Lead Investigator: Vipul Jairath, Alimentiv Inc
Title of Proposal Research: Responsiveness of the UC-100 Score in Patients with Moderately to Severely Active Ulcerative Colitis
Vivli Data Request: 7549
Funding Source: Alimentiv is providing funding in in the form of resources to conduct the analysis. Alimentiv is a contract research organization that uses the profits from delivery of commercial services to reinvest in academic research that aligns with its partners shared purpose to accelerate drug discovery.
Potential Conflicts of Interest: Vipul Jairath received consulting fees from AbbVie, Alimentiv Inc (formerly Robarts Clinical Trials), Arena pharmaceuticals, Bristol Myers Squibb, Celltrion, Eli Lilly, Ferring, Fresenius Kabi, GlaxoSmithKline, Genetech, Gilead, Janssen, Merck, Mylan, Pendopharm, Pfizer, Roche, Sandoz, Takeda, Topivert; speaker’s fees from, Abbvie, Ferring, Janssen, Pfizer, Shire, Takeda. The research outlined in this study is not receiving funding from any of the outlined companies with the exception of Alimentiv Inc which is the sponsor and funder of this study. Christopher Ma received consulting fees from AbbVie, Alimentiv Inc. (formerly Robarts Clinical Trials Inc.), Amgen, AVIR Pharma Inc, Ferring, Fresenius Kabi, Janssen, Mylan, Takeda, Pfizer, Roche; speaker’s fees from AbbVie, AVIR Pharma Inc, Janssen, Takeda, and Pfizer; research support from Pfizer. The research outlined in this study is not receiving funding from any of the outlined companies with the exception of Alimentiv Inc which is the sponsor and funder of this study.
Summary of the Proposed Research:
Ulcerative colitis (UC) is a disease that causes swelling and pain, most commonly in the large bowel and rectum which are located at the end of the digestive tract (the tube that carries food and its byproducts through the body), closest to the anus. Diarrhea and feces with blood in it are frequent symptoms. While there is no cure for UC, there are several approved drugs to help control the symptoms. Unfortunately, no one drug works in all patients and even when a drug does work, it may become less effective over time. Therefore, there is a need for other drugs that work and that are safe to use.
To learn whether a new drug works to treat UC, it must be tested in research studies. One way to do this is to compare the drug to a placebo, which is a “fake” copy of the real drug. The placebo looks like the real drug, but it contains none of the medicine. In such a study, some UC patients are given drug while others are given placebo. Usually this is decided randomly (like the flip of a coin) and neither the patients nor their doctors know what they are taking until the study is over. If a drug works, the group treated with the real drug should do better (have less disease activity) than the group that received the placebo. But how is this measured? In studies for UC, disease activity is often measured using the Mayo Clinic Score (MCS) but other measures, including the partial MCS (pMCS), adapted MCS, and Robarts Histopathology Index (RHI) also exist. The ability of a measure to detect a difference, when one exists, between the effect of a drug as compared to placebo is referred to as “sensitivity”. A more sensitive measure makes it easier to tell if a drug is working than a less sensitive measure does. It could also mean that fewer participants will be needed for a research study before we can tell if a drug is working, and this could, in turn, reduce the cost of drug development.
Early research results suggest that a measure of UC disease activity called the UC-100 is more sensitive than the MCS or other existing measures. The goal of the current study is to confirm whether this is true. To do this, we will use existing data from previous drug studies in UC and compare the ability of different measures to detect changes in disease activity. If this study finds that the UC-100 is a more sensitive measure of UC disease activity than other measures across several research study datasets, this could change the way we conduct future research studies in UC. With greater sensitivity, it could be possible to detect even small effects of a drug that might be missed with less sensitive measures. This might make it less likely for a potentially effective drug to be mistakenly labeled as a failure during the early drug development process. The ultimate aim is to get better at identifying when a drug is working in the hopes that this will lead to the approval of more safe and effective drugs for UC and provide a larger number of treatment options for patients whose disease is not adequately controlled by currently available drugs.
Statistical Analysis Plan:
This study is a retrospective analysis of existing clinical, endoscopic, and histologic data from completed phase 2 and phase 3 induction trials where an investigational biologic or small molecule drug was shown to be more effective than placebo for the treatment of moderately to severely active UC. Subject-level demographics, disease characteristics, and clinical data at baseline and inflammatory biomarker concentrations, MCS, MCS subscores, and Geboes and/or RHI scores at baseline and end of induction will be collected from trial databases.
Statistical analyses will only be performed on the data received, using a complete-case analysis basis. No methods for imputation of missing data will be used.
We will evaluate the responsiveness of the UC-100, the MCS, adapted 9-point MCS, partial MCS (pMCS), and RHI, to determine which index is most responsive. Responsiveness will be evaluated in two ways: longitudinal validity and treatment effect size.
Summary statistics for demographic and baseline characteristics will be calculated and presented. These will include age, disease duration, and sex, depending on availability.
The longitudinal validity will be assessed by determining the correlation between changes in the UC-100 score, MCS, adapted 9-point MCS, pMCS, and RHI and changes in fecal calprotectin (FC) and C-reactive protein (CRP) from baseline to the end of induction. The correlations will be measured using Spearman’s correlation (depending on the distribution of the data), with the associated 95% confidence intervals calculated using Fisher’s z-transformation. The Cohen benchmarks, where the small, medium, and large effects correspond to the correlations of 0.1, 0.3, and 0.5, will be applied to interpret the strength of correlation. Parallel line plots will be used to depict baseline, follow-up, and changes in scores.
For assessing the treatment effect size of the UC-100 score, MCS, adapted 9-point MCS, pMCS, and RHI, the criteria for meaningful change will be determined by the treatment assignment, with the active drug group in each trial considered “changed” and the placebo group considered “unchanged”. Treatment effect sizes will be assessed by two approaches. First, assuming that the indices at the end of induction are all normally distributed, the extent of responsiveness will be qualified using the standardized mean difference between the changed and the unchanged groups. The standardized mean difference is defined as the mean changes in the induction phase divided by the pooled estimate of standard deviation for the induction phase changes. Associated 95% confidence intervals will be calculated using the exact method.5 Interpretation of standardized mean difference will be based on Cohen’s benchmarks of .2, .5, and .8 suggesting “small”, “moderate”, and “large” effect sizes respectively.
The second approach uses the non-parametric estimate of the probability to assess the treatment effect size, equal to the area under the receiver operating characteristic (ROC) curve. This approach could help to qualify the ability of the indices to discriminate subjects with and without meaningful changes and doesn’t need any assumption about the indices. Associated 95% confidence intervals for the area under the ROC curve will be obtained based on the log-odds transformation. These estimates will be interpreted according to the benchmark of 0.56, 0.64, and 0.71 correspond to “small”, “medium”, and “large” respectively. The difference of the estimate of the area under the ROC curve between the UC-100 score and the other 4 indices and the associated 95% confidence intervals will be presented. The test of whether the difference is equal to zero will be performed using Fisher’s z-transformation with a simple variance estimator.\
To combine the results from different platforms the estimates from all studies will be pooled as conventional meta-analysis using Two-stage meta-analyses for both the treatment effect sizes and the correlation. In the two-stage meta-analysis, correlations from all studies will be Fisher’s z-transformed before pooling. Then correlations will be pooled using the generic inverse variance pooling method. A random-effects model with the restricted maximum likelihood estimator will be employed, with R package “meta” and function metacor. Treatment effect sizes will be pooled, and a random-effect model will be employed with Knapp-Hartung adjustments and the restricted maximum likelihood estimator, where “metacont” in the R package “meta” will be applied.
In order to maintain the structure and independence of the studies being requested, each study will be kept as a separate file. When combining the results of the studies into a single dataset for the one-stage analysis, we will create a variable that uniquely identifies each study.
Requested Studies:
A Multicenter, Randomized, Double-blind Placebo-controlled Study of the Human Anti-TNF Monoclonal Antibody Adalimumab for the Induction of Clinical Remission in Subjects With Moderately to Severely Active Ulcerative Colitis
Data Contributor: AbbVie
Study ID: NCT00385736
Sponsor ID: M06-826
A Multicenter, Randomized, Double-blind, Placebo-controlled Study of the Human Anti-TNF Monoclonal Antibody Adalimumab for the Induction and Maintenance of Clinical Remission in Subjects With Moderately to Severely Active Ulcerative Colitis
Data Contributor: AbbVie
Study ID: NCT00408629
Sponsor ID: M06-827
A Multi-Center, Randomized, Double-Blind, Placebo-controlled Study of Adalimumab in Japanese Subjects With Moderately to Severely Active Ulcerative Colitis.
Data Contributor: AbbVie
Study ID: NCT00853099
Sponsor ID: M10-447
A Phase 3, Randomized, Placebo-Controlled, Blinded, Multicenter Study of the Induction and Maintenance of Clinical Response and Remission by MLN0002 in Patients With Moderate to Severe Ulcerative Colitis
Data Contributor: Takeda
Study ID: NCT00783718
Sponsor ID: C13006
Phase III, Multicenter, Randomized, Double-blinded, Placebo-controlled, Parallel-group Study to Examine the Efficacy, Safety, and Pharmacokinetics of Intravenous MLN0002 (300 mg) Infusion in Induction and Maintenance Therapy in Japanese Patients With Moderately or Severely Active Ulcerative Colitis
Data Contributor: Takeda
Study ID: NCT02039505
Sponsor ID: MLN0002/CCT-101
A Phase 2/3 Multicenter, Randomized, Placebo-controlled, Double-blind Study to Evaluate the Safety and Efficacy of Golimumab Induction Therapy, Administered Subcutaneously, in Subjects With Moderately to Severely Active Ulcerative Colitis
Data Contributor: Johnson & Johnson
Study ID: NCT00487539
Sponsor ID: CR014176
A Phase 3 Multicenter, Randomized, Placebo-controlled, Double-blind Study to Evaluate the Safety and Efficacy of Golimumab Maintenance Therapy, Administered Subcutaneously, in Subjects With Moderately to Severely Active Ulcerative Colitis
Data Contributor: Johnson & Johnson
Study ID: NCT00488631
Sponsor ID: CR014179
A Phase 3 Multicenter, Placebo-controlled, Double-blind, Randomized-withdrawal Study to Evaluate the Safety and Efficacy of Golimumab Maintenance Therapy, Administered Subcutaneously, in Japanese Subjects With Moderately to Severely Active Ulcerative Colitis
Data Contributor: Johnson & Johnson
Study ID: NCT01863771
Sponsor ID: CR100937
A Phase 2/3 Multicenter, Randomized, Placebo-controlled, Double-blind Study to Evaluate the Safety and Efficacy of Golimumab Induction Therapy, Administered Intravenously, in Subjects With Moderately to Severely Active Ulcerative Colitis
Data Contributor: Johnson & Johnson
Study ID: NCT00488774
Sponsor ID: CR014188
A Phase 3, Multicenter, Randomized, Double-Blind, Placebo-Controlled Study Evaluating the Efficacy and Safety of Infliximab in Chinese Subjects With Active Ulcerative Colitis
Data Contributor: Johnson & Johnson
Study ID: NCT01551290
Sponsor ID: CR018769
A Phase 3, Randomized, Double-blind, Placebo-controlled, Parallel-group, Multicenter Protocol to Evaluate the Safety and Efficacy of Ustekinumab Induction and Maintenance Therapy in Subjects With Moderately to Severely Active Ulcerative Colitis
Data Contributor: Johnson & Johnson
Study ID: NCT02407236
Sponsor ID: CR106920
Public Disclosures:
Solitano, V., Panaccione, R., Sands, B.E., Wang, Z., Zou, G., Peyrin-Biroulet, L., Danese, S., Cornfield, L.J., Feagan, B.G., Singh, S. and Jairath, V., 2024. P280 Comparative responsiveness of disease activity indices in moderately-to-severely active Ulcerative Colitis: a post-hoc analysis of the UNIFI trial. Journal of Crohn’s and Colitis, 18(Supplement_1), pp.i647-i648. Doi : 1093/ecco-jcc/jjad212.0410