Gastroenterology 2015;148:344–354

Comparative Effectiveness of Immunosuppressants and

Biologics for Inducing and Maintaining Remission in Crohn’s
Disease: A Network Meta-analysis
Glen S. Hazlewood,1,* Ali Rezaie,3,* Meredith Borman,1 Remo Panaccione,1 Subrata Ghosh,1
Cynthia H. Seow,1,2 Ellen Kuenzig,2 George Tomlinson,4 Corey A. Siegel,5 Gil Y. Melmed,3 and
CLINICAL AT

Gilaad G. Kaplan1,2
1
Department of Medicine, 2Departmet of Community Health Sciences, University of Calgary, Calgary, Alberta, Canada;
3
Department of Medicine, Cedars-Sinai Medical Center, Los Angeles, California; 4University Health Network, University
of Toronto, Toronto, Ontario, Canada; 5Department of Medicine, Dartmouth-Hitchcock Medical Center, Lebanon,
New Hampshire

This article has an accompanying continuing medical education activity on page e14. Learning Objective: Upon completion of the
CME activity, successful learners will be able to compare different treatment strategies in managing Crohn’s disease and
understand the purpose of conducting a network meta-analysis.

Keywords: Network Meta-analysis; IBD; Anti-TNF Therapy;

See Covering the Cover synopsis on page 266. Immunosuppressive Agents.

BACKGROUND & AIMS: There is controversy regarding the

best treatment for patients with Crohn’s disease because of the
lack of direct comparative trials. We compared therapies for
induction and maintenance of remission in patients with
Crohn’s disease, based on direct and indirect evidence.
METHODS: We performed systematic reviews of MEDLINE,
EMBASE, and Cochrane Central databases, through June 2014.
We identified randomized controlled trials (N ¼ 39) comparing
methotrexate, azathioprine/6-mercaptopurine, infliximab,
adalimumab, certolizumab, vedolizumab, or combined thera-
pies with placebo or an active agent for induction and main-
tenance of remission in adult patients with Crohn’s disease.
Pairwise treatment effects were estimated through a Bayesian
random-effects network meta-analysis and reported as odds
ratios (OR) with a 95% credible interval (CrI). RESULTS:
Infliximab, the combination of infliximab and azathioprine
(infliximab þ azathioprine), adalimumab, and vedolizumab
were superior to placebo for induction of remission. In pair-
wise comparisons of anti–tumor necrosis factor agents,
infliximab þ azathioprine (OR, 3.1; 95% CrI, 1.4–7.7) and
adalimumab (OR, 2.1; 95% CrI, 1.0–4.6) were superior to
certolizumab for induction of remission. All treatments were
superior to placebo for maintaining remission, except for the
combination of infliximab and methotrexate. Adalimumab,
infliximab, and infliximab þ azathioprine were superior to
azathioprine/6-mercaptopurine: adalimumab (OR, 2.9; 95%
CrI, 1.6–5.1), infliximab (OR, 1.6; 95% CrI, 1.0–2.5),
infliximab þ azathioprine (OR, 3.0; 95% CrI, 1.7–5.5) for
maintenance of remission. Adalimumab and infliximab þ
azathioprine were superior to certolizumab: adalimumab (OR,
2.5; 95% CrI, 1.4–4.6) and infliximab þ azathioprine (OR, 2.6;
95% CrI, 1.3–6.0). Adalimumab was superior to vedolizumab
(OR, 2.4; 95% CrI, 1.2–4.6). CONCLUSIONS: Based on a
network meta-analysis, adalimumab and infliximab þ azathi-
oprine are the most effective therapies for induction and
maintenance of remission of Crohn’s disease.
C rohn’s disease is a chronic inflammatory condition of
the intestinal tract that affects individuals in the
prime of their lives, subjecting them to social stigma,
impinging on their ability to attend work or school, and
reducing their quality of life.1 The United States spends
approximately $6.1 billion annually on direct inflammatory
bowel disease health care costs, with the major drivers of
cost being hospitalizations, surgeries, and medications.2
Over the past few decades the rate of intestinal resections
for Crohn’s disease has been reduced after the introduction
of immunosuppressant drugs (ie, azathioprine/6-
mercaptopurine and methotrexate) and then anti–tumor
necrosis factor (TNF) therapy (infliximab, adalimumab, and
certolizumab).3
However, the choice of induction and maintenance
strategy remains challenging. Although multiple trials exist,
most are placebo-controlled, with a lack of head-to-head
trials between active treatments. The paucity of head-to-
head clinical trials has raised controversial therapeutic de-
cisions including the choice between immunosuppressants
vs anti-TNF therapy, choosing within a class of medication,
and deciding whether to prescribe anti-TNF monotherapy vs
concomitant anti-TNF with immunosuppressants.4,5 More-
over, gastroenterologists now must contend with the posi-
tioning of vedolizumab in the therapeutic paradigm of
*Authors share co-first authorship.
Abbreviations used in this paper: CDAI, Crohn’s Disease Activity Index;
CrI, credible interval; OR, odds ratio; TNF, tumor necrosis factor; WDAE,
withdrawals due to adverse events.
© 2015 by the AGA Institute
0016-5085/$36.00
http://dx.doi.org/10.1053/j.gastro.2014.10.011
February 2015
Crohn’s disease.6 Numerous treatment algorithms have
been proposed by clinical experts in an attempt to synthe-
size the literature because direct comparative efficacy trials
are lacking.4 In keeping with this, comparisons of biologic
treatment strategies for Crohn’s disease were listed by the
Institute of Medicine as one of the top priorities for
comparative effectiveness research.7,8
Where direct head-to-head evidence is lacking, indirect
evidence may help inform decision making that aims to
maximize efficacy, minimize toxicity, and optimize costs. An
indirect treatment comparison can be made between 2
treatments if each treatment has been compared with a
common comparator. Comparisons are made between
treatment effects, not individual treatment arms, thereby
preserving randomization.9 For example, if one trial com-
pares treatments A and B and another trial compares
treatments B and C, an indirect comparison between treat-
ments A and C can be determined by comparing the relative
effects (eg, odds ratios) of the 2 trials. A Bayesian network
meta-analysis (or mixed treatment comparison) considers
all indirect and direct evidence, to determine the relative
treatment effects between all interventions that can be
linked through shared comparators.9 Considering indirect
evidence adds strength to the estimation of treatment ef-
fects, even where head-to-head trials are available.9
The primary objective of this study was to compare the
efficacy of therapies for induction and maintenance of
remission including azathioprine/6-mercaptopurine, metho-
trexate, approved anti-TNF therapies (infliximab, adalimu-
mab, and certolizumab), vedolizumab, or their combination
in adult patients with Crohn’s disease, based on direct and
indirect evidence from randomized controlled trials.
Methods
Eligibility Criteria
We included all randomized controlled trials that assessed
treatments (azathioprine/6-mercaptopurine, methotrexate,
infliximab, adalimumab, certolizumab, and vedolizumab) alone
or in combination in adult patients with Crohn’s disease. We
included trials assessing induction of remission of immuno-
suppressants between 12 and 17 weeks. We included trials
assessing the induction of remission of biologic therapy be-
tween 4 and 17 weeks because the onset of action of biologic
therapies is more rapid. Trials assessing maintenance of
remission had to be at least 24 weeks in duration.10
Studies assessing natalizumab were excluded because pre-
scription of natalizumab is restricted to individuals failing
immunosuppressive and anti-TNF therapy.11 We also excluded
trials studying only pediatric (age, <18 y) or postoperative
patients, studies in which the treatment was not fixed (eg,
standard of care), studies with a randomized withdrawal
design, trials with a cross-over design, studies exclusively
assessing fistulizing Crohn’s disease, and studies that did not
report remission as an outcome. After identification of eligible
studies, trials that could not be linked within the network
through a shared comparator were excluded.
The primary outcome was remission, which was defined as
a Crohn’s Disease Activity Index (CDAI) less than 150. When the
Comparative Effectiveness of Therapies in CD 345
CDAI was not reported, we used the remission criteria defined
in the study. Secondary end points included total withdrawals
and withdrawals due to adverse events (WDAEs). Total with-
drawals were defined as the total number of patients who were
withdrawn from the study for any reason after randomization.
WDAEs were collected as defined by the study.
Literature Search and Study Selection
Trials were identified through existing Cochrane systematic
reviews and a technical report from the American Gastroenter-
ology Association.10,12–16 We updated the database search from
CLINICAL AT
January 2007 (ie, the earliest search date of the Cochrane re-
views) to June 2014 in MEDLINE, Embase, and the Cochrane
Central register of controlled trials. The database search strategy
was adapted from the systematic reviews (the full search
strategy is available in the Supplementary Materials and
Methods). We supplemented this with a search of trial regis-
tries (www.clinicaltrials.gov) and by screening all American
College of Gastroenterology, Digestive Disease Week, United
European Gastroenterology Week, and European Crohn’s and
Colitis Organization conference proceedings published from
January 2007 through June 2014. For vedolizumab, a separate
literature search was performed from 1966 through June 2014
on the earlier-noted research databases. Search results were
screened by 2 independent reviewers (A.R., M.B.) first by title
and abstract and then by full text. Disagreements were resolved
through consensus and discussion with a third reviewer (G.G.K.).
Selected studies were reviewed independently by 5 experts in
the inflammatory bowel disease literature (R.P., S.G., C.H.S.,
G.Y.M., and C.A.S.) to confirm the inclusion and exclusion of trials.
Data Collection and Quality Appraisal
Data were abstracted for relevant study characteristics
(Supplementary Table 1) and for all primary and secondary
outcomes. For induction, remission was extracted at the time of
the primary outcome specified in the trial, with the following
exception: we used data at or closest to 12 weeks when the
time point of the primary outcome was not specified or in-
duction was not the primary goal of the study. For maintenance,
remission was extracted at the end of the trial. Total with-
drawals and WDAEs were extracted at trial end for both in-
duction and maintenance trials. Data were extracted on a basis
of intention-to-treat analysis.
In each trial arm, we abstracted the total number of patients
randomized and the total number of patients who experienced
the outcome. If only percentages were reported, the number of
patients with the outcome was calculated and rounded to the
nearest whole number. For data available only in graphic
format, images in the highest resolution available were digi-
tized and extracted using the software program Graphclick
(version 3.0.2; Arizona Software; www.arizona-software.ch/
graphclick/). Any disagreements were resolved through dis-
cussion and repeat extraction. The quality of trials was rated
through the Cochrane Risk of Bias tool.
Synthesis of Results
For the main analysis we excluded trials with a high risk of
bias. Treatment effects for remission and total withdrawals
were determined using a random-effects Bayesian network
meta-analysis (mixed treatment comparison). A random-effects
 346 Hazlewood et al
model was preferred based on the clinical heterogeneity across
trials. For WDAEs (a secondary outcome), a fixed-effects model
was used because a random-effects model did not produce
meaningful results because of the rarity of events.
Statistical heterogeneity was measured by the between-
study standard deviation (SD) in log odds ratio (OR). There is
no definite threshold for values of SD of the log(OR) that in-
dicates minimal or important heterogeneity, although values
between 0.1 to 0.5 are considered small, values between 0.5
and 1.0 are considered fairly high, and values greater than 1.0
indicate extreme heterogeneity.17 Statistical analyses were
performed using R statistical software version 2.15.2 with the
CLINICAL AT
rjags package version 3.3.0 (www.r-project.org). The Just
Another Gibbs Sampler (JAGS) code for the Bayesian network
meta-analysis has been published and is provided in the
Supplementary Materials and Methods.18
Uninformative prior probability distributions were used for
all variables. All chains were run with 10,000 burn-in iterations
followed by 10,000 monitoring iterations. Convergence was
assessed by running 3 chains, inspecting the sampling history
plots, and calculating Gelman–Rubin–Brooke statistics.
Summary Measures
For each pair-wise comparison we calculated the odds ratio
(OR) with a 95% credible interval (CrI) and the probability that
each treatment was superior to the other. We considered a
treatment as showing superiority (or inferiority) if the 2-sided
95% CrI of the OR excluded 1, which equates to a 97.5%
probability that the treatment is superior.
Additional Analyses
Several sensitivity analyses were performed for induction
and maintenance of remission in which we performed the
following: (1) included trials with a high risk of bias; (2)
Gastroenterology Vol. 148, No. 2
excluded trials with 6-mercaptopurine, low azathioprine dosing
(<2 mg/kg), and oral methotrexate; and (3) excluded trials in
which patients had prior exposure to anti-TNF therapy. For
maintenance of remission we performed the following additional
sensitivity analyses: (1) included only those trials whose pri-
mary outcome was maintenance of remission at 1 year or longer;
(2) included only trials that randomized patients after successful
induction (in remission or after achieving a decrease in CDAI of
70 or 100 points); and (3) included only treat-through trials.
Treat-through trials were defined as randomized controlled
trials that randomized active patients at baseline and continued
to treat patients for 24 weeks or longer. Also, a network
meta-analysis was developed using a fixed-effects model rather
than a random-effects model.
To assess whether the effect of treatment depended on the
underlying response rate, we performed meta-regression
where treatment effects relative to placebo were allowed to
depend on the placebo response rates through a common
regression coefficient. The meta-regression analysis was
restricted to placebo-controlled trials because there were too
few head-to-head trials to estimate regression coefficients for
each pair-wise comparison. In addition, we performed a
random-effects network meta-analysis without meta-
regression, using only placebo-controlled trials. By comparing
results from this analysis with the meta-regression, we evalu-
ated whether differences in the effect of a treatment were
owing to the meta-regression or the study selection.
To examine the consistency of the evidence, we compared
the estimated treatment effects from the network meta-analysis
with standard direct treatment effects where head-to-head
trials existed. Direct treatment effects were estimated using a
random-effects meta-analysis of odds ratios, with the between-
study variance estimated around the Mantel–Haenszel fixed
effect, using Review Manager 5 (Version 5.3. Copenhagen: The
Nordic Cochrane Centre, The Cochrane Collaboration, 2014).
Figure 1. Flow-chart of
search results.
February 2015 Comparative Effectiveness of Therapies in CD 347

Results

Infliximab þ methotrexate

0.75 (0.17–2.7), 33%

Search Results, Trial Characteristics,
and Risk of Bias

–
We included 39 trials from 35 articles (Figure 1).6,19–52
Four articles contained 2 trials each that were considered
separately.6,49–51 Excluded studies and the rationale for
exclusion are shown in Supplementary Table 2. Character-

Infliximabþ azathioprine
istics of included trials are shown in Supplementary Table 1.

0.62 (0.18–2.4), 23%

0.47 (0.18–1.1), 4%
Fifteen trials evaluated anti-TNF therapy, 4 trials evaluated
vedolizumab, 15 trials evaluated immunosuppressants, and

CLINICAL AT
–
5 trials evaluated combination therapy. Twenty-four trials

An OR greater than 1 favors the intervention (row) over the comparator (column), indicating a greater odds of induction of remission.
provided data on induction of remission and 24 trials pro-
vided data on maintenance of remission. Of the 24 trials that
provided data on maintenance of remission, 14 were treat-
through trials and 10 were maintenance trials that ran-

1.4 (0.59–4.2), 79%

0.90 (0.22–4.4), 44%

0.67 (0.33–1.5), 15%

Comparator, OR (95% CrI),a probability intervention superior to comparator (greater odds of remission)
domized individuals if they responded to induction therapy.

Adalimumab
Seven trials compared active treatments and did not include

–
a placebo arm19,22,27,31–33,47 (Supplementary Table 1 and
Figure 2). CDAI was used to define remission in all except 4
of the earlier trials.20,34,37,50 The risk of bias was judged to
be high in the 4 trials that were not double-blind.19,32,33,47

1.0 (0.32–2.4), 53%

1.5 (0.61–3.1), 85%

0.93 (0.34–2.6), 44%

0.70 (0.25–1.5), 20%

These 4 trials were excluded from the primary analysis. A
detailed assessment of the risk of bias is presented in

Infliximab
Supplementary Table 3. Supplementary Table 4 summarizes

–
the studies that were included in the main analysis (ie, in-
duction and/or remission) and the sensitivity analysis.

3.1 (1.4–7.7), >99%

2.1 (0.98–5.5), 97%

1.9 (0.56–8.2), 84%

1.4 (0.77–2.7), 89%

2.1 (1.0–4.6), 98%
Synthesis of Results
Certolizumab

Induction of remission. Azathioprine/6-mercap-

–

topurine and methotrexate were not different from

placebo for inducing remission relative to placebo (azathi-
oprine/6-mercaptopurine: OR, 1.2; 95% CrI, 0.76–2.1; and
methotrexate: OR, 1.5; 95% CrI, 0.72–3.2) (Table 1). Inflix-
1.7 (0.43–8.9), 76%

1.3 (0.53–3.2), 71%

(0.40–2.1), 38%
(0.69–6.4), 89%
(0.76–4.8), 92%
(1.0–8.5), 98%

imab, infliximab þ azathioprine, adalimumab, and vedoli-

Methotrexate

zumab had a more than 99% probability of being superior

–
Table 1.Pair-Wise Comparisons for Induction of Remission

1.8
1.9
2.7
0.89
(1.3–5.0), >99%

(1.9–6.3), >99%
(0.49–2.9), 71%

2.1 (0.67–7.9), 90%

1.6 (0.78–3.2), 91%

(0.58–2.0), 63%
6-mercaptopurine

(1.0–4.9), 98%
Azathioprine/

1.3

3.4
1.1
2.3
2.4

2.0 (1.2–3.3), >99%

(1.4–7.2), >99%
(1.6–5.5), >99%
(2.0–9.8), >99%
1.2 (0.76–2.1), 81%

(0.72–3.2), 88%
(0.95–2.0), 96%

2.6 (0.81–11), 94%

Placebo

1.5
1.4
2.8
2.9
4.3
6-mercaptopurine

methotrexate
Intervention

azathioprine

Figure 2. Network diagram of (A) induction trials, and (B)

Methotrexate
Certolizumab
Azathioprine/

Vedolizumab
Adalimumab
Infliximab þ

Infliximab þ

maintenance trials. Each line represents one treatment

Infliximab

comparison, with the line thickness reflecting the sample size.

Dashed lines indicate trials at high risk of bias that were
excluded from the primary analysis.
a
 348 Hazlewood et al Gastroenterology Vol. 148, No. 2

0.85 (0.29–2.5), 38%

to placebo for induction of remission (Table 1). Infliximab,
infliximab þ azathioprine, and adalimumab showed a more

methotrexate
Infliximab þ
than 98% probability of superiority to azathioprine/6-

–
mercaptopurine relative to azathioprine/6-mercaptopurine
(infliximab: OR, 2.3; 95% CrI, 1.3–5.0; infliximab þ azathi-
oprine: OR, 3.4; 95% CrI, 1.9–6.3; adalimumab: OR, 2.4; 95%
CrI, 1.0–4.9) (Table 1). In pair-wise comparisons between

Infliximab þ azathioprine

0.42 (0.17–0.92), 2%
anti-TNF agents, significant superiority was observed when

0.48 (0.16–1.3), 8%
infliximab þ azathioprine (OR, 3.1; 95% CrI, 1.4–7.7) and
adalimumab (OR, 2.1; 95% CrI, 1.0–4.6) were compared

–
with certolizumab (Table 1). The SD of the log(OR) for
CLINICAL AT

between-study variability was 0.25 (95% CrI, 0.03–0.56),

indicating an acceptable level of statistical heterogeneity.17

An OR greater than 1 favors the intervention (row) over the comparator (column), indicating a greater odds of maintaining remission.
For all sensitivity analyses of induction of remission,

1.0 (0.48–2.5), 53%

0.77 (0.39–1.5), 22% 0.42 (0.22–0.85), 1%

0.91 (0.41–2.1), 41% 0.51 (0.17–1.4), 9%
treatment comparisons relative to placebo and azathio-

Adalimumab
prine/6-mercaptopurine are shown in Supplementary

Comparator, OR (95% CrI),a probability intervention superior to comparator (greater odds of remission)
Figure 1A and B. Point estimates of the treatment effects

–
were similar for the other comparisons in sensitivity ana-
lyses. In the meta-regression model, for every 10% increase
in the placebo response rate, the OR decreased by a factor of

1.8 (0.94–3.4), 96%

1.8 (1.0–3.8), 98%
0.83 (95% CrI, 0.7–1.0). The meta-regression for the placebo
response rate is reported in Supplementary Table 5.

Infliximab
Treatment effects from the direct comparisons, when data

–
were available, were similar to the network meta-analysis
(Supplementary Figure 2A, with the original Forest plots
showing trial-level data in Supplementary Figure 3A).

97% 2.5 (1.4–4.6), >99%

Maintenance of remission. All treatments except

65% 1.4 (0.77–2.6), 87%

1.1 (0.35–3.6), 54% 1.3 (0.44–3.5), 68%

0.91 (0.39–2.3), 42% 1.1 (0.57–2.1), 59%

96% 2.6 (1.3–6.0), 99%
infliximab þ methotrexate had a greater than 98% proba-
Certolizumab

bility of being superior to placebo for maintenance of

–
remission (Table 2). Methotrexate monotherapy was not
different from other treatments; however, the credible in-
tervals were wide (Table 2). Adalimumab, infliximab, and
infliximab þ azathioprine had a greater than 98% proba-
34%
Methotrexate

bility of being superior to azathioprine/6-mercaptopurine

(0.39–2.1),
(0.51–2.8),
(0.96–5.0),
(0.90–6.1),

relative to azathioprine/6-mercaptopurine (adalimumab:

–

OR, 2.9; 95% CrI, 1.6–5.1; infliximab: OR, 1.6; 95% CrI,
1.0–2.5; infliximab þ azathioprine: OR, 3.0; 95% CrI,
0.85
1.2
2.1
2.2
Table 2.Pair-Wise Comparisons for Maintenance of Remission

1.7–5.5) (Table 2). In pair-wise comparisons between

anti-TNF agents, both adalimumab and infliximab þ
azathioprine had a 99% probability of being superior to
(1.6–5.1), >99%
(1.7–5.5), >99%
(0.58–2.8), 78%
(0.65–1.9), 72%

1.5 (0.57–3.7), 79%

1.3 (0.65–2.3), 76%

(1.0–2.5), 98%

certolizumab (Table 2). Vedolizumab was inferior to adali-

mercaptopurine
Azathioprine/6-

mumab (vedolizumab vs adalimumab: OR, 0.42; 95% CrI,

0.22–0.85) (Table 2). The SD of the log(OR) was 0.12 (95%
–

CrI, 0.01–0.50), indicating an acceptable level of statistical

heterogeneity.17
1.4
1.2
1.6
2.9
3.0

For all sensitivity analyses of maintenance of remission,

treatment comparisons relative to placebo and azathio-
1.7 (1.3–2.6), >99%

(1.4–3.0), >99%
(1.8–4.5), >99%
(3.3–8.1), >99%

2.2 (1.3–3.7), >99%

prine/6-mercaptopurine are shown in Supplementary

2.6 (0.96–6.6), 97%
(2.8–11), >99%
(1.1–4.8), 98%

Figure 1C and D. Results of the sensitivity analyses were

Placebo

similar regardless of the comparator chosen. Point estimates

of the treatment effects were similar for most comparisons
in sensitivity analyses; however, CrI for the sensitivity an-
2.4
2.0
2.8
5.1
5.2

alyses generally were wider because they were based on

6-mercaptopurine

fewer trials. In the meta-regression model, for every 10%

increase in the placebo response rate, the OR decreased by a
methotrexate
Intervention

azathioprine

factor of 0.94 (95% CrI, 0.79–1.1). The meta-regression for

Methotrexate
Certolizumab
Azathioprine/

Vedolizumab
Adalimumab
Infliximab þ

Infliximab þ

the placebo response rate is reported in Supplementary

Infliximab

Table 5. Treatment effects from the direct comparisons,

when data were available, were similar to the network
a
February 2015
meta-analysis (Supplementary Figure 2B, with the original
Forest plots showing trial-level data in Supplementary
Figure 3B).
Withdrawals. Total withdrawals were significantly less
common for adalimumab and infliximab þ azathioprine as
compared with placebo (relative to placebo: infliximab þ
azathioprine: OR, 0.27; 95% CrI, 0.08–0.72; and adalimumab;
OR, 0.43; 95% CrI, 0.26–0.69) (Table 3). Azathioprine/6-
mercaptopurine, methotrexate, infliximab, and infliximab þ
azathioprine were associated with more WDAEs than placebo
(Table 3). Azathioprine/6-mercaptopurine and methotrexate
were associated with more WDAEs compared with all
anti-TNF monotherapies except for azathioprine/6-
mercaptopurine vs infliximab (OR, 0.71; 95% CrI, 0.44–1.2).
Adalimumab was associated with fewer WDAEs than all
treatments except for vedolizumab and infliximab þ metho-
trexate. Vedolizumab had significantly fewer WDAEs than
azathioprine, methotrexate, infliximab, and infliximab þ
azathioprine (Table 3).
Discussion
We conducted a systematic review and network meta-
analysis of randomized controlled trials of immunosup-
pressants and biologics for induction and maintenance of
remission in adults with Crohn’s disease. Our systematic
review identified a paucity of head-to-head trials in the
Crohn’s disease literature, which are necessary to inform
clinicians on the appropriate efficacious and safe treatment
regimens for Crohn’s disease. By applying a network meta-
analysis approach, our study integrated direct head-to-
head data with indirect evidence to provide the most
robust data available for inducing and maintaining remis-
sion in Crohn’s disease. The purpose of comparative
effectiveness research is to “assist consumers, clinicians,
purchasers, and policy makers to make informed decisions
that will improve healthcare at both the individual and
population levels.”8 Until more head-to-head trials are
completed, these data are the best we have available to
guide clinical decision making within and between classes
of drug therapies.
Azathioprine/6-mercaptopurine and methotrexate were
not different from placebo for induction of remission in
Crohn’s disease. This finding is consistent with clinical
practice, whereby immunosuppressants are not prescribed
as monotherapy for induction of remission. Infliximab,
infliximab þ azathioprine, adalimumab, and vedolizumab
were superior to placebo in inducing remission. In pairwise
comparisons, infliximab þ azathioprine and adalimumab
showed superiority to certolizumab. However, induction
studies should be compared carefully because of intrinsic
differences in study designs. For example, the time point in
defining remission for induction trials varied from 1 to 4
months.
For several decades, azathioprine/6-mercaptopurine
have been the primary treatment for maintenance of
remission of moderate to severe Crohn’s disease. Although
this network meta-analysis confirms that azathioprine/6-
mercaptopurine are superior to placebo for maintenance
Comparative Effectiveness of Therapies in CD 349
of remission, its efficacy was inferior to adalimumab,
infliximab, and infliximab þ azathioprine. In this network
meta-analysis, methotrexate was not different from azathi-
oprine/6-mercaptopurine or any anti-TNF agent. Although
these data suggest that methotrexate may be an efficacious
primary treatment option for Crohn’s disease, the results
should be interpreted cautiously because the credible in-
tervals were wide. The uncertainty around the estimates
was the result of the small sample size of methotrexate
clinical trials. Furthermore, 219,33 of the 5 trials were at high
risk of bias and therefore were excluded from our primary
CLINICAL AT
analysis. In addition, among the 3 trials included in the
primary analysis, the dose of methotrexate used was mixed
(oral vs subcutaneous dose range, 12.5–25 mg). Nonethe-
less, this network meta-analysis at least supports a direct
head-to-head trial of conventionally dosed methotrexate vs
azathioprine/6-mercaptopurine with or without an addi-
tional anti-TNF arm to further inform this important ther-
apeutic choice.
The introduction of anti-TNF therapy has had a signifi-
cant impact on the management of Crohn’s disease by
improving quality of life and reducing hospitalization and
surgery.3 Although all 3 approved anti-TNF agents available
in the United States were superior to placebo for mainte-
nance of remission, important differences were observed
between agents when used as monotherapy (eg, adalimu-
mab was superior to certolizumab). Thus, the indirect evi-
dence suggests that the efficacy of the earlier-described
agents varies despite being in the same therapeutic class.
This may not be surprising because 2 previous anti-TNF
agents (etanercept and CDP571) were shown to be ineffi-
cacious in the treatment of Crohn’s disease.53,54
A previous network meta-analysis comparing anti-TNF
monotherapies in induction and maintenance of remission
showed superior efficacy of adalimumab compared with
certolizumab for induction; however, anti-TNF therapies
had similar efficacy for maintenance of remission.55 In
contrast, our analysis showed the superiority of adalimu-
mab vs certolizumab for both induction and maintenance.
The difference in findings likely is explained by the greater
number of trials included in our network meta-analysis,
which increases the certainty of indirect comparisons.
Our results contribute to the evolving understanding of
the role of combining immunosuppressants with anti-TNF
therapy.5 In 2010, the Study of Biologic and Immunomod-
ulator Naive Patients in Crohn’s Disease (SONIC) study
showed that infliximab þ azathioprine was superior to each
drug in isolation.22 In our analysis, the combination of
infliximab þ azathioprine was superior to certolizumab and
infliximab monotherapy, but was not different from adali-
mumab for maintenance of remission. Although studies
have not combined immunosuppressants with certolizu-
mab or adalimumab, this combination is presumed to be
beneficial owing to reduced immunogenicity.56 However, a
meta-analysis of placebo-controlled trials with individual
patient-level data showed that concomitant immunosup-
pressants with infliximab, but not adalimumab or certoli-
zumab, was associated with improved clinical remission
rates.57 Thus, replicating the SONIC study design with
 CLINICAL AT

350
Hazlewood et al
Table 3.Pair-Wise Comparisons for Total Withdrawals and Withdrawals Resulting From Adverse Events

Comparator, OR (95% CrI),a probability intervention superior to comparator (lower odds of withdrawal)

Azathioprine/ Infliximabþ
Intervention Placebo 6-mercaptopurine Methotrexate Certolizumab Infliximab Adalimumab Infliximabþ azathioprine methotrexate

Total withdrawals
Azathioprine/ 0.69 (0.41–1.1), 95% –
6-mercaptopurine
Methotrexate 0.96 (0.47–1.9), 55% 1.4 (0.62–3.3), 21% –
Certolizumab 0.85 (0.56–1.4), 76% 1.2 (0.67–2.6), 27% 0.89 (0.39–2.2), 61% –
Infliximab 0.71 (0.39–1.3), 88% 1.0 (0.55–2.1), 47% 0.75 (0.29–1.9), 74% 0.84 (0.37–1.7), 68% –
Adalimumab 0.43 (0.26–0.69), >99% 0.62 (0.32–1.2), 92% 0.45 (0.18–1.0), 97% 0.50 (0.24–0.96), 98% 0.60 (0.27–1.3), 91% –
Infliximab þ 0.27 (0.08–0.72), 99% 0.39 (0.14–0.98), 98% 0.29 (0.07–0.93), 98% 0.32 (0.09–0.94), 98% 0.39 (0.11–1.1), 96% 0.64 (0.18–1.9), 79% –
azathioprine
Infliximab þ 0.76 (0.20–2.9), 66% 1.1 (0.29–4.5), 44% 0.80 (0.17–3.6), 62% 0.90 (0.21–3.5), 56% 1.1 (0.33–3.5), 45% 1.8 (0.44–7.4), 20% 2.8 (0.59–16), 10% –
methotrexate
Vedolizumab 0.89 (0.51–1.6), 66% 1.3 (0.65–2.9), 24% 0.93 (0.38–2.3), 56% 1.0 (0.50–2.1), 45% 1.3 (0.57–2.9), 28% 2.1 (1.0–4.6), 2% 3.3 (1.1–12), 2% 1.2 (0.28–5.1), 41%
WDAEs
Azathioprine/ 3.9 (2.4–6.4), <1% –
6-mercaptopurine
Methotrexate 13 (3.2–109), <1% 3.3 (0.78–29), 6% –
Certolizumab 0.88 (0.64–1.2), 78% 0.23 (0.13–0.40), >99% 0.07 (0.01–0.28), >99% –
Infliximab 2.7 (1.6–4.7), <1% 0.71 (0.44–1.2), 92% 0.21 (0.02–0.93), 98% 3.1 (1.7–5.8), <1% –
Adalimumab 0.48 (0.31–0.74), >99% 0.12 (0.06–0.24), >99% 0.04 (0.00–0.16), >99% 0.55 (0.32–0.93), 99% 0.18 (0.09–0.34), >99% –
Infliximab þ 3.2 (1.6–6.1), <1% 0.81 (0.47–1.4), 77% 0.24 (0.03–1.1), 96% 3.6 (1.7–7.5), <1% 1.1 (0.64–2.0), 32% 6.5 (3.0–14), <1% –
azathioprine
Infliximab þ 6.6 (0.47–201), 8% 1.7 (0.12–52), 35% 0.49 (0.02–18), 66% 7.5 (0.53–226), 7% 2.4 (0.18–71), 25% 14 (0.95–411), 3% 2.1 (0.15–63), 29% –
methotrexate
Vedolizumab 0.68 (0.40–1.1), 93% 0.17 (0.09–0.35), >99% 0.05 (0.01–0.23), >99% 0.77 (0.41–1.4), 80% 0.24 (0.12–0.51), >99% 1.4 (0.72–2.8), 16% 0.21 (0.09–0.50), >99% 0.10 (0.00–1.5), 95%

Gastroenterology Vol. 148, No. 2

a
An OR less than 1 favors the intervention (row) over the comparator (column), indicating a lower odds of withdrawal.
February 2015
adalimumab and/or certolizumab would provide valuable
clinical direction.
With the recent approval of vedolizumab for induction
and maintenance therapy for Crohn’s disease, gastroenter-
ologists will need to decide the sequence of prescribing
vedolizumab relative to immunosuppressants and anti-TNF
therapies. In the network meta-analysis, vedolizumab was
significantly superior to placebo for induction and mainte-
nance of remission; however, the indirect comparison
showed that vedolizumab was not different from immuno-
suppressants or anti-TNF therapies. Thus, the position of
prescribing vedolizumab will need to balance its compara-
tive efficacy against its potential to reduce adverse events
through its gut-selective immunosuppressing properties.58
Crohn’s disease patients were significantly more likely to
have an adverse event leading to trial withdrawal with
azathioprine/6-mercaptopurine or methotrexate relative to
placebo, most anti-TNF monotherapies, and vedolizumab. In
contrast, we observed the fewest total withdrawals and
withdrawals secondary to adverse events with adalimumab
and infliximab þ azathioprine. Vedolizumab was associated
with fewer WDAEs when compared with infliximab and
infliximab þ azathioprine, but not compared with adali-
mumab and certolizumab. However, these data should be
reviewed cautiously because randomized controlled trials
have insufficient power to detect small but important
differences between drugs for rare adverse events.
Furthermore, because adverse events were rare we used a
fixed-effects model, which can underestimate the degree of
uncertainty. Finally, some withdrawals may have been sec-
ondary to Crohn’s disease flares rather than true adverse
events of treatment, thus reflecting a lack of efficacy rather
than increased toxicity.
In network meta-analyses, comparisons are made be-
tween differences of comparators (ie, odds ratios) and, thus,
the results may depend on the baseline placebo response.
This is important in Crohn’s disease clinical trials because of
the variability of placebo response that has been observed
between clinical trials.59 We conducted a meta-regression
to adjust for the baseline placebo response rate. The meta-
regression was inconsistent with our primary analysis
in some situations (eg, infliximab vs azathioprine/6-
mercaptopurine). However, the meta-regression should be
interpreted cautiously because our meta-regression included
few trials, and in any meta-regression (traditional or
network) the variable evaluated in the model (eg, baseline
placebo response) is not distributed randomly across the
trials and therefore is prone to confounding by other trial-
specific characteristics.
Our results should be interpreted in the context of lim-
itations associated with network meta-analyses. The paucity
of head-to-head trials and reliance on only indirect evidence
(eg, for infliximab þ methotrexate comparisons) resulted in
less precise credible intervals. Also, the assumption made
when pooling studies in a traditional meta-analysis is that
effect modifiers are balanced across trials. With a network
meta-analysis, the same assumption is made, but it is
extended across different treatment comparisons.60 Another
concern with any meta-analysis is heterogeneity across
Comparative Effectiveness of Therapies in CD 351
trials. The trials in our study differed in several aspects,
including the disease severity of the patient populations,
risk of bias, disease severity at the time of randomization,
prior exposure to anti-TNF therapy, and primary end points.
Although heterogeneity between clinical trials merits
cautious interpretation, sensitivity analyses only showed
marginal differences in the results. We also found consis-
tency between the treatment effects in the network
meta-analysis and those observed in a direct (traditional)
meta-analysis, when direct evidence was available, although
there were few closed loops in the evidence network
CLINICAL AT
(Figure 2).
The generalizability of our results is limited to the
eligible populations enrolled in the included trials. For
example, randomized controlled trials typically recruit
Crohn’s disease patients with moderate to severe disease
activity and, thus, our interpretations do not necessarily
relate to patients with mild Crohn’s disease or severe flares
requiring hospitalizations. Furthermore, treatment ap-
proaches for Crohn’s disease may differ based on age, dis-
ease location, disease behavior, prior surgery, and smoking,
which could not be explored using the data available in the
clinical trials.61 Similarly, we were not able to evaluate
randomized controlled trials that compared strategies of
treating Crohn’s disease such as comparing early use of anti-
TNF therapy (ie, top-down approach) with incremental use
of immunosuppressants and anti-TNF agents (step-up
approach).
By integrating direct and indirect evidence, this network
meta-analysis serves as a guide for clinicians making
complicated decisions on the medical management of
moderate-to-severe Crohn’s disease. Infliximab with and
without azathioprine, adalimumab, and vedolizumab were
effective at inducing remission when compared with pla-
cebo. Furthermore, azathioprine/6-mercaptopurine, meth-
otrexate, all of the anti-TNF therapies, and vedolizumab
were superior to placebo for maintenance of remission.
Azathioprine/6-mercaptopurine were inferior to anti-TNF
therapy in maintaining remission. Furthermore, our data
suggest that anti-TNF therapies are not equally efficacious
for induction or maintenance. Overall, this network meta-
analysis indicates that adalimumab and infliximab þ
azathioprine were the most efficacious strategies for in-
duction and maintenance of remission of Crohn’s disease.
These data call for randomized controlled head-to-head
trials between commonly prescribed medication regimens
for moderate-to-severe Crohn’s disease.
Supplementary Material
Note: To access the supplementary material accompanying
this article, visit the online version of Gastroenterology at
www.gastrojournal.org, and at http://dx.doi.org/10.1053/
j.gastro.2014.10.011.
References
1. Molodecky NA, Soon IS, Rabi DM, et al. Increasing
incidence and prevalence of the inflammatory bowel
 352 Hazlewood et al
diseases with time, based on systematic review.
Gastroenterology 2012;142:46–54.e42; quiz e30.
2. Kappelman MD, Rifas-Shiman SL, Porter CQ, et al. Direct
health care costs of Crohn’s disease and ulcerative co-
litis in US children and adults. Gastroenterology 2008;
135:1907–1913.
3. Frolkis AD, Dykeman J, Negron ME, et al. Risk of surgery
for inflammatory bowel diseases has decreased over
time: a systematic review and meta-analysis of
population-based studies. Gastroenterology 2013;145:
996–1006.
4. Feagan BG, Lemann M, Befrits R, et al. Recommenda-
CLINICAL AT
tions for the treatment of Crohn’s disease with tumor
necrosis factor antagonists: an expert consensus report.
Inflamm Bowel Dis 2012;18:152–160.
5. Melmed GY, Spiegel BM, Bressler B, et al. The appro-
priateness of concomitant immunomodulators with anti-
tumor necrosis factor agents for Crohn’s disease: one
size does not fit all. Clin Gastroenterol Hepatol 2010;
8:655–659.
6. Sandborn WJ, Feagan BG, Rutgeerts P, et al. Vedoli-
zumab as induction and maintenance therapy for
Crohn’s disease. N Engl J Med 2013;369:711–721.
7. Cheifetz AS, Melmed GY, Spiegel B, et al. Setting pri-
orities for comparative effectiveness research in inflam-
matory bowel disease: results of an international provider
survey, expert RAND panel, and patient focus groups.
Inflamm Bowel Dis 2012;18:2294–2300.
8. Institute of Medicine (IOM). Initial national priorities for
comparative effectiveness research. Washington, DC:
The National Academies Press, 2009.
9. Jansen JP, Fleurence R, Devine B, et al. Interpreting in-
direct treatment comparisons and network meta-analysis
for health-care decision making: report of the ISPOR Task
Force on Indirect Treatment Comparisons Good Research
Practices: part 1. Value Health 2011;14:417–428.
10. Dassopoulos T, Sultan S, Falck-Ytter YT, et al. American
Gastroenterological Association Institute technical
review on the use of thiopurines, methotrexate, and anti-
TNF-alpha biologic drugs for the induction and mainte-
nance of remission in inflammatory Crohn’s disease.
Gastroenterology 2013;145:1464–1478. e1–5.
11. Juillerat P, Wasan SK, Fowler SA, et al. Efficacy and
safety of natalizumab in Crohn’s disease patients treated
at 6 Boston academic hospitals. Inflamm Bowel Dis
2013;19:2457–2463.
12. Behm BW, Bickston SJ. Tumor necrosis factor-alpha
antibody for maintenance of remission in Crohn’s dis-
ease. Cochrane Database Syst Rev 2008;1:CD006893.
13. Chande N, Tsoulis DJ, MacDonald JK. Azathioprine or 6-
mercaptopurine for induction of remission in Crohn’s
disease. Cochrane Database Syst Rev 2013;
4:CD000545.
14. McDonald JW, Tsoulis DJ, Macdonald JK, et al. Metho-
trexate for induction of remission in refractory Crohn’s
disease. Cochrane Database Syst Rev 2012;12:
CD003459.
15. Patel V, Macdonald JK, McDonald JW, et al. Metho-
trexate for maintenance of remission in Crohn’s disease.
Cochrane Database Syst Rev 2009;4:CD006884.
Gastroenterology Vol. 148, No. 2
16. Prefontaine E, Macdonald JK, Sutherland LR. Azathio-
prine or 6-mercaptopurine for induction of remission in
Crohn’s disease. Cochrane Database Syst Rev 2010;
6:CD000545.
17. Spiegelhalter DJ, Abrams KR, Myles JP. Bayesian ap-
proaches to clinical trials and health-care evaluation.
Chichester, England: John Wiley & Sons Ltd, 2004.
18. Ades AE, Sculpher M, Sutton A, et al. Bayesian methods
for evidence synthesis in cost-effectiveness analysis.
Pharmacoeconomics 2006;24:1–19.
19. Ardizzone S, Bollani S, Manzionna G, et al. Comparison
between methotrexate and azathioprine in the treat-
ment of chronic active Crohn’s disease: a randomised,
investigator-blind study. Dig Liver Dis 2003;35:
619–627.
20. Arora S, Katkov W, Cooley J, et al. Methotrexate in
Crohn’s disease: results of a randomized, double-blind,
placebo-controlled trial. Hepatogastroenterology 1999;
46:1724–1729.
21. Candy S, Wright J, Gerber M, et al. A controlled double
blind study of azathioprine in the management of Crohn’s
disease. Gut 1995;37:674–678.
22. Colombel JF, Sandborn WJ, Reinisch W, et al. Inflix-
imab, azathioprine, or combination therapy for Crohn’s
disease. N Engl J Med 2010;362:1383–1395.
23. Colombel JF, Sandborn WJ, Rutgeerts P, et al. Adali-
mumab for maintenance of clinical response and remis-
sion in patients with Crohn’s disease: the CHARM trial.
Gastroenterology 2007;132:52–65.
24. Ewe K, Press AG, Singe CC, et al. Azathioprine com-
bined with prednisolone or monotherapy with predniso-
lone in active Crohn’s disease. Gastroenterology 1993;
105:367–372.
25. Feagan BG, Fedorak RN, Irvine EJ, et al. A comparison of
methotrexate with placebo for the maintenance of remis-
sion in Crohn’s disease. North American Crohn’s Study
Group Investigators. N Engl J Med 2000;342:1627–1632.
26. Feagan BG, Greenberg GR, Wild G, et al. Treatment of
active Crohn’s disease with MLN0002, a humanized
antibody to the alpha4beta7 integrin. Clin Gastroenterol
Hepatol 2008;6:1370–1377.
27. Feagan BG, McDonald JW, Panaccione R, et al. Meth-
otrexate in combination with infliximab is no more
effective than infliximab alone in patients with Crohn’s
disease. Gastroenterology 2014;146:681–688.e1.
28. Feagan BG, Rochon J, Fedorak RN, et al. Methotrexate
for the treatment of Crohn’s disease. The North American
Crohn’s Study Group Investigators. N Engl J Med 1995;
332:292–297.
29. Hanauer SB, Feagan BG, Lichtenstein GR, et al. Main-
tenance infliximab for Crohn’s disease: the ACCENT I
randomised trial. Lancet 2002;359:1541–1549.
30. Hanauer SB, Sandborn WJ, Rutgeerts P, et al. Human
anti-tumor necrosis factor monoclonal antibody (adali-
mumab) in Crohn’s disease: the CLASSIC-I trial.
Gastroenterology 2006;130:323–332.
31. Lemann M, Mary JY, Duclos B, et al. Infliximab plus
azathioprine for steroid-dependent Crohn’s disease pa-
tients: a randomized placebo-controlled trial. Gastroen-
terology 2006;130:1054–1061.
February 2015
32. Mantzaris G, P. P, A. K, et al. A prospective, randomized
trial of infliximab and azathioprine for the induction and
maintenance of remission of steroid-dependent Crohn’s
disease. Gastroenterology 2004;126:A54.
33. Mate-Jimenez J, Hermida C, Cantero-Perona J, et al. 6-
mercaptopurine or methotrexate added to prednisone
induces and maintains remission in steroid-dependent
inflammatory bowel disease. Eur J Gastroenterol Hep-
atol 2000;12:1227–1233.
34. Oren R, Moshkowitz M, Odes S, et al. Methotrexate in
chronic active Crohn’s disease: a double-blind, ran-
domized, Israeli multicenter trial. Am J Gastroenterol
1997;92:2203–2209.
35. Panes J, Lopez-Sanroman A, Bermejo F, et al. Early
azathioprine therapy is no more effective than placebo
for newly diagnosed Crohn’s disease. Gastroenterology
2013;145:766–774.e1.
36. Reinisch W, Panes J, Lemann M, et al. A multicenter, ran-
domized, double-blind trial of everolimus versus azathio-
prine and placebo to maintain steroid-induced remission in
patients with moderate-to-severe active Crohn’s disease.
Am J Gastroenterol 2008;103:2284–2292.
37. Rosenberg JL, Levin B, Wall AJ, et al. A controlled trial of
azathioprine in Crohn’s disease. Am J Dig Dis 1975;
20:721–726.
38. Rutgeerts P, D’Haens G, Targan S, et al. Efficacy and
safety of retreatment with anti-tumor necrosis factor
antibody (infliximab) to maintain remission in Crohn’s
disease. Gastroenterology 1999;117:761–769.
39. Rutgeerts P, Van Assche G, Sandborn WJ, et al. Adali-
mumab induces and maintains mucosal healing in pa-
tients with Crohn’s disease: data from the EXTEND trial.
Gastroenterology 2012;142:1102–1111.e2.
40. Sandborn WJ, Feagan BG, Stoinov S, et al. Certolizumab
pegol for the treatment of Crohn’s disease. N Engl J Med
2007;357:228–238.
41. Sandborn WJ, Hanauer SB, Rutgeerts P, et al. Adali-
mumab for maintenance treatment of Crohn’s disease:
results of the CLASSIC II trial. Gut 2007;56:1232–1239.
42. Sandborn WJ, Rutgeerts P, Enns R, et al. Adalimumab
induction therapy for Crohn disease previously treated
with infliximab: a randomized trial. Ann Intern Med 2007;
146:829–838.
43. Sandborn WJ, Schreiber S, Feagan BG, et al. Certolizu-
mab pegol for active Crohn’s disease: a placebo-
controlled, randomized trial. Clin Gastroenterol Hepatol
2011;9:670–678.e3.
44. Sands BE, Feagan BG, Rutgeerts P, et al. Effects of
vedolizumab induction therapy for patients with Crohn’s
disease in whom tumor necrosis factor antagonist
treatment had failed. Gastroenterology 2014;147:
618–627.e3.
45. Schreiber S, Khaliq-Kareemi M, Lawrance IC, et al.
Maintenance therapy with certolizumab pegol for
Crohn’s disease. N Engl J Med 2007;357:239–250.
46. Schreiber S, Rutgeerts P, Fedorak RN, et al.
A randomized, placebo-controlled trial of certolizumab
pegol (CDP870) for treatment of Crohn’s disease.
Gastroenterology 2005;129:807–818.
Comparative Effectiveness of Therapies in CD 353
47. Schroder O, Blumenstein I, Stein J. Combining inflix-
imab with methotrexate for the induction and mainte-
nance of remission in refractory Crohn’s disease: a
controlled pilot study. Eur J Gastroenterol Hepatol 2006;
18:11–16.
48. Targan SR, Hanauer SB, van Deventer SJ, et al. A short-
term study of chimeric monoclonal antibody cA2 to tu-
mor necrosis factor alpha for Crohn’s disease. Crohn’s
Disease cA2 Study Group. N Engl J Med 1997;
337:1029–1035.
49. Watanabe M, Hibi T, Lomax KG, et al. Adalimumab for
the induction and maintenance of clinical remission in
CLINICAL AT
Japanese patients with Crohn’s disease. J Crohns Colitis
2012;6:160–173.
50. Willoughby JM, Beckett J, Kumar PJ, et al. Controlled
trial of azathioprine in Crohn’s disease. Lancet 1971;
2:944–947.
51. Winship DH, Summers RW, Singleton JW, et al. National
Cooperative Crohn’s Disease Study: study design and
conduct of the study. Gastroenterology 1979;
77:829–842.
52. Winter TA, Wright J, Ghosh S, et al. Intravenous CDP870,
a PEGylated Fab’ fragment of a humanized antitumour
necrosis factor antibody, in patients with moderate-to-
severe Crohn’s disease: an exploratory study. Aliment
Pharmacol Ther 2004;20:1337–1346.
53. Sandborn WJ, Hanauer SB, Katz S, et al. Etanercept for
active Crohn’s disease: a randomized, double-blind,
placebo-controlled trial. Gastroenterology 2001;
121:1088–1094.
54. Sandborn WJ, Feagan BG, Radford-Smith G, et al.
CDP571, a humanised monoclonal antibody to tumour
necrosis factor alpha, for moderate to severe Crohn’s
disease: a randomised, double blind, placebo controlled
trial. Gut 2004;53:1485–1493.
55. Stidham RW, Lee TC, Higgins PD, et al. Systematic
review with network meta-analysis: the efficacy of anti-
TNF agents for the treatment of Crohn’s disease.
Aliment Pharmacol Ther 2014;39:1349–1362.
56. Moss AC, Brinks V, Carpenter JF. Review article:
immunogenicity of anti-TNF biologics in IBD - the role of
patient, product and prescriber factors. Aliment Phar-
macol Ther 2013;38:1188–1197.
57. Jones J, Kaplan G, Peyrin-Biroulet L, et al. Impact of
concomitant immunomodulator treatment on efficacy
and safety of anti-TNF therapy in Crohn’s disease: a
meta-analysis of placebo controlled trials with individual
patient-level data. United European Gastroenterology
Week 2013;1:A16.
58. Wyant T, Leach T, Sankoh S, et al. Vedolizumab affects
antibody responses to immunisation selectively in the
gastrointestinal tract: randomised controlled trial results.
Gut 2014 Epub ahead of print.
59. Su C, Lichtenstein GR, Krok K, et al. A meta-analysis of
the placebo rates of remission and response in clinical
trials of active Crohn’s disease. Gastroenterology 2004;
126:1257–1269.
60. Ades A. ISPOR states its position on network meta-
analysis. Value Health 2011;14:414–416.
 354 Hazlewood et al
61. Moran GW, Dubeau MF, Kaplan GG, et al. Phenotypic
features of Crohn’s disease associated with failure of
medical treatment. Clin Gastroenterol Hepatol 2014;12:
434–442.
Author names in bold designate shared co-first authors.
Received February 21, 2014. Accepted October 6, 2014.
Reprint requests
Address requests for reprints to: Gilaad G. Kaplan, MD, MPH, FRCPC,
Departments of Medicine and Community Health Sciences, University of
CLINICAL AT
Calgary, 3280 Hospital Drive NW, 6D56, Calgary, Alberta, T2N 4N1 Canada.
e-mail: ggkaplan@ucalgary.ca; fax: (403) 592-5090.
Conflicts of interest
These authors disclose the following: Gilaad Kaplan has served as a speaker
for Jansen, Merck, Schering-Plough, Abbott, and UCB Pharma, has
participated in advisory board meetings for Jansen, Abbott, Merck, Schering-
Plough, Shire, and UCB Pharma, and has received research support from
Merck, Abbott, and Shire; Remo Panaccione has served as a speaker, a
consultant, and an advisory board member for Abbott Laboratories, Merck,
Schering-Plough, Shire, Centocor, Elan Pharmaceuticals, and Procter and
Gamble, has served as a consultant and speaker for Astra Zeneca, has
served as a consultant and an advisory board member for Ferring and UCB,
has served as a consultant for Glaxo-Smith Kline and Bristol Meyers Squibb,
has served as a speaker for Byk Solvay, Axcan, Jansen, and Prometheus,
Gastroenterology Vol. 148, No. 2
has received research funding from Merck, Schering-Plough, Abbott
Laboratories, Elan Pharmaceuticals, Procter and Gamble, Bristol Meyers
Squibb, and Millennium Pharmaceuticals, and has received educational
support from Merck, Schering-Plough, Ferring, Axcan, and Jansen; Subrata
Ghosh has served as a speaker for Merck, Schering-Plough, Centocor,
Abbott, UCB Pharma, Pfizer, Ferring, and Procter and Gamble, has
participated in ad hoc advisory board meetings for Centocor, Abbott, Merck,
Schering-Plough, Proctor and Gamble, Shire, UCB Pharma, Pfizer, and
Millennium, and has received research funding from Procter and Gamble,
Merck, and Schering-Plough; Glen Hazlewood has received fellowship
funding from UCB Pharma and honoraria and travel expenses from UCB
Pharma and Abbott, and has participated in an advisory board meeting for
Amgen; Ali Rezaie has received honoraria from Ferring Pharmaceuticals;
Cynthia Seow has served as a speaker for Janssen, Merck, Schering-
Plough, and Warner Chilcott, and has participated in advisory board
meetings for Janssen, Abbott, Merck, and Schering-Plough; Corey Siegel
has served on advisory boards for AbbVie, Given Imaging, Janssen,
Prometheus, Salix, Takeda, and UCB, has received grant funding from
AbbVie, Janssen, Salix, and UCB, and has presented CME activities for
AbbVie, Merck, and Santarus; Gil Melmed has served as a speaker for
Abbott and Prometheus Labs, has participated in ad hoc advisory board
meetings for Luitpold, Janssen, Given Imaging, UCB, and AbbVie, and has
received research funding from Pfizer and Prometheus. The remaining
authors disclose no conflicts.
Funding
This research was supported by The Alberta IBD Consortium, which was
funded by an AHFMR Interdisciplinary Team Grant (AHFMR is now Alberta
Innovates–Health Solutions). The funders had no role in the study design,
data collection and analysis, decision to publish, or preparation of the
manuscript.
Supplementary Table 1.Characteristics of Included Trials

February 2015
Time of primary
Number of outcome Baseline disease Prior
Location; number randomized measurement, severity at the time of exposure
Study Trial type of centers Study arms patients Definition of remission wk randomization Concomitant therapy to anti-TNF

Adalimumab
CLASSIC-I,30 Induction North America and Adalimumab 40 mg/ 299 CDAI < 150 4 CDAI (220–450) Steroids, 33%; 0%
2006 Europe; 55 20 mg, 40 mg/80 immunosuppressants,
mg, or 80 mg/ 29%
160 mg sc at
weeks 0 and 2
Placebo
GAIN,42 2007 Induction North America and Adalimumab 160/80 325 CDAI < 150 4 CDAI (220–450) Steroids, 39%; 100%
Europe; 52 mg sc at weeks immunosuppressants,
0 and 2 49%
Placebo
Watanabe Induction Japan; 2 Adalimumab 160/80 90 CDAI < 150 4 CDAI (220–450) Steroids, 21%; 58%
et al,49 2012 mg or 80/40 mg immunosuppressants,
sc at weeks 32%
0 and 2
Placebo
EXTEND,39 2012 Treat-through (no North America and Adalimumab 40 mg 129 Mucosal healing 52 70 point decrease in Steroids, 26%; 52%
induction data) Europe; 19 sc EOW baseline CDAI immunosuppressants,
Placebo (220–450) after 41%
induction therapy
CHARM,23 2007 Maintenance North America, Adalimumab 40 mg/ 778 CDAI < 150 26 and 56 Reduction of at least Steroids, 44%; 50%
Europe, South wk sc, or 40 mg 70 points from the immunosuppressants,
Africa, sc EOW after baseline CDAI 47%
Australia; 92 induction (220–450) after
Placebo induction therapy
CLASSIC II,41,a Maintenance North America, Adalimumab 40 mg/ 55 CDAI < 150 56 CDAI < 150 after Steroids, 49%; 0%
2007 Europe; 53 wk sc, or 40 mg induction therapy immunosuppressants,

Comparative Effectiveness of Therapies in CD 354.e1

sc EOW after 22%
induction
Placebo
Watanabe Maintenance Japan; 2 Adalimumab 40 mg 43 CDAI < 150 52 70 point decrease in Steroids, 16%; 54%
et al,49 2012 sc EOW baseline CDAI immunosuppressants,
Placebo (220–450) after 4 36%
weeks of induction
therapy
Certolizumab
Sandborn Induction Multinational; 120 Certolizumab 400 439 CDAI < 150 6 CDAI (220–450) Steroids, 45%; 0%
et al,43 2011 mg sc at weeks immunosuppressants,
0, 2, and 4 33%
Placebo
Schreiber et al,46 Induction North America, Certolizumab 100, 292 CDAI < 150 12 CDAI (220–450) Steroids, 36%; 22%
2005 Europe, and 200, or 400 mg immunosuppressants,
South Africa; sc at weeks 0, 4, 37%
58 and 8
Placebo
Supplementary Table 1. Continued

354.e2
Time of primary
Number of outcome Baseline disease Prior

Hazlewood et al
Location; number randomized measurement, severity at the time of exposure
Study Trial type of centers Study arms patients Definition of remission wk randomization Concomitant therapy to anti-TNF

Winter et al,52 Induction Israel, Europe, and Certolizumab 5, 10, 92 CDAI < 150 4 CDAI (220–450) Steroids, 28%; 24%
2004 South Africa; or 20 mg/kg IV at immunosuppressants,
24 week 0 45%
Placebo
PRECISE 1,40 Treat-through (with Multinational; 171 Certolizumab 400 662 CDAI < 150 26 CDAI (220–450) Steroids, 39%; 28%
2007 induction data) mg sc at weeks immunosuppressants,
0, 2, and 4, then 37%
every 4 weeks
Placebo
PRECISE 2,45 Maintenance Multinational; 147 Certolizumab 400 428 CDAI < 150 26 Reduction of at least Steroids, 36%; 24%
2007 mg sc every 4 100 points from the immunosuppressants,
weeks after baseline CDAI 40%
induction (220–450) after
Placebo induction therapy
Infliximab
Targan et al,48 Induction North America and Infliximab 5, 10, or 108 CDAI < 150 4 and 12 CDAI (220–400) Steroids, 59%; 0%
1997 Europe; 18 20 mg/kg IV at immunosuppressants,
week 0 37%
Placebo
ACCENT I,29 Maintenance North America, Infliximab 5 or 10 335 CDAI < 150 30 and 54 Reduction of at least Steroids, 52%; 0%
2002 Europe, and mg/kg IV every 8 70 points and 25% immunosuppressants,
Israel; 55 weeks after from the baseline 27%
induction CDAI (220–400)
Placebo after induction
therapy
Rutgeerts Maintenance North America and Infliximab 10 mg/kg 73 CDAI < 150 44 Reduction of at least 0%
et al,38 1999 Europe; 17 every 8 weeks IV 70 points from the
after single-dose baseline CDAI
induction therapy (220–400) after
Placebo induction therapy
Vedolizumab
Feagan et al,26 Induction Canada; 21 Vedolizumab 0.5 or 2 185 CDAI < 150 8 CDAI (220–450) Steroids, 0%; 0%
2008 mg/kg IV at immunosuppressants, 0%
weeks 0 and 4
Placebo
GEMINI 2a,6 Induction Multinational; 285 Vedolizumab 300 mg 368 CDAI < 150 6 CDAI (220–450) Steroids, 34%; 62%

Gastroenterology Vol. 148, No. 2

2013 IV at weeks 0 immunosuppressants,
and 2 17%
Placebo
GEMINI 3,44 Induction Multinational; 107 Vedolizumab 300 mg 416 CDAI < 150 6 CDAI (220–450) Steroids, 54%; 76%
2014 IV at weeks 0, 2, immunosuppressants,
and 6 34%
Placebo
Supplementary Table 1. Continued

February 2015
Time of primary
Number of outcome Baseline disease Prior
Location; number randomized measurement, severity at the time of exposure
Study Trial type of centers Study arms patients Definition of remission wk randomization Concomitant therapy to anti-TNF

GEMINI 2b,6 Maintenance Multinational; 285 Vedolizumab 300 mg 461 CDAI < 150 52 CDAI < 150 after Steroids, 36%; 54%
2013 IV every 4 or induction therapy immunosuppressants,
every 8 weeks 17%
Placebo
Immunosuppressants
Ewe et al,24 Induction Germany; 1 Azathioprine 2.5 mg/ 42 CDAI < 150 and at 16 CDAI > 150 Steroid taper for all patients; 0%
1993 kg/day least a 60-point immunosuppressants
Placebo reduction from other than azathioprine not
baseline allowed
Feagan et al,28 Induction North America; 7 Intramuscular MTX 141 CDAI < 150 and 16 Steroid-dependent Immunosuppressants other 0%
1995 25 mg/wk steroid free Crohn’s disease for than MTX not allowed
Placebo at least 3 months
Ardizzone Treat-through (with Italy; 1 Azathioprine 2 mg/ 54 Steroid free and CDAI 24 CDAI  200 and Only tapering dose of steroids 0%
et al,19 2003 induction data) kg/day < 150 steroid dependent was allowed
MTX 25 mg/wk IV
first 3 months
then orally
Arora et al,20 Treat-through (with United States; 2 Oral MTX 15–22.5 33 Failure equaled no 52 CDAI > 150 and Immunosuppressants other 0%
1999 induction data) mg/wk improvement in steroid dependent than MTX not allowed
Placebo CDAI and no or CDAI < 150 on
reduction in >15 mg/day
steroids at 3 prednisone
months or
hospitalization
AZTEC,35 2013 Treat-through (no Spain; 31 Azathioprine 2.5 mg/ 131 Steroid free and CDAI 76 Diagnosed CD within 8 Not allowed except for steroids 0%
induction data) kg/day < 150 weeks of inclusion
Placebo

Comparative Effectiveness of Therapies in CD 354.e3

Candy et al,21 Treat-through (with South Africa; 1 Azathioprine 2.5 mg/ 63 CDAI < 150 60 CDAI > 200 Steroid taper for all patients 0%
1995 induction data) kg/day before maintenance
Placebo therapy;
Immunosuppressants, 0%
Mate-Jimenez Treat-through (no Spain; 1 6-MP 1.5 mg/kg/day 38 Failure to achieve 106 Steroid-dependent Only tapering dose of steroids 0%
et al,33 2000 induction data) MTX 15 mg/wk remission within 36 disease was allowed
Mesalamine 3 g/day weeks or flare
(CDAI>150)
NCCDS part I,51 Treat-through (with United States; 14 Azathioprine 2.5 mg/ 295 CDAI < 150 52 CDAI < 150 and no Not allowed (32% were being 0%
1979 induction data) kg/day progression of treated with steroids within
Placebo disease on barium 2 weeks of randomization)
Sulfasalazine 1 g/15 radiographs
kg/day
Prednisone 0.25–
0.75 mg/kg/day
Oren et al,34 Treat-through (with Israel; 12 Oral MTX 12.5 mg/wk 84 (35 entered HBI  3 and steroid- 36 HBI  7 Steroids, 77%; 0%
1997 induction data) 6-MP 50 mg/day clinical free immunosuppressants, 0%
Placebo remission)
Supplementary Table 1. Continued

354.e4
Time of primary
Number of outcome Baseline disease Prior

Hazlewood et al
Location; number randomized measurement, severity at the time of exposure
Study Trial type of centers Study arms patients Definition of remission wk randomization Concomitant therapy to anti-TNF

Reinisch et al,36 Treat-through (with Multinational; 38 Azathioprine 2.5 mg/ 138 Steroid free and CDAI 28 CDAI (220–450) All patients received tapering 0%
2008 induction data) kg/day < 150 does of steroids;
Placebo immunosuppressants were
Everolimus 6 mg/day not allowed
Rosenberg Treat-through (no United States; 1 Azathioprine 2 mg/kg 20 Composite clinical 26 Requiring at least 10 Steroids, 100%; 0%
et al,37 1975 induction data) Placebo score mg of prednisone immunosuppressants, 0%
for maintenance of
symptoms
Willoughby Treat-through (no United Kingdom; 1 Azathioprine 2 mg/ 12 Composite clinical/ 24 Steroid-dependent Steroids, 100% 0%
et al,50 1971 induction data) kg/day laboratory score remission
Placebo
Feagan et al,25 Maintenance Canada and the Intramuscular MTX 76 Steroid free and 40 CDAI < 150 and Steroids, 0%; 0%
2000 United States; 15 mg/wk increase in CDAI of steroid free after immunosuppressants, 0%
7 Placebo no more than 100 induction therapy
points with MTX
NCCDS part II,51 Maintenance United States; 14 Azathioprine 274 CDAI < 150 52 CDAI < 150 and no Not allowed (32% were being 0%
1979 1 mg/kg/day progression of treated with steroids within
(maximum dose disease on barium 2 weeks of randomization)
of 75 mg) radiographs
Placebo
Sulfasalazine
0.5 g/15 kg/day
Prednisone 0.25 mg/
kg/day
Willoughby Maintenance United Kingdom; 1 Azathioprine 10 Composite clinical/ 24 Steroid-dependent Steroids, 100% 0%
et al,50 1971 2 mg/kg/day laboratory score remission
Placebo
Combination therapy
Lemann et al,31 Induction France; 20 Infliximab 5 mg/kg IV 115 CDAI < 150 and 24 Patients with active All patients on tapering dose of 0%
2006 at weeks 0, 2, steroid free disease despite 6 steroids;
and 6 þ months of steroids immunosuppressants
azathioprine or 6- other than azathioprine/6-
MP MP were not allowed
Azathioprine (2–3
mg/kg) or 6-MP
(1–1.5 mg/kg) þ

Gastroenterology Vol. 148, No. 2

placebo
Mantzaris Induction Not reported Infliximab 5 mg/kg IV 47 CDAI < 150 and 6 CDAI > 150 and Not allowed 0%
et al,32 2004 at weeks 0, 2, steroid free steroid dependent
and 6
Infliximab þ
azathioprine
2.5 mg/kg/day
Supplementary Table 1. Continued

February 2015
Time of primary
Number of outcome Baseline disease Prior
Location; number randomized measurement, severity at the time of exposure
Study Trial type of centers Study arms patients Definition of remission wk randomization Concomitant therapy to anti-TNF

Schroder et al,47 Induction Germany; 1 Infliximab 5 mg/kg IV 19 CDAI < 150 12 Steroid or azathioprine Steroids, 79%; 0%
2006 at weeks 0 and 2 refractory disease immunosuppressants
plus other than MTX not
methotrexate 20 allowed
mg/wk for 48
weeks
Infliximab 5 mg/kg IV
at weeks 0 and 2
COMMIT,27 Treat-through (with Canada; 15 Infliximab 5 mg/kg IV 126 CDAI < 150 and 14 and 50 Patients who had Steroid was tapered by week 0%
2014 induction data) every 8 weeks steroid free initiated steroid 14; immunosuppressants
after induction þ induction in the other than MTX not
placebo preceding 6 weeks allowed
Infliximab 5 mg/kg IV irrespective of
every 8 weeks CDAI score
after induction þ
MTX 25 mg sc
weekly
SONIC,22 2012 Treat-through (with North America, Infliximab 5 mg/kg IV 508 CDAI < 150 26 CDAI (220–450) Steroids, 87%; 0%
induction data) Europe, and at weeks 0, 2, 6, immunosuppressants, 0%
Israel; 92 14, and 22
Azathioprine 2.5 mg/
kg/day
Combination therapy
(infliximab þ
azathioprine)

Comparative Effectiveness of Therapies in CD 354.e5

NOTE. Only patients from the CLASSIC I study who were in remission (CDAI < 150) were included in the study.
6-MP, 6-mercaptopurine; CDAI, Crohn’s Disease Activity Index; EOW, every other week; HBI, Harvey-Bradshaw Index; IV, intravenous; MTX, methotrexate; sc, subcu-
taneous; wk, week.
a
Classic II study.