Defining Clinically Meaningful Responses with the CDPRO/SS Instrument in Crohn's Disease

Presentation of data defining clinically meaningful responses for the Bowel Domain and Functional Domain in the CDPRO/SS short Patient Reported Outcomes Instrument (SS is for Signs and Symptoms, not short). This is a shortened 6 Q version of the full 5 domain CDPRO instrument. This has the support of the EMA and is under review by the FDA for qualification as the gold standard PRO for use in Crohn's disease clinical trials.

Transcript

1. © UEG. 2019 Presentation by Responder Definitions for the Crohn’s Disease Patient-Reported Outcomes Signs and Symptoms (CD-PRO/SS) Instrument Using Patients with Crohn’s Disease Treated with Etrolizumab Peter D. R. Higgins, Kendra DeBusk, Rhian Jacob, Azra Hassanali, Zaineb Sharafali, Young S. Oh, Alysha Kadva, Alison Matsui Peter D. R. Higgins University of Michigan, Ann Arbor, MI, USA @ibddoctor
2. © UEG. 2019 Disclosure of Conflicts of Interest Peter D.R. Higgins is a site investigator for the etrolizumab Phase 3 clinical studies Funding for this research was provided by Roche • Third-party medical writing support was provided by ApotheCom (San Francisco, CA) and was funded by Roche UEG copyright only applies to the format of slide template and has no copyright or ownership interest in the content.
3. © UEG. 2019 PROs Are Important for Evaluating Treatment Effects on the Clinical Course of IBD • Available PROs have limitations in quantifying symptoms and capturing the patient perspective • The CD-PRO/SS and the UC-PRO/SS instruments are the first IBD tools: o Developed with input from patients, clinical experts, and health authorities1-3 o Designed to comprehensively evaluate signs, symptoms, and impacts of Crohn’s disease and UC that are most important to patients • We sought to numerically define a clinically meaningful change in the CD-PRO/SS | Presentation by Peter D. R. Higgins 3 CD-PRO/SS, Crohn’s Disease Patient-Reported Outcomes Signs and Symptoms; EMA, European Medicines Agency; IBD, inflammatory bowel disease; PRO, patient-reported outcome; UC-PRO/SS, Ulcerative Colitis Patient-Reported Outcomes Signs and Symptoms; UC, ulcerative colitis. 1. Higgins PDR et al. Presented at 14th Congress of ECCO (European Crohn’s and Colitis Organisation); March 6-9, 2019; Copenhagen, Denmark. P224. 2. Higgins PDR et al. J Patient Rep Outcomes. 2018;2:26. 3. Higgins PDR et al. J Patient Rep Outcomes. 2018;2:24. 4. US Food and Drug Administration. https://www.fda.gov/drugs/drug-development-tool-qualification-programs/clinical-outcome-assessments-coa-qualification-submissions. Accessed September 25, 2019. 5. European Medicines Agency. https://www.ema.europa.eu/en/documents/other/letter-supportdevelopment-patient-reported-outcomes-tools-use-endpoint-inflammatory-bowel-disease_en.pdf. Published April 2019. Accessed August 15, 2019. The CD-PRO/SS and the UC-PRO/SS have received EMA support and are undergoing the FDA qualification process for use in IBD clinical trials4,5
4. © UEG. 2019 The CD-PRO/SS Instrument • The CD-PRO/SS instrument consists of 2 separately scored scales: a 3-item Bowel Domain and 3-item Functional Domain* o Based on qualitative and quantitative studies, 6 sign and symptom items were identified and validated1 • The CD-PRO/SS instrument has both electronic and paper versions o Both were used in the development and validation analyses1 | Presentation by Peter D.R. Higgins 4 BM, bowel movement; CD-PRO/SS, Crohn’s Disease Patient-Reported Outcomes Signs and Symptoms. *Item scores for the domains were calculated as an average of the most recent of ≥ 4 out of 7 days before Week 0 (baseline) and Week 14. 1. Higgins PDR et al. J Patient Rep Outcomes. 2018;2:24. Domain (Score Range) Item Score Bowel (0-16) Number of BM 0 – 8 (categorical) Liquid BM 0 (no) – 4 (always) BM right away 0 (no) – 4 (very severe) Functional (0-12) Pass gas 0 (no) – 4 (very often) Pain in belly 0 (no) – 4 (very severe) Bloating in belly 0 (no) – 4 (very severe) The CD-PRO/SS Instrument Item Description
5. © UEG. 2019 Objective • To determine responder definitions for the CD-PRO/SS using adult patients with moderate-to-severe CD participating in BERGAMOT, a Phase 3 clinical trial evaluating etrolizumab | Presentation by Peter D.R. Higgins 5
6. © UEG. 2019 Methods: Study Design and Patients | Presentation by Peter D.R. Higgins 6 • The electronic CD-PRO/SS instrument was administered to all patients before every induction visit • Data from all treatment arms were pooled for this analysis aTNF-α, anti-tumour necrosis factor alpha; CD-PRO/SS, Crohn’s Disease Patient-Reported Outcomes Signs and Symptoms; Etro, etrolizumab; OLI, open-label induction; SC, subcutaneous; q4w, every 4 weeks; wk, week. *Patients who had inadequate response, refractory response, or intolerance to aTNF-αs. †Cohort 1 (n=215) and Cohort 2 (n=264) were pooled for data analysis. BERGAMOT (NCT02394028) is an ongoing Phase 3 study evaluating etrolizumab in aTNF-α–naive and aTNF-α–experienced* adult patients with moderate-to-severe Crohn’s disease Induction 14 weeks Maintenance 52 weeks Safety F/U 12 weeks 2:3:3 1:2:2 1:1 Cohort 1 Exploratory (n = 300†) Cohort 2 OLI (n = 350†) Cohort 3 Pivotal (n = 500) RANDOMISATION Etro 210 mg SC at wk 0, 2, 4, 8, 12 Etro 105 mg SC q4w Placebo Etro 210 mg SC at wk 0, 2, 4, 8, 12 Etro 105 mg SC q4w Etro 210 mg SC at wk 0, 2, 4, 8, 12 Etro 105 mg SC q4w Placebo Etro 105 mg SC q4w Placebo 1:1 RERANDOMISATION Etro Responders Weeks
7. © UEG. 2019 Methods: Responder Definitions • Per FDA PRO guidance,1 minimum clinically meaningful changes in the bowel and functional domains were calculated using distributional (using baseline values) and anchor (IBDQ and CDAI) definitions of response • The CD-PRO/SS responder definitions were determined by triangulation of distributional- based methods and the differences in mean CD-PRO/SS scores between anchor responders and nonresponders using patients from both treatment arms • Cumulative distribution function curves were used to assess proportion of patients in combined treatment arms based on the CD-PRO/SS for a range of responder cutoff values | Presentation by Peter D.R. Higgins 7 CDAI, Crohn’s Disease Activity Index; CD-PRO/SS, Crohn’s Disease Patient-Reported Outcomes Signs and Symptoms; IBDQ, Inflammatory Bowel Disease Questionnaire; PRO, patient-reported outcome. 1. US Department of Health and Human Services. Food and Drug Administration. https://www.fda.gov/downloads/drugs/guidance/ucm193282.pdf. Published December 2009. Accessed August 13, 2019.
8. © UEG. 2019 Baseline Demographics and Disease Characteristics | Presentation by Peter D.R. Higgins 8 Baseline Parameters Patients (N=479a) Age, mean (SD), years 38.4 (13.2) Sex, male, % 51.2 Race, white, % 82.9 BMI, mean (SD), kg/m2 25.1 (6.0) Disease duration, median (SD), years 8.3 (9.1) Corticosteroids use at baseline, % 41.3 Immunosuppressants use at baseline, % 33.8 Prior aTNF-α use, % 67.4 Disease location, % Colon 25.9 Ileo-colonic 52.4 Ileum 21.7 CDAI, mean (SD) 314.2 (61.2) SES-CD, mean (SD) 13.9 (7.3) IBDQ, mean (SD) 115.8 (28.6) aTNF-α, anti-tumour necrosis factor alpha; CDAI, Crohn’s Disease Activity Index; IBDQ, Inflammatory Bowel Disease Questionnaire; SD, standard deviation; SES-CD, Simple Endoscopic Score for Crohn’s Disease. aCohort 1, n=215; Cohort 2, n=264.
9. © UEG. 2019 Baseline, Week 14, and Change from Baseline in CD-PRO/SS Bowel Domain Scores | Presentation by Peter D.R. Higgins 9 Bowel Domaina (Score Range, 0-16) CD-PRO/SS, Crohn’s Disease Patient-Reported Outcomes Signs and Symptoms. aData present pooled cohort for patients randomly assigned to etrolizumab and placebo. Baseline (N=479) Week 14 (n=421) Change from Baseline (n=334) Minimum 1 0.43 −10.43 Maximum 13 13.75 4.72 SD 2.56 3.07 2.89 Mean 8.23 5.83 −2.45 Week 14 Baseline CD-PRO/SS Bowel Domain Score 7.5 5.0 2.5 0.0 15.0 10.0 12.5
10. © UEG. 2019 Baseline, Week 14, and Change from Baseline in CD-PRO/SS Functional Domain Scores 10 Functional Domaina (Score Range, 0-12) | Presentation by Peter D.R. Higgins CD-PRO/SS, Crohn’s Disease Patient-Reported Outcomes Signs and Symptoms. aData present pooled cohort for patients randomly assigned to etrolizumab and placebo. Baseline (N=479) Week 14 (n=421) Change from Baseline (n=334) Minimum 1.43 0 −9.86 Maximum 11.43 11.43 6.01 SD 2.13 2.35 2.55 Mean 6.41 4.21 −2.28 Week 14 Baseline CD-PRO/SS Functional Domain Score 12.5 10.0 7.5 5.0 2.5 0.0
11. © UEG. 2019 Response Using the CD-PRO/SS Defined as ≥ 3.0-point Decrease in Bowel Domain | Presentation by Peter D.R. Higgins 11 Bowel Domaina IBDQ Anchor CDAI Anchor Non- responder (n=84) Responder (n=210) Non- responder (n=134) Responder (n=183) Minimum −9.00 −9.67 −6.00 −10.43 Maximum 4.71 3.71 4.71 4.63 SD 2.43 2.65 2.04 2.71 Mean −0.47 −3.33 −0.69 −3.81 CDAI Anchor IBDQ Anchor Worsening Improvement −2.86 −3.12 CDAI, Crohn’s Disease Activity Index; CD-PRO/SS, Crohn’s Disease Patient-Reported Outcomes Signs and Symptoms; IBDQ, Inflammatory Bowel Disease Questionnaire; SD, standard deviation. aData present pooled cohort for patients randomly assigned to etrolizumab and placebo. Change from Baseline in CD-PRO/SS Bowel Domain 5 0 -5 -10 -15 Nonresponder Responder Nonresponder Responder
12. © UEG. 2019 | Presentation by Peter D.R. Higgins 12 Response Using the CD-PRO/SS Defined as ≥ 2.5-point Decrease in Functional Domain IBDQ Anchor CDAI Anchor Non- responder (n=84) Responder (n=210) Non- responder (n=134) Responder (n=183) Minimum −6.31 −9.86 −7.43 −9.86 Maximum 6.01 2.72 6.01 2.72 SD 2.11 2.42 2.08 2.25 Mean −0.72 −2.90 −0.79 −3.48 CDAI, Crohn’s Disease Activity Index; CD-PRO/SS, Crohn’s Disease Patient-Reported Outcomes Signs and Symptoms; IBDQ, Inflammatory Bowel Disease Questionnaire; SD, standard deviation. aData present pooled cohort for patients randomly assigned to etrolizumab and placebo. Change from Baseline in CD-PRO/SS Functional Domain Score 10 5 0 -5 -10 Functional Domaina CDAI Anchor IBDQ Anchor Worsening Improvement −2.18 −2.68 Nonresponder Responder Nonresponder Responder
13. © UEG. 2019 40% of Patients Were Responders Based on the Bowel Domain of the CD-PRO/SS Instrument | Presentation by Peter D.R. Higgins 13 CD-PRO/SS, Crohn’s Disease Patient-Reported Outcomes Signs and Symptoms. aData present pooled cohort for patients randomly assigned to etrolizumab and placebo. Cutoff Value Responders, % (n=334) 2.8 43% 2.9 41% 3.0 40% 3.1 40% 3.2 38% Bowel Domain Cutoff Value Proportion of Responders in CD-PRO/SS Bowel Domain, % Bowel Domaina
14. © UEG. 2019 44% of Patients Were Responders Based on the Functional Domain of the CD-PRO/SS Instrument | Presentation by Peter D.R. Higgins 14 Cutoff Value Responders, % (n=334) 2.1 50% 2.2 47% 2.3 46% 2.4 46% 2.5 44% CD-PRO/SS, Crohn’s Disease Patient-Reported Outcomes Signs and Symptoms. aData present pooled cohort for patients randomized to etrolizumab and placebo. Functional Domain Cutoff Value Proportion of Responders in CD-PRO/SS Functional Domain, % Functional Domaina
15. © UEG. 2019 Conclusions • This is the first analysis to calculate a responder definition for each domain of the CD-PRO/SS for use in clinical trials and practice • Preliminary definitions for a clinically meaningful response to treatment using the CD-PRO/SS are a decrease of ≥ 3.0 points in the Bowel Domain or ≥ 2.5 points in the Functional Domain • Using these definitions, 40% and 44% of all patients were responders by week 14, according to the Bowel Domain and the Functional Domain, respectively • The cutoffs for clinically meaningful improvements for the CD-PRO/SS and the UC-PRO/SS instruments are being validated in the ongoing etrolizumab Phase 3 clinical studies evaluating adult patients with IBD | Presentation by Peter D. R. Higgins 15 CD-PRO/SS, Crohn’s Disease Patient-Reported Outcomes Signs and Symptoms; IBD, inflammatory bowel disease; UC-PRO/SS, Ulcerative Colitis Patient-Reported Outcomes Signs and Symptoms.
16. © UEG. 2019 | Presentation by Peter D. R. Higgins 16 Additional Information on the CD-PRO/SS Instrument • Translations, cost structure, licensing information, paper or electronic versions, and user manuals of the CD-PRO/SS instrument are available through Evidera. Please contact UCandCDPRO@evidera.com
17. Backup Slides
18. Distributional-Based Methods Method Cohort Bowel Functional Cohen’s Coefficient 1 (n=215) 1.2 1.1 2 (n=264) 1.3 1.1 Pooled (n=479) 1.3 1.1 SEM 1 (n=215) 1.5 1.3 2 (n=264) 1.5 1.3 Pooled (n=479) 1.5 1.3 SEM, standard error of the mean.
19. Joint Response Evaluation N=334 Bowel Domain Functional Domain Nonresponder Responder Nonresponder 148 (44%) 39 (12%) Responder 53 (16%) 94 (28%) Joint Response in All Patientsa Joint Response in All Respondersa aResponders and nonresponders on either domain. N/A, not applicable. aResponders only on either domain. N=186 Bowel Domain Functional Domain Nonresponder Responder Nonresponder N/A 39 (21%) Responder 53 (28%) 94 (51%)