Antitumor necrosis factor treatment for pediatric inflammatory bowel disease


  • Charlotte I. de Bie MD,

    1. Department of Pediatric Gastroenterology, Erasmus MC-Sophia Children's Hospital, Rotterdam, the Netherlands
    Search for more papers by this author
  • Johanna C. Escher MD, PhD,

    1. Department of Pediatric Gastroenterology, Erasmus MC-Sophia Children's Hospital, Rotterdam, the Netherlands
    Search for more papers by this author
  • Lissy de Ridder MD, PhD

    Corresponding author
    1. Department of Pediatric Gastroenterology, Erasmus MC-Sophia Children's Hospital, Rotterdam, the Netherlands
    • Erasmus MC-Sophia Children's Hospital, Pediatric Gastroenterology, Department of Pediatrics, Dr. Molewaterplein 60, 3015 GJ, Rotterdam, the Netherlands
    Search for more papers by this author

  • Conflict of interest: the second and last author have participated in the REACH study by inclusion of patients (initiated by Centocor). The company did not play any role in the writing of this article.


Infliximab, adalimumab, and certolizumab are monoclonal antibodies against tumor necrosis factor-α (TNFα), a proinflammatory cytokine with an increased expression in the inflamed tissues of inflammatory bowel disease (IBD) patients. Currently, infliximab is the only anti-TNF drug that has been approved for use in refractory pediatric Crohn's disease (CD). Nevertheless, adalimumab and certolizumab have been used off-label to treat refractory pediatric IBD. Over the past 10 years, anti-TNF treatment has been of great benefit to many pediatric IBD patients, but their use is not without risks (infections, autoimmune diseases, malignancies). Despite the growing experience with these drugs in children with IBD, optimal treatment strategies still need to be determined. The purpose of this review is to summarize the current knowledge on the use of anti-TNF drugs in pediatric IBD and to discuss the yet-unsolved issues. (Inflamm Bowel Dis 2011;)

Inflammatory bowel disease (IBD) may present during childhood or adolescence in up to 20%–30% of all patients.1 There seems to be a worldwide trend toward increasing incidence rates of pediatric IBD.2 Unique to pediatric-onset disease is the occurrence of linear growth impairment and delay in puberty, which can be present at diagnosis and may be a sign of ongoing inflammation during the course of disease. As in adults, the phenotypic spectrum of IBD in children and adolescents is variable. Nevertheless, specific demographic and phenotypic differences characterize early versus later-onset disease.3, 4 For instance, Crohn's disease (CD) occurring prior to puberty affects a preponderance of males, whereas adult females are more commonly affected; most pediatric CD patients have ileocolonic disease, while adults more often present with isolated terminal ileal disease or isolated colonic disease; and ulcerative colitis (UC) presents with more extensive disease in children than in adults.

The treatment paradigm for pediatric IBD is quite similar to adult treatment, with induction and maintenance of remission as the main treatment goals. In children, special considerations in treatment are needed regarding optimal growth and development.4 In pediatric CD, both exclusive enteral nutrition (EEN) and corticosteroids are effective for induction of remission.5 However, EEN has significant advantages over steroids due to its beneficial effect on growth and fewer side effects. As in adult patients, thiopurines such as azathioprine (AZA) and 6-mercaptopurine (6-MP) are used for maintenance of remission, and are often introduced at the time of remission induction. Methotrexate (MTX) is an alternative to thiopurines when these drugs are ineffective or not tolerated. Pediatric UC is usually managed with aminosalicylates and corticosteroids, depending on disease extent and severity. Maintenance treatment consists of aminosalicylates or thiopurines for patients with relapsing disease.6

Over the last decade the introduction of antitumor necrosis factor (TNF) drugs has dramatically changed the treatment and management of IBD, especially in patients who are refractory to or intolerant of the conventional treatment regimens. Currently, there are three anti-TNF drugs that are licensed by the U.S. Food and Drug Administration (FDA) to treat CD in adult patients: infliximab (IFX), adalimumab (ADA), and certolizumab pegol. IFX is also registered for use in adult UC. In pediatric IBD, IFX is the only anti-TNF drug that has been approved by the FDA, and only for use in refractory CD.7 Nevertheless, all anti-TNF drugs have been used off-label to treat refractory pediatric IBD. In recent years, long-term follow-up data on IFX use in pediatric CD, increasing data on IFX use in pediatric UC, and on ADA use in pediatric CD have become available. This review will focus on the current evidence supporting the use of anti-TNF drugs in pediatric IBD patients. We will also address and discuss the yet-unsolved issues regarding anti-TNF treatment in pediatric IBD.


TNFα, a cytokine produced by activated macrophages, monocytes, and T cells,8 is a key mediator in immune responses, and has an increased expression in the mucosa of inflamed intestine.9, 10 This cytokine exerts its activity as either the cell surface-associated prepeptide (transmembrane TNFα), or as the cleavage product (soluble TNFα) generated by TNFα converting enzyme. Soluble TNFα mediates its biological activities through type 1 and 2 TNF receptors of remote tissues, while the biological activities of transmembrane TNFα are supposed to be mediated mainly through type 2 TNF receptors. Both receptors can bind intracellular adaptor proteins leading to the activation of a complex signaling cascade.11 This cascade is responsible for a wide range of cellular responses, including cell death, survival, differentiation, proliferation, and migration.8 Transmembrane TNFα also functions as a receptor that transmits outside-to-inside (reverse) signals back into TNFα-producing cells.11

Currently, three anti-TNF antibodies are approved by the FDA for the treatment of adult IBD. IFX (Remicade; Centocor, Malvern, PA) is a monoclonal chimeric anti-TNF antibody (75% human, 25% murine),12 ADA (Humira; Abbott Laboratories, Chicago, IL) is a fully humanized monoclonal anti-TNF antibody, and certolizumab (Cimzia; UCB, Brussels, Belgium) is a humanized monoclonal anti-TNF antibody Fab' fragment which is chemically linked to polyethylene glycol.13 All three anti-TNF agents can bind and neutralize soluble and transmembrane TNFα.14, 15 However, neutralization of TNFα is probably not the sole mechanism of action of anti-TNF treatment in IBD, as etanercept and onercept (both human soluble TNF receptors) were ineffective in inducing remission in CD patients.16, 17 Several working mechanisms of anti-TNF treatment have been proposed that are independent of their TNFα neutralizing capacity, such as induction of apoptosis in T cells and monocytes by binding of transmembrane TNFα,18–20 antibody-dependent cell-mediated cytotoxicity, and complement-dependent cytotoxicity.11, 14, 21 Certolizumab does not have these functional properties, suggesting that they are not essential for the efficacy of anti-TNF drugs.14 Recently, the importance of binding to Fc-receptors in the mechanism of action of anti-TNF antibodies was demonstrated by Vos et al.22 In vitro, IFX, ADA, and certolizumab-IgG induced formation of a distinct macrophage subset in an Fc-region-dependent manner. These macrophages had immunosuppressive capacities, including the production of antiinflammatory cytokines and inhibition of T cell proliferation.

Understanding the exact mechanisms of action of anti-TNF drugs and their relationship with clinical effects will contribute to the optimization of these treatments and to the development of new anti-TNF agents.


IFX as Induction Therapy

After the first use of IFX in 1992 in a 13-year-old girl with severe colonic CD,23 randomized clinical trials in adult CD patients have confirmed the clinical benefit of IFX induction therapy.24, 25 In children, mostly small, nonrandomized studies have reported on the outcome of IFX induction treatment.26–36 The results of these studies are summarized in Table 1. It is somewhat difficult to compare the results of these studies because of different infusion schedules, different definitions for response and remission, and variable timing of evaluating treatment response, as well as differences in concomitant immunomodulator therapy. Overall, these studies demonstrated that IFX was very effective in the majority of pediatric CD patients, but relapse of disease was common after discontinuation of IFX treatment (despite continuous use of thiopurines or MTX). Additionally, IFX induction treatment proved to be successful in achieving mucosal healing.28, 31 Results of pharmacokinetic assessments indicated that serum IFX concentrations in pediatric patients were similar to those in adults.28

Table 1. Overview of Studies on the Efficacy of Infliximab Induction Therapy in Pediatric Crohn's Disease
StudyNAge (yr)Concomitant ISInfliximabTreatment OutcomeDefinition Used for Treatment Outcome
  1. IS: immunosuppression at start of infliximab treatment (use of thiopurines or methotrexate). PCDAI: Pediatric Crohn's Disease Activity Index. CDAI: Crohn's Disease Activity Index. HBI: Harvey-Bradshaw Index. ESR: erythrocyte sedimentation rate.

Hyams et al2619Mean 14.474%1 to 3 infusions (5 mg/kg)Wk 12: significant decrease in PCDAI 
Kugathasan et al2715Mean 12.887%1 infusion (5 mg/kg)Wk 4: 93% responseResponse: PCDAI improvement of ≥ 25
     Wk 10: 67% remissionRemission: PCDAI ≤ 15
     Wk 52: 21% prolonged response 
Baldassano et al2821Median 1571%1 infusion (1, 5, or 10 mg/kg)100% response, 48% remission at some point during a 12-wk follow-upResponse: PCDAI improvement ≥ 10 or modified CDAI decrease ≥ 70 Remission: PCDAI < 10 or modified CDAI < 150.
Cezard et al2921Mean 1595%3 infusions (5 mg/kg on days 0, 15, and 45)Day 45: 90% remission Wk 52: 10% prolonged responseHBI ≤ 4
Lionetti et al3022Mean 1364%mean 3.3 infusions with an interval range from 2 – 12 wkWk 18: significant decrease in PCDAI 
Borrelli et al3118Median 13100%3 infusions (5 mg/kg on wk 0, 2, and 6)Wk 8: 56% remissionPCDAI ≤ 10
Lamireau et al3288Median 1482%median 4 infusionsDay 90 (± 7): 34% remission, 53% improvement in symptoms, 13% relapseRemission: HBI ≤ 4 and ESR ≤ 20 mm, or fistula closure
Afzal et al3324Median 13100%3 infusions (5 mg/kg on wk 0, 2, and 6)After induction: 71% remission, 82% relapse within 4 months of 3rd infusionHBI < 4
Hyams et al34112Mean 13.3100%3 infusions (5 mg/kg on wk 0, 2, and 6)Wk 10: 88% response, 59% remissionResponse: PCDAI improvement ≥ 15, with a total PCDAI score of 30 or less
      Remission: PCDAI ≤ 10
Wynands et al3538Mean 13.884%3 infusions (5 mg/kg on wk 0, 2, and 6)Wk 12: 95% remissionHBI < 5
Ruemmele et al3640Mean 13.9100%3 infusions (5 mg/kg on wk 0, 2, and 6)Wk 10: 85% remissionHBI < 5 and/or complete fistula closure

IFX for Fistulizing CD

Although the presence of fistulas (in combination with luminal disease) is often an indication to start IFX therapy, data on the efficacy of IFX in children with fistulizing CD are based on a small number of patients.29, 30, 36–38 Table 2 summarizes the results of these studies. Recently, Crandall et al38 performed post-hoc analyses on the effect of IFX on concurrent perianal disease in a subpopulation of 31 pediatric CD patients of the REACH study (28%). Twenty-two patients had perianal disease at baseline. Two weeks after a single IFX infusion, nine patients (41%) attained partial or complete response. At week 54, complete response was achieved in 15 patients (68%), and partial response in one patient (5%). Nine patients without perianal disease at baseline developed fistulas during IFX treatment: seven had complete response, and two had no response at week 54.

Table 2. Overview of Pediatric Studies on the Efficacy of Infliximab in Fistulizing Crohn's Disease
StudyNInfliximabEffect on Fistula
  • a

    Post-hoc analyses of the REACH study.34

  • b

    Partial response: initial perirectal subscore of 10 (active fistula, drainage, tenderness or abscess) decreasing to 5 (1–2 indolent fistula, scant drainage, no tenderness); Complete response: initial perirectal subscore of 5 or 10 decreasing to 0 (no symptoms or asymptomatic tags).

Cezard et al29123-dose induction days 0, 15, 45Day 90: 100% fistula closure
Lionetti et al3013mean 3.3 infusions, interval range 2 – 12 wksWk 18: 54% fistula closure, 23% partial response, 23% minimal response
Ruemmele et al36133-dose induction wk 0, 2, 6Wk 10: 69% complete fistula closure
De Ridder et al3716Various treatment schedules Mean 7 infusionsAfter a mean follow-up of 25 months: 56% fistula closure or drainage cessation
Crandall et ala38223-dose induction wk 0, 2, 6Wk 2: 41% partial or complete responseb
  Randomization wk 10: 
  infliximab every 8 or 12 wkWk 54: 73% partial or complete responseb

IFX for treatment of enterovesicular fistulas in children has been described in only two case series.39, 40 In total, eight children received a three-dose induction schedule with a variable outcome. Despite closure of the enterovesicular fistulas, surgery of severe underlying bowel disease was still required in two patients. In three patients IFX failed in closing the fistulas, and the remaining three patients had reduction of their symptoms.

Overall, the reported data suggest that IFX is effective in the treatment of children and adolescents with fistulizing CD.

IFX as Maintenance Therapy

Placebo-controlled trials have demonstrated that IFX is effective in maintaining remission in adult CD patients.25, 41 In pediatric CD, there are no placebo-controlled, double-blind prospective studies on IFX maintenance therapy. The largest randomized study was the REACH study: an industry-driven prospective, open-label investigation of a three-dose induction with IFX in moderate to severely active pediatric CD, followed by doses at either 8- or 12-week intervals for maintenance.34 At baseline, 112 patients on concomitant immunosuppression (IS) received an induction regimen of 5 mg/kg IFX at weeks 0, 2, and 6. At week 10, 103 patients were randomized to receive IFX maintenance treatment every 8 weeks (n = 52) or every 12 weeks (n = 51). At the endpoint visit of 54 weeks, 29 patients (56%) in the 2-month interval group were in clinical remission (PCDAI [Pediatric Crohn's Disease Activity Index] ≤10) and not requiring dose adjustments compared with 12 patients (24%) in the 3-month interval group (P < 0.001). Response to IFX was associated with improved quality of life. During the study, 32 patients (31%) needed dose adjustments due to loss of response: dosage increase to 10 mg/kg (n = 9), reduction in dose interval to 8 weeks (n = 10), or both adjustments (n = 13). These adjustments were successful in 75% (24/32) of patients. Allowing for dose intensification in case of relapse, remission rates at week 54 were also superior with every-8-week dosing compared with every 12-week dosing (71% vs. 47%, P = 0.02). In conclusion, the REACH study demonstrated that maintenance therapy every 8 weeks was superior to every 12 weeks in maintaining clinical response and remission in pediatric CD.

There is another randomized, open-label study (n = 40) on the efficacy of IFX as maintenance therapy for active pediatric CD, which demonstrated the superiority of scheduled IFX maintenance treatment over episodic IFX treatment.36 In addition, several observational studies, both prospective and retrospective, have been published on the efficacy of repeated use of IFX in pediatric CD.35, 42–49 The results of all randomized and observational studies are summarized in Table 3.

Table 3. Overview of Studies on the Efficacy of Repeated Infliximab Infusions in Pediatric Crohn's Disease
StudyNAge (yr)Concomitant ISInfliximabTreatment OutcomeDefinition Used for Treatment Outcome
  • a

    Expanded patient cohort of the study by de Ridder et al.45

  • b

    Open-label extension of the REACH study.34

  • IS: immunosuppression at start of infliximab treatment (use of thiopurines or methotrexate). IFX: infliximab. PCDAI: Pediatric Crohn's Disease Activity Index.

  • HBI: Harvey-Bradshaw Index. PGA: Physician's Global Assessment.

Hyams et al34112Mean 13.3100%3-dose induction wk 0, 2, 6 Randomization wk 10: IFX every 8 wk (group A) IFX every 12 wk (group B)Wk 54: 56% remission (group A), 24% remission (group B)PCDAI ≤ 10
Ruemmele et al3640Mean 13.9100%3-dose induction wk 0, 2, 6 Randomization wk 10: IFX every 8 wk (group A) IFX on demand (group B)Wk 60: 83% remission (group A), 61% remission (group B)HBI < 5 and/or complete fistula closure
Wewer et al4224Median 15.492%IFX on demand, scheduled IFX infusions Median 4 infusions90 days after intended cessation of IFX: 29% prolonged response, 42% IFX dependency, 25% no responseProlonged response: (total) regression of symptoms. IFX dependency: necessity of repeated infusions to maintain clinical response. No response: no regression of symptoms.
Duricova et al43828 – 1891%Induction only, IFX on demand, scheduled IFX infusions Median 7 infusionsAfter 3 to 75 mo follow-up: 22% prolonged response, 66% IFX dependency, 12% no responseSee study by Wewer et al.
Wynands et al3538Mean 13.884%3-dose induction wk 0, 2, 6 (group A) 3-dose induction wk 0, 2, 6 followed by IFX every 8 wk for 1 year (group B)Wk 52: 25% remission (group A), 58% remission (group B) Wk 104: 27% remission (after withdrawal of IFX in group B)HBI < 5
Sinitsky et al4416Mean 13.094%3-dose induction wk 0, 2, 6 followed by IFX every 8 wkYr 1: 83% clinical remission; 58% relapse at some time during 1st yrPCDAI < 15
de Ridder et al4566Mean 14.598%IFX on demand, scheduled IFX infusions Mean 15 infusionsAfter mean follow-up of 41 mo: 15% prolonged response, 56% IFX dependency, 29% loss of responseProlonged response: main tenance of good clinical response minimally 90 days after last IFX infusion. IFX dependency: necessity of repeated infusions to maintain clinical response. Loss of response: initially good clinical response, but finally withdrawal of IFX and switch of medical therapy or surgery.
de Bie et al.a46152Median 15.095%3-dose induction wk 0, 2, 6 (usually) followed by scheduled IFX infusions Median 10.5 infusionsAfter median follow-up of 25 mo: 10% prolonged response, 61% IFX dependency, 26% loss of response, 3% no response Yr 1: 87% continuing IFX Yr 3: 60% continuing IFX Yr 5: 50% continuing IFXSee study by de Ridder et al.
Hyams et al47202Mean 12.790%IFX on demand, scheduled IFX infusions Median 10 infusionsYr 1: 93% continuing IFX, 54% inactive disease Yr 2: 78% continuing IFX, 67% inactive disease Yr 3: 67% continuing IFX, 57% inactive diseasePGA=1, without requiring concomitant corticosteroids in the preceding quarter or surgery in the previous year
Crombé et al48120Median 1869%IFX on demand, scheduled IFX infusionsAfter median follow-up of 32 mo: 54% IFX efficacy Yr 1: 82% continuing IFX Yr 3: 55% continuing IFXRemission according to PGA
Hyams et alb4960Mean 13.2100%5 mg/kg every 8 or 12 wk, or 10 mg/kg every 8 wkModified intention-to-treat analysis: Yr 2: 63% no disease Yr 3: 62% no disease Yr 4: 50% no diseasePGA=1

Adjustments in treatment schedule (dose escalation, reduction in dose interval, or both) were frequently needed to maintain clinical remission, with rates varying from 27%–49%.35, 46–48 The median time to any dose adjustment was 6–9 months.46, 47 Intestinal surgery after initiation of IFX was performed in 25%–35% of pediatric CD patients.42, 45, 48 Crombé et al48 estimated that the cumulative risk of intestinal surgery in children after initiation of IFX was 14% at year 1, 24% at year 3, and 30% at year 5.

Taken together, IFX has proven to be an effective maintenance therapy for pediatric CD, but a substantial number of patients lose initial response and require dose adjustments to maintain clinical response. Current international treatment guidelines recommend IFX for induction and maintenance of remission in pediatric CD patients with moderate to severe disease, refractory to or intolerant of conventional treatment.5 Current clinical practice is to administer IFX infusions (5 mg/kg) at 0, 2, and 6 weeks (induction therapy). When this three-dose induction schedule has been effective, patients continue with scheduled maintenance infusions every 8 weeks. Dose escalation or reduction in dose interval may (temporarily) be required to prolong the duration of remission.

IFX and Growth Retardation

Growth retardation is a common complication of pediatric CD, and restoration of normal growth is considered a marker of therapeutic success.50 Chronic undernutrition and direct effects of proinflammatory cytokines (such as TNFα, interferon γ, IL-6) are the two major and interrelated factors responsible for growth impairment in children with CD.51

IFX treatment has been reported to improve both height velocity and height for age SD scores (SDS) in children with refractory CD, especially in patients who were treated prior to or in early puberty.52, 53 The REACH study and the open-label extension study also reported on the positive effect of IFX on linear growth.34, 49 Mean baseline height for age SDS of patients with at least 1 year delay in bone age improved significantly at both week 30 (mean improvement in SDS of 0.3) and week 54 (mean improvement in SDS of 0.5). Improvement in height status continued throughout the open-label extension, with median changes from baseline of the main study of 0.82 (n = 15) at the end of 2 years of IFX therapy, 1.01 (n = 10) at the end of year 3, and 1.56 (n = 4) at the end of year 4. Significant increases in mean height for age SDS or height velocity SDS were also observed in four other studies.29, 31, 36, 48 In contrast, three retrospective studies found no significant improvement of growth, but details on pubertal status were not available in these patients.42, 44, 54 Other discouraging data have been reported by Pfefferkorn et al55: despite frequent use of immunomodulators and IFX, growth delay persisted in many children with CD in the first 2 years following diagnosis.

IFX induction therapy can improve biomarkers of bone formation, as was demonstrated by Thayu et al.56 Serum bone-specific alkaline phosphatase (BSAP), N-terminal propeptide of type 1 collagen (P1NP), urine C-telopeptide of collagen cross-links (CTX-1), and deoxypyrodinoline increased significantly during induction, and were associated with increases in height for age SDS at 54 weeks.

In conclusion, the majority of data suggest that IFX treatment has a beneficial effect on growth in children with growth impairment due to CD, especially in the prepubertal and early-pubertal patients.


In adult patients with moderate or severe UC, placebo-controlled studies have proven that IFX can be effective in inducing and maintaining remission, and in preventing colectomy in the short term.57–60 In children with refractory UC, data on the efficacy of IFX treatment are limited to two prospective cohort studies,61, 62 and several small, retrospective case series.63–69 Turner et al61 described a prospective cohort of 128 children hospitalized for severe acute UC. Thirty-three of 37 patients failing intravenous corticosteroids were treated with IFX. Short-term response (Pediatric Ulcerative Colitis Activity Index [PUCAI] <35) was seen in 76% (25/33) of these patients, while 52% (13/25) had a sustained response during a 1-year follow-up period. Seven patients (28%) who initially responded to IFX and continued treatment after discharge underwent colectomy within 1 year.

In the second prospective study, 52 of 332 (16%) pediatric UC patients (mean age 13.3 years, 54% male) were treated with IFX after a median disease duration of 9 months.62 The indication to initiate IFX was steroid-refractory disease in 63% of patients and steroid-dependent disease in 35% of patients. Most patients (65%) received a three-dose induction schedule followed by scheduled maintenance treatment. Inactive disease (as assessed by the Physician's Global Assessment [PGA]) without use of corticosteroids was noted in 12/47 (26%) patients at 3 months, 12/44 (27%) at 6 months, 15/39 (38%) at 12 months, and 6/28 (21%) at 24 months. Dose adjustments were needed in 53% of patients on scheduled maintenance therapy. The likelihood of remaining colectomy-free after initiation of IFX treatment was 75% after 6 months, 72% after 12 months, and 61% after 24 months.

Retrospective case series have reported their experience with IFX treatment in 9 to 40 pediatric UC patients who were unresponsive to conventional treatment (i.e., steroid-dependent colitis, steroid-refractory colitis, acute severe colitis).63–69 Using various definitions, response rates varied between 55% and 77% after differing periods of follow-up (range: 2 weeks to 27 months). Colectomy rates of 29%–39% were reported.

A systematic review of the treatment of acute severe colitis in children pooled the results of six pediatric studies on the use of IFX in refractory severe UC.70 The pooled short-term IFX success rate (in most cases indicating discharge from the hospital without the need for colectomy) was 75% (95% confidence interval [CI]: 67%–83%), and the long-term success rate (avoidance of colectomy) was 64% (95% CI: 56%–72%).

In summary, IFX has a role in the management of children with moderate or severe UC, but appears to be less effective than in pediatric CD. In the short term, colectomy seems to be prevented in a significant proportion of children with UC, but long-term data are not available. Large multicenter trials are needed to determine the long-term benefit of IFX in different subgroups of pediatric UC patients.


ADA has been shown to induce and maintain response in adult CD patients who are naïve to anti-TNF therapy, intolerant of IFX, or have lost response to IFX.71–74 Recently, a randomized controlled trial demonstrated that ADA was also effective for induction of remission in adult UC patients who failed treatment with corticosteroids and/or IS.75 Although ADA is not (yet) licensed for use in pediatric IBD, this drug has been used off-label to treat children with refractory disease. Data on the efficacy of ADA treatment in pediatric CD are summarized in Table 4.76–85 Taken together, ADA treatment seems to be efficacious for inducing and maintaining remission in pediatric CD patients, but follow-up periods have been relatively short. Preliminary results suggest that anti-TNF naïve patients benefit most from ADA treatment. As with IFX maintenance treatment, a significant number of patients lose response to ADA over time and require dose adjustments to maintain response. The optimal dosing scheme for children with CD is yet to be determined, but high-dose ADA for maintenance treatment seems to be more effective than low-dose ADA.

Table 4. Overview of Studies on the Efficacy of Adalimumab Therapy in Pediatric Crohn's Disease
StudyNAge (yr)Prior IFXAdalimumabTreatment OutcomeDefinition Used for Treatment Outcome
  • In addition to the studies listed, there have been several case reports on the positive effect of adalimumab treatment in pediatric Crohn's disease.83–85

  • IFX: infliximab. eow: every other week. PGA: Physician's Global Assessment. PCDAI: Pediatric Crohn's Disease Activity Index. HBI: Harvey-Bradshaw Index.

  • a

    Inclusion of patients > 18 years.

  • b

    Five patients already in remission at start of adalimumab treatment.

Hyams et al7619264% ≥ 13 yr44%Induction: 160/80 mg (when ≥40 kg), 80/40 mg (when <40 kg) Maintenance, double-blind, randomization wk 4: Low-dose (per body weight) 20 or 10 mg eow High-dose (per body weight) 40 or 20 mg eowWk 4: 80% response (160/80 mg), 87% response (80/40 mg) Wk 26: 28% remission (low-dose), 39% remission (high-dose); higher remission rates in IFX naïve patientsResponse: PCDAI decrease ≥ 15 from baseline Remission: PCDAI ≤ 10
Rosh et al77115Mean 15.895%Induction: 160/80 mg (19%), 80/40 mg (44%), 40/40 mg (15%), other (8%), unknown (14%) Maintenance: 80 mg (5%), 40 mg (88%), 30 mg (2%), 20 mg (4%); 85% eow, 12% weekly3 mo: 21/66 (32%) remission 6 mo: 18/42 (43%) remission 12 mo: 16/33 (49%) remission Dose adjustment: 25%Inactive disease according to PGA
Russell et al7870Median 14.894%Induction: 160/80 mg (4%), 80/40 mg (59%), 24 mg/m2 (23%), other (14%) Maintenance: 80 mg (1%), 40 mg (91%), 24 mg/m2 (4%), other (4%); 97% eow, 3% weekly1 mo: 17/70 (24%) remission 6 mo: 32/55 (58%) remission 12 mo: 12/29 (41%) remission Dose adjustment: 35%PGA
Viola et al7923Median 16.1a61%160/80 mg and 80 mg eow (57%) 120/80 mg and 80 mg eow (9%) 80/40 mg and 40 mg eow (35%)Wk 12: 31% remission Wk 24: 50% remission Wk 48: 65% remission Dose adjustment: 48%PCDAI ≤ 10
Wyneski et al8015Median 17.9a100%160 mg and 80 mg eow (7%) 80 mg eow (7%) 80 mg and 40 mg eow (73%) 40 mg eow (7%) 40 mg and 20 mg eow (7%)After median follow-up of 6.5 mo: 50% complete response, 14% partial response, 36% no response Dose adjustment: 13%Complete response: completely weaned from steroids, or steroid- free interval > 3 mo Partial response: partially weaned from steroids, or some recurrence of symptoms No response: no weaning from steroids, or need for surgery
Rosenbach et al8114Median 13.9a71%First dose: 160 mg/1.73 m3. Second dose: 80 mg/1.73 m3. Maintenance: 40 mg/1.73 m3 eowAfter median follow-up of 17.3 mo: 50% complete response 50% partial response Dose adjustment: 57%Complete response: HBI < 4, weight recovery, improvement of inflammatory markers Partial response: HBI < 8
Noe et al827Median 16100%80 mg eow (14%), 80 mg and 40 mg eow (43%), 40 mg eow (43%)After 3 – 6 mo: 86%bPCDAI ≤ 10

The effect of ADA on refractory pediatric UC has only been described in a few patients:78, 82 three of four children with UC were successfully treated with ADA, with the nonresponder requiring surgery. Multicenter trials are needed to determine the benefit of ADA in pediatric UC.

Data on the use of certolizumab in IBD are only available for adult patients. Placebo-controlled randomized trials have demonstrated that certolizumab is effective in inducing and maintaining remission in adult patients with moderate to severe CD, whether or not they had previously been treated with IFX.86–88 Pediatric studies are currently under way to determine the safety and efficacy of certolizumab in children with CD.


Approximately 30% of children with IBD will develop at least one extraintestinal manifestation after diagnosis, such as musculoskeletal, dermatological, ophthalmologic, and/or hepatobiliary manifestations.89, 90 Anti-TNF treatment for children with extraintestinal symptoms has only been described in case reports and small case series. IFX seemed to be effective in cases of pyoderma gangrenosum, orofacial involvement, erythema nodosum, cutaneous metastatic CD of the penile and scrotal skin, primary lung involvement in IBD, primary sclerosing cholangitis in combination with pancreatitis, and clavicular osteomyelitis.91–95 Additionally, both IFX and ADA were effective for pediatric patients with peristomal pyoderma gangrenosum or uveitis.96–102

Evidence from adult studies also suggests that both IFX and ADA can be effective for the treatment of extraintestinal manifestations.103


Although anti-TNF therapy is effective for the majority of pediatric IBD patients, about 10% of patients do not benefit from induction therapy (primary nonresponders),34, 36, 79 while other patients eventually lose their initial therapeutic response.46, 47, 62, 78 This interindividual variation in response is likely caused by multiple host factors, such as disease and immune phenotype, and genetic background.

In children with CD, episodic IFX treatment has been associated with higher relapse rates compared with scheduled maintenance treatment.36 Other clear risk factors have not yet been identified, although some interesting data have emerged. Two pediatric studies have suggested that anti-TNF therapy was more effective when therapy was initiated early in the disease course,27, 30 but other studies failed to demonstrate this association in pediatric CD.29, 32, 46, 77, 78 Duricova et al43 reported that stricturing/penetrating disease behavior and intestinal surgery prior to IFX treatment were significantly associated with treatment failure. Additionally, retrospective studies have found that young male patients and patients on concomitant IS were more likely to respond to ADA.77, 78 A recent study by Dubinsky et al104 found six known susceptibility loci that were associated with primary IFX nonresponse in pediatric CD, as well as the presence of perinuclear antineutrophil cytoplasmic antibodies (pANCA). In pediatric UC, data on potential predictors of response are scarce. Fanjiang et al67 found that IFX was less effective in patients with chronic steroid-dependent UC than in acutely ill patients, but this association was not confirmed by other studies.62, 66, 68

For adult IBD patients, an overview on examined predictive factors for successful anti-TNF therapy has been published in a recent review.105 Factors possibly associated with a better efficacy in CD were: short disease history, younger age at diagnosis, objective evidence of inflammation at start of treatment (increased CRP and/or mucosal lesions at endoscopy), concomitant IS, isolated colonic disease, no previous abdominal surgery, no strictures, and not smoking. In adult UC, a pANCA+/ASCA− serotype and older age were associated with a suboptimal early clinical response to IFX.106

To conclude, in current clinical practice there is no clear guidance as to which type of IBD patient may benefit most from anti-TNF treatment. It is important to find reliable predictors of response to optimize treatment and to improve the benefit/risk profile of anti-TNF drugs.


Several pediatric studies have demonstrated short-term corticosteroid-sparing effects of IFX and ADA.26, 29, 31, 32, 34, 77, 79, 107, 108 IFX was also associated with prolonged corticosteroid withdrawal over a 3-year period.47 By 1, 2, and 3 years, less than 10% of patients continuing on IFX maintenance treatment were receiving corticosteroids.

In the first years after introduction of IFX, immunosuppressive therapy with thiopurines or MTX was usually continued during IFX maintenance therapy. The main rationale for this combined treatment was to improve short-term and long-term clinical outcomes by preventing the formation of antibodies to IFX (ATI, see IMMUNOGENICITY section), and achieving higher IFX serum levels.109, 110 However, this protective effect may not be present when patients are treated with scheduled maintenance infusions, as was demonstrated in adult studies.111, 112 In addition, results from a randomized trial in adult CD patients showed that withdrawing IS after 6 months of combination therapy did not affect clinical outcome during a 2-year follow-up, although median IFX trough levels were generally lower after withdrawal of IS.113 In patients naïve to thiopurines or MTX, the additional value of combination therapy may be different, as was shown in the SONIC trial: a large randomized double-blind trial in 508 adult CD patients who were not previously treated with immunosuppressive or biological therapy.114 Patients were randomly assigned to receive IFX and placebo, AZA and placebo, or both IFX and AZA. Steroid-free remission rates at week 26 were significantly higher in patients receiving both IFX and AZA than in patients receiving IFX monotherapy (57% vs. 44%, P = 0.02), or patients receiving AZA monotherapy (57% vs. 30%, P < 0.001). Similar trends were found at week 50. In addition, the total disappearance of mucosal ulcers was highest in the combined IFX and AZA group at week 26 (44% vs. 30% IFX mono vs. 17% AZA mono, P < 0.001).

The downside of combination therapy may be an increased risk of toxicity. With the reports on hepatosplenic T-cell lymphoma (HSTCL, see SAFETY section) in predominantly young male patients on combination therapy, many pediatric gastroenterologists converted to IFX monotherapy after a short duration of combination therapy, or combined therapy with MTX (instead of a thiopurine). In children with IBD, studies on the potential benefit of combination therapy are scarce, and more data on this important issue are therefore mandatory.


The great majority of IBD patients included in pediatric studies on IFX or ADA were treated according to a step-up strategy, indicating that conventional treatment had failed before anti-TNF therapy was initiated (steroid-dependency, steroid-resistance, intolerance or insufficient response to immunosuppressive therapy). However, it could be more effective to use anti-TNF drugs early in the disease course, as the early stages of immune-mediated disease may be more susceptible to immunomodulation.115, 116 Preliminary evidence from adult studies suggests that a top-down strategy may alter the natural history of IBD. As the accumulation of tissue damage is a key factor of IBD and often leads to strictures and/or fistulas, irreversible destruction of the digestive tract requiring surgery might be prevented. In 133 adult CD patients, an open-label randomized trial compared top-down treatment (three-dose IFX induction schedule and AZA) with step-up treatment (corticosteroids, and AZA in case of flaring or steroid-dependency).117 After 1 year, steroid-free remission rates were significantly higher in patients treated with early combined IS (62% vs. 42%, P = 0.03). In addition, median time to relapse was significant longer in the “top-down group.”

Experience with a top-down approach is limited in pediatric IBD. In two case reports, IFX was used as first-line therapy in a 14-year-old boy and 12-year-old girl with CD. An impressive clinical improvement was seen after two IFX infusions, and they were still in clinical remission on combination therapy of 6-MP and Asacol after 5 months, and combination therapy of AZA and IFX after 7 months, respectively.118, 119 Recently, a South Korean study retrospectively compared three treatment strategies in 36 newly diagnosed pediatric CD patients with a minimal follow-up of 2 years.120 Group A (n = 10) received induction treatment with oral prednisolone and mesalamine maintenance treatment; group B (n = 13) was treated with oral prednisolone and AZA; and group C (n = 13) received induction with IFX, followed by IFX and AZA maintenance treatment for 1 year, and AZA monotherapy after that year. At 1 year follow-up, there were significant differences in relapse rates between group A and C (80% vs. 23%, P = 0.012), and group B and C (62% vs. 23%, P = 0.047). Relapse was defined by a PCDAI score >10. After 2 years, relapse rates were 90%, 77%, and 39% in group A, B, and C, respectively. This study was limited by the small number of patients, retrospective data collection, and potential biased assessment of treatment efficacy.

Although these limited data support the view that early use of potent immunosuppressive therapy can change disease course, they need to be confirmed in large, prospective trials to determine the benefit/risk ratio of this approach, as there are concerns about potential dangerous long-term side effects of anti-TNF medication (see SAFETY section).


IFX therapy is associated with a risk of formation of ATI, which can partly be explained by the murine component of IFX. ADA treatment, being fully humanized, is associated with the same phenomenon, but to a lesser extent. The formation of antibodies to anti-TNF drugs may lead to acute infusion reactions (AIR), delayed hypersensitivity reactions, and decreased serum drug levels leading to a shorter duration of response.109, 110, 121 Concomitant use of IS may reduce the risk of antibody formation,110 whereas episodic treatment is associated with an increased risk of antibody formation.121

In three small pediatric studies, ATI were detected in about one-third of CD patients.35, 110, 121 In contrast, the REACH study observed ATI in only 3% of patients,34 which should be interpreted with caution, as most patients (77%) had inconclusive test results for ATI. Adult studies have reported ATI rates varying from 14%–61%.25, 109, 122

Data on formation of antibodies to ADA or certolizumab are only available from adult studies. Antibodies to ADA have been reported in 3%–17% of CD patients with refractory disease,72, 123, 124 and antibodies to certolizumab were present in about 9% of CD patients in the PRECISE studies.86, 87

The most common symptoms of AIR are shortness of breath, flushing, nausea, headache, hypoxemia, and tachycardia. Pooling of 18 pediatric studies showed AIR in 168 of 1100 IFX-treated patients (15%), and in 228 of 7137 infusions (3%).26, 28–30, 32, 34–36, 42, 45, 49, 107, 110, 121, 125–128 Most reactions were mild and responded rapidly to treatment, temporarily stopping the infusion, and/or reducing the flow rate of the infusion. Premedication (antihistamines, antipyretics, or corticosteroids) did not seem to prevent the development of AIR.125 In general, the rate of infusion reactions in children is similar to that in adults.25, 129 However, a small study observed a significant difference in the rate of severe systemic reactions between adult (11/52, 21%) and pediatric patients (1/34, 3%, P < 0.02).128

Delayed hypersensitivity reactions are usually defined as joint pain and swelling, associated with fever and/or rash, occurring more than 1 day postinfusion. These reactions occurred in 0%–8% of IFX-treated children with IBD,32, 34, 36, 42, 126, 127 which is similar to the frequencies that have been reported in adults.25, 129

Formation of autoimmune antibodies has been described in pediatric CD patients treated with IFX. Positive antinuclear antibodies (ANA), without any clinical symptoms, were detected in 20%–29% of patients.29, 34, 36, 49 The incidence of formation of antibodies to double-stranded DNA varied from 0%–10% of patients.28, 29, 34, 49 Development of a systemic lupus erythematosus-like syndrome is rare in the pediatric population, and has been described in case reports only.130, 131 In adults, ANA formation has been reported in 56% of IFX-treated CD patients132 and in 19% of ADA-treated CD patients.72 The clinical relevance of ANA induction by anti-TNF drugs is still unclear.

Another autoimmune disorder that has been described sporadically in pediatric IBD patients on IFX is vasculitis.32, 44, 133



Pooling of pediatric IBD studies shows serious or unusual infections in 49 of 1483 IFX-treated patients (3.3%): sepsis (n = 5), Listeria monocytogenes meningitis (n = 1), herpes zoster or varicella infections (n = 12), severe reactivation of Epstein–Barr virus (EBV, n = 1), pneumonia (n = 5), abscess (n = 14), cutaneous tinea infections (n = 3), opportunistic fungal infection (n = 1), osteomyelitis (n = 1), cellulitis (n = 1), Pseudomonas infection of a gastrostomy site (n = 1), pseudomembranous colitis (n = 1), gastroenteritis (n = 1), appendicitis (n = 1), and in one patient appendicitis and pancreatitis.26–37, 42–48, 61, 62, 64, 66–69, 107, 127 There was one sepsis-related death in an 11-year-old boy with severe refractory CD who was also treated with parenteral nutrition, corticosteroids, and AZA. The sepsis may have originated from an abscess located near a stenosis in the colon.37

In the relatively small pediatric studies on ADA use, serious or unusual infections were seen in 7 of 237 (3.0%) patients: sepsis (n = 2), a severe case of Clostridium difficile, and abscess formation (n = 4).77–82, 85 There were two sepsis-related deaths in patients who were also on IS and home parental nutrition (coagulase-negative staphylococcal central venous catheter sepsis complicated by an invasive pulmonary aspergillosis; E. coli and Candida central venous catheter sepsis).78

Several case reports have also reported on serious or unusual infections in pediatric patients treated with IFX or ADA: Listeria moncytogenes meningitis, EBV-associated hemophagocytic lymphohistiocytosis, opportunistic fungal skin infection with Pityrosporum folliculitis, histoplasmosis, flare-up of an intramyocardial inflammatory process, and a fatal case of disseminated cytomegalovirus.134–139

IBD patients have an increased risk of opportunistic infections (e.g., invasive fungal infections, reactivation of latent tuberculosis), especially patients on a combination of immunomodulator therapies and those with malnutrition.140 Testing for tuberculosis (chest radiograph, skin test of purified protein derivative tuberculin) prior to anti-TNF therapy is recommended.

Neutropenia, occurring in association with anti-TNF therapy, has been described in a few pediatric IBD patients.61, 78, 141


There are concerns that anti-TNF treatment may increase the likelihood of tumor development. Adult studies have yielded conflicting results. In contrast to clinical trial data with IFX and ADA in rheumatoid arthritis,142 several large IBD cohort studies did not find an increased risk of malignancy in patients treated with anti-TNF treatment compared with anti-TNF-naïve patients.129, 143 However, a meta-analysis on 8905 adult CD patients demonstrated that use of anti-TNF drugs with immunomodulators was associated with an increased risk of non-Hodgkin's lymphoma (NHL).144

One particular serious type of lymphoma, HSTCL, has been reported in IBD patients treated with anti-TNF therapy.145 HSTCL is a rare form of NHL and presents with hepatosplenomegaly (in the absence of peripheral lymphadenopathy) and general symptoms such as fever, weight loss, and fatigue. The tumor is extremely aggressive, and overall median survival with maximal therapy is less than 1 year.146 As of June 2008, 15 cases of HSTCL in IBD patients (13 CD, 2 UC) receiving combined IFX and thiopurine therapy have been confirmed.145 Two of these cases were diagnosed after the patients had switched from IFX to ADA. HSTCL occurred predominantly in male adolescents and young adults (age 12–39 years; 14M/1F) who received 1–24 IFX infusions. All but one patient had a fatal outcome. Recent data show that the number of reported HSTCL cases under combination therapy has increased to 20.147 In addition, there have been at least 16 cases reported among IBD patients treated only with AZA or 6-MP.145, 147 At present, it remains unclear whether the development of HSTCL is particularly associated with thiopurine monotherapy or combination therapy.

In children treated with anti-TNF drugs, 48 cases of malignancy were identified by the FDA as of April 2008 (31 following IFX use, two following ADA use, and 15 following etanercept use).148 In early onset IBD patients receiving IFX (n = 24), HSTCL accounted for most cases (n = 9), followed by NHL (n = 4), and Hodgkin's lymphoma (n = 2). The other cases included: leukemia, leiomyosarcoma, nephroblastoma, malignant melanoma, basal cell carcinoma, hepatic malignancy (undifferentiated), metastatic hepatocellular cancer, thyroid cancer, and colorectal cancer. All patients (16M/8F, age 4–22 years) received concomitant IS. One case of HTSCL was reported in a 20-year-old UC patient who was treated with ADA for 8 months. He had previously received IFX and 6-MP before switching to ADA. The reporting rate for malignancies and lymphomas in all children treated with IFX was higher when compared with background rates in the general U.S. pediatric population.

The recent pediatric literature has reported on four additional cases of malignancies in children treated with anti-TNF drugs. A 17-year-old female developed a bowel-associated Hodgkin's lymphoma ≈1.5 years after discontinuation of IFX.47 A second patient died of metastatic undifferentiated right colon carcinoma 7.5 years after discontinuation of IFX.48 Third, a case of basal cell carcinoma after 27 months of IFX therapy was reported.46 The fourth patient (13-year-old, IBD-unclassified) died of an EBV-positive natural killer T cell lymphoma with associated hemophagocytic lymphohistiocytosis after 1 year of IFX maintenance treatment.149

Large population-based studies and accurate registries are necessary to determine whether the development of malignancies is related to the severity and chronicity of the inflammatory disease, to the prolonged use of IS, or to a combination of both.


Anti-TNF treatment has been associated with rare cases of neurological disorders in adults, such as optic neuritis, seizures, and demyelinating disorders. In children, two cases of posterior reversible encephalopathy syndrome following IFX infusion have been described in a CD and UC patient.150, 151 In addition, a case of progressive multifocal leucoencephalopathy (PML) was described in a third patient.152 This 16-year-old CD patient developed a severe sepsis after the second IFX infusion. MRI examination after 3 months showed brain lesions related to PML. Six months after discontinuation of IFX, neurological symptoms and brain lesions had disappeared completely. Kachko et al153 reported on a 16-year-old CD patient who developed complex regional pain syndrome type I after his first IFX infusion. To our knowledge, no other neurological disorders have been reported in pediatric patients.

Anti-TNF therapy has been associated with adverse outcomes in adult IBD patients with congestive heart failure.154 Four pediatric studies have reported on the occurrence of cardiac symptoms during IFX treatment: one CD patient who suffered from a cardiac arrest secondary to cardiac arrhythmia associated with a long QT interval,47 one CD patient who developed cardiac insufficiency,48 one UC patient with a pericardial effusion with cardiomyopathy resolving after discontinuation of IFX,61 and one UC patient who developed familial cardiomyopathy.67 In addition, a pilot study found asymptomatic cardiac involvement, as assessed by Doppler echocardiography, in 7 of 12 pediatric IBD patients receiving IFX.155

Dermatological symptoms such as eczema, or psoriasiform lesions, are an emerging observation in pediatric IBD patients treated with anti-TNF drugs. IFX-induced psoriasis was observed in 8% (6/73) of pediatric IBD patients,156 whereas another study reported a wide variety of skin eruptions in 8% (12/152) of pediatric CD patients.46 In adults, skin eruptions occurred in 20% (150/734) of IBD patients.129 Most lesions responded well to topical steroids.

Psychiatric side effects of anti-TNF drugs are reported to be rare. Recently, the open-label extension of the REACH study reported on suicide attempts in two pediatric CD patients on IFX maintenance treatment (intentional overdose with acetaminophen and alprazolam).49 In adults, a case report described a 43-year-old UC patient who developed a depression with psychotic symptoms during IFX treatment and made a suicide attempt 4 months after initiation of the infusions.157 Another case of a suicide attempt during IFX treatment was reported in a 30-year-old CD patient who experienced panic attacks 2 hours after each infusion. After the fifth infusion, she intoxicated herself with paroxetine.158

In conclusion, severe adverse events of anti-TNF treatment are reported in low frequency, but they may be severe and even fatal. By FDA and EMA (European Medicines Agency) mandate, Centocor initiated in 2009 a prospective long-term (20 years) observational registry of 5000 pediatric IBD patients in Europe and North America. This registry intends to collect information on all serious adverse events associated with IFX, as well as other medical therapies for IBD.


IFX, ADA, and certolizumab are classified as category B agents by the FDA, indicating that there is no evidence of human toxicity in pregnancy. Limited data from two safety registries,159, 160 two prospective cohort studies,161, 162 and a retrospective case series163 suggest that the rates of miscarriage and neonatal complications in women exposed to IFX or ADA during pregnancy are not different from those in nonexposed women. To our knowledge, there are no published data on the effect of certolizumab on pregnancy outcome in CD patients.

Placental transfer of IFX and ADA is unlikely during the early stages of pregnancy, but these immunoglobulins easily pass the placental barrier in the second and third trimester.164 Case series have reported clinically significant IFX and ADA levels in the cord blood of the infant when these drugs were administered at the end of the second trimester or during the third trimester.165–167 In all 13 cases, IFX or ADA levels on the day of birth were higher in the child than in the mother, and remained detectable for 2–6 months. The effects of these high drug levels on the developing immune system of the child are unknown, but the results of two case series showed that children with high IFX levels did not seem to have an increased risk of infections in their first year of life, and that they had normal responses to inactivated vaccines.166, 168 However, a recent case report described a fatal case of a disseminated BCG infection in an infant vaccinated at 3 months of age.169 His mother was treated with IFX throughout the pregnancy. The recent consensus of the World Congress of Gastroenterology therefore recommended that live-virus vaccines should not be given to infants exposed in utero to IFX or ADA, unless serum levels are undetectable.170 Discontinuing IFX or ADA at the beginning of the second trimester (when possible) will probably prevent intrauterine and postnatal exposure,166 but further studies are needed to determine the exact timing of discontinuation of IFX or ADA during pregnancy.

Certolizumab, a Fab' fragment, is assumed to cross the placenta by passive diffusion, resulting in less placental transfer compared with IFX and ADA. This was confirmed in four infants from mothers who used certolizumab during pregnancy with the last dose 1–4 weeks before delivery.170, 171 Mothers' levels on the day of birth ranged from 4.9–59.6 μg/mL, and infant levels varied from 0.4–1.0 μg/mL.

Taken together, IFX or ADA treatment during pregnancy seems to be relatively safe, but these results are based on small patient numbers. The long-term effects of the intrauterine and postnatal exposure to anti-TNF drugs remain uncertain.


Over the past 10 years, IFX has been of great benefit to many pediatric CD patients. This biological drug has proven to be efficacious in inducing and maintaining remission, achieving mucosal healing, inducing perianal fistula closure, reducing corticosteroid exposure, promoting growth, and improving quality of life. IFX also has a role in the management of refractory pediatric UC, but appears to be less effective than in pediatric CD. ADA, although data are limited, seems to be effective for pediatric CD. The role of certolizumab in pediatric IBD is yet to be determined. Unfortunately, the use of anti-TNF therapies may not be without risks, such as opportunistic and serious infections, autoimmune diseases, and malignancies. To improve the benefit/risk profile of anti-TNF drugs, selecting patients with a high likelihood of response to these treatments is essential. However, reliable predictors of response are currently lacking. It is mandatory to continue the search for clinical parameters (such as genetic polymorphisms, serological markers, and cytokine profiles) that can identify those patients who are most likely to benefit from anti-TNF treatment.

Another problem is the secondary loss of response to IFX and ADA, which occurs in a substantial number of patients. Concomitant use of immunomodulators may improve the clinical outcomes, but this potential benefit needs to be weighed against a possibly increased risk of malignancies, especially HSTCL. A randomized controlled trial examining the efficacy and safety of anti-TNF drugs as monotherapy versus combination therapy in pediatric IBD patients is needed to answer this important question.

A third point of discussion is the optimal timing for introduction of anti-TNF treatment. Early introduction of anti-TNF drugs in the disease course may prevent complications and surgery in the future, but, again, the long-term benefit/risk ratio of this approach needs to be determined. Future research has to focus on genetic and serologic markers that can stratify newly diagnosed IBD patients according to their risk of developing active disease or complications. This will enable the selection of patient who will benefit most from early biological treatment, leading to individually tailored therapeutic management.

Other issues that need to be sorted out are the optimal dosing schemes for IFX and ADA, the need for discontinuing anti-TNF treatment in patients with a long-term response, the benefit of therapeutic drug monitoring by measuring serum drug levels and antibodies to IFX and ADA, and the need to achieve mucosal healing in every patient. Finally, the optimal choice between IFX and ADA in anti-TNF-naïve IBD patients may benefit a significant number of patients, and also reduce costs. A head-to-head comparison of the efficacy, safety, and cost-effectiveness of IFX and ADA in pediatric IBD is therefore essential.