*Nationwide Children's Hospital, Columbus, OH, USA
†Connecticut Children's Center, Hartford, USA
‡Medical College of Wisconsin, Milwaukee, USA
§Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada
||Centocor Inc, Malvern, PA, USA
¶RW Johnson Pharmaceutical Research and Development, La Jolla, CA, USA
#Addenbrooke's Hospital, Cambridge, UK
**North Shore-Long Island Jewish Health System, New Hyde Park, NY, USA
††Children's Center for Digestive Health Care, Atlanta, GA
‡‡MassGeneral Hospital for Children, Boston, MA, USA
§§Queen Paola Children's Hospital, Antwerp, Belgium
||||Texas Children's Hospital, Houston, USA
¶¶Children's Hospital of Philadelphia, Philadelphia, PA, USA
##RW Johnson Pharmaceutical Research and Development, Titusville, NJ, USA
Received 16 September, 2008
Accepted 23 March, 2009
Address correspondence and reprint requests to Dr Wallace Crandall, Nationwide Children's Hospital, The Ohio State University School of Medicine, 700 Children's Drive, Columbus, OH 43205 (e-mail: Wallace.Crandall@NationwideChildrens.org).
Centocor Inc (Malvern, PA) provided support for this study.
Wallace Crandall received research funding in conjunction with the conduct of this study, served as a consultant to Centocor, and received an honorarium from Centocor. Jeffrey Hyams received research funding in conjunction with the conduct of this study and other studies sponsored by Centocor; and served as a consultant to Centocor. Subra Kugathasan received research funding in conjunction with the conduct of this study, served as a consultant to Centocor, and received an honorarium from Centocor. Anne Griffiths received research funding in conjunction with the conduct of this study. Julie Zrubek is an employee of Centocor Inc. Allan Olson and Grace Liu, currently employed at Robert Wood Johnson Pharmaceutical Research and Development, are former employees of Centocor Inc. Robert Heuschkel received research funding in conjunction with the conduct of this study and also received a grant from Schering-Plough. James Markowitz received research funding in conjunction with this and other studies, served as a consultant to Centocor; and received an honorarium from Centocor. Stanley Cohen received research funding in conjunction with the conduct of this study, received other research grants, and chaired conferences sponsored by Centocor. Harland Winter received research funding in conjunction with the conduct of this study and served as a consultant to Centocor. Gigi Veereman-Wauters received research funding in conjunction with the conduct of this study. George Ferry received research funding in conjunction with the conduct of this study and served as a consultant to Centocor. Robert Baldassano received research funding in conjunction with the conduct of this study.
Crohn disease (CD) is a chronic inflammatory bowel disease characterized by transmural inflammation involving any part of the gastrointestinal tract, often in a discontinuous fashion. Common symptoms of CD include abdominal pain, diarrhea, anorexia, weight loss, fever, and fatigue. In children and adolescents, CD may also negatively affect growth, education, and social and emotional development (1–5). Perianal disease is common and ranges from mild lesions (eg, small, non-inflamed skin tags and fissures) requiring only conservative therapy to severe complications (eg, ulcerations, abscesses, fistulas, strictures) requiring aggressive therapy, including surgery. Often, a combination of medical and surgical therapies is used (6). The cumulative frequency of perianal fistulas ranged from 14% to 38% in adults with CD at referral centers (7) and was 13% in children and adolescents with CD evaluated at a single referral center (8). It has also been reported that anal fissures, perianal fistula, or abscesses precede or present simultaneously with the diagnosis of intestinal disease in 36% to 81% of patients with CD who develop perianal signs and symptoms (7).
Antibiotics (most commonly ciprofloxacin or metronidazole), immunomodulators, and exclusive enteral or parenteral nutrition are used in the treatment of perianal CD in children, but randomized controlled trials of efficacy have not been conducted (9–13).
Successful treatment of luminal and fistulizing CD in adults has been reported with infliximab (Centocor Inc, Malvern, PA), a chimeric monoclonal antibody, which neutralizes the biological activity of circulating and membrane-bound tumor necrosis factor-α (TNF-α) (14–17). It has also been shown to be beneficial in the treatment of luminal CD in children (18) and is approved for the treatment of fistulizing CD in adults and luminal disease in adults and children.
This report extracts data on CD-related perianal signs and symptoms from a previously reported prospective trial designed to evaluate the efficacy and safety of infliximab in the treatment of children with active, luminal CD (18). The Pediatric Crohn Disease Activity Index (PCDAI) perirectal subscore, a subscore that assesses disease-related perianal signs and symptoms (including but not limited to fistula activity), was used to measure perianal CD outcomes. This report presents the results of post hoc analyses evaluating the effects of infliximab on concurrent perianal CD signs and symptoms in a subgroup of patients from the REACH trial.
MATERIALS AND METHODS
Patient eligibility criteria and study procedures for REACH have been described previously (18). Eligible patients were 6 to 17 years of age, had CD of at least 3 months duration, and had a baseline PCDAI score greater than 30 (19,20). Concomitant therapies included azathioprine, 6-mercaptopurine, methotrexate, aminosalicylates, and oral corticosteroids (18). Patients who received previous treatment with infliximab or other agents targeting TNF-α were ineligible. Patients with active luminal disease and concomitant perianal disease were included, but those with isolated active perianal disease were excluded.
The institutional review boards at participating sites approved the protocol. Written informed consent was obtained from all parents or legal guardians, and assent was obtained from children on the basis of individual institutional review board guidelines.
Eligible patients received an induction regimen of infliximab 5 mg/kg at weeks 0, 2, and 6. At week 10, patients in clinical response were randomized to receive infliximab 5 mg/kg either every 8 weeks (q8w) beginning at week 14, or every 12 weeks (q12w) beginning at week 18 (18). Patients who were not in clinical response at week 10 were considered treatment failures and discontinued from the trial. All randomized patients were followed for efficacy and safety through week 54.
After randomization, patients who lost clinical response were eligible for a one-time crossover to a more frequent or higher infliximab dose (18).
Signs and symptoms of perianal disease were assessed at each visit using the PCDAI perirectal subscore. Patients with no symptoms or asymptomatic tags received a score of 0; those with “1–2 indolent fistula, scant drainage, no tenderness” received a score of 5; and those with “active fistula, drainage, tenderness or abscess” received a score of 10. Perianal disease signs or symptoms that did not precisely fulfill the predefined score criteria (eg, deep perianal fissuring, perianal ulceration, inflamed perianal skin tags) were scored at the discretion of the investigator. A partial response was defined as an initial perirectal subscore of 10 decreasing to 5. Complete response was defined as an initial perirectal subscore of 5 or 10 decreasing to 0.
Patients with perianal signs or symptoms were summarized over time by treatment group and PCDAI perirectal subscore. The numbers and proportions of patients in partial or complete response were summarized overall and by treatment group. Patients who discontinued study participation or who had missing PCDAI perirectal subscores had their last recorded perirectal subscore carried forward. Patients who had a prohibited CD-related surgery or concomitant medication change, or who crossed over to a more frequent or higher infliximab dose had their last recorded PCDAI perirectal subscore before the event carried forward. For patients who did not have baseline perianal symptoms, but developed them during the trial, response was determined by using the first nonzero postbaseline perirectal subscore as the baseline score.
Categorical variables were summarized by number of observations and percentage; continuous variables were summarized by number of observations, mean, and standard deviation.
Safety results reported here are limited to key findings in this REACH patient cohort with concurrent perianal CD. Adverse events, infections, and serious adverse events were summarized through week 54 for patients with signs and symptoms of perianal disease at baseline. Safety results for the entire REACH study population have been described previously (18).
At baseline, 23 patients had signs and symptoms of perianal disease, of whom 22 were randomized at week 10 to an infliximab treatment maintenance regimen (10 received infliximab 5 mg/kg q8w; 12 received infliximab 5 mg/kg q12w; Fig. 1A). One patient discontinued infliximab treatment after week 2 because of worsening luminal CD.
With few exceptions, baseline demographic and disease characteristics of those patients with signs and symptoms of perianal disease at baseline were similar to those without perianal signs and symptoms and to the overall REACH study population (Table 1). No formal statistical testing of demographic or disease characteristics was prespecified for this post hoc subgroup analysis; however, a numerically greater proportion of patients with signs and symptoms of perianal disease were male (19/23 patients, 82.6% vs 47/89 patients, 52.8%, respectively) and were taking oral corticosteroids at baseline (10/23 patients, 43.5% vs 29/89 patients, 32.6%, respectively) compared with patients without signs and symptoms of perianal disease at baseline.
Response in Patients With Baseline Perianal Signs and Symptoms
Decreases in signs and symptoms of perianal disease were seen as early as week 2. At week 2, after a single infusion of infliximab 5 mg/kg, 40.9% (9/22) of randomized patients with signs and symptoms of perianal disease at baseline achieved response (4 partial and 5 complete). Response was achieved by 16 (72.7%) patients at both week 6 (3 partial and 13 complete) and week 10 (2 partial and 14 complete) (Fig. 2).
After randomization to the infliximab 5 mg/kg q8w or q12w regimen, the proportion of patients with signs and symptoms of baseline perianal disease responding to infliximab treatment remained consistent during maintenance treatment, and did not appear to be affected by the frequency of infliximab infusions. At week 30, response was achieved in 15 (68.2%) patients, 7 (7 complete) randomized to the q8w regimen, and 8 (2 partial and 6 complete) randomized to the q12w regimen. At week 54, response was achieved in 16 (72.7%) patients overall, 8 (8 complete) randomized to the q8w regimen and 8 (1 partial and 7 complete) randomized to the q12w regimen.
Among 22 randomized patients with signs and symptoms of baseline perianal disease, 15 patients received an initial perirectal subscore of 5, of whom 11 had complete response (7 and 4 randomized to the q8w and q12w regimens, respectively), 2 had no response (both randomized to the q12w regimen), and 2 had an increase in symptom severity at week 54 (1 randomized to each regimen). Additionally, 7 patients received an initial subscore of 10, of whom 4 had complete response (1 and 3 randomized to the q8w and q12w regimens, respectively), 1 had partial response (randomized to the q12w regimen), and 2 had no response at week 54 (1 randomized to each regimen). Nine patients with signs and symptoms of perianal disease at baseline crossed over to a higher or more frequent dosing regimen (2 and 7 randomized to the q8w and q12w regimens, respectively) (Fig. 1A).
Development of Perianal Signs and Symptoms During Infliximab Induction and Maintenance Therapy
Infliximab treatment did not prevent development of perianal signs and symptoms because 9 patients developed perianal signs and symptoms during treatment (Fig. 1B). Three of these patients had a history of perianal abscesses, including 1 who also had an abscess incised and drained.
Among these 9 patients, 3 developed signs and symptoms before randomization. Two of these patients were randomized to the q8w regimen and had complete response at week 54 and 1 patient was randomized to the q12w regimen and had no response at week 54.
After randomization, 5 of 39 patients (12.8%) who received the q12w regimen and 1 of 42 patients (2.4%) who received the q8w regimen developed signs and symptoms of perianal disease. The patient who received the q8w regimen had no response at week 54 and all 5 patients who received the q12w regimen had complete response at week 54, of whom 3 responded after crossing over to a higher or more frequent infliximab dose.
Treatment with corticosteroids, antibiotics (ciprofloxacin or metronidazole), or immunosuppressants was examined to assess whether concomitant use of these CD medications in patients with perianal symptoms augmented the therapeutic benefit of infliximab maintenance therapy (ie, concomitant use after week 30). Among 31 patients with concurrent perianal CD, 3 received corticosteroids during infliximab maintenance therapy; at week 54, 1 patient had complete response and 2 had no response. Overall, 10 patients received antibiotics during infliximab maintenance therapy; at week 54, 5 had complete response, 1 had partial response, and 4 had no response. Additionally, almost 90% of this REACH cohort received concomitant immunosuppressants, and by protocol, doses were to remain stable. The use of corticosteroids or antibiotics did not appear to augment the therapeutic benefit of infliximab therapy.
Among randomized patients with baseline perianal signs and symptoms in the infliximab 5 mg/kg q8w (N = 11) and q12w (N = 11) groups, the mean duration of treatment was 43.8 weeks versus 44.2 weeks, respectively, and the mean duration of follow-up was 51.1 weeks versus 53.4 weeks, respectively. The duration of treatment and follow-up were similar among randomized patients regardless or the presence or absence of baseline perianal signs and symptoms (data not shown). Among patients with baseline perianal signs and symptoms and those in the overall REACH population, similar proportions of patients had at least 1 adverse event (22 of 23 patients, 95.7% vs 106 of 112 patients, 94.6%). The proportion of patients with 1 or more infection (as identified by the investigator) was numerically greater among patients with concurrent perianal disease at baseline (16 of 23 patients, 69.6%) than in the overall REACH population (61 of 112 patients, 54.5%).
The only serious adverse event among patients with baseline concurrent perianal CD occurred in the patient who discontinued after week 2 because of CD exacerbation leading to subtotal colectomy and ileostomy.
Three patients who developed concurrent perianal CD after randomization to an infliximab regimen developed serious adverse events. One patient who received infliximab q8w was hospitalized at week 54 for treatment of CD exacerbation and an anal fissure. Two patients who received infliximab q12w developed perianal disease and serious infections during therapy. One patient developed a labial abscess approximately 9 weeks after the week 30 infliximab infusion and was hospitalized for surgical drainage of a rectovaginal fistula with abscess. Another patient was hospitalized at approximately week 46 for CD exacerbation and suspected infection. Serious infections in these patients resolved with intravenous antibiotic therapy and both had complete response at week 54.
REACH demonstrated the efficacy and safety of infliximab in children with moderately to severely active luminal CD; however, REACH was not specifically designed to investigate the effect of infliximab on concurrent perianal disease. Although no formal statistical analyses were prespecified for this subanalysis, it is noteworthy that when comparing patients with and without concurrent perianal CD at baseline, greater proportions of patients with perianal symptoms were male (82.6% vs 52.8%, respectively) and were taking oral corticosteroids (43.5% vs 32.6%, respectively). Keljo et al (21) have recently reported the results of an inception cohort study of the natural history of pediatric fistulizing perianal disease treated with a variety of conventional therapies or infliximab. Although the different design characteristics and study objectives do not allow a direct comparison of these results with those of our clinical trial, there are notable comparisons in population demographic and disease characteristics. Unlike the inception cohort of Keljo et al, the subgroup of patients represented in our study does not comprise patients with newly diagnosed CD (mean disease duration of 1.8 years). Similar to the inception cohort, we also found a predominance of males with signs and symptoms of perianal CD at baseline, and when compared with patients without baseline perianal CD, a larger proportion of patients with baseline perianal symptoms had colonic involvement.
Among 22 children with concurrent baseline perianal CD who were randomized to an infliximab regimen, 41%, 73%, 73%, 68%, and 73% of patients were in perianal response at weeks 2, 6, 10, 30, and 54, respectively. Complete response of perianal symptoms was achieved in 68% of patients at week 10 and 73% of patients at week 54. When comparing concomitant use and nonuse of corticosteroids, antibiotics, or immunosuppressants during infliximab maintenance therapy, there was no evidence of an incremental benefit with respect to perianal symptom relief.
Among 9 patients who developed perianal signs and symptoms during the study, 3 patients developed perianal CD before randomization. Two of these patients received the q8w regimen and had complete response at week 54 and 1 received the q12w regimen and had no response at week 54. Of the 6 patients who developed perianal signs and symptoms after randomization, 5 received infliximab q12w and responded at week 54 (3 of whom responded after crossing over to a higher or more frequent infliximab dose), and 1 received infliximab q8w and had no response. Thus, 5 of 39 patients (12.8%) randomized to the q12w regimen and 1 of 42 patients (2.4%) randomized to the q8w regimen developed perianal signs and symptoms. Although there is a suggestion that development of perianal disease occurred more frequently with the q12w regimen (12.8%) when compared with the q8w regimen (2.4%), these results should be interpreted with caution because of the observational nature of this finding and the small number of patients (n = 6) who developed concurrent perianal disease.
It should be noted that this analysis does not demonstrate the efficacy of infliximab in the treatment of fistulizing pediatric CD. Rather, it suggests that in children with moderately to severely active luminal CD and a variety of concurrent perianal symptoms, infliximab is effective at treating perianal complaints. Additionally, retrospective studies in children with CD have shown diminished fistula and perianal symptoms in response to infliximab therapy (22,23).
The efficacy of infliximab in adults with fistulizing CD was assessed in ACCENT II, a prospective, randomized, placebo-controlled clinical study in which greater than 80% of patients had at least 1 perianal fistula (17). In this trial, among patients who responded to induction therapy and were randomized to maintenance therapy, 54%, 78%, 96%, 90%, and 64% of patients were in fistula response at weeks 2, 6, 10, 30, and 54, respectively. In ACCENT II, 36% of patients achieved complete fistula response at week 54. Parsi et al (24) have shown that in adults with fistulizing CD, external fistula and perianal fistula in particular have a higher rate of closure to infliximab compared with other fistula types.
For the treatment of perianal and fistulizing CD, uncontrolled case series have shown that antibiotics are useful as short-term therapies to decrease or reduce draining, but relapse is immediate on discontinuation. Meta-analyses have demonstrated that immunosuppressive therapy is effective but slow and often incomplete in the treatment of perianal and fistulizing CD (8,25). Studies have demonstrated the efficacy and safety of immunosuppressive therapy in pediatric inflammatory bowel diseases for maintaining remission and minimizing corticosteroid dependency (26–28). These conventional treatments have a slow onset of efficacy in fistulizing CD (11,12).
This study is limited by the observational nature of results that were not prespecified in the protocol, small sample size, which was not powered to evaluate this endpoint, and by the use of the PCDAI perirectal subscore to measure perianal CD activity. By using the PCDAI perirectal subscore, perianal CD activity does not reflect fistula activity exclusively, but may be confounded by the activity of abscesses, severe fissuring or ulcerations, inflamed skin tags, or other perianal CD symptoms. Thus, comparisons to the use of infliximab in adults with fistulizing CD should be interpreted with caution. Nevertheless, these observations warrant further investigation given the paucity of data existing in the literature on the treatment of children with concurrent perianal CD.
In summary, approximately 70% of children with signs and symptoms of perianal CD will have a clinical response when treated with a standard induction regimen followed by 1 year of either q8w or q12w infliximab therapy.
The authors thank Grace Lang and Hongyan Zhang for statistical analysis, James Barrett and Mary Whitman for writing assistance, and the REACH investigators (18) for their participation in and contribution to this work.
1. Kim SC, Ferry GD. Inflammatory bowel diseases in pediatric and adolescent patients: clinical, therapeutic, and psychosocial considerations. Gastroenterology 2004; 126:1550–1560.
2. Griffiths AM. Specificities of inflammatory bowel disease in childhood. Best Pract Res Clin Gastroenterol 2004; 18:509–523.
3. Sawczenko A, Ballinger AB, Savage MO, et al
. Clinical features affecting final adult height in patients with pediatric-onset Crohn's disease. Pediatrics 2006; 118:124–129.
4. Mackner LM, Crandall WV. Psychological factors affecting pediatric inflammatory bowel disease. Curr Opin Pediatr 2007; 19:548–552.
5. Mackner LM, Crandall WV, Szigethy EM. Psychosocial functioning in pediatric inflammatory bowel disease. Inflamm Bowel Dis 2006; 12:239–244.
6. Griggs L, Schwartz DA. Medical options for treating perianal Crohn's disease. Dig Liver Dis 2007; 39:979–987.
7. Sandborn WJ, Fazio VW, Feagan BG, et al
. American Gastroenterological Association Clinical Practice Committee. AGA technical review on perianal Crohn's disease. Gastroenterology 2003; 125:1508–1530.
8. Tolia V. Perianal Crohn's disease in children and adolescents. Am J Gastroenterol 1996; 91:922–926.
9. Griffiths AM. Inflammatory bowel disease in childhood. In: Satsangi J, Sutherland LR, editors. Inflammatory Bowel Diseases. Edinburgh: Churchill Livingstone; 2003. pp. 699–715.
10. Day AS, Whitten KE, Sidler M, et al
. Systematic review: nutritional therapy in paediatric Crohn's disease. Aliment Pharmacol Ther 2007; 27:293–307.
11. Jeshion WC, Larsen KL, Jawad AF, et al
. Azathioprine and 6-mercaptopurine for the treatment of perianal Crohn's disease in children. J Clin Gastroenterol 2000; 30:294–298.
12. Markowitz J, Rosa J, Grancher K, et al
. Long-term 6-mercaptopurine treatment in adolescents with Crohn's disease. Gastroenterology 1990; 99:1347–1351.
13. Kirschner BS. Safety of azathioprine and 6-mercaptopurine in pediatric patients with inflammatory bowel disease. Gastroenterology 1998; 115:813–821.
14. Hanauer SB, Feagan BG, Lichtenstein GR, et al
. Maintenance infliximab for Crohn's disease: the ACCENT I randomised trial. Lancet 2002; 359:1541–1549.
15. Rutgeerts P, Feagan BG, Lichtenstein GR, et al
. Comparison of scheduled and episodic treatment strategies of infliximab in Crohn's disease. Gastroenterology 2004; 126:402–413.
16. Present DH, Rutgeerts P, Targan S, et al
. Infliximab for the treatment of fistulas in patients with Crohn's disease. N Engl J Med 1999; 340:1398–1405.
17. Sands BE, Anderson FH, Berstein CN, et al
. Infliximab maintenance therapy for fistulizing Crohn's disease. N Engl J Med 2004; 350:876–885.
18. Hyams J, Crandall W, Kugathasan S, et al
. Induction and maintenance infliximab therapy for the treatment of moderate-to-severe Crohn's disease in children. Gastroenterology 2007; 132:863–873.
19. Hyams JS, Ferry GD, Mandel FS, et al
. Development and validation of a pediatric Crohn's disease activity index. J Pediatr Gastroenterol Nutr 1991; 12:439–447.
20. Hyams JS, Mandel F, Ferry GD, et al
. Relationship of common laboratory parameters to the activity of Crohn's disease in children. J Pediatr Gastroenterol Nutr 1992; 14:216–222.
21. Keljo DJ, Markowitz J, Langton C, et al
. Course and treatment of perianal disease in children newly diagnosed with Crohn's disease. Inflamm Bowel Dis 2009; 15:383–387.
22. Teitelbaum JE, Saeed S, Triantafyllopoulou M, et al
. Infliximab in pediatric Crohn disease patients with enterovesicular fistulas. J Pediatr Gastroenterol Nutr 2007; 44:279–282.
23. de Ridder L, Escher JC, Bouquet J, et al
. Infliximab therapy in 30 patients with refractory pediatric Crohn disease with and without fistulas in The Netherlands. J Pediatr Gastroenterol Nutr 2004; 39:46–52.
24. Parsi MA, Lashner BA, Achkar JP, et al
. Type of fistula determines response to infliximab in patients with fistulous Crohn's disease. Am J Gastroenterol 2004; 99:445–449.
25. Rutgeerts P. Review article: treatment of perianal fistulizing Crohn's disease. Aliment Pharmacol Ther 2004; 20(Suppl 4):106–110.
26. Markowitz J, Grancher K, Mandel F, et al
. Immunosuppressive therapy in pediatric inflammatory bowel disease: results of a survey of the North American Society for Pediatric Gastroenterology and Nutrition. Am J Gastroenterol 1993; 88:44–48.
27. Uhlen S, Belbouab R, Narebski K, et al
. Efficacy of methotrexate in pediatric Crohn's disease: a French multicenter study. Inflamm Bowel Dis 2006; 12:1053–1057.
28. Jaspers GL, Verkade HJ, Escher JC, et al
. Azathioprine maintains first remission in newly diagnosed pediatric Crohn's disease. Inflamm bowel Dis 2006; 12:831–836.
© 2009 Lippincott Williams & Wilkins, Inc.