median (IQR) of 5.0 (0.0, 15.0) adolescent patients and 25.0 (10.0, 50.0) adult patients with EoE overall. Most physicians worked in a private practice (71.5% (133/186)), an academic hospital (21.5% (40/186)), or a nonacademic hospital (5.9% (11/186)); the four regions of the USA were well represented.
Patient baseline demographics and clinical characteristics
Overall, 74 adolescents (11–17 years old) and 325 adults (at least 18 years old) with EoE were included in this study. The mean (SD) age of adolescents and adults was 14.9 (1.8) and 32.9 (11.0) years, respectively; most patients were male (adolescents: 71.6%; adults: 74.5%) and White (adolescents: 75.7%; adults: 81.2%; Table 1). Most patients had at least one comorbidity, of which the most frequently reported in adolescents and adults were allergic rhinitis (50.0% and 31.7%, respectively) and asthma (41.9% and 33.8%, respectively; Table 1). The presence of allergic rhinitis and food allergy in adults increased significantly with increased EoE disease severity (p < 0.05).
Baseline demographics and clinical characteristics of adolescents and adults with EoE, stratified by physician-assessed EoE disease severity.
Statistical comparisons between severity groups were performed using ANOVA tests for continuous variables and Chi-squared tests for categorical variables. For categorical variables with expected counts less than 10, Fisher’s exact tests were used instead of Chi-squared. Statistically significant values are in bold (p < 0.05).
ANOVA, analysis of variance; EoE, eosinophilic esophagitis; NA, not applicable; SD, standard deviation.
EoE diagnostic process
The mean (SD) duration from the first EoE symptom to the index date was 11.3 (16.1) and 17.4 (31.8) months for adolescents and adults, respectively (Table 2); this increased significantly with increasing EoE disease severity (p < 0.001). When EoE was first suspected, reports of food impactions in adolescents and reports of dysphagia, food impactions, regurgitation, and vomiting in adults significantly increased with increasing EoE disease severity (all, p < 0.01; Figure 2(a) and (b)). At the index date, reports of dysphagia and food impactions numerically increased with increasing EoE disease severity in adolescents (Figure 3(a)). However, in adults, reports of food impactions, regurgitation, and weight loss at the index date significantly increased with EoE disease severity (all p < 0.01; Figure 3(b)). The proportion of adults with endoscopic findings indicative of EoE and the proportion of adolescents and adults with marked/elevated eosinophilia (determined at the physician’s discretion, based on information from the patients’ medical charts) or a peak eosinophil count of ⩾15 eos/hpf generally increased with increasing EoE disease severity. Endoscopic findings were generally based on biopsies from multiple sites (distal and proximal), irrespective of EoE disease severity (Table 2).
Patient EoE diagnosis history and endoscopic findings in support of EoE diagnosis, stratified by physician-assessed EoE disease severity.
Statistical comparisons between severity groups were performed using ANOVA tests for continuous mean variables and Chi-squared tests for categorical variables. For categorical variables with expected counts less than 10, Fisher’s exact tests were used instead of Chi-squared. Statistically significant values are in bold (p < 0.05).
Patients with a histologically confirmed EoE diagnosis, but no eosinophil count recorded in their medical chart, were included and thus not all patients who enrolled had a peak eosinophil count of ⩾15 eos/hpf; marked/elevated eosinophilia was determined at the physician’s discretion, based on information from the patients’ medical charts.
ANOVA, analysis of variance; EoE, eosinophilic esophagitis; eos/hpf, eosinophils per high power field; IQR, interquartile range; SD, standard deviation.
Symptoms present when EoE was first suspected in (a) adolescent and (b) adult patients with EoE, stratified by physician-assessed EoE disease severity. Statistical comparisons between severity groups were performed using Chi-squared tests; for categorical variables with expected counts less than 10, Fisher’s exact tests were used instead of Chi-squared. Statistically significant values are in bold (p < 0.05).
Symptoms present at the index date in (a) adolescent and (b) adult patients with EoE, stratified by physician-assessed EoE disease severity. Statistical comparisons between severity groups were performed using Chi-squared tests; for categorical variables with expected counts less than 10, Fisher’s exact tests were used instead of Chi-squared. Statistically significant values are in bold (p < 0.05).
EoE disease severity classification
Patient demographics were generally similar across each of the EoE disease severity groups (Table 1). Physician-assessed EoE disease severity was classified as mild, moderate, and severe in 40.5%, 55.4%, and 4.1% of adolescents and in 32.9%, 56.3%, and 10.8% of adults, respectively (Figure 4). Adapted I-SEE scores also demonstrated that most patients in the study had mild or moderate disease (mild: 70.7% (282/399); moderate: 25.1% (100/399)); only a small proportion of patients were classified as having severe disease (4.3% (17/399)). Overall, although the physician-assessed EoE disease severity was generally higher than the adapted I-SEE-assessed severity (moderate vs mild; Figure 4), the proportion of patients with severe EoE was similar across the two severity classification methods. The distribution of EoE disease severity (both physician- and I-SEE-assessed) was consistent for adolescent and adult patients.
Proportion of patients with mild, moderate, or severe disease severity based on a physician-assessed and an adapted I-SEE severity score.
A mild EoE phenotype was observed in 43.2% and 40.3%, a refractory phenotype in 32.4% and 34.5%, and a fibrostenotic phenotype in 4.1% and 10.2% of adolescents and adults, respectively (Supplemental Table 2). The EoE phenotype correlated significantly with physician-assessed EoE disease severity across both populations (p < 0.001).
Severity assessment
The mean (SD) duration from the index date to the first severity assessment was 4.0 (6.4) and 4.7 (10.9) months for the adolescent and adult populations, respectively. Overall, patients with severe disease experienced the shortest duration from the index date to the first severity assessment; this result was not statistically significant in either patient population (Table 2). The most used severity assessment methods were presence of symptoms, endoscopic findings/reference scores, histopathology reports, and treatment failure with PPIs (Figure 5(a) and (b)). The use of histopathology reports and treatment failure with PPIs increased significantly (p < 0.001) with increasing EoE disease severity in the adult population (Figure 5(b)).
Severity assessment methods used for (a) adolescent and (b) adult patients with EoE, stratified by physician-assessed EoE disease severity. Statistical comparisons between severity groups were performed using Chi-squared tests; for categorical variables with expected counts less than 10, Fisher’s exact tests were used instead of Chi-squared. Statistically significant values are in bold (p < 0.05).
Patient treatment patterns
In this study, a line of treatment can include medications, dietary modifications, procedures, surgeries, or a combination of these; no US FDA-approved medications were available at the time this study was conducted. The most frequently used treatments following the index date (regardless of EoE disease severity) were PPIs as first-line treatment (adolescents: 79.7%; adults: 74.2%) and topical corticosteroids as both second-line treatment (adolescents: 39.5%; adults: 49.4%) and third-line treatment (adolescents: 36.4%; adults: 23.1%; Supplemental Table 3). Dietary restrictions or modifications were most used as a first line of treatment (adolescents: 31.1%; adults: 27.4%); the use of dietary restrictions or modifications did not correlate with EoE disease severity across any treatment line. Following the index date, most patients with mild disease severity were receiving one line of treatment (adolescents: 60.0%; adults: 68.2%). Most patients with moderate or severe disease were receiving a second line of treatment (adolescents: 48.8% (moderate), 66.7% (severe); adults: 50.8% (moderate), 45.7% (severe); Figure 6) owing to treatment failure or discontinuation of the first line of treatment.
Proportion of patients receiving 1, 2, 3, or >3 lines of treatment after the index date, stratified by physician-assessed EoE disease severity. Statistical comparisons between severity groups were performed using Chi-squared tests.
EoE-related HCRU
Overall, annualized rates of HCRU were lower in adolescent than in adult patients with EoE from the index date to the first severity assessment (Supplemental Table 4). In general, HCRU was lower in both adolescents and adults from the first severity assessment to the end of follow-up, and from the index date to the end of follow-up compared with the index date to the first severity assessment. During the period from the index date to the first severity assessment, outpatient office-based visits and upper endoscopies were the most frequently used healthcare resources for both age groups (Supplemental Table 4). There were no significant trends in HCRU based on EoE disease severity from the index date to the first severity assessment in the adolescent population; however, the mean number of inpatient admissions (hospitalizations) per year increased significantly with increasing EoE disease severity in the adult population. During this same period, the mean numbers of food impaction removals, emergency room visits, and dilation procedures per year were higher in adults with moderate and severe EoE than in those with mild EoE, although these differences were not statistically significant. The mean (SD) number of upper endoscopies per year from the index date to the first severity assessment was lower for adolescents than for adults (5.3 (6.8) vs 13.2 (33.9); Supplemental Table 4). In contrast, from the first severity assessment to the end of follow-up, the mean (SD) number of upper endoscopies per year among adolescents and adults was similar; however, the total number of upper endoscopies was lower (0.5 (0.9) vs 0.7 (1.0)) than during the index date to the first severity assessment. The corresponding data from the index date to the end of follow-up were 1.0 (1.1) for adolescents and 1.2 (1.1) for adults. From the first severity assessment to the end of follow-up, the mean number of upper endoscopies and dilation procedures for the adolescent population and the mean number of food impaction removals and dilation procedures for the adult population increased significantly with increasing EoE disease severity (Supplemental Table 4). A similar pattern across severity groups was observed from the index date to the end of follow-up for the number of upper endoscopies and dilation procedures in adolescents and the number of food impaction removals in adults (Supplemental Table 4).
Discussion
Despite the growing prevalence and understanding of EoE, 25 disease severity is not typically reported in clinical practice. 11 In this retrospective study, physician-assessed EoE disease severity, based on clinical severity and markers of severity, indicated that fewer than 10% of adolescents and adults with EoE had severe disease. Overall, the physician-assessed EoE disease severity was generally higher than the adapted I-SEE-assessed severity, 11 considering the differences observed between mild and moderate disease severity. Presence of symptoms, endoscopic findings, and histopathology reports were most frequently used by physicians to assess EoE disease severity. The duration between the first symptom and the index date, the number of treatment lines received by patients, and certain HCRU increased with increasing EoE disease severity.
The symptoms of EoE are nonspecific, which often leads to delayed diagnosis.26,27 In this study, the mean (SD) duration from the first EoE symptom to the index date was shorter in adolescents (11.3 (16.1) months) than in adults (17.4 (31.8) months). It took patients with more severe EoE significantly longer to receive a diagnosis following the first symptom presentation compared with patients with less severe EoE; this finding should prompt earlier medical intervention following symptom onset. However, these data align with the current understanding of EoE disease progression; persistent, untreated eosinophil-predominant inflammation can lead to fibrostenosis and tissue remodeling, resulting in exacerbated esophageal dysfunction and a more severe presentation.28–31 These findings could suggest a lack of awareness of the natural history of EoE among physicians in the US, or that physicians may not be fully aware of patients’ symptoms, perhaps as a result of dietary elimination and adaptive behaviors. The presence of some symptoms was found to increase with increasing EoE disease severity, so it can also be hypothesized that patients with more severe EoE may have been misdiagnosed or may have delayed seeking medical advice from their provider, owing to increased symptom tolerance, or the presence of symptoms that were not disease-specific.32,33
Following diagnosis, first-, second-, and third-line treatment options for EoE were independent of EoE disease severity. The treatments used aligned with the AGA/JTF 2020 clinical guidelines for EoE management, which strongly recommended topical corticosteroids over no treatment and conditionally recommended PPIs as primary pharmacologic choices. 6 However, a greater proportion of patients received PPIs than topical corticosteroids first-line, despite this difference in strength of recommendation in the clinical practice guidelines available at the time. 6 This could be related to the ease of administration, access to and/or familiarity with PPIs, or a lack of availability of compounding pharmacies for preparation of topical corticosteroids. 34 Furthermore, although not mentioned in the guidelines at the time, 6 histamine-2-receptor blockers were used by approximately 15% of adolescents and adults first-line. However, this result is surprising as there is no evidence to support that this medication class is effective in treating EoE. 35 Systemic corticosteroids were not used first-line but were used in approximately 10%–15% of patients as a second- or third-line treatment option. This may also be considered relatively high when taking into account that topical corticosteroids were recommended over systemic corticosteroids in the clinical practice guidelines. 6 Reasons for this may include patient preference for administration method, concomitant treatment of non-EoE related conditions or, as mentioned, the lack of availability of compounding pharmacies. 34 It should be noted that an update to the EoE clinical guidelines has recently been published. 35 A greater number of treatment lines were used by patients with moderate or severe disease than by those with mild disease, perhaps indicating that initial treatment strategies may need to be tailored to EoE disease severity. Current literature indicate that dupilumab could be used to treat patients with refractory or severe EoE.8,36 However, neither dupilumab nor BOS was examined because they were approved by the US FDA after this study was conducted, and after the publication of the current clinical EoE guidelines6,8–10; further studies are needed to assess where on the EoE disease spectrum these treatments are used in the real world.
HCRU trends were generally similar in adolescents and adults during follow-up and somewhat unexpectedly did not consistently increase with increasing EoE disease severity. Although a trend toward increased HCRU was observed in adults with moderate and severe EoE compared with those with mild EoE (during the period from the index date to the first severity assessment), these findings were not statistically significant. Given the short median duration from the index date to the first severity assessment (adolescents: 2.0 months; adults: 1.1 months), and the likelihood that most patients may have had at least one physician visit during this period, annualizing these HCRU rates may be an overestimation. Patients may have also undergone initial procedures that did not need to be repeated after severity assessment. Annualized HCRU data recorded during this period should therefore be interpreted with caution. During the period from the first severity assessment to the end of follow-up, the mean number of dilation procedures increased significantly with increasing EoE disease severity for adolescents and adults. Current EoE guidelines recommend esophageal dilation in adult patients with dysphagia associated with an esophageal stricture over no dilation. 6 The literature suggests that dilation is typically used for patients with refractory and persistent EoE.37–39 The data from this study align with this and suggest that dilation is used more commonly in patients with severe EoE because these patients typically have a greater degree of fibrostenosis, potentially due to longer symptom durations before diagnosis. The HCRU data also suggested that adults were more likely than adolescents to undergo a dilation procedure or food impaction removal. These procedures were less commonly reported following a disease severity assessment during the follow-up period compared with the period before first severity assessment, which could be attributed to symptom improvement owing to disease management following a diagnosis.
To address the lack of consistent metrics to determine EoE disease severity, the I-SEE was developed by a multidisciplinary team of experts during an AGA-sponsored consensus conference in 2021 (published after this study was completed). 11 An adapted I-SEE scoring system was applied post hoc to the data set in this study, and we found that most patients had mild or moderate EoE disease severity (adolescents: 94.6%; adults: 96.1%). This aligns with the distribution of physician-assessed EoE disease severity in this study (adolescents: 95.9%; adults: 89.2%). However, the adapted I-SEE identified a substantially higher proportion of patients with mild disease (adolescents: 74.3%; adults: 69.9%) compared with the physician-assessed EoE disease severity (adolescents: 40.5%; adults: 32.9%). Data were not available on esophageal perforation or fibrostenotic histology, so the adapted scoring system used here may have slightly underestimated the severity scores.
This general disease severity distribution during this study, in which most patients had mild or moderate disease, aligns with recent literature in which the I-SEE metric has been applied to other clinical data sets and indicates that approximately 79%–86% of patients with EoE have mild or moderate disease.18,19 However, the physician- and I-SEE-assessed EoE disease severity measured in this study, which was largely reflective of community practice settings, was lower than previously reported in academic settings.18,19
The I-SEE scoring system used in this study was adapted from the original instrument, 11 which should be considered when interpreting the data. The three domains remain the same in the two systems (i.e., symptoms and complications, inflammatory features, and fibrostenotic features), consistent with the core pathological features of EoE (symptoms, endoscopy, and histology). 35 The I-SEE was adapted to reflect the data available from this study (Supplemental Table 1). The original instrument awarded points based on the frequency of symptoms; however, this information was not available in our data set, so the adapted I-SEE captured the different types of symptoms. Changes were also made to acknowledge differences in the available data for inflammatory, endoscopic, and histological features (see footnotes to Supplemental Table 1 for full details). Despite these amendments, the adapted I-SEE remains a robust, clinically useful tool for the assessment of EoE disease severity in a broad range of patients.
As with any trial, this study is subject to several limitations. Even though this study explored disease characteristics and HCRU in both adolescents and adults with EoE, it did not include younger pediatric patients (aged <11 years) or pediatricians, limiting the generalizability of the results. Pediatric patients with EoE can present with distinct symptoms (e.g., feeding/eating difficulties, failure to thrive) and require different management strategies (e.g., a greater focus on diet restriction and elemental diets) than older patients.40,41 There remains the need for a companion study to focus on the real-world severity patterns among pediatric patients with EoE (aged <11 years). Biases due to the retrospective design may have introduced heterogeneity into the data. As a study that used real-world evidence, data collection was limited by the availability of clinical data in patients’ medical charts. For example, PPIs are available as over-the-counter medications in the USA. Therefore, some patients may have been taking PPIs in addition to their prescribed medications, and this potentially confounding information would not have been available in the medical records. Further studies are needed to assess the impact of the US FDA approval of dupilumab and BOS on treatment patterns, and the effects of these newly approved treatments on EoE in clinical practice.9,10 Selection bias might exist, and the sample may not be representative of the entire EoE population, owing to the voluntary participation of physicians. However, the study participants were selected from a nationwide online panel and included physicians from private, academic, and community-based practices, which aimed to provide a representative sample of physicians and patients across the USA. Importantly, most patients were not from academic centers, meaning that vital data from a community setting were captured. Molecular and biological components were not captured and thus EoE endotypes were not assigned. 24 Predicting disease prognosis based on endotypes for individual patients remains a significant clinical challenge. Better understanding of EoE endotypes may facilitate better disease management, improve outcomes, and optimize HCRU. 24 Therefore, this would be an important area for future studies, particularly given the increasing role of the personalization of treatments for EoE 24 ; however, it was not possible to examine in this present study owing to the lack of access to esophageal biospecimens. Finally, as mentioned, the I-SEE scoring system had to be adapted and was applied to this data set post hoc at the index date only, owing to the limited availability of the necessary data parameters. Hence, the I-SEE results may not provide a detailed characterization of patient disease severity and should be interpreted with caution.
A strength of this study was that a wide range and a large number of geographically diverse providers, mostly from community practice settings, with standardized data collection methods, were utilized. The patient population enrolled in this study was also more racially diverse than is typically observed in other real-world studies of EoE; even though EoE is more commonly reported in White patients, 42 during this study, 8.8% of patients were Asian and 8.0% of patients were Black or African American.36,43 Additionally, patients’ medical charts were selected based on clear inclusion and exclusion criteria to ensure a broad EoE patient population and to allow for generalizability of data. Therefore, this study contributes to the limited literature on real-world clinical severity in EoE.
Conclusion
Based on physician assessment, approximately 90% of patients with EoE in a real-world, community setting had mild or moderate disease. Physician-assessed EoE disease severity was generally greater than severity assessed by the adapted I-SEE scoring system. Physicians typically used a combination of the presence of symptoms, endoscopic findings, and histopathology reports to determine EoE disease severity. A greater number of treatment lines were taken by patients with moderate or severe disease than by those with mild disease, owing to previous treatment failure or treatment discontinuation. This may suggest that initial treatment should be tailored to perceived EoE disease severity and that improving our understanding of the real-world severity distribution of EoE, as well as treatment responses following different therapies for each EoE disease severity class, has the potential to help guide management strategies.
Supplemental Material
sj-docx-1-tag-10.1177_17562848251347361 – Supplemental material for Assessment of real-world disease severity in patients with eosinophilic esophagitis in the United States
Supplemental material, sj-docx-1-tag-10.1177_17562848251347361 for Assessment of real-world disease severity in patients with eosinophilic esophagitis in the United States by Evan S. Dellon, Glenn T. Furuta, Paul Feuerstadt, Priya Bansal, Todor I. Totev, Oscar Patterson-Lomba, Hongjue Wang, Sanjana Sundaresan, Rajeev Ayyagari, Tao Fan, Jeanne Jiang, Mena Boules and Brian Terreri in Therapeutic Advances in Gastroenterology
Footnotes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
