Biologics in pediatric Crohn’s disease: is it time to move to an earlier therapeutic approach?

Abstract

The treatment of Crohn’s disease in children has undergone a revolution in the past decade following studies that have demonstrated the efficacy of anti-TNFα agents in producing durable clinical response/remission as well as reversal of growth delay in many patients. The positioning of biologic therapy continues to be debated. Should it be reserved for children failing conventional therapy including immunomodulators or should it be used as primary therapy shortly after diagnosis in children with more severe disease likely to suffer a more complicated disease course? Risk stratification will be crucial to any therapeutic decisions and emerging data hold promise that identification of those most likely to benefit will be available in the near future.

Keywords::

• anti-TNFα
• biological therapy
• children
• Crohn’s disease
• immunomodulators
• risk stratification

Crohn’s disease is a chronic inflammatory destructive disorder of the gastrointestinal tract that is thought to result from an abnormal response of the intestinal immune system to the enteric microbiome in genetically predisposed individuals [1]. Those affected with Crohn’s disease often have an impaired quality of life with gastrointestinal and extra-intestinal symptoms of variable severity and a frequent need for intestinal surgery. Growth failure is common in children. Cure is not currently possible, and management has historically been directed toward control of symptoms. However, with approval and subsequent increasingly common use of biologic therapy with anti-TNFα monoclonal antibody therapy starting in 1998, it has become apparent that not only is symptom control more likely, but actual mucosal healing is possible, raising the prospect of altering the usual natural history that can result in bowel stricturing and perforation.

Clinicians caring for those with Crohn’s disease are now faced with an important challenge as they consider therapeutic alternatives and outline benefits and risks for their patients. Should they use ‘conventional’ therapy such as 5-aminosalicylates, corticosteroids and immunomodulators first and reserve more expensive anti-TNFα therapy for those failing initial therapy (step-up approach)? Conversely, does it make more sense to use more effective therapy with anti-TNFα agents first, possibly preventing stricturing, fistulization and abscess formation, and then perhaps consider de-escalation to other agents down the road (top-down approach)? The answer to these questions often gets muddled in a misunderstanding of benefits and risks associated with each approach. Moreover, the patient’s or family’s fear of anti-TNFα as ‘stronger’ (i.e., more dangerous) medicine frequently clouds the issue. Missing from the debate is knowledge that helps to predict disease course. To provide the very best care, would we not want to use our most efficacious therapy for those at high risk of disabling disease and perhaps our safest (albeit less effective) therapies for those with very little chance of disease progression?

Risk stratification

An ability to predict disease course would be important in devising treatment strategies. Younger age (<40 years), extensive disease, early use of corticosteroids (perhaps simply a surrogate for more severe disease), ileal disease location, upper gastrointestinal tract involvement, highly elevated levels of serum C-reactive protein, increased serologic reactivity to microbial antigens, deep ulceration noted at endoscopy and specific genetic polymorphisms have been shown to be associated with a greater likelihood of complicated disease [2–5]. Proximal bowel involvement, evaluated with the most sensitive method of small-bowel capsule endoscopy should also be considered to stratify the risk and consequently the impact of the chosen therapeutic approach [6]. Variants in NOD2, IBD5, DLG5, ATG16L1, IL23R and IL12B have been associated with increased risk of fibrostenosing or penetrating disease, but it is not clear this association extends beyond ileal involvement associated with these polymorphisms [7]. For now at least, the power of genetic mutations to predict disease course remains limited. Attention has also turned toward examining the enteric microbiome as a possible predictor of disease course. In treatment-naïve children with Crohn’s disease, a recent study showed an increased abundance in bacteria that included Enterobacteriaceae, Pasteurellaceae, Veillonellaceae and Fusobacteriaceae and decreased abundance in Erysipelotrichales, Bacteroidales and Clostridiales that correlated with disease status [8]. Intestinal dysbiosis (decreased Firmicutes) has been shown to be predictive of relapse following infliximab therapy [9]. Further work characterizing the relationship of intestinal microbiome to course and response to therapy is needed. Finally, gene expression signatures are being examined as predictors of response to therapy [10].

Evidence-base for clinical decisions

In children, disease presentation involves the ileum in 80% of cases (ileocolonic or ileal alone), is most often moderate-to-severe at diagnosis, frequently includes deep ulceration and in 25% of cases includes growth failure. In this population, in particular, as well as in adults with more severe disease, there is a strong temptation to use anti-TNFα therapy at or shortly following diagnosis. An open-label study in adults in 2008 demonstrated superior outcome in patients initially treated with combination therapy of azathioprine and a three-dose induction of infliximab followed by episodic infliximab as needed compared to those receiving standard step-up therapy with azathioprine alone initially and then anti-TNF therapy as needed [11]. At 1 year, 62% of the former group was in corticosteroid-free remission without surgery compared to 42% of the latter group. In 2010, the SONIC trial (Study of Biologic and Immunomodulator Naïve Patients in Crohn’s Disease) was published demonstrating that combination therapy with infliximab induction/maintenance plus azathioprine was superior to infliximab induction/maintenance alone or azathioprine alone in achieving corticosteroid-free remission at 26 and 52 weeks [12]. However, these adult patients were not newly diagnosed, having a mean disease duration of 2.3 years. In both these adult studies, infliximab infusions were protocolized and dose and interval adjustment not performed.

More recently, a study of newly diagnosed children with Crohn’s disease has provided the first ‘real-life’ attempt at comparative effectiveness of early therapy with anti-TNFα compared to traditional immunomodulator-based early therapy with subsequent step-up to anti-TNFα if needed [13]. Using propensity scoring methodology to carefully match early anti-TNFα-exposed children (<3 months from diagnosis) to those receiving early immunomodulator only or those receiving no early immunosuppression, striking differences at 1 year were noted. Corticosteroid-free remission without surgery at 1 year was noted in 85, 60 and 54% of the three groups, respectively. Moreover, in the early immunomodulator group, 20/68 needed to commence anti-TNFα therapy between 3 and 12 months, and in the no early immunosuppression group 47 needed to step-up to immunomodulators, anti-TNFα or both during that same period. C-reactive protein was still elevated in 24, 44 and 53% of the three treatment groups, respectively. While weight gain was noted in all groups, only the early anti-TNFα group had an improved height z score at 1 year.

Risk of therapy

The ultimate decision of a patient or family to pursue a specific therapeutic intervention will be influenced not just by its likelihood of being effective, but also by potential toxicity [14]. In particular, the fear of cancer first and life-threatening infections second is foremost on their minds. Physicians too are respectful of potentially severe or life-threatening complications of therapy and want to match their therapy to the ‘severity’ of disease. Unfortunately, with Crohn’s disease, the relentless progression of intestinal damage can occur even in the absence of severe symptoms [15]. Thiopurines have formed the basis of maintenance therapy for many patients since early publication of their efficacy [16], but more recent data have called into question whether they change the natural history of disease [17]. Moreover, a clear-cut increased risk of lymphoma and skin cancer has tempered enthusiasm in many corners for their use as monotherapy [18,19]. While ample evidence suggests that combination therapy of anti-TNFα with azathioprine is more clinically effective than either anti-TNFα alone or azathioprine alone, the use of combination therapy or thiopurine monotherapy in young males has been associated with a rare and often fatal cancer, hepatosplenic T-cell lymphoma (HSTCL) [20]. For physicians caring for children and young adults, this has dampened enthusiasm for therapy utilizing thiopurines. Methotrexate, often used in combination with anti-TNFα in the treatment of rheumatoid arthritis, is increasingly used in combination with anti-TNFα agents for the treatment of Crohn’s disease. However, a large prospective study comparing infliximab alone versus infliximab plus methotrexate in adult patients previously treated with corticosteroids showed no significant difference in clinical outcome at 1 year [21]. Nonetheless, combination therapy with methotrexate does appear to decrease infliximab immunogenicity.

All therapies have potential side effects that range from bothersome to serious and life-threatening. Anti-TNFα therapy is no exception, and concerns include acute and delayed infusion reactions, psoriasis, neutropenia, demyelinating disease, autoimmune disease and worsening of congestive heart failure. While infliximab use may increase the risk of serious infection in patients with Crohn’s disease, in adults risk is even higher with the use of prednisone or narcotics [22]. Cancer risk with either infliximab or adalimumab monotherapy is not greater than the general population, while combination therapy of either agent with a thiopurine does appear to increase the risk [23,24].

Why use early anti-TNFα therapy?

Recognizing the long-term risk of undertreated Crohn’s disease, particularly in young individuals with phenotypically severe disease, the author believes the time has come to adopt anti-TNFα therapy either at diagnosis or within the first 3 months for these individuals. Shorter disease duration has been shown to be associated with a better response to anti-TNFα therapy [25], and waiting until disease behavior becomes complicated (stricturing or penetrating) rather than when the disease is largely inflammatory is associated with a higher likelihood of surgery [26]. The term ‘conventional’ therapy should be discarded as it implies that therapy beyond corticosteroids, 5-aminosalicylates and immunomodulators such as thiopurine and methotrexate is unconventional. Symptom relief is no longer sufficient to define success, and the use of biologic therapy has raised the bar for expectations. Still unresolved is whether combination therapy of anti-TNFα plus an immunomodulator is more effective and durable than anti-TNFα monotherapy with close monitoring of drug levels to optimize dosing [27]. The latter approach would be presumed to be associated with less risk than with combination therapy. Better risk stratification models as well as well-controlled natural history studies will be invaluable in helping patients and their families assess benefits and risks of all approaches. These models can then be used to design prospective studies comparing long-term (5–10 years) outcomes using different initial treatment regimens.

Financial & competing interests disclosure

JS Hyams has received research support, been on the advisory board and part of the speaker’s bureau for Janssen. He has been a consultant for Soligenix, TNI Biotech and Takeda; and he has received research support and been on the advisory board for Abbvie. The author have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.