Study reference

Study characteristics

Patient characteristics ²

Intervention (I)

Comparison / control (C) ³

Follow-up

Outcome measures and effect size ⁴

Comments

Diamanti 2009

Type of study:

Retrospective observational study, study period: 1999-2005

Setting:

Tertiary center for IBD treatment

Country:

Italy

Source of funding:
Unknown

Inclusion criteria:

< 18 years, with CD

Exclusion criteria:

 (1) those who did not take 5-aminosalicylic-acid compounds and immunomodulators simultaneously; (2)

those who had had steroid therapy during 3 months

preceding the beginning of the study; (3) those who had

only a single infliximab infusion; (4) those who tried new therapies during the study period; and (5) those with fever, abscess, or draining fistula during the last week preceding the study start-up.

N total at baseline:

Intervention: 14

Control: 14 (matched for gender and age)

Groups comparable at baseline:

No significant differences were

found between the two groups as regards clinical features or

previously administered therapies.

But comparison regarding PCDAI at baseline in the two groups is not shown.

Describe intervention (treatment/procedure/test):

Patients treated with infliximab, group A: patients received mesalazine (30 mg/kg/day) and azathioprine (2 mg/kg/day) and 5 mg/kg infliximab.

I.V. infliximab was given at week 0, week 4 and then every 8 weeks.

Indications for adding infliximab to conventional therapy were refractoriness to steroids in eight patients and dependency on steroids in the remaining six.

Describe control (treatment/procedure/test):

Patients treated with traditional therapy, group B: received mesalazine (30 mg/kg/day) and azathioprine (2 mg/kg/day).

Length of follow-up:

Median treatment duration was 10 months (5.5-28) in group A. No further information available on duration of follow up.

Loss-to-follow-up:

Two patients from group A were lost to follow-up because they were moved to an adult IBD care centre.

Incomplete outcome data:

Not described

Outcome measures and effect size (include 95%CI and p-value if available):

Primary outcomes: changes in linear growth and nutritional status (weight, BMI):

Weight (kg): 39.7 ± 13.1 at baseline versus 42.6 ± 13.2 final in group A (p <0.03). 42.1 ± 11.8 at baseline vs 44.1 ± 12.0 at final in group B (NS).

BMI (kg/m2): 17.9 ± 3.3 at baseline vs 18.9 ± 3.1 at final in group A (p<0.05). 18.7 ± 2.6 at baseline vs 19.5 ± 2.4 at final (NS).

Differences in height were non-significant in both treatment groups.

Secondary outcomes:

Median PCDAI (IQR): 23.5 (10-56) at baseline vs 10 (5-17.5) at final in group A.

17.5 (12.5-26.25) at baseline vs 20 (13.75-22.5) at final in group B.

No significant differences for Resting energy expenditure, basal metabolic rate, caloric intake and protein intake.

Aim of this study: to assess the effects of IFX as compared with those of standard therapy on growth and nutritional status.

Comments: No p-values are described for the comparison of group A and B. So no statistical testing on the treatment effect in group A vs group B.

Hyams 2017

Type of study:

Prospective observational study

Setting:

Multicenter 2007- june 2016

Country:

56 sites in North America and 26 sites in the EU

(UK, Belgium, The Netherlands, France, Germany,

Denmark, and Italy).

Source of funding:

Janssen Biotech, Inc. (manufacturer of IFX)

Inclusion criteria:

< 17 years of age in North America, and 6 years to less than 17 years of age in the EU.

Exclusion criteria:

Patients were ineligible to participate in the study if they had a history of other IBD-like conditions associated with genetic diseases (ie, glycogen storage disease

1b)

N total at baseline: total 5766

Intervention:

IFX group: 2396
Biologics group: 2824

Control: 2942 (=non-biologic)

Groups comparable at baseline:

Significant differences for age, median duration of disease, modified PCDAI an partial Mayo score.

Describe intervention (treatment/procedure/test):

IFX group (exposure to IFX as the only biologic in the presence or absence of exposure to non-biologics)

Biologic group (exposure to any biologic, including IFX)

Describe control (treatment/procedure/test):

Non-biologic (exposure to 5-aminosalicylates, corticosteroids, thiopurines, methotrexate, antibiotics or other non-biologic IBD therapies, in the absence of exposure to any biologics).

Length of follow-up:

Median follow up: 4.7 years

Loss-to-follow-up:

Attrition rate of 20.1% for the total study population

Incomplete outcome data:

Not described

Outcome measures and effect size (include 95%CI and p-value if available):

Primary outcomes:

8/15 malignancies was leukemia or lymphoma.

5 HLH events were observed.

Unadjusted incidence rates: no increased risk of malignancy (0.46/1000 patientyears) or HLH (0.0/1000 patient-years) in patients exposed to IFX as the only biologic vs those unexposed to biologics (malignancy: 1.12/1000 patient-years; HLH: 0.56/1000 patient-years).

SIRs : no increased risk of malignancy among patients exposed to IFX (SIR, 1.69; 95% CI, 0.46–4.32) vs patients not exposed to a biologic agent (SIR, 2.17; 95% CI, 0.59–5.56), even when patients were stratified by thiopurine exposure.

Secondary outcomes:

Aimed to compare unadjusted incidence rates of

malignancy and HLH in pediatric patients with IBD exposed to

infliximab (IFX) with patients not exposed to biologics and

calculated standardized incidence ratios (SIRs).

Kang

2016

Type of study:

Prospective observational study

Setting:

Medical center, jan 2011 –dec 201.

Country:

Korea

Source of funding:

Unknown

Inclusion criteria:

Paediatric patients with moderate-to-severe luminal CD of non-penetrating, non-stricturing behaviour.

Moderate-to-severe CD at diagnosis was defined as

a score of 30 points or greater according to the Pediatric Crohn’s

Disease Activity Index (PCDAI)

Exclusion criteria:

Patients of indeterminate-

type IBD, mild disease activity at diagnosis, penetrating

or stricturing disease behaviour, previous history of bowel surgery,

IFX indicated for the treatment of refractory perianal fistulas, and

age at IFX baseline of 18 years or over were excluded.

N total at baseline: 76

Intervention: 48

Control: 28

Important prognostic factors²:

See logistic regression in outcome measures.

Groups comparable at baseline

There were no significant differences between the two groups except for the duration from initial diagnosis to IFX infusion (median 8.1 months vs 0.7 months; p < 0.001).

Describe intervention (treatment/procedure/test):

Early combined immunosuppression group (group B);

Treatment was initiated within 1 month from diagnosis with intravenous IFX together with oral AZA and mesalazine.

Mesalazine was given concomitantly with AZA to improve treatment

respons and reduce thiopurine-related toxicity by increasing the levels of 6-thioguanine nucleotides (6-TGN).

Partial exclusive nutrition (PEN) was given both during CS induction in those who were allocated to the step-up group, and also during induction with IFX in groups A and B.

IFX was administered according to the scheduled

induction regimen of 5 mg/kg at weeks 0, 2, and 6, and no dose adjustments were allowed during this period. Scheduled IFX was repeated every 8 weeks starting from week 14.

Describe control (treatment/procedure/test):

Escalated combined immunosuppression group (group A);

Started with oral CS at a dose of 1 mg/kg/day (maximum 60 mg/kg/day) and was tapered over 8 weeks. Oral AZA and mesalazine was also started with oral CS. When treatment respons was refractory to or dependent on CS, IFX

was started with no changes in the treatment with AZA and mesalazine.

Partial exclusive nutrition (PEN) was given both during CS induction in those who were allocated to the step-up group, and also during induction with IFX in groups A and B.

IFX was administered according to the scheduled

induction regimen of 5 mg/kg at weeks 0, 2, and 6, and no dose adjustments were allowed during this period. Scheduled IFX was repeated every 8 weeks starting from week 14.

Length of follow-up:

Unclear

Loss-to-follow-up:

Intervention (group B):

N =3 due to primary non-respons (10.7%) at week 14.
Plus N=1 before week 54 due to secondary loss of respons.

Control (group A):

N =1 due to primary non respons (0.02%)

Incomplete outcome data:

See above, no further description.

Outcome measures and effect size (include 95%CI and p-value if available):

* Mucosal healing was defined as a SES-CD score of zero.

Primary outcome:

Mucosal healing at week 14 from baseline IFX: MH rates were higher in group B than in group A, but NS (32% versus 51%, p=0.121).

Logistic regression analysis: treatment group B was positively associated with MH at week 14 from baseline IFX (OR = 4.271, 95% CI = 1.249–14.605, p = 0.02), and ‘upper gastrointestinal tract involvement’ was negatively associated with MH at week 14 from baseline IFX (OR = 0.156, 95% CI = 0.041–0.589, p = 0.006).

Mucosal healing at week 54 from baseline IFX: MH rates were significantly higher in group B than in group A (42% vs 74%; p=0.007).

Logistic regression analysis: ‘group B’ and a ‘healed mucosa status at week 14 from baseline IFX’

were positively associated with MH at week 54, and ‘upper gastrointestinal tract involvement’ and ‘presence of perianal fistulas’ were negatively associated with MH at week 54.

Secondary outcome:
No significant differences between groups in clinical remission rates and laboratory remission rates.

Tertiary outcome:

Delta Z-scores of weight, height and BMI for age between baseline and week 54:

Weight: 0.75 ± 0.89 for group A and 0.82 ± 0.64 for group B (p=0.248).

Height: 0.10 ± 0.37 for group A and 0.11 ± 0.23 for group B (p = 0.653)

BMI: 0.89 ± 0.88 for group A and 0.96 ± 0.74 for group B (p = 0.381)

Aimed to compare the efficacy of combined immunosuppression: escalated combined immunosuppression vs early combined immunosuppression.

Kim

2011

Type of study:

Retrospective study

Setting:

Two centers in Korea, period: march 2001-august 2008

Country:

Korea

Source of funding:

Unknown

Inclusion criteria:

Pediatric patients with CD (based on Porto criteria).

Exclusion criteria:

Infectious disease such as tuberculosis.

N total at baseline: 29

Intervention: 18

Control: 11

Important prognostic factors²:

I: includes patients with moderate or severe CD.

C: includes eight patients with steroid-resistant CD and 3 with steroid dependent CD.

Number of patient with incurable fistulas:

I: n=12 (66.6%)

C: n=11 (54.5%)

Groups comparable at baseline

No significant differences, except for the duration from initial diagnosis to IFX infusion (0.8 ± 0.6 versus 11.5 ± 7.4 months; p= <0.001).

Describe intervention (treatment/procedure/test):

Top-down group:

Patients had not been treated previously with other medications such as corticosteroids or other immunomodulators.

Describe control (treatment/procedure/test):

Step-up group: oral corticosteroids (prednisolone, 1–2 mg⁄ kg ⁄ day), were used for induction therapy. Mesalamine

(Pentasa, 50–80 mg⁄ kg ⁄ day) or azathioprine (Imuran,

2–3 mg⁄ kg ⁄ day) was provided for maintenance therapy as the conventional treatment. Infliximab (5 mg⁄ kg) was administered by intravenous infusion at weeks zero, two and six, in combination with daily azathioprine, and this

course was repeated every 8 weeks for 10 months thereafter.

Length of follow-up:

At least 12 months

Loss-to-follow-up:

Intervention:

N (%)

Reasons (describe)

Control:

N (%)

Reasons (describe)

Incomplete outcome data:

Intervention:

N (%)

Reasons (describe)

Control:

N (%)

Reasons (describe)

Outcome measures and effect size (include 95%CI and p-value if available):

Disease remission = PCDAI less than 10 points. Moderate to severe disease is > 30 points.

Remission rate at week 8: 3/11 in step-up group versus 16/18 in top-down group (p=0.001).

Remission rate at 1 year: 5/11 in step-up group versus 15/18 in top-down group (p=0.048).

ESR significantly higher in Step-up group at 8 weeks: 51.5 ± 38.4 versus 17.1 ± 15.0; p=0.002, and also at 1 year: 44.9 ± 38.1 versus 21.3 ± 17.8; p=0.031

Albumin significantly lower in step-up group at 8 weeks 3.8 ± 0.4 versus 4.2 ± 0.3; p=0.019, and also at 1 year: 3.9 ± 0.4 4.3 ± 0.3 0.013.

Weight gain: Step up group 9.4 ± 7.1 vs 11.7 ± 6.0 in the top-down group; no significant difference.

Fistula closure assessment: 12/12 (100%) fistulas completely closed in top-down group compared to 3/6 (50%) in step-up group after one year. At 8 weeks this was 58.3% and 16.7% respectively.

Severe adverse events; Patients treated with mesalamine had three adverse events (two patients with anorexia and one with pancreatitis). Of patients treated with azathioprine, seven

were found to have adverse events (one with anorexia, one with pancreatitis and five with leukopenia). For patients treated with infliximab, one patient in the ‘step-up’ group had an adverse event: dyspnea and tachycardia following the third infusion.

Aim: to evaluate the efficacy of top-down versus step-up strategies.

Comments:

Kugathasan 2017

Type of study:

Prospective inception cohort study

Setting:

Multicenter, 28 sites. Nov 2008 – June 2012.

Country:

USA, Canada

Source of funding: Crohn’s and Colitis Foundation of America, Cincinnati Children’s Hospital Research Foundation Digestive

Health Center.

Inclusion criteria:

Newly diagnosed CD patients < 18 years.

Exclusion criteria:

IBD other than CD. Patient with incomplete information on disease location, patients who experienced complications either at or within 90 days at diagnosis, an those who did not attend at least one follow up visit.

N total at baseline:

Intervention: 191 (180 IFX, 11 adalimumab)

Control: 191 (from the same cohort, matched by propensity-score matching).

Important prognostic factors²:

Describe intervention (treatment/procedure/test):

Patients with early anti-TNFa exposure.

Early anti-TNFa exposure was defined as complication free initiation of therapy within 90 days of diagnosis and successful completion of both induction doses (three doses of infliximab and two doses of adalimumab) and at least one maintenance dose

Describe control (treatment/procedure/test):

Patient who did not receive early anti-TNF treatment, matched

Propensity of early anti-TNFα exposure was modeled by regressing the effects of baseline factors affecting anti-TNFα use (age, race, sex, disease location, perianal disease, height, Z-score, weight Z-score, PCDAI score at diagnosis, any deep ulcer on colonoscopy, and interaction between age and PCDAI) on early anti-TNFα use. Scores representing the propensity of early anti-TNFα exposure were computed and used to match each of the patients who received early anti-TNFα with a patient who did not.

Length of follow-up:

Not specifically mentioned for the total cohort or for the propensity matched groups. But median (IQR) for the following groups:

B1 (n=835); 47 (36-55)
B2 (n=54); 41 (36-50)
B3 (n=24); 40 (34- 49)

Loss-to-follow-up:

Not described

Incomplete outcome data:

Not described

Outcome measures and effect size (include 95%CI and p-value if available):

Progression to stricturing behaviour (B2): 14 / 191 in the early anti-TNFa group versus 12 /191 in the controls (p > 0.75)

Progression to penetrating behaviour (B3): 4 / 191 versus 11 / 191 (p=0.07)

Following competing risk model: Patients who received early anti-TNFα therapy were less likely to have penetrating complications (hazard ratio (HR) 0.30, 95% CI 0.10–0.89; p=0.0296) but not stricturing complication (1.13, 0.51–2.51; 0.76) than were those who did not receive early anti-TNFα therapy.

Aim: To derive a risk-stratification model based on clinical, host biology and microbial factors defined at diagnosis, and treatments including anti-TNF therapy.

Lee

2010

Type of study:

Retrospective chart analysis

Setting:

Single center (Samsung Medical Center), march 2001 – feb 2007

Country:

Korea

Source of funding:

Unknown

Inclusion criteria:

Newly diagnosed CD patients based on Porto criteria. Patients had not been treated with agents such as corticosteroids, mesalamine or IFX. Patients were followed for at least 24 months.

Exclusion criteria:

Infectious diseases such as tuberculosis.

N total at baseline: 36

Intervention: 13

Control: 23 (subdivided in 2 groups of n=10 and n=13)

Groups comparable at baseline

Only median age, duration of follow up and mean PCDAI were compared between the 3 groups, no significant differences.

Describe intervention (treatment/procedure/test):

Patients received infliximab and azathioprine for induction and maintenance therapy for the first year, and were treated with azathioprine after 1 year (group C, n=13).

IFX was administered by intravenous infusion at weeks 0, 2 and 6, in combination

with daily azathioprine and subsequently every 8 wk

for 10 mo. This group was not previously treated with other medications.

Describe control (treatment/procedure/test):

Patients were treated with mesalamine after induction therapy with prednisolon (group A, n=10) or were treated with azathioprine after induction therapy with prednisolon (group B, n=13).

Length of follow-up:

Mean: 51.9 months (range 24-101)

Loss-to-follow-up:

Four lost to follow up, not specified in which group.

Incomplete outcome data:

Three patients did not respond to induction regimen, 2 did not stop corticosteroid and one patient did not respond to three infusions of infliximab for induction.

Outcome measures and effect size (include 95%CI and p-value if available):

Opvlamming rate (opvlamming = PCDAI >10): The opvlamming rate at 1 year was 23.1% (95% confidence

interval (CI): 0.2%-46.0%) in group C (3 /13 patients), 61.5% (95% CI: 35.1%-88.0%) in group B (8 /13 patients) and 80.0% (95% CI: 55.2%-100.0%) in group A (8 / 10 patients). At 1 year, the absolute difference between groups C and B was 38.5% (95% CI: 3.5%-73.4%) and 56.9% (95% CI: 23.2%-90.7%) between groups C and A. Significant difference between each of the groups (P = 0.047 for B vs C, P = 0.012 for A vs C).

This remained significant at 2 years of follow up.

Adverse events:

Patients treated with mesalamine had three adverse events. Five patients treated with azathioprine were found to have adverse events. Only one patient had dyspnea and tachycardia after the third infusion of infliximab

Aim: to evaluate the efficacy of the early use of IFX, compared to conventional treatments, at 1 and 2 years of follow-up.

Periods in which patients were treated differed (group A mainly 2001-2003 and group C mainly 2005-2007)

Partly the same data as in Kim 2011 (adverse events are exactly the same)

YS Lee

2012

Type of study:

Retrospective chart analysis

Setting:

Single centre (Samsung Medical Centre), 1998-2009.

Country:

Korea

Source of funding:

Unknown

Inclusion criteria:

Patients diagnosed with CD according to Porto criteria.

Exclusion criteria:

Patients with infectious disease, such as tuberculosis.

N total at baseline: 28

Intervention: 10

Control: 18

Important prognostic factors²:

Median age

I: mean 10.2 ± 5.4

C: mean 14.4 ± 1.3

p-value: 0.010

PCDAI at prior 1^st IFX infusion:

I: mean 40.4 ± 10.2

C: mean 30.5 ± 14.9

p-value: 0.069

Groups comparable at baseline?

No, as mentioned above. Also, duration from the initial diagnosis to infliximab infusion is significantly different (p 0.001)

Describe intervention (treatment/procedure/test):

Step-up group: 10 patients

refractory to conventional therapy, including eight

with steroid-resistant CD and two with steroid dependent

CD who received infliximab.

They received infliximab and

azathioprine for induction and maintenance therapy

for the first year, and were treated with azathioprine

after 2 years.

Describe control (treatment/procedure/test):

Top-down group: Eighteen patients with moderate-to-severe CD were assigned to the ‘top-down’ group.

They had not been previously treated with other medication. Infliximab (5 mg/kg) was administered by intravenous infusion at weeks 0, 2, and 6 in combination with daily azathioprine, and this course was repeated every 8 weeks for 10-12 months thereafter.

Length of follow-up:

At least 36 months

Loss-to-follow-up:

Not described

Incomplete outcome data:

Not described

Outcome measures and effect size (include 95%CI and p-value if available):

Opvlamming rate at 1 year: 16.7% in

‘top-down’ group (3 /18 patients), 50% in

‘step-up’ group (5/10 patients). At 1-year, the

absolute difference between two groups was

33.3%. There was not a significant difference between the groups (p=0.091).

Opvlamming rate at 2-year follow-up: in ‘top-down’ group 50% (9 /18 patients) and 90% in the ‘step-up’ group (9/ 10 patients). At 2 years, the absolute difference between two groups was 40%. There was a significant difference between the groups (p=0.048).

Opvlamming rate at 3-year follow-up: 61.1% and 90% respectively. There was no significant difference (p=0.194)

Adverse events: Two patients treated with mesalamine had anorexia. Eleven patients treated with azathioprine experienced adverse events (leukopenia in nine and pancreatitis in two). For patients treated with infliximab, only one patient in the ‘step-up’ group had an adverse event, consisting of dyspnea and tachycardia that occurred after the third infusion.

Aim: To evaluate the efficacy of the top-down strategy compared to the step-up strategy at 1,2 and 3 years of follow-up.

Comments:

Again the same adverse events, so probably same cohort. This time also the same outcome measure.

Lee D

2015

Type of study:

Prospective observational cohort study

Setting:

multicenter

Country:

Canada, USA

Source of funding:

Inclusion criteria:

Children and young adults < 22 years of age. Needing treatment of active CD (PCDAI >10)

Exclusion criteria:

Presence of an ostomy, treatment with probiotics within 2 weeks of initiating EN, treatment with anti-TNF therapy within 8 weeks of starting EN, or treatment with EN within 1 week of initiating anti-TNF therapy

N total at baseline: 90

Intervention: n=52

Control: n= 16 PEN, 22 EEN

Important prognostic factors²:

All baseline variables with a p-value <0.25 were included in a disease risk score using a backward selection procedure. Adjustments for the disease risk score were made in the analysis.

Groups comparable at baseline?

Median disease duration was

shorter in the EN groups (P = 0.001; PEN 0.11 yr, EEN 0.03

yr, and anti-TNF 0.72 yr), and percentage male was also greater in the EN therapy groups. Baseline PCDAI was lowest in the anti-TNF group but similar in the PEN and EEN

groups (30.2 ± 12.0, 37.6 ± 19.3, and 38.8 ± 11.0, respectively; P = 0.01).

Describe intervention (treatment/procedure/test):

Patients receiving anti-TNFa (all infliximab, one adalimumab)

Describe control (treatment/procedure/test):

Patients receiving PEN or EEN.

Length of follow-up:

8 weeks

Loss-to-follow-up:

Intervention:

N=0 (0%)

Control:

N =4 (10%); one in PEN group, 3 in EEN group.

Three due to poor tolerance of therapy and 2 either because of loss of interest or the decision to stop EEN as therapy.

Incomplete outcome data:

Not described

Outcome measures and effect size (include 95%CI and p-value if available):

Fecal calprotectine at week 8 after start of therapy: The percentage of patients with FCP ≤50 mg/g was significantly higher in the anti-TNF group compared to the EEN group (30 versus 5%; p = 0.03). The proportion of subjects achieving FCP ≤50 mg/g, FCP ≤250 mg/g, and >50% reduction in FCP again displayed a trend toward increasing rates of success from PEN to EEN to anti-TNF (P = 0.01, P = 0.001, and P = 0.03, respectively)

Over 8 weeks, all 3 treatment groups had

a significant decrease in PCDAI and FCP (P , 0.05 for all comparisons). After adjusting for the disease risk score, a trend of increasing proportion of subjects achieving clinical respons was seen from PEN to EEN to anti-TNF (adjusted P = 0.08).

Weight Z-scores were significantly higher in the PEN group (0.76 ± 0.66 versus 0.38 ± 0.40 in anti-TNF group; p=0.04) and in the EEN group (0.70 ± 0.75 versus 0.38 ± 0.40 in anti-TNF group; p=0.03).

BMI Z-scores were significantly higher in the PEN group (1.24 ± 1.36 versus 0.48 ± 0.56 in anti-TNF group; p=0.07) and in the EEN group (1.13 ± 0.96 versus 0.48 ± 0.56 in anti-TNF group; p=0.002)

With additional adjustment for clinical remission status at week 8, systemic symptoms improved more with EEN than with anti-TNF

(P = 0.01);

Aim: CD. As part of a larger study on the

impact of EEN, PEN, and anti-TNF therapy on the composition

of the gut microbiota, clinical outcomes, mucosal

healing, and QOL were assessed for the different treatment groups.

Comments:

EEN was used at the 2 Canadian sites, and PEN with an unrestricted

diet was used in Philadelphia

Outcomes are adjusted for disease risk score.

Yoo Min Lee

2015,

Type of study:

Retrospective review of a prospective cohort

Setting:

Single center, sept 2008- may 2012 (Samsung Medical Center)

Country:

Korea

Source of funding:

Unknown

Inclusion criteria:

Paediatric patients diagnosesd as having moderate to severe CD (based on Porto criteria, PCDAI > 30), who received infliximab.

Exclusion criteria:

N total at baseline:

Intervention: 31

Control: 20

Groups comparable at baseline?

No significant difference between the 2 groups other

than the duration from the initial diagnosis to infliximab infusion (10.8 ± 9.0 months vs 1.0 ±0.5 months; P<0.0001)

Describe intervention (treatment/procedure/test):

Top-down (group A): early infliximab therapy. Patients received treatment with infliximab,

azathioprine, and mesalazine for induction and maintenance therapy

for the first year. Oral corticosteroids, azathioprine, and mesalazine were started after diagnosis, and oral corticosteroids were stopped

after a tapering period of 8 weeks.

Describe control (treatment/procedure/test):

Step-up (group B): patients receiving infliximab after proved to be refractory to conventional treatment. No changes in treatment with azathioprine and mesalazine were made.

Length of follow-up:

3 years

Loss-to-follow-up:

Not described

Incomplete outcome data:

Not described

Outcome measures and effect size (include 95%CI and p-value if available):

After 1 year of treatment of infliximab: 26 patients in group A (84%) and 9 patients in group B (45%) were in clinical remission. 25 patients in group A (81%) and 8 patients in group B (40%) had achieved deep remission (P=.0.0006) based on colonoscopy.

The opvlamming-free rates at 1, 2, and 3 years were 83.9% (95% CI 0.655–0.930), 58.1% (95% CI 0.390–0.731), and 35.5% (95% CI 0.194–0.519), respectively, in group A and 45.0% (95% CI

0.231–0.647), 20.0% (95% CI 0.062–0.393), and 15.0% (95% CI 0.037–0.335), respectively, in group B (P.0.0094).

The overall remission period in group A (33.2 ± 2.6 months) was longer than in group B (28.2 ± 6.0 months) (P=0.005).

Z-scores of weight-, height-, and body mass index for-age, and weight-for-height at the start of infliximab treatment and 3 years after infliximab initiation did not show statistical significance between the 2 groups. Delta z scores of weight-for height between the 2 periods, however, were significantly higher in group A compared with those in group B (1.190 ± 0.956 vs 0.424 ± 1.422; P=0.0424)

Among the adverse events, leukopenia was most frequently observed with 6 patients in group A and 3 patients in group B. Nausea and vomiting was observed in 4 patients in group A and 1

patient in group B. Pancreatitis was observed in 3 patients: 1 in group A and 2 in group B. Rash was observed in 1 patient in group A. There was no significant difference in adverse events between

the 2 groups (P=0.9718).

Aim: to compare the efficacy of remission maintenance

between infliximab ‘‘top-down’’ and ‘‘step-up’’ strategies in moderate to

severe pediatric Crohn disease during 3 years.

Comments:

At one year colonoscopy was performed and patients were allocated to treatment again based on the findings. Not described how they took this into account in the analysis.

Walters

2014

Type of study:

Retrospective analysis of Prospective, observational cohort study

Setting:

Data from RISK trial (Kugathasan 2017), 28 centers, nov 2008-jan 2012

Country:

USA, Canada

Source of funding:

See above

Inclusion criteria:

Children < 17 year of age, newly diagnosed with inflammatory CD.

Exclusion criteria:

Stricturing or penetrating disease.

Patients receiving both anti-TNFa and IM in the first 3 months (combination therapy; n = 12) were not included in this analysis.

N total at baseline:

Intervention: 68 (early anti TNFa)

Control: 248 (early IM mono), 236 (late therapy)
In propensity matched groups: n=68 in each group (total n=204)

Important prognostic factors²:

Patients receiving early anti-TNFa therapy tended to

have a greater likelihood of PCDAI greater than 30 at presentation (p=0.0047) and

a greater likelihood of deep ulcers at endoscopy (p=0.0018).

They were less likely to receive CS in the first 3 months (p=0.001).

Patients < 10 years of age were less likely to receive early

anti-TNFa therapy compared with other approaches (0.017).

Propensity matched controls were used (for every patient with early anti-TNF) to increase comparability.

Describe intervention (treatment/procedure/test):

Three groups:

First group: Early anti-TNFa monotherapy; defined as the treatment administered within 3 months of the initial diagnosis without the other, respectively.

Patients could receive various other treatments (CS, mesalamine, exclusive enteral nutrition) at the

discretion of their physician.

Describe control (treatment/procedure/test):

Second group: Early IM monotherapy; defined as the treatment administered within 3 months of the initial diagnosis without the other, respectively.

Third group: Late therapy (no early immunotherapy) was defined as therapy started between 3 and 12 months.

Patients could receive various other treatments (CS, mesalamine, exclusive enteral nutrition) at the

discretion of their physician.

Length of follow-up:

Minimum of 1 year

Loss-to-follow-up:

Not described

Incomplete outcome data:

Not described

Outcome measures and effect size (include 95%CI and p-value if available):

Primary outcome:

Looking at the whole cohort:

CS-free clinical remission (PCDAI ≤10) at 1 year after diagnosis: 58 of 68 (85%) patients treated with early anti- TNFa only, 152 of 248 (61%) patients treated with early IM only, and 129 of 236 (55%) patients not treated with either anti-TNFa or IM within the first 3 months (P < .0001).

When looking at the propensity matched groups:

The remission rate with early anti-TNFa therapy

(85%) was significantly greater than with either early IM (60%) or no early immunotherapy (54%) (P = 0.0003).

Remission at 1 year did not differ when comparing patients receiving early IM or no early immunotherapy (P = .49).

Relative risk for 3 treatment groups to achieve CS-free remission:

Anti-TNF only versus IM only: RR 1.41 (1.14-1.76), p-value 0.0017.

Anti-TNF vs no early immunotherapy: RR 1.157 (1.23-1.99) p-value 0.0002.

*Including late additional therapy as a covariate and accounting for its effect did not change the results.

Secondary outcome in propensity matched group by one year:

Height: Within the early anti-TNFa–exposed group, the mean height z-score increased by 0.14, whereas for the other 2 groups the height z-score remained unchanged (-0.02 in IM group and -0.06 in the late therapy group; P =0.039).

Weight: Within the early anti-TNFa–exposed group, the mean weight z-score increased by 0.85, which was a higher increase than in the other two groups (0.60 in IM group and 0.62 in the late therapy group; P =0.044).

BMI: Within the early anti-TNFa–exposed group, the mean BMI z-score increased by 1.11, which was a higher increase than in the other two groups (0.87 in IM group and 0.91 in the late therapy group; P =0.3).

Aims: to compare

the effectiveness of early (≤ 3 mo after diagnosis) treatment with an anti–tumor necrosis factor TNFa with that of an immunomodulator in attaining clinical remission and facilitating growth of pediatric patients.

Comments:

Study reference

Study characteristics

Patient characteristics ²

Intervention (I)

Comparison / control (C) ³

Follow-up

Outcome measures and effect size ⁴

Comments

Lee, 2016

Type of study:

Retrospective cohort study

Setting:

Data obtained from Truven Health MarketScan

Commercial Claims and Encounters (CCAE) databases containing health care claims for about 180 million people, 2009-2013

Country:

US

Source of funding:

Unknown

Inclusion criteria:

Children and young adults < 24 years of age, newly diagnosed with IBD. At least 6 months of enrolment after the date of diagnosis in the registry and patients had have received at least one IBD medication during the follow-up period.

Exclusion criteria:

Patients with a history of tuberculosis and/or medications for tuberculosis.

N total at baseline:

Intervention: 1298

Control: 2002

Important prognostic factors²:

Not assessed

Groups comparable at baseline?

Not assessble

Describe intervention (treatment/procedure/test):

“top-down”: was defined as TNFa inhibitor initiation within 30 days of the first medication prescription for IBD (i.e., 5-aminosalicylates, systemic corticosteroids,

thiopurines, and/or immunosuppressants)

Describe control (treatment/procedure/test):

The conventional “step-up” approach was defined as TNFa inhibitor initiation more than 30

days after the first IBD medication prescription.

Length of follow-up:

Not clearly described

Loss-to-follow-up:

Not described

Incomplete outcome data:

Not described

Outcome measures and effect size (include 95%CI and p-value if available):

Patients who received top-down treatment were less likely to use corticosteroids (32.5% versus 94.2%, P= 0.0001), 5-aminosalicylates (17.3% versus 75.1%, P = 0.0001), methotrexate (7.2% versus 14.6%, P = 0.0001), or thiopurines (13.5% versus 54.8%, P = 0.0001) compared

with step-up patients. No difference in switching between different types of TNFa (5,7% vs. 7.4%, P = 0.0581).

Aims:

Comments:

Not real clinically relevant outcomes for our research question

Study reference

Study characteristics

Patient characteristics ²

Intervention (I)

Comparison / control (C) ³

Follow-up

Outcome measures and effect size ⁴

Comments

Grossi
2015

Type of study:

Prospective observational cohort study

Setting:

Multi center

Country:

Not specified (the pediatric IBD registry by the IBD Collaborative Research Group)

Source of funding:

Unknown

Inclusion criteria:

Newly diagnosed Crohn’s disease < 16 years of age.

Exclusion criteria:

Patients who received treatment with another biologic agent or resectional surgery before initiating infliximab, as well as episodic infliximab therapy, were excluded.

Patients for whom infliximab was used primarily for

perirectal disease only were excluded.

N total at baseline: 502

Intervention: 338
(144 stopped IM within 6 months, 194 continued IM for > 6 months)

Control: 135

Groups comparable at baseline?

There were no significant clinical differences between

patients who did not receive concomitant IM, patients

with concomitant IM for 6 months or less, or patients

with concomitant IM for more than 6 months with the

exception that non-B1 behavior at infliximab initiation was 13%, 20%, and 8% in the 3 groups, respectively (P < .01).

Describe intervention (treatment/procedure/test):

Combination therapy: infliximab plus azathioprine or methotrexate, subdivided in :
- concomitant IM for ≤6 mo

- concomitant IM for > 6 mo

A total of 231 subjects received only thiopurines (126 male, 105 female), and 81 received only

methotrexate (60 male, 21 female). Twenty-six patients

were excluded from the IM-specific analysis because they

used both methotrexate and a thiopurine at different

times during their infliximab therapy.

Describe control (treatment/procedure/test):

Monotherapy: infliximab only.

Length of follow-up:

The median follow-up

period from diagnosis was 4.5 years, and was 2.75 years from infliximab initiation.

Loss-to-follow-up:

Not described

Incomplete outcome data:

Not described

Outcome measures and effect size (include 95%CI and p-value if available):

Kaplan–Meier analysis comparing the

concomitant IM use groups for all patients

showed that patients in the >6 m group were more likely to remain on infliximab compared with those in the >6 m or no-IM groups (P < .001 overall for males, and P < .01 for females).

The probability of remaining on infliximab at 5 years after infliximab start in patients

with more than 6 months of concomitant IM use was 0.70 ± 0.04 (0.73 ±0.05 for males, 0.65 ± 0.07 for females) compared with 0.48 ± 0.08 (0.41± 0.11 for males, 0.59± 0.08 for females) in patients with no IM and 0.55± 0.06 (0.59± 0.07 for males, 0.49± 0.09 for females) for those in the <6 month group (p <0.01). IM use before infliximab therapy had no effect on subsequent

durability independent of IM use during infliximab therapy. No change in results was noted when non-B1 patients were excluded from this analysis.

Cox proportional hazards analysis :

longer-term use (>6 mo) of any IM was associated significantly with increased durability of infliximab when compared with patients who did not receive concomitant

IM therapy (hazard ratio, 0.45; 95% CI, 0.30–0.69; P <0.001). The hazard ratio for thiopurines was 0.62 (95% CI, 0.40–0.97; P <0 .05), and for methotrexate was 0.09 (95% CI, 0.02–0.37; P < 0.001).

There was no significant difference between the 3 treatment groups in the probability of requiring escalation of therapy by either dose or frequency by 1 or 3 years. However, among those patients with dose or frequency

escalation there was a significant difference between treatment groups in the time (years) to the first escalation: no IM, 1.00 ± 1.17 years; IM <6 months, 0.99± 0.92 years; and IM >6 months, 1.41 ± 1.41 years (P <0 .05).

Safety: There was 1 malignancy and 2 deaths (1x following HLH & sesis, 1x cardiac arrhythmia considered unrelated to therapy in a patient with a history of prolonged QT interval) in the study population during the follow-up period. The frequency of infusion reactions was similar in all 3 IM groups.

Aims: To investigate

how immunomodulatory treatment affects the durability of infliximab therapy.

Kansen 2017

Type of study:

Retrospective observational study

Setting:

Multicenter (five tertiary paediatric university hospitals and one general hospital)

Country:

The Netherlands

Source of funding:
Unknown

Inclusion criteria:

Patients diagnosed with CD based on clinical, endoscopic, radiologic and histologic standards and had started infliximab (IFX; Remicade®) before the age of 18 years.

Exclusion criteria:

Patients were ineligible if they were diagnosed as IBD-unclassified or if follow-up was less than 3 months after initiation of infliximab treatment.

N total at baseline: 229

Intervention: 201 (115 on early combined immunosuppression (ECI), 86 on continuous combined immunosuppression (CCI))

Control: 28

ATI measurement in 162 (81 ECI, 62 CCI, 19 IFX mono)

Groups comparable at baseline?

* There were 21

patients treated with alternating combination and monotherapy and those patients were included in the analysis of group ECI.

*Comparison of baseline characteristics performed but results of statistical testing not described.

Describe intervention (treatment/procedure/test):

Patients in the CCI group continued IM during IFX treatment throughout the observation period.

In the ECI group IFX had been combined with IM for a median of 6.2 months (IQR 4-15).

118 patients received thiopurines (115 azathioprine, 3 mercatpopurine), 25 received methotrexate.

Describe control (treatment/procedure/test):

IFX monotherapy

Length of follow-up:

Median: 32 months (IQR 18.3 – 53.9)

Loss-to-follow-up:

Not described.

Incomplete outcome data:

Outcome measures and effect size (include 95%CI and p-value if available):

Primary endpoint:

Six out of 62 patients in the CCI group developed ATIs (10%), 11 of 81 in the ECI group (14%), and 8 of 19 in the IFX group (42%). Kaplan-Meier analysis comparing the incidence of ATI formation showed significantly higher probability of developing ATIs in patients in the IFX monotherapy group compared

with those in the CCI (P=0.003) and the ECI group (P=0.008), log-rank overall 0.002.

In the ECI group, all 11 positive ATI patients developed ATIs in the IFX monotherapy period as immunomodulatory therapy had been stopped at least 4 months before positive titres were measured.

Kaplan-Meier estimates of the probability of remaining ATI free at 12, 24, and 36 months after introduction of IFX in the CCI group were 93.4% (confidence interval (CI) 87.1–99.7), 91% (CI 83.4–98.6), and 91% (CI 83.4–98.6), respectively.
In the ECI group respectively 97.5% (CI 94–101), 93% (CI 87.1–98.9), and 85.6% (CI 76.8–94.4) and in the IFX group respectively 72.6% (CI 52–93.2), 57.7% (CI 33–82.4), and 48.1% (CI 21.3–

75).

The need for treatment escalation was not different between the groups, with respectively

40 out of 86 patients (46.5%) in CCI, 65 out of 115 (56.5%) in ECI, and 18 out of 28 (64.3%) in the IFX group.

Aim: The primary study end point was defined as the formation of positive antibodies (>12 AE/mL). Loss of respons was a secondary outcome parameter. Additional secondary end points included need for treatment escalation, time to both antibody formation and loss of respons and adverse events.

Kierkus

2015

Type of study:

Randomized open-label trial

Setting:

Multicenter (15 tertiary academic institutions)

Country:

Poland

Source of funding:
Unknown

Inclusion criteria:

7 to 17 year old patients.

Patients with moderate to severely active CD (defined by PCDAI > 30), and lacked or lost respons to previously given paharmacotherapy other than biological therapy.

Exclusion criteria:

Patients who had received any biological agent within 8 weeks before or during the trial.

N total at baseline: 99

Intervention: 45

Control: 39

Groups comparable at baseline?

No significant differences at baseline between groups. Also no significant difference at moment of randomization.

Describe intervention (treatment/procedure/test):

Group I: Patients in which IFX and immunomodulatory agent were continued up to week 54

*Induction: all patients (in both groups) received IFX (5 mg/kg) along with an immunomodulatory agent (azathioprine 1.5-3 mg/kg, methotrexate 10-25 mg/week).

Describe control (treatment/procedure/test):

Group 2: Patients in which immunomodulatory agent was discontinued after 26 weeks.

Length of follow-up:

54 weeks

Loss-to-follow-up:

Intervention:

N = 0

Control:

N = 0

Incomplete outcome data:

Intervention:

N=2 (4%) due to loss of clinical respons.

Control:

N=2 (5%) due to loss of clinical respons.

Outcome measures and effect size (include 95%CI and p-value if available):

Primary endpoint:

Clinical remission defined as PCDAI ≤ 10 points after one year of maintenance therapy.

Mean PCDAI in group I at 54 weeks was 8.48 (median 5, IQR 0-15) compared with 8.31 (median 5, IQR 0-15) in group II; non-significant.

Mean SES-CD at 54 weeks was 5.56 (median 1, IQR 0-9) in group I compared with 3.82 (median 3, IQR 0-6) in group 2; non-significant.

No significant differences were documented

between mean PCDAI and SES-CD scores determined at the end of induction (10th week) and maintenance (54th week) phases in both groups.

SAE’s: 5 SAE’s in group 2 )3x activation of EBV infection, 1x psoriatic lesions, 1x diarrhea.

Secondary endpoints:

Necessity of intensifying/modifying maintenance therapy, such as surgical treatment: surgery (4/45 in group I vs 1/39 in group II); NS

Increased dose of the immunomodulatory agent or IFX: 13 of 45 (29%) and 11 of 39 (28%) patients of groups I and II, respectively; NS

Steroid reimplementation: 3 of 45 (6.7%) and 3 of 39 (7.7%) patients of groups I and II, respectively

Aim: to compare the efficacy and safety of maintenance therapy with IFX and continuation of the immunomodulator versus withdrawal of the immunomodulator.

Comments: Four SAE’s in group I occurred during induction phase, those were withdrawn from the study.

Nuti

2016

Type of study:

Prospective observational cohort study

Setting:

Single center

Country:

Italy

Source of funding:

Unknown

Inclusion criteria:

Diagnosed with CD at 6-18 years of age, who started therapy with an anti-TNF agent between Jan 2009 and Oct 2012.

Exclusion criteria:

The need for immediate surgery, symptomatic stenosis or ileal or colonic strictures with prestenotic dilatation, and the presence of bacterial or viral infection.

N total at baseline: 37

Intervention: 23

Control: 14

Groups comparable at baseline?

Unclear; this was a subanalysis, no characteristics per group described.

Describe intervention (treatment/procedure/test):

Combination therapy: anti-TNFa plus azathioprine or methotrexate

The duration of such combinations was 8 ± 3.8 months (range 1–17 months).

Describe control (treatment/procedure/test):

Monotherapy: anti-TNFa only.

Length of follow-up:

9-12 months

Loss-to-follow-up:

Not described

Incomplete outcome data:

Not described

Outcome measures and effect size (include 95%CI and p-value if available):

In both groups SES-CD values decreased significantly at follow-up, but reduction was more significant in the combination therapy

group (p = 0.0002 vs p = 0.033).

No difference in the achievement of complete MH (p = 0.835). For complete plus partial MH as the outcome, the combination therapy group had significantly better results (p = 0.035)

The rates of complete MH were similar in the two groups, combination therapy vs monotherapy, (37.5 vs 33%) but the rate of no MH was considerably higher in the monotherapy one (18.75 vs 46.6%).

*complete MH = SES-CD of 0 or 1

The rates of clinical remission according to PCDAI values at follow-up did not differ statistically (combination, 74%; monotherapy, 64%).
*clinical remission was defined as the absence of symptoms related to CD and PCDAI ≤10.

The rate of non-respons at follow up was 13% in the combination group and 21% in the

monotherapy one (not significant).

Aims: To assess the efficacy of biologics in obtaining mucosal healing in paediatric CD. The secondary aims were: (1) to assess respons based on early or late treatment introduction and on combination therapy with IMM versus

biologics alone; and (2) to evaluate clinical outcome 2 years after the second endoscopy.

Comments: Limited data because of subgroupanalysis.

Study reference

Study characteristics

Patient characteristics ²

Intervention (I)

Comparison / control (C) ³

Follow-up

Outcome measures and effect size ⁴

Comments

Falaiye, 2014

Type of study:

Retrospective cohort

Setting: hostilatized patients

Country: U.S.

Source of funding: none

Inclusion criteria:

anti-TNF naïve patients

2–20 years with

refractory colitis-predominant IBD (UC, CD, or IBD-U),

initiated on infliximab

Exclusion criteria:

N total at baseline:

Intervention: 9

Control:20

Important prognostic factors²:

Not specified

Groups comparable at baseline? unknown

Describe intervention (treatment/procedure/test):

immunomodulator + IFX

Describe control (treatment/procedure/test):

IFX

Length of follow-up:

923 days (range 133-1952)

Loss-to-follow-up:

Not specified

Incomplete outcome data:

Not specified

Outcome measures and effect size (include 95%CI and p-value if available):

Dose escalation of IFX (increase IFX dose or reduction in the infusion interval from the standard (5 mg/kg at 0, 2, and 6 weeks induction;

maintanace 8-week): associated with monotherapy (combi 2/9 (22%) vs mono 16/20 (80%)) (OR 0.71 [95% CI 0.01,0.49], p=0.01)

IFX failure (cessation of infliximab for inefficacy or adverse reaction, or total

colectomy): not associated with monotherapy (combi 2/9 (22%) vs mono 16/20 (80%))

(combi 78% vs. mono 55%; OR 2.86 [95% CI 0.47,17.35], P = 0.41)

Geen onderscheid tussen CD 15 (52%), UC 12 (41%) en IBD-U 2 (7%): all patients had inflammation of the colon

McGinnis, 2008

Type of study:

Retrospective cohort

Setting: outpatient clinic

Country: U.S.

Source of funding: none

Inclusion criteria:

children with

steroid-resistant

and dependent

UC who

received IFX in a 36-month interval

Exclusion criteria:

N total at baseline:

Intervention: 20

Control:19

Important prognostic factors²:

Not specified

Groups comparable at baseline? unknown

Describe intervention (treatment/procedure/test):

IFX + Azathioprine or mercaptopurine

Describe control (treatment/procedure/test):

IFX

Length of follow-up:

Mean 18.5 months (range 1-36)

Loss-to-follow-up:

Four patients (10%) were lost to follow-up. Not specific intervention versus control.

Incomplete outcome data:

See loss to follow-up

Outcome measures and effect size (include 95%CI and p-value if available):

Respons (posttreatment

recategorization into Truelove and Witts ‘‘mild’’

category or better):

No difference in respons combi (15/20) vs. mono (12/19), p=0.50

Hyams, 2010

Type of study: prospective, multicenter, inception cohort study

Setting: hospitalized and outpatient

Country: U.S.

Source of funding: none

Inclusion criteria:

Children < 16 years with UC started on IFX

Exclusion criteria:

N total at baseline:

Intervention: 32

Control: 20

Important prognostic factors²:

Not specified

Groups comparable at baseline? unknown

Describe intervention (treatment/procedure/test):

Azathioprine or mercaptopurine +IFX (at start IFX, at 3, 6, 12 ,24 months)

Describe control (treatment/procedure/test):

IFX

Length of follow-up:

Median 30 months

Loss-to-follow-up:

Not specified

Incomplete outcome data:

Not specified

Outcome measures and effect size (include 95%CI and p-value if available):

Lower colectomy rate in combi therapy at:

- combi vs mono 3 months
13% vs 38%

- combi vs mono 6 months

25% vs 36%

- combi vs mono 12 months

26% vs 50%

- combi vs mono 24 months , 47% vs 78%

Geen p-waarde gegeven voor mono vs combi op 3, 6, 12, 24 maanden

Hyams, 2012

Type of study: prospective, multicenter, inception cohort study

Setting: hospitalized and outpatient

Country: U.S.

Source of funding:

Janssen Research & Development

Inclusion criteria:

Patients (6–17 years old) with active UC (Mayo scores of 6–12; endoscopic subscores 2) not

responded to or tolerated conventional treatment start on IFX

Exclusion criteria:

N total at baseline:

Intervention: 28

Control: 32

Important prognostic factors²:

Not specified

Groups comparable at baseline? unknown

Describe intervention (treatment/procedure/test):

IFX + AZA, 6-MP, MTX at baseline

Describe control (treatment/procedure/test):

IFX

Length of follow-up:

8 weeks for nonreponders, 52 week for responders

Loss-to-follow-up:

7 (30.4%) discontinued treatment because of an

adverse event, 3 (13.0%) for unsatisfactory therapeutic effect,and 1 (4.3%) for other reasons. Groups not specified.

Incomplete outcome data:

Not specified

Outcome measures and effect size (include 95%CI and p-value if available):

clinical respons: decrease in

the Mayo score by 30% and 3 points, with a decrease in the rectal bleeding subscore of 1 or a rectal bleeding subscore of 0 or 1.

clinical remission: as defined by a mayo score ≤2 with no individualsubscore >1

mucosal healing: Mayo endoscopy subscore of 0 or 1

Week 8 combi (n=32) vs mono (n=28): no differences
- clinical respons

 23 (71.9%) vs 21 (75.0%)
- Clinical remission
13 (40.6%) vs 11 (39.3%)

- mucosal healing

21 (65.6%) vs 20 (71.4%)

Week 52 combi (n=11) vs mono (n=10): no differences
5 (45.5%) vs 3 (30.0%)

Geen p-waarde gegeven voor mono vs combi

Dan-Nielsen, 2014

Type of study:

Retrospective cohort study

Setting: outpatient

Country: denmark

Source of funding: none

Inclusion criteria:

Children with UC < 15 years at diagnosis started on IFX

Exclusion criteria:

N total at baseline:

Intervention: 29

Control: 15

Important prognostic factors²:

Not specified

Groups comparable at baseline? unknown

Describe intervention (treatment/procedure/test):

IFX + AZA/6MP initiated before first infusion

Describe control (treatment/procedure/test):

IFX + AZA/6MP initiated directly after first infusion

Length of follow-up:

Loss-to-follow-up:

8 of 45 (18%) patients did not complete the induction treatment. In 6 of 8 cases owing to lack of respons: 5 of those were colectomized, and 1 was bridged to AZA/6MP by leukocyte apheresis

Incomplete outcome data:

Intervention:

N (%)

Reasons (describe)

Control:

N (%)

Reasons (describe)

Outcome measures and effect size (include 95%CI and p-value if available):

Clinical respons (not defined):

AZA/6MP initiated before versus after initiation of IFX treatment) was associated with clinical remission (P = 0.49

AZA/6-MP voor of direct na 1e infusie, geen exacte definitie van clinical respons, geen exacte cijfers/percentages van monovs combi

Study reference

(first author, year of publication)

Bias due to a non-representative or ill-defined sample of patients?¹

(unlikely/likely/unclear)

Bias due to insufficiently long, or incomplete follow-up, or differences in follow-up between treatment groups?²

(unlikely/likely/unclear)

Bias due to ill-defined or inadequately measured outcome ?³

(unlikely/likely/unclear)

Bias due to inadequate adjustment for all important prognostic factors?⁴

(unlikely/likely/unclear)

Diamanti, 2009

Likely (patients were treated with IFX for a reason, so likely to be a different type of patient)

Likely (information on duration of follow up is scarce, only 10 months treatment duration mentioned in group A, which might be too short for assessment of growth parameters (especially height)

Unlikely

Likely (PCDAI at baseline not compared between the two groups).

Hyams, 2017

Unlikely

Unlikely

Unlikely

Unlikely

Kang, 2016

Unlikely

Unclear

Unlikely

Unlikely

Kim, 2011

Likely (patients in the step-up therapy group had a longer disease duration and were resistant to other therapies already)

Unlikely

Unlikely

Likely (no p-values for differences in groups at baseline mentioned, no MVA performed).

Kugathasan, 2017

Unlike, derived from the same cohort in the same period.

Unlikely

Unlikely

Unlikely, patients were matched pairwise using propensity scores.

Lee, 2010

Likely (patients in the step-up therapy group had a longer disease duration and were resistant to other therapies already)

Unlikely

Unlikely

Likely (only a few baseline characteristics were checked, not disease location, perianal disease or not etc.)

Lee, 2012

Likely (patients in the step-up therapy group had a longer disease duration and were resistant to other therapies already)

Unlikely

Unlikely

Likely (significant differences at baseline, so confounders)

Dave Lee, 2015

Unlikely

Unlikely

Unlikely

Unlikely (adjustment for confounders was performed)

Yoo Min Lee, 2015

Likely (patients in the step-up therapy group had a longer disease duration and were resistant to other therapies already)

Unlikely

Unlikely

Likely (no differences at baseline, but not described how reallocation of treatment was taken into account.

Walters, 2014

Unlikely

Unlikely

Unlikely

Unlikely, possible confounders were taken into account. Also propensity matched groups were created.

Study reference

(first author, year of publication)

Bias due to a non-representative or ill-defined sample of patients?¹

(unlikely/likely/unclear)

Bias due to insufficiently long, or incomplete follow-up, or differences in follow-up between treatment groups?²

(unlikely/likely/unclear)

Bias due to ill-defined or inadequately measured outcome ?³

(unlikely/likely/unclear)

Bias due to inadequate adjustment for all important prognostic factors?⁴

(unlikely/likely/unclear)

Lee, 2016

Unlikely

unclear

unlikely

likely

Study reference

(first author, year of publication)

Bias due to a non-representative or ill-defined sample of patients?¹

(unlikely/likely/unclear)

Bias due to insufficiently long, or incomplete follow-up, or differences in follow-up between treatment groups?²

(unlikely/likely/unclear)

Bias due to ill-defined or inadequately measured outcome ?³

(unlikely/likely/unclear)

Bias due to inadequate adjustment for all important prognostic factors?⁴

(unlikely/likely/unclear)

Nuti, 2016

Unlikely

Unlikely

Unlikely

Likely (no test for comparison of baseline characteristics used)

Grossi, 2015

Unlikely

Unlikely

Unlikely

Unlikely

Kansen, 2017

Unclear

Unlikely

Unlikely

Unlikely

Study reference

(first author, year of publication)

Bias due to a non-representative or ill-defined sample of patients?¹

(unlikely/likely/unclear)

Bias due to insufficiently long, or incomplete follow-up, or differences in follow-up between treatment groups?²

(unlikely/likely/unclear)

Bias due to ill-defined or inadequately measured outcome ?³

(unlikely/likely/unclear)

Bias due to inadequate adjustment for all important prognostic factors?⁴

(unlikely/likely/unclear)

Hyams, 2010

likely

unclear

unlikely

likely

Study reference

(first author, publication year)

Describe method of randomisation¹

Bias due to inadequate concealment of allocation?²

(unlikely/likely/unclear)

Bias due to inadequate blinding of participants to treatment allocation?³

(unlikely/likely/unclear)

Bias due to inadequate blinding of care providers to treatment allocation?³

(unlikely/likely/unclear)

Bias due to inadequate blinding of outcome assessors to treatment allocation?³

(unlikely/likely/unclear)

Bias due to selective outcome reporting on basis of the results?⁴

(unlikely/likely/unclear)

Bias due to loss to follow-up?⁵

(unlikely/likely/unclear)

Bias due to violation of

intention to treat analysis?⁶

(unlikely/likely/unclear)

Kierkus, 2015

Open label randomization.

Patients with the clinical respons on week 10 were centrally randomized into 2 different maintenance protocols.

Likely (open-label)

Likely (for clinical respons outcome measures like PCDAI because open-label, SES-CD not influenced)

Likely (open-label)

Likely (open-label)

Unlikely

Unlikely (Reported n=0 lost to follow up in both groups)

Unlikely (ITT analysis was performed)