Ulcer Size After Induction Therapy Performs Better Than Symptom Assessment for Prediction of One Year Endoscopic Remission in Crohn’s Disease: A Post Hoc Analysis

Abstract

Background

We evaluated whether postinduction ulcer size and patient-reported outcome (PRO) severity are associated with the achievement of 1-year endoscopic remission (ER) in patients with Crohn’s disease (CD).

Methods

This post hoc analysis combined data from several clinical trials including 283 patients with baseline ulcers ≥5 mm with repeat endoscopy after ustekinumab or adalimumab induction therapy. Patient-reported outcomes including stool frequency (SF) and abdominal pain (AP) were measured by the Crohn’s Disease Activity Index. Thresholds of SF ≥4 and/or AP ≥2 indicated moderately to severely active CD. Endoscopic remission was defined as Simple Endoscopic Score for CD (SES-CD) <3. Multivariate logistic regression models adjusted for confounders (including disease duration and treatment allocation) evaluated the relationships between postinduction ulcer size, PRO symptoms, and achievement of 1-year ER.

Results

Among the 131 CD patients who continued to have ulcers ≥5 mm after induction therapy, 48 (36.6%) achieved 1-year ER. Patients with postinduction ulcers ≥5 mm were approximately 5 times less likely to achieve 1-year ER than the 152 individuals who had small or no postinduction ulcers (odds ratio [OR], 0.20; 95% CI, 0.08-0.51, P = .001). In patients with ulcers ≥5 mm after induction, postinduction PRO scores (including PRO2 and PRO3) did not predict 1-year ER.

Conclusions

Crohn’s disease patients with ulcers ≥5 mm after induction therapy are less likely to achieve 1-year ER. Postinduction PRO severity does not offer additional prognostic information. This may suggest that objective measures of disease such as endoscopic ulcer size should be considered over symptom assessments for determining clinical response to therapy and utilized in trials for maintenance therapy.

Introduction

Crohn’s disease (CD) is an inflammatory bowel disease (IBD) that is characterized by a chronic progressive course and heterogeneous clinical manifestations.¹ Most patients present with an inflammatory phenotype at diagnosis; however, many develop stricturing and/or penetrating complications.^2–4 Individuals with CD present with a wide spectrum of symptoms, some being asymptomatic and others exhibiting debilitating symptoms including diarrhea, abdominal pain (AP), rectal bleeding, arthralgia, and perianal fistula. Unfortunately, symptoms often do not correlate with underlying disease activity, with some CD patients reporting substantial gastrointestinal symptoms with no active disease demonstrated on endoscopy. Similarly, some asymptomatic patients have endoscopically active disease.^4–10

Treatment paradigms have historically focused on symptom improvement as a primary therapeutic goal in CD; however, it has been shown that targeting clinical remission (CR) alone does not alter disease course or improve CD-related outcomes.^2,3,11,12 More recent guidelines such as STRIDE and STRIDE-II have shifted away from patient-reported symptoms and towards objective measures of mucosal inflammation such as endoscopic remission (ER) as measures of disease activity.^13,14 Endoscopic healing is associated with modification of subsequent disease course and improved long-term outcomes, with reduced risk of future relapse, hospitalization, and surgical resection.¹⁵ Patients who achieved ER during the CALM study were found to have less risk of disease progression and less surgery compared with patients who did not achieve ER.¹⁶

Until recently, clinical trials of medical therapy in CD used the Crohn’s Disease Activity Index (CDAI) as a measure of disease activity for participant inclusion and study outcomes.¹⁷ The CDAI is a composite index of patient-reported outcomes (PRO), physical findings, medications, laboratory tests, and extraintestinal findings. This disease activity index has been used to demonstrate efficacy of several CD therapies as the primary outcome measure in multiple clinical trials. However, the CDAI has been criticized for not including patients’ perspectives, demonstrating high interobserver variability and poor reproducibility, and inadequately correlating with objective measures of inflammation such as endoscopy and fecal calprotectin.^18,19 As a result, individual components of the CDAI, AP, and stool frequency (SF) have been proposed as patient-reported outcome measures for clinical trials.²⁰ The PRO2 includes AP and SF with proposed cut-offs for remission that performed comparably to predict CR (CDAI < 150).²⁰ Furthermore, a common rerandomization design in CD clinical trials uses the CDAI to assess eligibility for rerandomization to maintenance therapy. We have previously demonstrated that patient-reported symptom improvement after induction therapy does not predict who will achieve ER.²¹ Moreover, we also have shown that objective measures of disease activity in CD, such as the biomarker fecal calprotectin, perform better than PROs after completion of induction therapy in predicting long-term ER.²²

This analysis aims to evaluate whether the persistence of ulcers on endoscopy following induction therapy is associated with a decreased likelihood of ER at 1-year and whether PROs provide additional prognostic value among CD patients with persistent endoscopic inflammation.

Methods

Study Design

Data from the UNITI studies (ClinicalTrials.gov identifiers NCT01369329, NCT01369342, NCT01369355) were acquired through the Yale Open Data Access Project (YODA #2020-4529) and with permission from Janssen Inc.^23,24 The EXTEND study data (NCT00348283) was obtained through VIVLI (Protocol #00006496) and used with permission from Abbvie Inc.²⁵ Data can be made available through the YODA Project and VIVLI. As data were previously collected and deidentified, the Hamilton Integrated Research Ethics Board determined that local ethics review was not required. Thus, informed consent was not required.

UNITI Study Design

The study design and eligibility criteria of the UNITI studies have previously been published.²⁴ Data from UNITI included adult individuals with moderate to severe CD defined by a CDAI score of 220 to 450. Patients were enrolled in the induction studies including UNITI-1 (those who had primary or secondary nonresponse to antitumor necrosis factor [TNF] agents) and UNITI-2 (those who had poor response to conventional therapy such as immunomodulators or corticosteroids). To be enrolled, participants had at least 1 objective measure of active inflammation including C-reactive protein (CRP) >3 mg/L, fecal calprotectin >250 mg/kg, or endoscopic ulceration in the ileum and/or colon. Participants were randomized to intravenous (IV) placebo or ustekinumab (UST). Initial UST dosing was 130 mg IV or approximately 6 mg/kg IV of UST (weight-based dosing). Patients who responded to UST at 8 weeks were subsequently eligible for IM-UNITI. Those included in IM-UNITI were rerandomized to one of 3 subcutaneous maintenance regimens: placebo, 90 mg of UST every 8 weeks, or 90 mg of UST every 12 weeks. The induction portion of UNITI was 8 weeks, and the maintenance phase was 44 weeks, for a total study duration of 52 weeks.

Endoscopy was not required for participation in the UNITI trials; however, a colonoscopy was performed on a subset of included patients. Colonoscopies were performed at the completion of UST induction (week 8) and at the end of the maintenance phase (week 44). Those who did not meet criteria to participate in IM-UNITI were offered open label UST and continued in the endoscopy substudy. The endoscopic evaluation of the colonoscopies was centrally reviewed by a single blinded reviewer (Dr. Paul Rutgeerts).²⁶

EXTEND Study Design

The details of the EXTEND study including design and inclusion criteria have previously been published.²⁵ Those eligible for EXTEND included adult patients with moderate to severely active CD based on mucosal ulceration on colonoscopy and CDAI score of 220 to 450, in addition to previous inadequate response to conventional therapy (immunomodulators or glucocorticoids) or TNF antagonists. Primary nonresponders were excluded. Participants received adalimumab (ADA) induction therapy for 4 weeks, then were rerandomized to maintenance 40 mg of ADA every 2 weeks or placebo. The total study duration was 52 weeks. Colonoscopy was performed at baseline, week 12, and week 52; the endoscopic evaluation was centrally read. In EXTEND, a central review committee assessed endoscopic endpoints. In the event of disagreement from a central reader, an adjudication process occurred, which involved 1 or 2 additional independent blinded reviewers.²⁷

Participants

The UNITI studies included 334 patients in the endoscopy substudy analysis. Participants who crossed over between UST and placebo were excluded. One hundred fifty-nine patients received UST induction and maintenance throughout the 52-week period. EXTEND enrolled 129 patients with baseline endoscopic assessments; however, only 124 patients had complete data.

This post hoc analysis of UNITI and EXTEND included 283 patients from these studies who had complete baseline, postinduction endoscopic, and PRO assessment data. An intention to treat analysis was used to evaluate all 1-year outcomes. If participants did not undergo a 1-year colonoscopy due to loss of follow-up, they were deemed to not have achieved ER. Among the total of 283 participants, 131 had baseline ulcers ≥5 mm that persisted at the postinduction assessment.

Variables

Simple Endoscopic Score for Crohn’s Disease

Endoscopic disease activity was assessed using the Simple Endoscopic Score for Crohn’s disease (SES-CD) in both studies.²⁸ For the SES-CD, ulcer size is scored in each segment as 0 if no ulcers, 1 if diameter <5 mm, 2 if diameter between 5 mm and 2 cm, and 3 if diameter >2 cm.²⁸ The sum of all variables that are measured for the SES-CD for each of the 5 bowel segments (terminal ileum, right colon, transverse colon, left colon, rectum) provides the total SES-CD score. A score ≥16 reflects severely active disease, ≥7-15 moderately active, ≥3-6 mildly active, and <3 inactive disease.²⁸

The endoscopic variable of interest was the presence of ulcers ≥5 mm (SES-CD score of 2 or 3 for ulcer size) at the postinduction endoscopic assessment in at least one segment of the ileum or colon. Postinduction endoscopic assessment was done at week 8 in the UNITI studies, and at week 12 in EXTEND.

Patient-reported Outcomes

Patient-reported outcomes of interest included AP, SF, and general well-being (GWB), which are self-reported by the patient over 7 days prior to assessment. Abdominal pain is scored as 0 (none), 1 (mild), 2 (moderate), and 3 (severe). For SF, daily liquid stool count is averaged. Thresholds of SF ≥4 and/or AP ≥2 were used to indicate moderate to severely active PRO scores of CD or high PRO scores, as this combination was shown to correlate with moderate-severe CDAI.²⁹ These thresholds have also been used by recent CD trials to define active symptoms within the inclusion criteria (ClinicalTrials.gov identifiers NCT03440385 and NCT03870334).^30,31 General well-being is also reported in the CDAI and has been considered as a PRO of interest within the PRO3 score.²⁰ General well-being is scored between 0 and 4 and is self-reported by the patient over the 7 days prior to assessment as the following: generally well (0), slightly under par (1), poor (2), very poor (3), or terrible (4).

Remission

Endoscopic remission was defined as a total SES-CD score <3 (ER-1). A sensitivity analysis was conducted with ER defined alternatively as a total SES-CD score of 0 (ER-2). Endoscopic response was defined as total SES-CD reduction ≥50% from baseline. PRO2 remission was defined as a 7-day average daily SF of ≤1.5 and AP ≤1.0. These values have been determined to be optimal cut-offs in prior studies of moderate CD patients by using ROC curves for CDAI remission.²⁰ PRO3 remission was a 7-day average of daily SF of ≤1.5, AP ≤1.0, and GWB ≤1.0. PRO3 response was defined as ≥50% improvement in PRO3 score from baseline.

Outcomes

Postinduction Ulcers and Endoscopic Remission at 1 year

Our primary objective was to evaluate whether the presence of ulcers ≥5 mm in CD patients following induction therapy with UST or ADA was associated with ER-1 at 1 year in individuals who had baseline ulcers ≥5 mm. Secondary analyses were planned for outcomes using an alternative definition of ER-2 and endoscopic response at 1 year.

Postinduction Ulcers and PROs on Endoscopic Remission at 1 year

Patients with baseline ulcers ≥5 mm and at least 1 elevated baseline PRO score (SF >4 and/or AP >2) were analyzed to determine whether postinduction PRO scores (higher AP, SF, and GWB scores compared with lower scores) could predict ER-1 or endoscopic response. Baseline ulcer size and PRO scores are week 0 measurements. Prespecified analyses were planned to determine if patients who achieved complete resolution of PRO scores at postinduction (PRO2 or PRO3 remission) were more likely to achieve ER-1 and endoscopic response at 1 year compared with patients who did not. Sensitivity analyses were also planned to evaluate total SES-CD score and the affected surface SES-CD subscore to determine the relative prognostic importance of ulcer size.

Statistical Analyses

Descriptive statistics were used to summarize baseline demographics and disease characteristics of CD patients. Continuous variables are presented as means and standard deviations (SDs) or as medians and interquartile ranges (IQRs); dichotomous variables are presented as proportions or percentages. Proportions of patients achieving PRO2 remission, PRO3 remission, and PRO3 response outcomes at postinduction were calculated. Multivariate logistic regression models were used to evaluate the relationship between postinduction ulcer size and PRO symptoms (as described previously) and achievement of ER-1 and endoscopic response. Adjustments for known confounders of endoscopic healing were performed including disease duration and treatment allocation. Sensitivity analyses were performed to evaluate outcomes based on treatment and location of ulcers (ileum vs colon). For each analysis, results are presented as adjusted and unadjusted odds ratios (ORs) with 95% confidence intervals (CIs) and their associated P values.

The purpose of our analysis was to generate hypotheses. As such, rather than hypothesis-confirming, the results from this study should be construed as potential associations. The statistical significant threshold was set to 0.01—not the commonly accepted value of 0.05 to address the issue of multiple testing and false positive results. Data were analyzed using Stata/IC version 15.0.

Results

Demographics

Characteristics of the 283 CD patients analyzed in this study with baseline ulcers ≥5 mm and who underwent postinduction endoscopy are presented in Table 1. The median disease duration was 8.7 years (IQR, 3.4-10.3 years), the mean age was 39 years (SD, 12.4), and a total of 118 (41.7%) participants were male. Induction with UST was completed by 124 patients (43.8%), and ADA by 159 (56.2%). Among the 283 patients with baseline ulcers ≥5 mm, 122 (43.1%) patients had ulcers in the ileum, 59 (20.8%) in the ascending colon, 60 (21.2%) in the transverse colon, 103 (36.4%) in the descending colon, and 81 (28.6%) in the rectum. Details regarding the affected surface of segments at baseline are provided in Supplementary Table 1. The median baseline SES-CD score was 12.0 (IQR, 7.0-19.0). The majority of patients had ileocolonic disease (159 of 283, 56.2%), followed by isolated ileal disease (75 of 283, 26.5%) and colonic disease (49of 283, 17.3%). Furthermore, the median baseline AP score was 2.0 (IQR, 1.6-2.4), median baseline SF score was 5.3 (IQR, 3.4-7.3), and median baseline GWB score was 2.1 (IQR, 1.7-2.9).

Of the 283 individuals who underwent baseline and postinduction endoscopy, a total of 48 (17%) achieved 1-year ER-1, 39 (13.8%) achieved ER-2, and 65 (23%) achieved endoscopic response. There were 131 participants with ulcers ≥5 mm at baseline that persisted at postinduction assessment. At 1 year, 9 of 131 (6.9%) of these patients achieved ER-1, 6 of 131 (4.6%) patients achieved ER-2, and 22 of 131 (16.8%) achieved endoscopic response. Additionally, 57 of 131 (43.5%) had a PRO3 response, 31 of 131 (23.7%) had PRO3 remission, and 32 of 131 (24.4%) had PRO2 remission at the end of induction treatment with UST or ADA.

Postinduction Ulcer Size and 1-year Endoscopic Outcomes

Among the 131 CD patients with baseline and postinduction endoscopy ulcers ≥5 mm, ER-1 and ER-2 were approximately 5 times less likely to be achieved compared with patients without postinduction ulcers ≥5 mm (ER-1, OR 0.20, 95% CI, 0.08-0.51, P = .001; ER-2, OR 0.17, 95% CI, 0.06-0.53, P = .002; Table 2). Similarly, patients with postinduction ulcers ≥5 mm were less likely to achieve 1-year endoscopic response (OR 0.38; 95% CI, 0.18-0.38; P = .009). Findings were consistent in magnitude and direction across treatment groups (Supplementary Table 2) and in the ileum and colon separately (Supplementary Table 3). Postinduction ulcer size had a greater ability to predict ER-1 and ER-2 compared with postinduction affected surface and total SES-CD score. Among patients with postinduction affected surface ≥50% (ie, subscore of 2 or 3), odds of ER-1 and ER-2 were lower (adjusted OR, 0.23; 95% CI, 0.09-0.58; P = .002) and (adjusted OR, 0.26; 95% CI, 0.09-0.78; P = .003). Similar findings were observed when postinduction affected surface ≥75% (ie, subscore of 3) was evaluated (Table 3). Patients with a total postinduction SES-CD score ≥3 had a similar reduced odds of ER-1 and ER-2 (adjusted OR, 0.23; 95% CI, 0.11-0.48; P < 0.001) and (adjusted OR, 0.24; 95% CI, 0.11-0.54; P = .001; Table 4).

Postinduction PRO Severity and 1-year Endoscopic Outcomes

Among 283 patients with baseline ulcers ≥5 mm, those with at least 1 elevated baseline PRO score (SF >4 and/or AP >2) were analyzed to determine whether postinduction PRO scores (higher AP, SF, and GWB scores compared with lower scores) could predict 1-year ER-1 or endoscopic response. There were 281 patients with at least 1 elevated PRO score. No significant differences were observed when comparing patients with higher postinduction PRO severity scores to lower scores in the odds of achieving 1-year ER-1. Patients with moderate-severe AP postinduction were as likely to achieve 1-year ER-1 compared with those with mild or no AP (OR, 1.42; 95% CI, 0.76-2.66; P = .272). Similar findings were found for CD patients with elevated postinduction PROs compared with lower scores in achieving ER-1 for the following: SF, GWB, combined AP and SF, combined AP and GWB, combined SF and GWB, combined AP, SF and GWB compared with their respective lower scores (Table 5). Furthermore, there was no significant difference in 1-year ER-1 was observed between patients in postinduction PRO2 or PRO3 remission compared with patients not in PRO2 or PRO3 remission.

There was no difference in achieving 1-year endoscopic response in those with higher postinduction PRO scores compared with lower scores among patients with postinduction ulcers ≥5 mm. There was no difference in achieving 1-year endoscopic response among patients in postinduction PRO2 or PRO3 remission compared with those not in PRO2 or PRO3 remission (Table 6).

Discussion

In this post hoc analysis of the UNITI and EXTEND trials, we have established that CD patients with persistent postinduction mucosal ulcers ≥5 mm are less likely to achieve ER at 1-year. This finding was consistent in magnitude and direction for both ustekinumab and adalimumab. Sensitivity analyses demonstrated that postinduction ulcer size had the greatest ability to predict ER-1 and ER-2 at 1 year compared with postinduction-affected surface subscore and total SES-CD score. We also demonstrated that changes in PROs and PRO severity following induction therapy do not provide additional prognostic information. Specifically, individuals with persistent ulcers ≥5 mm were 5 times less likely to achieve 1-year ER and less likely to achieve endoscopic response compared with patients with smaller or no postinduction ulceration. Improvement in PRO scores (such as PRO2 or PRO3 remission) or subcomponents, including AP, SF, GWB, did not predict endoscopic improvement among those patients with baseline elevated PRO scores and ulcers ≥5 mm. These findings suggest that persistence of ulcers ≥5 mm following induction therapy is highly informative of likelihood to achieve ER at 1 year, whereas PRO improvement after induction does not provide additional prognostic utility.

The achievement of CR is important in CD patient care and is associated with improvement in patients’ quality of life.³² However, there has been a shift away from relying solely on PROs to monitor CD activity and response to therapy, instead moving towards objective measures like endoscopic findings and biomarkers. The recent STRIDE-II consensus statements have emphasized targeting objective improvements in endoscopy and biomarkers more than the previous STRIDE recommendations.¹⁴ Endoscopic remission is the long-term treatment target recommended by numerous international guidelines and consensus statements.^14,33–35 Our group has previously published data demonstrating that postinduction PROs including AP and SF are strongly predictive of 1-year CR but are not associated with 1-year ER.²¹ The current analysis supports those findings and reinforces the limited value of PRO improvements for predicting longer-term endoscopic improvement. Concomitant functional gastrointestinal syndromes in patients with CD could be one explanation for why PROs are unreliable in predicting 1-year ER. Symptoms of irritable bowel syndrome (IBS), small intestinal bacterial overgrowth, and bile salt malabsorption can mimic those of CD. A meta-analysis of CD patients in remission demonstrated that the prevalence of IBS-type symptoms was approximately 32%, and CDAI scores may be comparable in IBS and IBD.^36,37 This has been reinforced by a post hoc analysis from the SONIC trial that demonstrated clinical symptoms as measured by the CDAI is inaccurate at measuring underlying inflammation. Additionally, cross-sectional studies have not demonstrated adequate correlation between PROs and active inflammation in CD measured biochemically or endoscopically.^18,19,38

Our findings may impact both clinical trials and clinical practice. Clinical trials largely focus on CD symptoms as defined using the CDAI and other PRO inclusion criteria when assessing group allocation, rerandomization to maintenance therapy, and study outcomes.^24,25 We have demonstrated that PROs lack prognostic value; thus consideration should be given for clinical trials to minimize the reliance on these subjective measures and incorporate more objective, prognostically significant criteria such as postinduction mucosal ulceration size. This would allow for the minimization of confounding factors that may exist when relying on PROs to determine eligibility for rerandomization and to draw study conclusions. This shift towards incorporation of objective measures of CD therapy response is reflected in anticipated studies such as REACT2 (ClinicalTrials.gov identifier NCT01698307) that identified absence of mucosal ulcerations (>5 mm) as an end-point target.³⁹ Moreover based on our results, there may be a benefit in subclassifying patients into strata by ulcer size, with ≥5 mm representing a subgroup that may require more intensified therapy. For clinical practice, clinicians who identify patients with ongoing ulcers ≥5 mm may consider increased monitoring of this population by trending biomarkers, measuring therapeutic drug concentrations, or escalating pharmacotherapy; although this study does not demonstrate that such strategies would provide additional benefit for these patients. It is also worth mentioning there may be increased costs to this strategy for both health systems and patients to monitor postinduction CD endoscopic activity. Moving forward, trialists and clinicians may consider utilizing more objective endoscopic findings in conjunction with biochemical markers such as fecal calprotectin to predict the likelihood of endoscopic healing.²²

Our post hoc analysis of UNITI and EXTEND has several strengths. Firstly, high-quality data sets from 2 phase 3/4 clinical trials were utilized, and all endoscopic interpretations were centrally read. However, our analysis is not without its limitations. There are several possible sources of selection bias in our analysis. First, the UNITI and EXTEND trials included patients with a baseline CDAI of 220 to 450, and patients with mild or very severe symptoms were not enrolled. Second, the timing of postinduction endoscopic assessments was not consistent in our analysis (week 8 in EXTEND and week 12 in UNITI). Third, a cohort of patients in the UNITI study participated in the endoscopic substudy, and it is unclear how this cohort was selected. Moreover, patients were followed for 52 weeks, and there is some uncertainty regarding how long biologic therapy should be used for before evaluating for ER, and whether this time frame varies based on the biologic used.¹⁴ Lastly, some of the statistically significant findings in our study may be due to chance alone due to multiple comparisons. However, this study is intended to be hypothesis-generating rather than confirmatory, and a lower P value threshold was used to account for this possibility.

Conclusions

These results highlight the importance of relying on objective measures such as endoscopy, rather than subjective PROs, to predict 1-year ER following CD induction therapy. Moving forward, prospective studies are needed to determine whether patients with ongoing ulcers ≥5 mm after completion of induction therapy would benefit from intensification of medical therapy. There also remains a need to explore relationships between endoscopic lesions and biomarkers for predicting ER after induction therapy. Consideration should be given to clinical trials that incorporate objective and prognostically significant measures of CD activity when assessing for response to therapy and lessen reliance on nonprognostic, subjective measures of disease.

Author Contributions

C.P.—study concept and design; data interpretation; drafting of the manuscript.

E.W.—acquisition and compilation of data; statistical analysis; drafting of the manuscript.

A.A.—study concept and design; data interpretation; drafting of the manuscript.

P.D.—study design; drafting of the manuscript.

J.M.—study design; drafting of the manuscript.

W.R.—study design; drafting of the manuscript.

N.N.—study concept and design; acquisition and compilation of data; statistical analysis; data interpretation; drafting of the manuscript.

All authors approved the final version of the manuscript.

Funding

This study received no funding and no authors have received support for the submitted manuscript.

Conflicts of Interest

N.N. holds a McMaster University Department of Medicine Internal Career Award. N.N. has received honoraria from Janssen, Abbvie, Takeda, Pfizer, Merck, and Ferring.

J.M. has received honoraria from Janssen, AbbVie, Allergan, Bristol-Meyer-Squibb, Ferring, Janssen, Lilly, Lupin, Merck, Pfizer, Pharmascience, Roche, Shire, Takeda and Teva.

P.D. is supported by a Digestive Diseases Research Center grant NIH DK120515 and a Research Scholar Award from the American Gastroenterology Association. Parmabir S.Dulai has received research support and consulting for Takeda, Abbvie, Janssen, Pfizer. W.R. has received support for the following:

Speaker for Abbott Laboratories, Abbvie, Aesca, Aptalis, Astellas, Centocor, Celltrion, Danone Austria, Elan, Falk Pharma GmbH, Ferring, Immundiagnostik, Mitsubishi Tanabe Pharma Corporation, MSD, Otsuka, PDL, Pharmacosmos, PLS Education, Schering-Plough, Shire, Takeda, Therakos, Vifor, Yakult,

Consultant for Abbott Laboratories, Abbvie, Aesca, Algernon, Amgen, AM Pharma, AMT, AOP Orphan, Arena Pharmaceuticals, Astellas, Astra Zeneca, Avaxia, Roland Berger GmBH, Bioclinica, Biogen IDEC, Boehringer-Ingelheim, Bristol-Myers Squibb, Cellerix, Chemocentryx, Celgene, Centocor, Celltrion, Covance, Danone Austria, DSM, Elan, Eli Lilly, Ernest & Young, Falk Pharma GmbH, Ferring, Galapagos, Genentech, Gilead, Grünenthal, ICON, Index Pharma, Inova, Janssen, Johnson & Johnson, Kyowa Hakko Kirin Pharma, Lipid Therapeutics, LivaNova, Mallinckrodt, Medahead, MedImmune, Millenium, Mitsubishi Tanabe Pharma Corporation, MSD, Nash Pharmaceuticals, Nestle, Nippon Kayaku, Novartis, Ocera, OMass, Otsuka, Parexel, PDL, Periconsulting, Pharmacosmos, Philip Morris Institute, Pfizer, Procter & Gamble, Prometheus, Protagonist, Provention, Robarts Clinical Trial, Sandoz, Schering-Plough, Second Genome, Seres Therapeutics, Setpointmedical, Sigmoid, Sublimity, Takeda, Therakos, Theravance, Tigenix, UCB, Vifor, Zealand, Zyngenia, and 4SC,

Advisory board member for Abbott Laboratories, Abbvie, Aesca, Amgen, AM Pharma, Astellas, Astra Zeneca, Avaxia, Biogen IDEC, Boehringer-Ingelheim, Bristol-Myers Squibb, Cellerix, Chemocentryx, Celgene, Centocor, Celltrion, Danone Austria, DSM, Elan, Ferring, Galapagos, Genentech, Grünenthal, Inova, Janssen, Johnson & Johnson, Kyowa Hakko Kirin Pharma, Lipid Therapeutics, MedImmune, Millenium, Mitsubishi Tanabe Pharma Corporation, MSD, Nestle, Novartis, Ocera, Otsuka, PDL, Pharmacosmos, Pfizer, Procter & Gamble, Prometheus, Sandoz, Schering-Plough, Second Genome, Setpointmedical, Takeda, Therakos, Tigenix, UCB, Zealand, Zyngenia, and 4SC

No other authors have any relevant conflicts of interest.

Data Availability Statement

Data can be made available upon request to third parties. This study, carried out under YODA Project #2020-4529, used data obtained from the Yale University Open Data Access Project, which has an agreement with JANSSEN RESEARCH & DEVELOPMENT, L.L.C.. The interpretation and reporting of research using this data are solely the responsibility of the authors and does not necessarily represent the official views of the Yale University Open Data Access Project or JANSSEN RESEARCH & DEVELOPMENT, L.L.C. This publication (Vivli protocol #00006496) is based on research using data from AbbVie that has been made available through Vivli, Inc. Vivli has not contributed to or approved and is not in any way responsible for the contents of this publication.

References