Lead Investigator: Neeraj Narula, Hamilton Health Sciences
Title of Proposal Research: A Comparison of the Rapidity of Clinical Response between Infliximab and Tofacitinib for Ulcerative Colitis.
Vivli Data Request: 7321
Funding Source: None
Potential Conflicts of Interest: Neeraj Narula has received a McMaster University Department of Medicine Internal Career Award, and has received honoraria from Janssen, AbbVie, Takeda, Pfizer, Merck, Sandoz, Novartis, and Ferring
Summary of the Proposed Research:
Active ulcerative colitis (UC) is a disabling condition and rapid resolution of symptoms is a main concern of both patients and their physicians. There are many choices for initial treatment but very few studies have compared the speed at which remission is achieved between different agents. In this study, we are planning to compare the rapidity of response to infliximab (IFX), a chimeric monoclonal anti-TNF antibody, and tofacitinib (TOF), an oral small molecule JAK inhibitor. We will be using data from the ACT and OCTAVE induction studies.
Statistical Analysis Plan:
Descriptive statistics will be used to summarize the proportion of patients achieving clinical outcomes. Logistic regression will be used to model the likelihood of achieving outcomes. Propensity score matching will be used to compensate for differences in baseline characteristics and prior and concomitant treatments between treatment groups.
As this is not a head-to-head RCT, we will use different methods to limit the potential for confounding. We will use multiple logistic regression models adjusted for significant baseline co-variates on univariate analyses. We will also calculate propensity scores to create a cohort of matched participants with a similar distribution of baseline scores.
Univariate analyses will be conducted to identify any association between baseline covariates and the outcomes of interest. A p-value threshold of < 0.10 will be used for the selection of variables from univariate analyses to include in the multivariate models. Potential confounders as determined by univariate analyses will be adjusted for in the multivariate analysis. Depending on the results, we expect to include age, gender, treatment received, baseline PMS ≥6, baseline Mayo endoscopic score of 3, concomitant immunomodulator use, concomitant steroid use, baseline CRP (mg/L), baseline albumin <3.4 g/dL, disease duration, and disease location.
As a sensitivity analysis, we will create a propensity score representing the conditional probability of receiving a treatment given the observed covariates. We will then select a matched cohort of infliximab and tofacitinib-treated patients with evenly distributed baseline propensity scores. The propensity scores will be created using a non parsimonious logistic regression model based on age, sex, disease duration, baseline PMS, Mayo endoscopic score, CRP, albumin, concomitant immunomodulator use, and concomitant corticosteroid use.
The region of common support will be assessed by visualization of the graphs of all the propensity scores for participants treated in the studies. Matching using a one-to-one basis using k-nearest neighbor without replacement. The subsequent matched pairs that are obtained will be used in the sensitivity analysis to assess the comparative effectiveness of infliximab and tofacitinib.
Subsequently, inverse probability weighted regression adjustment (IPWRA) will be used. The weights obtained from the inverse estimated probability that a participant received a treatment. This estimator is considered ‘double robust’ as the weights adjust the estimator if the treatment model is inappropriate, assuming the outcome model is correctly specified, and vice versa. Balance of covariates between baseline and post-IPTW adjustment will be assessed using means of standardized mean differences, with differences less than 10% considered to be well-balanced.
An additional sensitivity analysis will be performed comparing only standard FDA-approved doses for tofacitinib (20 mg/d) and infliximab (5mg/kg every 8 weeks).
Continuous variables will be presented as means and standard deviations or as medians and inter-quartile ranges, and dichotomous variables will be presented as proportions or percentages. Mann-Whitney U test or chi-squared will be used to compare differences in baseline characteristics. Frequencies of patients achieving the different endoscopic and clinical outcomes will be compared using the Chi-squared test. Unadjusted and adjusted odds ratios with 95% confidence intervals (CI) for achieving the outcomes of interest will be determined. Statistical significance is defined by p-value <0.05. Data will be analyzed using Stata version 15.0.
Requested Studies:
A Multicentre, Randomized, Double-blind, Placebo-controlled, Parallel-group Study Of Oral Cp-690,550 As An Induction Therapy In Subjects With Moderate To Severe Ulcerative Colitis
Data Contributor: Pfizer
Study ID: NCT01465763
Sponsor ID: A3921094
A Multicentre, Randomized, Double-blind, Placebo-controlled, Parallel-group Study Of Oral Cp-690,550 As An Induction Therapy In Subjects With Moderate To Severe Ulcerative Colitis.
Data Contributor: Pfizer
Study ID: NCT01458951
Sponsor ID: A3921095
A Randomized, Placebo Controlled, Double Blind, Parallel Group Multi-Center Study In Order To Investigate Safety And Efficacy Of CP- 690 550 In Subjects With Moderate To Severe Ulcerative Colitis.
Data Contributor: Pfizer
Study ID: NCT00787202
Sponsor ID: A3921063
A Randomized, Placebo-controlled, Double-blind Trial to Evaluate the Safety and Efficacy of Infliximab in Patients With Active Ulcerative Colitis
Data Contributor: Johnson & Johnson
Study ID: NCT00036439
Sponsor ID: CR004777
A Randomized, Placebo-controlled, Double-blind Trial to Evaluate the Safety and Efficacy of Infliximab in Patients With Active Ulcerative Colitis
Data Contributor: Johnson & Johnson
Study ID: NCT00096655
Sponsor ID: CR004783
Public Disclosures:
Wong ECL, Merat S, Monaco C, Dulai PS, Jairath V, Marshall JK, Reinisch W, Narula N. Comparative Efficacy of Infliximab Versus Tofacitinib for Inducing Remission in Biologic Naive Ulcerative Colitis: A Propensity Matched Study. Dig Dis Sci. 2023 Jun;68(6):2635-2646. doi: 10.1007/s10620-023-07956-8