Infliximab at Diagnosis: A Sledgehammer to Crack a Walnut?

Abstract

To the Editor: We were surprised to read the case report of de Ridder et al. in the September 2006 issue of the Journal(1). In this report the authors describe the dramatic response to infliximab and azathioprine of a 12-year-old girl with a new presentation of severe Crohn disease (CD). The report fails to address several important issues that require further discussion. The authors report mucosal healing at 8 weeks. Examples of an excellent clinical response to therapy (ie, Pediatric Crohn's Disease Activity Index <10) with mucosal healing in severe Crohn colitis are not unusual after 8 weeks of ‘conventional’ therapy. It was common practice both at the Royal Free and Necker-Enfants Malades Hospitals to assess mucosal healing after 8 weeks of induction therapy. Images of colonic mucosal healing after 8 weeks on exclusive Modulen IBD and azathioprine are by no means unusual. In addition, the induction of an initial disease remission is successful in >80% of children with CD using conventional therapy. In the case report of de Ridder et al, infliximab was used in view of ‘endoscopic severity.’ There is no published evidence that the nature, surface, or severity of endoscopic lesions in any way either predicts treatment response or correlates with clinical disease severity (2). The use of infliximab at the first sign of disease in a child may prejudice or even prevent its later use. Episodic use of infliximab appears to increase the risk of human anti-chimeric antibody and hence the risk of antibodies to infliximab (ATI) and thus the risk of significant adverse reactions and subsequent loss of efficacy (3,4). This is particularly important if colonic disease subsequently becomes azathioprine resistant. At present adolescents have the option of infliximab therapy to postpone/avoid otherwise inevitable subtotal colectomy and ileostomy. Prior infliximab use could therefore preclude effective infliximab use when truly indicated. In their landmark study, Markowitz et al (5) clearly documented a 91% remission rate 18 months after commencing azathioprine/6-mercaptopurine at diagnosis in children with moderate to severe CD. This well-designed, robust, randomized study had sufficient power to conclude that early azathioprine use significantly reduced disease relapse; however, this study also showed that almost 50% of children receiving placebo were also in remission at 18 months, and thus would have received azathioprine unnecessarily. Despite a large number of studies looking at predictors of disease progression (eg, age at presentation, disease severity, disease distribution, genotyping), it remains impossible even to easily predict the individual at high risk for surgery (6). Until more accurate predictors become a reality, indiscriminate use of potent therapies based on subjective measures of severity in our view is irresponsible. Biological therapies such as infliximab are a recent addition to the paediatric CD treatment algorithm. There is not yet any robust long-term follow-up data on this biological therapy in children. The concerns about malignancies and atypical T cell lymphomas continue to cause grave concern (7). Although isolated, there are reports of fatalities secondary to severe infectious complications as a result of infliximab (8). Despite all reported serious adverse events, none can detract from the clear and substantial benefit infliximab has for children with severe, debilitating CD. Such children frequently face the prospect of indefinite ileostomies, extensive fistulizing disease, or permanent short stature. The timely intervention with such an efficacious agent to prevent irreversible consequences of uncontrolled disease has transformed the lives of children with the most severe CD. In these cases most children and families clearly agree that the potential benefits far outweigh the risks. We believe this is currently not the case at diagnosis. This report of de Ridder et al, unlike the first publication by Derkx et al in 1993 (9), provides no new insight into disease mechanism or natural history, nor does it report a previously unattainable clinical response. The authors report this case to highlight the potential benefits of a top-down approach. The fundamental hypothesis underlying this approach is that early aggressive therapy may alter the long-term natural history of the disease. Of course it would be a monumental achievement to be able to obtain such a result in a chronic disease, yet this case report is published with a follow-up of only 7 months. Surely an intervention that the authors are suggesting may alter the natural history of the disease requires substantially longer follow-up. To attempt to prove that natural history can truly be altered, infliximab would have to have been given in isolation and a clinical remission then be sustained off therapy for several years. Alteration of natural history is not synonymous with maintenance therapy with infliximab. Before such potentially toxic therapies are offered to children at diagnosis, it is essential to clearly prove such an effect even exists in adults and to perform randomized studies with long-term follow-up in children.