Portal de Periódicos da CAPES

A Summary of the Meetings of the Development of a Core Outcome Set for Therapeutic Studies in Eosinophilic Esophagitis (COREOS) International Multidisciplinary Consensus

2021; Elsevier BV; Volume: 161; Issue: 3 Linguagem: Inglês

10.1053/j.gastro.2021.04.081

1528-0012

Christopher Ma, Alain Schoepfer, Ekaterina Safroneeva, Christopher Ma, Alain Schoepfer, Evan S. Dellon, Albert J. Bredenoord, Mirna Chehade, Margaret H. Collins, Brian G. Feagan, Glenn T. Furuta, Sandeep K. Gupta, Ikuo Hirano, Vipul Jairath, David A. Katzka, Rish K. Pai, Marc E. Rothenberg, Alex Straumann, Seema S. Aceves, Jeffrey A. Alexander, Nicoleta C. Arva, Dan Atkins, Luc Biedermann, Carine Blanchard, Antonella Cianferoni, Constanza Ciriza de los Ríos, Frederic Clayton, Carla M. Davis, Nicola de Bortoli, Jorge Amil Dias, Gary W. Falk, Robert M. Genta, Gisoo Ghaffari, Nirmala Gonsalves, Thomas Greuter, Russell J. Hopp, Karen S. Hsu Blatman, Elizabeth T. Jensen, Doug Johnston, Amir F. Kagalwalla, Helén Larsson, John Leung, Hubert Louis, Joanne C. Masterson, Calies Menard‐Katcher, Paul Menard‐Katcher, Fouad J. Moawad, Amanda B. Muir, Vincent A. Mukkada, Roberto Penagini, Robert D. Pesek, Kathryn A. Peterson, Philip E. Putnam, Alberto Ravelli, Edoardo Savarino, Christoph Schlag, Philipp Schreiner, Dagmar Simon, Thomas C. Smyrk, Jonathan M. Spergel, Tiffany Taft, Ingrid Terreehorst, Tim Vanuytsel, Carina Venter, Mário C. Vieira, Michael Vieth, Berber Vlieg‐Boerstra, Ulrike von Arnim, Marjorie M. Walker, Joshua B. Wechsler, Philip Woodland, John T. Woosley, Guang‐Yu Yang, Noam Zevit, Ekaterina Safroneeva,

Delphi Technique in Research

The Core Outcome Set for Therapeutic Studies in Eosinophilic Esophagitis (COREOS) collaborators are a group of more than 70 gastroenterologists, pathologists, allergists, researchers, dietitians, psychologists, and methodologists who convened in a series of in-person and virtual meetings between 2018 and 2020 with the aim of developing a core outcome set (COS) for use in therapeutic studies of pharmacologic and dietary therapies for the treatment of eosinophilic esophagitis (EoE). Given heterogeneity in reported outcomes and uncertainties regarding the most appropriate end points for use in both randomized controlled trials (RCTs) and observational studies involving EoE patients, the EoE experts launched the COREOS exercise in 2018 to standardize outcome definitions using methods established by the Core Outcome Measures in Effectiveness Trials (COMET) initiative.1Williamson P.R. Altman D.G. Bagley H. et al.The COMET Handbook: version 1.0.Trials. 2017; 18: 280Crossref PubMed Scopus (847) Google Scholar,2Kirkham J.J. Davis K. Altman D.G. et al.Core Outcome Set-STAndards for Development: The COS-STAD recommendations.PLoS Med. 2017; 14e1002447Crossref PubMed Scopus (302) Google Scholar The COS was developed using a multiphase approach, which is summarized in Figure 1. In the first phase, systematic reviews of the literature and patient engagement surveys were conducted to identify candidate outcomes that have been previously measured and are important to patients with EoE. Next, this information was used to build a framework of different outcome domains, and working groups for each domain were assembled to review the literature for relevant end points.3Warners M.J. Hindryckx P. Levesque B.G. et al.Systematic review: disease activity indices in eosinophilic esophagitis.Am J Gastroenterol. 2017; 112: 1658-1669Crossref PubMed Scopus (37) Google Scholar, 4Ma C. van Rhijn B.D. Jairath V. et al.Heterogeneity in clinical, endoscopic, and histologic outcome measures and placebo response rates in clinical trials of eosinophilic esophagitis: a systematic review.Clin Gastroenterol Hepatol. 2018; 16: 1714-1729.e3Abstract Full Text Full Text PDF PubMed Scopus (24) Google Scholar, 5Schoepfer A.M. Schürmann C. Trelle S. et al.Systematic Review of Outcome Measures Used in Observational Studies of Adults with Eosinophilic Esophagitis.Int Arch Allergy Immunol. 2021; (Ahead of print)Google Scholar, 6Safroneeva E. Balsiger L. Hafner D. et al.Adults with eosinophilic oesophagitis identify symptoms and quality of life as the most important outcomes.Aliment Pharmacol Ther. 2018; 48: 1082-1090Crossref PubMed Scopus (20) Google Scholar The relative importance of these domains was categorized in a Delphi survey as core, important, and research agenda domains, and discussed in a moderated in-person meeting on May 17, 2019 at Digestive Disease Week (San Diego, CA). In phase 3, a comprehensive list of outcome measures within each of the core domains was evaluated by the COREOS collaborators in a 2-round Delphi survey and, finally, outcomes were ratified in a virtual meeting on December 8, 2020. In this meeting summary, we highlight the major points of discussion that occurred during the development of the EoE COS. The COREOS steering committee first convened in person on June 5, 2018 at Digestive Disease Week 2018 in Washington, DC. This was followed by a teleconference on July 13, 2018. These first kick-off meetings aimed to establish timelines, primary objectives, strategic partnerships, participant recruitment, and roles and responsibilities for the group. The committee agreed that the aim of the COREOS collaboration would be to identify core outcomes in EoE to improve the efficiency of drug development and the quality of comparative research and ultimately shape clinical practice by selecting end points relevant to patients and physicians. The scope of the COS was set to include RCTs and observational studies, in both adult and pediatric populations, for dietary and pharmacologic therapies for EoE. Dilation was considered, although not included, as end points for endoscopic procedures (eg, technical success) are substantively different from those for anti-inflammatory therapies. The Committee agreed that the collaboration would have a global and multidisciplinary scope. Expert panelists were required to have established experience in EoE, as demonstrated by at least 5 peer-reviewed publications in the preceding 10 years; participation in at least 1 RCT in EoE in the preceding 5 years; or specific methodological expertise with respect to conducting COS exercises, trial design, or instrument development. Invitations to participate in the COREOS collaboration were distributed to members of the Consortium of Eosinophilic Gastrointestinal Disease Researchers (CEGIR); European Society of Eosinophilic Esophagitis (EUREOS); and Eosinophil Gastrointestinal Disorders Committee of The American Academy of Allergy, Asthma, and Immunology. The final group of collaborators included gastroenterologists, allergists, pathologists, dietitians, psychologists, and methodologists. A total of 11 core outcome domains were identified; in a Delphi survey, 66 panelists voted the following 4 outcome domains as critical for inclusion: patient-reported symptoms, EoE-specific quality of life, histopathology, and endoscopy. Domains considered important but optional for inclusion included genetic profiling, biomarkers, esophageal distensibility, immunologic dissection, and patient perception of health. Secondary impact on family/caregivers and resource utilization were voted as research agenda domains. The COREOS collaborators met in person on May 17, 2019 at Digestive Disease Week in San Diego, CA to review these results and identify important areas for COS development moving into phase 3. A total of 29 COREOS members attended the meeting, with broad geographic representation. The meeting was chaired by Drs Christopher Ma and Ekaterina Safroneeva. Dr Safroneeva opened the meeting by reviewing the aims of the COREOS collaboration and reviewing the results of the outcome domain Delphi survey. Each subsequent session was focused on 1 of the core domains (ie, histology, endoscopy, and symptoms and quality of life), with a presentation by expert opinion leaders in that domain followed by an open forum discussion. Each presentation focused on key outcome measures within the domain and considerations for COS selection. Drs Rish Pai (Mayo Clinic Arizona) and Margaret Collins (Cincinnati Children's Hospital) led the discussion regarding histopathology in EoE. Given the central role of eosinophilic inflammation in the diagnosis of EoE, there was agreement that histology should be a core outcome domain in the COS. Topics of discussion included use of the peak eosinophil count vs the EoE Histology Scoring System,7Collins M.H. Martin L.J. Alexander E.S. et al.Newly developed and validated eosinophilic esophagitis histology scoring system and evidence that it outperforms peak eosinophil count for disease diagnosis and monitoring.Dis Esophagus. 2017; 30: 1-8PubMed Google Scholar considerations for biopsy acquisition (including capturing subepithelial tissue and standardization of microscope field size), and advantages and disadvantages of different measures of histologic activity (including reliability, responsiveness, correlation with patient symptoms and quality of life, feasibility for measurement, and regulatory considerations for RCTs).8Warners M.J. Ambarus C.A. Bredenoord A.J. et al.Reliability of histologic assessment in patients with eosinophilic oesophagitis.Aliment Pharmacol Ther. 2018; 47: 940-950Crossref PubMed Scopus (41) Google Scholar,9Hirano I. Collins M.H. Assouline-Dayan Y. et al.RPC4046, a monoclonal antibody against IL13, reduces histologic and endoscopic activity in patients with eosinophilic esophagitis.Gastroenterology. 2019; 156: 592-603.e10Abstract Full Text Full Text PDF PubMed Scopus (144) Google Scholar The experts discussed concepts important for COS development, which included defining appropriate measurement tools and measurement conventions for histologic activity, and identifying thresholds for histologic remission. Drs Ikuo Hirano (Northwestern University), Amanda Muir (Children's Hospital of Philadelphia), and Christopher Ma (University of Calgary) presented on endoscopic assessment. The EoE Endoscopic Reference Score (EREFS)10Hirano I. Moy N. Heckman M.G. et al.Endoscopic assessment of the oesophageal features of eosinophilic oesophagitis: validation of a novel classification and grading system.Gut. 2013; 62: 489-495Crossref PubMed Scopus (490) Google Scholar was identified as the most likely tool to be used for assessing endoscopic disease activity, with demonstrated content validity, inter- and intra-rater reliability, and responsiveness.3Warners M.J. Hindryckx P. Levesque B.G. et al.Systematic review: disease activity indices in eosinophilic esophagitis.Am J Gastroenterol. 2017; 112: 1658-1669Crossref PubMed Scopus (37) Google Scholar For purposes of COS development, the experts discussed the most appropriate method to calculate the EREFS (ie, simple score vs weighted scoring and inflammatory vs fibrostenotic subscores11Dellon E.S. Cotton C.C. Gebhart J.H. et al.Accuracy of the eosinophilic esophagitis endoscopic reference score in diagnosis and determining response to treatment.Clin Gastroenterol Hepatol. 2016; 14: 31-39Abstract Full Text Full Text PDF PubMed Scopus (125) Google Scholar,12Schoepfer A.M. Hirano I. Coslovsky M. et al.Variation in endoscopic activity assessment and endoscopy score validation in adults with eosinophilic esophagitis.Clin Gastroenterol Hepatol. 2019; 17: 1477-1488.e10Abstract Full Text Full Text PDF PubMed Scopus (12) Google Scholar), appropriate definitions of endoscopic response and remission, and differentiation between impassable rings vs strictures. There was also a discussion around the use of distensibility measurements; this domain was voted to be important but not essential in the Delphi panel. Potential advantages of impedance planimetry as an esophageal "function test" that better measures esophageal caliber and distensibility,13Lin Z. Kahrilas P.J. Xiao Y. et al.Functional luminal imaging probe topography: an improved method for characterizing esophageal distensibility in eosinophilic esophagitis.Therap Adv Gastroenterol. 2013; 6: 97-107Crossref PubMed Scopus (56) Google Scholar is correlated with dysphagia symptoms, and has shown responsiveness to anti-inflammatory therapy and in a phase 2 trial of dupilumab14Hirano I. Dellon E.S. Hamilton J.D. et al.Efficacy of dupilumab in a phase 2 randomized trial of adults with active eosinophilic esophagitis.Gastroenterology. 2020; 158: 111-122.e10Abstract Full Text Full Text PDF PubMed Scopus (206) Google Scholar were reviewed. However, considerations of cost, availability, and incompletely characterized operating properties were identified as reasons for excluding distensibility from a core outcome that needs to be measured in every trial. Patient-reported symptoms and quality of life (QoL) measures were reviewed by Drs Evan Dellon (University of North Carolina) and Alain Schoepfer (Lausanne University Hospital) for adult patients, and by Drs Mirna Chehade (Mount Sinai, New York) and Sandeep Gupta (Indiana University) for pediatric populations. For adult populations, the experts reviewed advantages and disadvantages of symptom-based patient-reported outcome measures, including the EoE Activity Index15Schoepfer A.M. Straumann A. Panczak R. et al.Development and validation of a symptom-based activity index for adults with eosinophilic esophagitis.Gastroenterology. 2014; 147: 1255-1266.e21Abstract Full Text Full Text PDF PubMed Scopus (181) Google Scholar and Dysphagia Symptoms Questionnaire.16Dellon E.S. Irani A.M. Hill M.R. et al.Development and field testing of a novel patient-reported outcome measure of dysphagia in patients with eosinophilic esophagitis.Aliment Pharmacol Ther. 2013; 38: 634-642Crossref PubMed Scopus (102) Google Scholar There was considerable conversation regarding appropriate terminology relating to dysphagia and previous qualitative work to best elucidate this symptom. Several QoL measures were also reviewed; the EoE-QoL-A questionnaire was identified as the only validated measure of health-related QoL for adults with EoE with internal consistency, test–retest reliability, and validity.17Taft T.H. Kern E. Kwiatek M.A. et al.The adult eosinophilic oesophagitis quality of life questionnaire: a new measure of health-related quality of life.Aliment Pharmacol Ther. 2011; 34: 790-798Crossref PubMed Scopus (92) Google Scholar Identifying the most appropriate tool for assessing symptoms (with consideration of feasibility, scientific merit, and regulatory requirements), defining the recall period (daily vs 7 days), and defining thresholds for response/remission that are correlated with biologic activity were judged by experts to be important concepts for COS development in adults. In pediatric populations, experts identified several potential barriers to accurate patient-reported outcome measure assessment, including the large variability in symptom presentations, the variance in presentations across different age groups from infancy to adolescence, and the accuracy of self-reporting vs parental/proxy-reporting. The broad range of symptoms was identified as potentially problematic, particularly in the RCT setting. While some panelists proposed to include only dominant symptoms for the purposes of streamlining trial recruitment, others argued that this would lead to the exclusion of a sizable proportion of pediatric patients. Although different symptoms scores were reviewed, the Pediatric EoE Symptom Score, version 2.0, was identified as the most likely candidate for inclusion into COS, as this instrument captures multiple symptom complexes (including dysphagia, reflux, nausea/vomiting, and pain) and demonstrates responsiveness to elimination diet.18Franciosi J.P. Hommel K.A. DeBrosse C.W. et al.Development of a validated patient-reported symptom metric for pediatric eosinophilic esophagitis: qualitative methods.BMC Gastroenterol. 2011; 11: 126Crossref PubMed Scopus (77) Google Scholar Discussions from this meeting, in combination with the systematic review and patient engagement surveys, were used to inform the development of the COS statements for voting in phase 3. A total of 122 statements across the 4 core outcome domains were developed for the first round of the phase 3 Delphi surveys. Before the survey was distributed, an introductory videoconference was held March 19, 2020 to review the aims of the survey, instructions for survey completion, and the 9-point Likert scoring system (with outcomes scored in the 7–9 range being considered critical for inclusion in a COS, outcomes scored in the 4–6 range being considered important but not critical, and outcomes scored in the 1–3 range being considered of limited importance).1Williamson P.R. Altman D.G. Bagley H. et al.The COMET Handbook: version 1.0.Trials. 2017; 18: 280Crossref PubMed Scopus (847) Google Scholar This videoconference was recorded and distributed to all panelists, including those who were unable to attend. After the round 1 Delphi survey, all responses were reviewed by the lead and senior investigators. A priori–defined rules were used to determine which outcomes were carried forward to round 2: outcomes scored in the 7–9 range by ≥50% of panelists and 1–3 range by <15% of panelists were carried forward. A total of 59 outcomes were included in the round 2 survey, which was distributed to panelists with a report containing their own scoring from the first round, comments, and the group scores. Outcomes that were voted by ≥70% of panelists in the 7–9 range and <15% of panelists in the 1–3 range were defined as meeting consensus for inclusion. The final outcomes to be included in the COS were ratified in a moderated videoconference held December 8, 2020. Although this was initially planned as a face-to-face meeting with all stakeholder groups, this was amended to a virtual meeting due to COVID-19 public health restrictions. The meeting was chaired by Drs Christopher Ma and Ekaterina Safroneeva. All panelists who participated in any round of the Delphi surveys were invited to attend. The meeting date and time were chosen to accommodate the maximum number of participants: a total of 27 experts attended the ratification meeting, representing a broad range of disciplines, practice settings, and geographic regions. Dr Christopher Ma opened the meeting by summarizing the results from round 2 of the Delphi survey, outlining the objectives of the ratification meeting, and reviewing the voting procedures. All items reaching consensus in round 2 were discussed, as were all items that had a reasonable likelihood of potentially meeting the criterion for inclusion (defined as outcomes for which the upper 95% confidence interval of the proportion of panelists voting in the 7–9 category exceeded 70%). A total of 42 items were reviewed in the ratification meeting. Discussion on the merits and pitfalls of each outcome was encouraged, and panelists subsequently voted anonymously on each item to either "include in the COS," "do not include in the COS," or "unsure." Items receiving ≥70% of votes in the "include in the COS" category and <15 of votes in the "do not include in the COS" category were ratified for inclusion in the final COS. Advantages, disadvantages, and points of consideration raised by the ratification panel for each outcome domain are summarized in Table 1. The major topics of discussion captured in the meeting included:1.Histology end points: whether the peak eosinophil count should be reported using high-power field vs adjusted per mm2, appropriate thresholds for histologic remission (≤6 eosinophils/high-power field vs <15 eosinophils/high-power field), and utility of the EoE Histology Scoring System in RCTs vs observational studies2.Endoscopy end points: most appropriate scoring conventions for the EREFS (scoring from 0 to 8 vs from 0 to 9), thresholds for endoscopic remission based on total score or inflammatory vs fibrotic subscore3.Symptom-based and QoL end points: most appropriate instrument for measuring patient-reported symptoms in RCTs vs observational studies, instrument recall period and consistency with regulatory requirements, appropriate language that should be used to query dysphagia, use of disease-specific vs general QoL measures, and use of pediatric symptom severity scores and applicability across pediatric populationsTable 1Considerations Discussed at the COREOS Core Outcome Set Ratification MeetingOutcome domainConsiderationsDiscussion points and major conclusionsHistologyThe peak eosinophil count expressed either as eosinophils/hpf or as eosinophils/mm2The field should move toward standardizing assessment using per mm2 to account for differences in microscope ocular field sizeExpressing per hpf improves comparability against historical studies, which have almost exclusively used hpfBoth measures should be presented in RCTs and observational studies, although expressing eosinophils/mm2 may not be feasible in all observational studiesHistologic remission defined either as ≤6 or ≤15 eosinophils/hpfThe threshold of <15 eosinophils/hpf was preferred due to comparability with historical studiesThe threshold of <15 eosinophils/hpf is discordant with recommendations from the US Food and Drug Administration, but the threshold of ≤6 eosinophils/hpf may be too stringent to achieve, especially for mechanisms of action that are not eosinophil-specificThe use of EoEHSS to assess histologic disease activityThe EoEHSS measures other items beyond peak eosinophil density and has demonstrated reliability and responsiveness. It should be used in RCTs to capture the full spectrum of histologic manifestationsThe EoEHSS may not be feasible to evaluate in all observational studiesEndoscopyThe use of EREFS in RCTs and observational studiesThe EREFS captures the endoscopic features of the disease and has been shown to be reliably assessed by both expert and nonexpert endoscopists. It should be used in both RCTs and observational studiesThe EREFS scoring as either 0–8 or 0–9, based on the most severe grade of featuresThe difference lies in scoring of furrows (as absent vs present for 0–8 scoring or as 0 (none), 1 (mild), 2 (severe) for 0-9 scoring); 0–8 scoring was preferred in the Delphi. Although the data are lacking, 0–9 scoring may hypothetically be more responsive to change and can be collapsed to 0–8 scoring in post-hoc analysesEndoscopic remission is defined by an EREFS score ≤2Current studies suggest that this is the most appropriate threshold for endoscopic remission (either scored as the total EREFS or scoring using only the inflammatory sub-score)Symptoms and Quality of LifeThe use of DSQ and EEsAIBoth instruments have undergone an important degree of validation. The DSQ uses daily recall, which is congruent with US Food and Drug Administration recommendations, whereas the EEsAI uses a 7-d recall period.None of the current instruments captures the majority of symptoms or strategies for dealing with or avoiding dysphagia episodesThe DSQ and EEsAI should be used in RCTsBoth instruments are proprietary, which may limit their use, and were likely chosen from the multitude of PRO instruments, because they were the first such instrument to be validated and historically most usedThese instruments for measuring symptoms may not be feasible to apply in all observational studiesAdditional validation of PRO measures in EoE is a research priorityThe language used to define dysphagiaBoth trouble swallowing and delayed/slow passage of food were considered appropriate to query dysphagia in adults with EoEWhile food being stuck was considered appropriate in the Delphi panel, it was excluded during ratification due to concerns that it more accurately represents long- rather than short-lasting episode of food impaction. Evidence for this based on qualitative work is limitedThe meaning of specific terms for capturing dysphagia symptoms likely varies by languageThe use of the PEESS, version 2.0The PEESS, version 2.0 is the only instrument that has undergone some degree of validation in pediatric EoEAlthough the diverse symptom capture reflects the broad nature of symptom presentation in children, capturing many types of symptoms poses a challenge in RCT design (resulting in potentially heterogeneous populations being enrolled in an RCT, and uncertainty of generalizability across all pediatric age groups). Appropriateness of 30-d recall period for use in RCTs remains a cause for concernAdditional research in validating pediatric EoE PROs is requiredThe use of validated measures to assess QoLThe EoE-QoL-A for adult patients and PedsQL (both child self-report and proxy/parent-report) should be used in RCTsThese QoL instruments may not be feasible to apply in all observational studiesAdditional research is required to compare general QoL measures vs disease-specific QoL measures in EoEDSQ, Dysphagia Symptoms Questionnaire; EoEHSS, EoE Histology Scoring System; EoE-QoL-A, Adult Eosinophilic Oesophagitis Quality of Life Questionnaire; EEsAI, EoE Activity Index; hpf, high-power field; PEESS, Pediatric EoE Symptom Score; PedsQL, Pediatric Quality of Life; PRO, patient-reported outcome. Open table in a new tab DSQ, Dysphagia Symptoms Questionnaire; EoEHSS, EoE Histology Scoring System; EoE-QoL-A, Adult Eosinophilic Oesophagitis Quality of Life Questionnaire; EEsAI, EoE Activity Index; hpf, high-power field; PEESS, Pediatric EoE Symptom Score; PedsQL, Pediatric Quality of Life; PRO, patient-reported outcome. After discussion, 33 outcomes were ratified in the COS and are presented in the accompanying consensus recommendations. In summary, the COREOS collaborators developed an internationally guided, multidisciplinary COS for use in pharmacologic and dietary therapeutic studies in pediatric and adult patients with EoE. The ratification meeting was concluded with a discussion around strategies to disseminate these recommendations. The aim was to encourage adoption of the COS to reduce heterogeneity in end-point assessment, minimize risk of reporting bias, and improve the quality of evidence synthesis by facilitating valid cross-study comparisons. The COREOS collaborators also recognized that the end points used in EoE studies have evolved rapidly over the past 2 decades. Although this is the first version of a COS in EoE, the ongoing work in developing and validating instruments for measuring disease activity will likely shape future iterations. As the COREOS collaborators agreed on only the minimum core set of outcomes that should be measured, investigators are encouraged to continue exploring additional end points of interest. Finally, the COREOS collaborators acknowledged that the field is on the cusp of many exciting advances, particularly with multiple agents targeting different mechanisms of action in development. Taken together, the group decided to reconvene to update the COS in 2024.