Infliximab improves inflammation and anthropometric measures in pediatric Crohn's disease

Authors

  • Daniel M Sinitsky,

    1. Department of Gastroenterology, Sydney Children's Hospital, Sydney, Australia and
    Search for more papers by this author
  • Daniel A Lemberg,

    1. Department of Gastroenterology, Sydney Children's Hospital, Sydney, Australia and
    Search for more papers by this author
  • Steven T Leach,

    1. School of Women's and Children's Health, University of New South Wales, Sydney, Australia
    Search for more papers by this author
  • Timothy D Bohane,

    1. Department of Gastroenterology, Sydney Children's Hospital, Sydney, Australia and
    2. School of Women's and Children's Health, University of New South Wales, Sydney, Australia
    Search for more papers by this author
  • Reuben Jackson,

    1. Department of Gastroenterology, Sydney Children's Hospital, Sydney, Australia and
    Search for more papers by this author
  • Andrew S Day

    Corresponding author
    1. Department of Gastroenterology, Sydney Children's Hospital, Sydney, Australia and
    2. School of Women's and Children's Health, University of New South Wales, Sydney, Australia
      Associate Professor Andrew Day, Department of Gastroenterology, Sydney Children's Hospital, High Street, Randwick, NSW 2031, Australia. Email: andrew.day@unsw.edu.au
    Search for more papers by this author

Associate Professor Andrew Day, Department of Gastroenterology, Sydney Children's Hospital, High Street, Randwick, NSW 2031, Australia. Email: andrew.day@unsw.edu.au

Abstract

Background and Aim:  Infliximab (IFX) is a monoclonal antibody licensed to treat medically refractory Crohn's disease (CD). Our aim was to elucidate the effects of IFX therapy on clinical, growth and serum parameters in children with CD in a single pediatric center in Sydney, Australia.

Methods:  A retrospective case series review of children treated with IFX for CD at Sydney Children's Hospital, Australia was undertaken, with a review of outcomes after starting IFX. Main outcome measures were response and remission (as measured according to improvements in Pediatric Crohn's Disease Activity Index scores and Physician Global Assessment), laboratory markers (C-reactive protein, erythrocyte sedimentation rate, hemoglobin, white cell count, lymphocytes, neutrophils, platelets, albumin) and growth (Z scores).

Results:  The 16 patients included had a mean age at first infusion of 13.0 years (1.25–17.5 years). Six of 12 patients (with adequate data available) were in remission at 2 weeks following the first infusion. At 1 year, 10 of 12 patients (83%) were in remission. Mean C-reactive protein and erythrocyte sedimentation rate had fallen significantly (P < 0.05) at 2 weeks (from 29 to 7 mg/L and 40 to 19 mm/h, respectively). Positive trends were observed for all other parameters, excluding lymphocytes and white cell count. At 1 year, mean Z score for body mass index improved significantly from −0.9 to −0.1 (P < 0.01).

Conclusions:  Disease activity subsides in most children treated with IFX for CD. IFX therapy also improves some growth parameters. The pattern of improvement requires further elucidation, as the results in the present study suggest differing dosing frequency of infusion may achieve better efficacy.

Introduction

Crohn's disease (CD) causes chronic relapsing and remitting inflammation of the digestive tract. Although the peak incidence of CD is in the second and third decade of life, symptoms can begin at any age. In children, CD commonly affects growth, which can result in long-term effects, such as impaired acquisition of adult height, and its social consequences.

Growth and normal development of children with CD may be further affected by the treatments currently used to induce and maintain remission. For instance, corticosteroids, conventionally used for remission induction, can impact adversely upon linear growth1 as well as other short- and long-term adverse effects, including facial appearance, mood changes and osteoporosis.2–4 These side-effects often contribute to diminished quality of life and self-esteem.5 Furthermore, patients often develop corticosteroid dependence (36%) or resistance (20%).6 In addition, corticosteroids are not proven to have any role in maintenance of remission.7 Current proven non-biological treatments available for the maintenance of remission of CD include 6-mercaptopurine, azathioprine and methotrexate. These immunomodulators have slow onset of action, but are also associated with adverse events, such as pancreatitis, hepatocellular damage and bone marrow suppression.8

Over the last decade, infliximab (IFX), an anti-tumor necrosis factor (TNF)-α chimeric monoclonal antibody, has become a further therapeutic option for the induction and maintenance of remission in children with severe CD. IFX is given intravenously in three doses over 6 weeks (induction), and continued every 8 weeks thereafter (maintenance) if indicated. Response rates, as defined by falls in Pediatric Crohn's Disease Activity Index (PCDAI) scores, can be as high as 71%–100% for children treated with IFX5,9 with many achieving clinical remission.1,5,9–13 IFX can lead to improvements in serum inflammatory markers (C-reactive protein [CRP], erythrocyte sedimentation rate [ESR]),1,13,14 and is steroid-sparing, allowing withdrawal in more than half of patients by 10–13 weeks.10–15 Furthermore, growth velocity is believed to be improved by IFX therapy.1,11,16

Although the short-term benefits of IFX appear positive, these benefits may not be sustained and need to be considered in the context of potential risks. One series suggest that around one-fifth of patients will eventually become unresponsive to IFX within a year.11 In addition, there are continuing and increasing concerns about potential adverse effects of IFX. These include infusion-related events and consequences of immunosuppression. Especially worrying have been recent reports of rare but almost invariably lethal hepatosplenic T-cell lymphoma in adolescents.17–19 The aim of this retrospective review was to illustrate the experience with IFX in a single pediatric center, with focus upon not only short-term, but also medium-term, clinical outcomes, as well as delineation of changes in serum inflammatory markers, growth and adverse events.

Methods

Data collection

Children with CD who had received IFX at Sydney Children's Hospital (SCH) between January 2000 and 31 July 2007 were identified and their medical records reviewed retrospectively. Patient demographic and clinical data were extracted from medical charts, laboratory and pathological reports, and growth charts. Data collected included gender, date of birth, age at diagnosis, comorbidities, prior or other medical/surgical interventions, prior and current medication, age at, and date of, first IFX infusion, dates of subsequent IFX infusions, disease type (luminal/fistulizing/stricturing) and location, indication for IFX, IFX dose, IFX pre-medication, and adverse events. Whenever available, the following items were noted: PCDAI scores, height, weight, serum albumin, hemoglobin (Hb), CRP, ESR, white cell count (WCC), neutrophil, lymphocyte and platelet counts. Not all parameters were available at each time point for each patient.

Data interpretation

Clinical response, remission and relapse were recorded according to physician global assessment (PGA) and PCDAI scores20,21 where available, as originally assigned by the treating clinician. Remission was defined as PCDAI < 15.22 Response was defined as reduction in PCDAI score by ≥ 15 and/or documentation of physician-assessed clinical improvement. Relapse was defined as an increase of PCDAI by ≥ 15 after previously being in remission. Height and weight measurements were used to calculate body mass index (BMI), and all three parameters were converted to Z scores using Center for Disease Control Epi Info 2000 (Atlanta, GA, USA). In order to delineate changes in key parameters over time, patient data were grouped at relevant time points (baseline, 6 months, and 10 or 12 months for biochemical, hematological and growth parameters, respectively). Baseline data varied from 1 day to 3 weeks before IFX. Six-, 10- and 12-month data varied between 2 and 5 weeks either side of these time points.

Statistical analysis

All statistics were generated using Prism 4.0 (GraphPad, San Diego, CA, USA). Two-tailed Wilcoxon matched pairs test was carried out on PCDAI, serum and growth parameters. The two-tailed Mann–Whitney U-test compared pre-infusion data between those who did or did not achieve remission at 2, 26 and 52 weeks. Results are presented as median (range) unless otherwise stated and were considered significant if P < 0.05.

Ethical approval

This study was approved by the South Eastern Sydney and Illawarra Area Health Service Research Ethics Committee.

Results

Patients

Sixteen children with CD were identified as having received IFX over the 7.5-year time period. Of the cohort of 16 patients, all had luminal disease and six also had fistulizing disease. One patient had upper gastrointestinal tract (GT) involvement alone, two had lower GT involvement alone, whereas 13 had lower and upper GT disease (Table 1). All patients were treated with IFX for medically refractory disease and none received IFX as the sole initial therapy to induce remission (‘top-down'). The mean age at first infusion was 13.0 years (1.25–17.5 years) and the median number of infusions per patient was five (2–24). The mean duration of disease prior to the first IFX dose was 3.0 years (0.3–9.5 years) and the mean duration of follow up was 2.3 years from the first infusion (4 weeks–5.4 years) (Table 1).

Table 1.  Characteristics of patients with CD who had received IFX at Sydney Children's Hospital
  • Data not available for all patients at all time points.

  • Numbers reflect multiple areas of involvement in each patient.

  • 8 prednisone, 2 budesonide.

  • §

    1 patient did not receive any concurrent medication.

  • 5-ASA, 5-aminosalicylic acid; CD, Crohn's disease; SSZ/MSZ, sulfasalazine/mesalazine; TI, terminal ileum.

No. subjects 
 Total16
 Male13
 Female3
Age at first infusion 
 Mean (SD) (years)13.0 (4.2)
 Range (years)1.25–17.5
Duration of disease 
 Mean (SD) (years)3.0 (2.9)
 Range (years)0.3–9.5
Disease location 
 Upper gastrointestinal tract14
 Terminal ileal5
 Pancolitis (excluding TI)3
 Patchy colitis12
 Perianal8
Disease behavior 
 Luminal16
 Fistulizing6
Number of infusions 
 Total119
 Median (range)5 (2–24)
Previous medication 
 Oral steroid10
 Azathioprine15
 5-ASA (SSZ/MSZ)8
 Oral tacrolimus1
 6-Mercaptopurine1
 Methotrexate0
Concurrent medication§ 
 Oral steroid7
 Azathioprine14
 5-ASA (SSZ or MSZ)9
 Tacrolimus1
 6-Mercaptopurine1
 Methotrexate2
Duration of follow up 
 Mean (SD) (years)2.3
 Range (years)4–5.4

All patients were premedicated with intravenous (i.v.) hydrocortisone, i.v. promethazine and oral paracetamol, followed by 5 mg/kg dose IFX (Centocor, Philadelphia, PA, USA) infused over 2 h. Following induction (IFX infusions at 0, 2 and 6 weeks), the maintenance regimen (8 weekly thereafter) continued according to the patient's response.

Three patients required a surgical procedure for the management of their CD during IFX treatment. One of these children had not responded to IFX. Data are included in these patients up to the time of their operation, but not beyond this time, as subsequent changes were considered likely to be consequent to the surgical intervention.

Response and remission

The median PCDAI score prior to IFX therapy was 31.3 (ranging from 17.5 to 57.5). At seven weeks following the start of IFX the median PCDAI was recorded as 13.8 (range 0–45) (P = 0.125). When recorded between 37 and 45 weeks after the start of IFX, the median PCDAI was 10.0 (range 0–23) (P = 0.06, compared to baseline).

A clinical response was seen in 14 (88%) of the 16 patients (Table 2). Six (50%) were in remission at 2 weeks, 10 of 14 at 8 weeks (71%), nine of 13 at 26 weeks (69%) and 10 of 12 at 1 year (83%) (Table 3). However, seven of 14 patients with available data at 26 weeks had relapsed at some point within this time (50%), and seven of 12 (58%) by 1 year. At the latest follow up of the six patients with fistulizing disease, two had complete resolution (33%) and one patient's fistula had much improved (17%). Two patients showed no response (33%) and records were incomplete for the sixth patient.

Table 2.  Course of events with infliximab therapy
Patient
ID
Induction infusionsMaintenance infusionsResponseRelapseIFX stoppedEvents
 130NoN/AYesColectomy at week 10.
 235YesYesNoRelapsed after 2nd maintenance infusion. AZA switched to MTX before 8th infusion, IFX ongoing. Disease continuing to relapse/remit. Fistula unresponsive.
 331YesNoNoOngoing remission.
 431YesNoNoOngoing remission.
 5310YesYesYesRemission induced, but response lost by 12th infusion. IFX discontinued after 13 infusions.
 630YesNoYesSignificant improvement of severe perianal disease. No maintenance IFX. Long-term remission (2 years).
 737YesYesYesIFX ceased after 9th infusion (as in remission). One further infusion 4 months later for relapse, with reinduction of remission. Subsequent small bowel resection for jejunal strictures.
 832NoN/AYesNo response after five infusions.
 920YesNoNoInsufficient follow up to assess fistula response.
1039YesYesYesTendency to deteriorate between infusions. IFX stopped after 9th infusion (> 1 year). After relapse 6 months later, IFX maintenance regimen restarted with attenuated response. IFX ceased due to adverse events (Table 5).
1130YesNoYesResolution of severe perianal disease.
12321YesNoNoFistula closed by month 4. Tendency to deteriorate between infusions.
1339YesYesYesInitial response lost, patient defunctioned with ileostomy after 3rd infusion. Maintenance therapy restarted at week 56 but colectomy after 5th infusion as response inadequate. Fistula static. IFX discontinued.
1432YesYesYesRemission induced, but response lost by 5th infusion. IFX discontinued.
1533YesYesYesRemission induced, but relapse by 4th infusion. IFX discontinued after 6 infusions.
1632YesNoNoOngoing remission.
Table 3.  Response to infliximab
Time following first infusionNo. studiedNo. in remission (%)
2 weeks126 (50%)
8 weeks1410 (71%)
26 weeks139 (69%)
1 year1210 (83%)

Growth parameters following IFX treatment

Prior to giving IFX, the median BMI Z score was −0.9 (range −6.1–1.6), whereas the weight Z score was −1.2 (range −4.7–1.0) and height Z score was −0.5 (range −2.5–0.6) (Table 4). In recordings at 1 year following the start of IFX, BMI Z scores had improved to −0.1 (range −2.5–1.8) (P = 0.008), median Z score for weight improved to −0.5 (range −3.6–1.2) (P = 0.07). Z score for height at 1 year was −0.8 (range −2.8–1.4) (P = 0.95).

Table 4.  Anthropometric measures before and after infliximab in 16 children
 Z scores
Baseline6 months1 year
  • Median figures quoted.

  • *

    P = 0.008 compared to baseline.

  • BMI, body mass index.

BMI−0.9−0.2−0.1*
(−6.1–1.6)(−3.9–1.4)(−2.5–1.8)
Height−0.5−0.5−0.8
(−2.5–0.6)(−2.2–0.1)(−2.8–1.4)
Weight−1.2−1.0−0.5
(−4.7–1.0)(−3.7–1.0)(−3.6–1.2)

Changes in laboratory parameters with IFX

At baseline, 11 of 14 patients had CRP > 5 mg/L; seven of 10 patients had ESR > 20 mm/h; and 12 of 13 patients had albumin < 38 g/L. Two weeks after the first IFX infusion, the median CRP level had fallen from 24 mg/L (range 1–89 mg/L) to 3 mg/L (range 1–23 mg/L: P < 0.05) (Fig. 1a). ESR decreased from a median of 38 mm/h (range 2–102 mm/h) to 19 mm/h (range 0–14 mm/h) at 2 weeks (P < 0.05) (Fig. 1b). In addition, significant improvements were found for elevated platelets (P < 0.01 at 10 months, P < 0.05 at 12 weeks) compared to baseline. Positive trends were found for albumin and hemoglobin following IFX treatment (Table 5).

Figure 1.

Mean (a) CRP and (b) ESR following infliximab (IFX) therapy in children with Crohn's disease. C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) dropped significantly after first IFX infusion.

Table 5.  Clinical, biochemical and hematological parameters before and after IFX in 16 children
 Baseline6 months10 months
  • Median figures quoted, range given in parentheses.

  • *

    P = 0.04,

  • **

    P = 0.008 compared to baseline, and

  • P = 0.03 compared to value at 6 months.

  • CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; Hb, hemoglobin; IFX, infliximab; PCDAI, Pediatric Crohn's Disease Activity Index; WCC, white cell count.

PCDAI31.116.38.9
(18–58)(0–33)(0–23)
CRP (mg/L)24.116.83.0
(1–89)(1–44)(1–39)
ESR (mm/h)382214
(2–102)(2–76)(2–38)
Hb (g/L)106.5131.0129.0
(85–145)(89–152)(110–157)
WCC (×109/L)7.97.26.6
(3.5–23.8)(5.3–11.9)(3.3–9.3)
Lymphocytes (×109/L)1.21.51.4*
(0.6–2.8)(0.4–2.3)(0.4–3.0)
Neutrophils (×109/L)5.85.04.5
(3.8–19.3)(3.3–8.6)(2.8–7.8)
Platelets (×109/L)470398345
(182–762)(156–625)(171–444)**
Albumin (g/L)303537
(20–45)(30–42)(28–44)

The initial improvements in laboratory parameters were followed by transient deterioration and subsequent sustained improvement during the period of follow up, as illustrated for CRP and ESR (Fig. 1).

Adverse events

Eleven adverse events were recorded in seven patients (Table 6). Only one event was at the time of an IFX infusion, although this related to the pre-medication and not to IFX. There were no malignancies or deaths in the patient cohort during the follow-up period.

Table 6.  Adverse events following infliximab
Patient IDEventDetails
  1. ANCA, anti-neutrophil cytoplasmic antibodies; IBD, inflammatory bowel disease; IFX, infliximab; LFT, liver function test; R, right.

 2Recurrent perianal abscess
Pneumonia
Seton introduced, mostly resolved.
Required admission and i.v. antibiotics, resolved.
 3Molluscum contagiosum
Episodic knee arthralgia
Probable IFX-related opportunist, resolved;
episodes also occurred pre-IFX.
 5Bilateral lower limb and R arm venous thromboses + R fifth finger vasculitisResolved. ANCA + ve, believed to be IBD-related, not polyarteritis.
 7Episodic knee/ankle arthralgia2nd and 10th week following first infusion, and 1 year following last infusion. Similar episodes occurred pre-IFX.
 8Non-specific illnesses before and following 3rd infusionSecond illness took 10–14 days to recover, symptoms included fever, diarrhea/vomiting, lethargy, anorexia. IFX continued.
10Pseudomonas infection at gastrostomy site
Intermittent ankle arthralgia
Abnormal LFT
Possible IFX-related opportunist. IFX stopped.
12Fluctuating perianal abscessPresent pre-IFX.

Discussion

This retrospective study delineates the treatment outcomes and adverse effects, as well as changes in inflammatory markers and growth parameters over time, in a group of children with CD treated with IFX. Fourteen of these 16 patients had a positive clinical response to IFX and 10 of 12 children were in remission 1 year after starting IFX. These figures are comparable with recent published data of the response to IFX in pediatric CD.1,5,9–13

One of the critical aims of management of pediatric CD is optimization of growth, with both weight gain and linear growth being important. Poor disease control compromises growth in children, and this can lead to long-term adverse outcomes, such as impaired adult height. For these reasons, an assessment of the benefits upon growth is essential in evaluating the outcomes of specific therapies. Recent studies suggest significant improvements in growth parameters after IFX in children. However, there remains uncertainty as to why some clinically responsive children have a greater restoration of growth velocity than others.1,10,11,23–25

The children in the current study had depressed weight Z scores (median within 1 standard deviation of the population mean). Nonetheless, positive trends were identified in weight (P = 0.07 at final follow up) but not in height. However, mean Z score for BMI increased significantly (Table 4). The pubertal status of this cohort of children was not available, but, as many were teenagers at the onset of disease, one might assume that a number were post-pubertal. Clearly, pubertal status will influence any effect of improved disease control and nutrition upon further height gains. Indeed, a significant inverse correlation has been shown between Tanner stage of puberty and improvement in height Z score following IFX therapy in children with CD.24 The sample size of the current study also influences interpretation of these growth data. Further studies involving larger groups of children treated with IFX are required to further elucidate the impact on growth beyond 1 year.

IFX leads to histological and endoscopic improvements as well as improvements in inflammatory markers.26 Histological and endoscopic outcomes were not available in this retrospective study, but data of systemic inflammatory markers were available in most patients. This cohort had significant improvements in standard markers of gut inflammation (ESR, CRP, albumin and platelets). Furthermore, the PCDAI scores, which incorporate the results of albumin, ESR and hematocrit testing, also improved significantly with IFX.

Of interest, all parameters appeared to be variable with initial improvements followed by transient deteriorations, even in children with good responses to IFX (Fig. 1). This likely reflects variations in the effects of IFX between patients, depending on dosage and frequency of infusions. In particular, it may reflect the longer period (8 weeks) between the third and fourth infusions, than between the earlier infusions. Prospective determination of serial levels of these markers in a larger group of children treated with IFX would be required to further elucidate and confirm these patterns. If confirmed, such results could support differing infusion dosage and/or frequency in younger patients with CD to achieve optimal efficacy.

One of the increasing concerns about IFX in children and adolescents is the side-effect profile of the drug. The most common adverse event of IFX is an immediate infusion reaction, occurring in 20% of pediatric patients and in up to 9.7% of all infusions.8,9,11,13,27,28 Whereas published studies document the efficacy of pre-medication in the reduction of infusion reaction risk, only one study has separated those who have previously experienced an infusion reaction from those who have not. In a retrospective review of 243 adult patients, it was found that pre-medication only reduced the incidence of infusion reactions in patients who have had a previous infusion reaction.29 Interestingly, no immediate infusion reactions to IFX were observed in the current cohort; however, each child had received all three pre-medications prior to each infusion.

One important adverse effect of IFX is an increased susceptibility to opportunistic infection. IFX therapy has been associated with several cases of fatal sepsis.17 None of the children in the current cohort had sepsis or major infection. The only infectious agents identified in this series were Molluscum contagiosum (one child) and pseudomonal infection of a gastrostomy site (one child). IFX may have been a contributing factor in these infections, however, both are commonly seen in children outside the context of IFX. Other infectious illnesses were pneumonia in one child and an apparent viral illness in a second child (Patient 8). Again the role of IFX in these illnesses is unclear.

Three of the current subjects complained of episodic arthralgias of the large joints. This has been reported elsewhere in up to 3% of IFX-treated pediatric CD patients8,27,30,31 and it has been suggested that this may be a delayed-type hypersensitivity reaction to IFX.8,13,15,32 However, two of the three patients in the current series reported these symptoms before starting IFX. Furthermore, as 20% of patients with CD report joint pains,33 the relevance of these symptoms in the current study is unclear.

Other longer-term adverse events associated with IFX include demyelinating disorders, ANA formation with lupus-like syndromes, and TB reactivation. There is also concern over the incidence of malignancies, in particular hepatosplenic T-cell lymphoma, following long-term IFX therapy, especially in teenage boys.18,32 Fortunately, none of these events occurred in the current cohort within the period of observation of up to 5 years. Nevertheless, it is clear that IFX therapy warrants close vigilance for potential adverse events, including infection. Furthermore, the risk profile needs to be outlined carefully in counseling discussions prior to considering IFX in any child.

This series of responses and outcomes of IFX therapy in children is limited by its retrospective nature, the sample size of the cohort and some incomplete data. The prominence of male gender in this cohort may reflect the sample size. Although representative of children elsewhere, with similar distribution of disease and patterns of disease behavior, the data are from a single tertiary pediatric center. However, the children were managed with a consistent protocol with their progress followed closely in a multi-disciplinary IBD clinic.

In conclusion, this report provides additional important information to the currently available literature on the use and usefulness of IFX in children and adolescents with CD. Specifically, this report reinforces the notion that monitoring not only serum inflammatory markers but also growth parameters in the short and medium term provide clinical information relevant to the further management of these children. Furthermore, the pattern of improvement requires further elucidation as the results in this study suggest differing dosing frequency of infusion may achieve better efficacy. Ongoing assessment of the outcomes of IFX treatment in children is essential, as is ongoing surveillance for side-effects and adverse outcomes.

Ancillary