“Look before you leap, but remember, one who hesitates is lost.”
The inflammatory bowel diseases (IBD) Crohn disease and ulcerative colitis are chronic, idiopathic, autoinflammatory conditions. An era of “designer” therapy has emerged with the increasing ability to target various cytokines that are involved in the associated immune dysregulated state. Tumor necrosis factor-α (TNF-α) is now understood to be a key cytokine involved in the associated inflammatory pathways and infliximab, a chimeric immunoglobulin G1 monoclonal antibody directed against TNF-α, has emerged as an effective therapeutic agent for IBD (1).
Infliximab was approved by the US Food and Drug Administration (FDA) in 1998 for adult patients with moderate to severe Crohn disease who were unresponsive to conventional therapy. Pediatric approval for the same indication occurred in May 2006. In October 2006 the FDA expanded its approval further to the treatment of adults with moderate to severe ulcerative colitis. Along with these approvals, infliximab become the first specifically targeted, FDA-approved drug for IBD. Infliximab also became the first vehicle used to widely introduce the concept of biological therapy to clinicians and their patients with IBD. It is safe to say that in many ways, infliximab has forever changed the therapeutic landscape regarding the management of IBD. Now, instead of being satisfied with short-term improvement of clinical symptoms, concepts that were considered theoretical such as mucosal healing are becoming potential endpoints for clinical trials. Although some of our senior colleagues may remember a time when pediatric Crohn disease resulted in a premature mortality of 13% (2), current therapeutic emphasis is moving toward the improvement of our patients' quality of life while reducing complications such as hospitalization and surgery (3).
It is instructive to understand why the use of infliximab in pediatric IBD has been embraced relatively quickly. Simply stated, pediatric gastroenterologists recognized that there was a therapeutic gap for many of their patients with IBD. Although “conventional therapy” often worked well, there was still that group of children and adolescents with IBD who went on to have a disease course complicated by the need for recurrent hospitalizations and early surgical interventions while growth failure persisted (4). For this group of patients, a highly effective therapeutic agent had been long awaited. Almost from the onset of its prescription availability, the benefit of infliximab administration was perceived as being great, whereas the risk of withholding the medication seemed even greater, especially in severely affected individuals. Nothing underscores this fact more than the manufacturer's estimates that by the time infliximab received pediatric approval in the spring of 2006, more than 10,000 children and adolescents had already been exposed to this drug. Whether all of these pediatric patients actually needed infliximab can be debated. However, as an emerging agent for whom the ideal strategy of use has yet to be fully determined, infliximab use became more of a question of the philosophy and art of practice rather than evidence-based medicine.
The story of biologics in IBD treatment should be viewed as a cautionary one. We should not be surprised that high potential efficacy comes at a cost because there is never a therapeutic “free lunch.” Thus, what lessons can we learn from the unfolding tale? Lesson 1 is having respect for the unknown. When one practices on the cutting edge, there needs to be the recognition of the possibility of getting cut. Although a new drug can show marked efficacy, the limited size and controlled nature of premarketing clinical trials means that only the most common adverse events will be observed, and rare side effects will only be discovered over time after the drug becomes available in the marketplace and is used widely. For example, not long after drugs against TNF-α started being used in the clinical setting, it was noted that the inhibition of TNF-α had an uncanny ability to unleash infections, especially tuberculosis (5). More recently, warnings were issued by the manufacturer of infliximab regarding an association with a rare and all but uniformly fatal form of lymphoma known as hepatosplenic T cell lymphoma (HSTCL) (6). In this issue of the Journal, Mackey et al (7) describe the first 8 cases reported to the FDA of HSTCL occurring in patients with IBD who had received infliximab and who, notably, were also exposed to a purine analogue.
As a result of these emerging reports and warnings, the decision of who should receive a new drug becomes an important one for the practicing clinician and his or her patients. The decision-making process that precedes treatment initiation must include the reality of known and as yet unknown side effects. In an ideal world, we should be able to provide patients and their families with the necessary data to make their own risk–benefit analysis when it comes to starting a recently approved agent (8). It is well known that this is typically not the reality, however, because the data themselves are often difficult, even for clinicians, to decipher. None of us has the crystal ball that will enable us to predict how a particular patient will be affected by a medication. Therefore, the importance of remembering the potential for unknown side effects of a new or emerging therapy is pivotal to lesson 1.
As in the sports world, timing is everything when discussing new therapies. Lesson 2 is the realization that the perception of risk for an agent is completely dependent on what is already known about the medication. Had cases of HSTCL occurred during the clinical trials of infliximab, a very different story may have evolved concerning its use. A comparison with the emerging agent natalizumab provides an instructive example. Three cases of progressive multifocal leukoencephalopathy occurred during the clinical trials of this promising agent leading to its voluntary recall (9). Subsequent study has shown no additional cases in the clinical trial cohort (10). Because of its marked efficacy in multiple sclerosis, natalizumab has returned to the market with a greater respect for its potential toxicity. Translating this experience to infliximab, the question that emerges is whether infliximab would have survived early reports of a rare and fatal cancer. This question would especially loom large if such reports occurred before the full efficacy of the drug was appreciated by a wide professional and lay audience. Although we do not know what would have happened, speculation is such that HSTCL during the clinical trials may have relegated infliximab to a similar place in history as the use of thalidomide during pregnancy and the Ford Pinto for gas tank design.
Lesson 3 deals with the importance of reporting—the need for the rapid and accurate collection and distribution of adverse event data once a new drug is commercially available. This lesson calls out loudly from the pages of the article by Mackey et al. Certainly, the most concerning aspect of the HSTCL association is the all-but-incurable nature of this malignancy (11). The association between this emerging therapy and a rare yet dire malignancy leaves us wondering: what is the etiology? What is the true degree of risk? Is the underlying etiology linked to the infliximab, the purine analogues, the combination of these therapies, or some as yet unidentified factor? Although we do not know more, the simple reporting of these adverse events has allowed the pursuit of answers to begin.
In the United States, MedWatch was launched by the FDA in 1993 (http://www.fda.gov/medwatch) to educate health professionals about the importance of monitoring for adverse events and to facilitate direct, voluntary reporting of adverse events to the FDA. Once a significant adverse event occurs, the clinician needs to complete a MedWatch form and report the event to the FDA. The appropriate pharmaceutical company is then alerted by the FDA. Accordingly, every practitioner must see him- or herself as an integral part of the “pharmacovigilance system.” Suspicion of a possible association rather than proof of causality is sufficient reason to report. When a drug manufacturer is made aware of an adverse event, they are required by federal law to notify the FDA within 15 d. During that time, due diligence must be performed to confirm the event and the surrounding details. Of course, the FDA is not necessarily interested in obtaining a report on every adverse event encountered because this could lead to reporting volumes that would be difficult for the organization to handle. Rather, the FDA is particularly interested in serious adverse events such as life-threatening events, death, congenital anomaly, disability, hospitalization, and surgical intervention. Even now, when issues of patient confidentially are highly visible on all of our radar screens, this is a recognized process that must occur so that true risks can be determined using the added power of data generated by the postmarketing use of the agent. Thus, lesson 3 is the take-home message that we can glean from the article by Mackey et al—health providers need to put aside any fears of letting the world know that they had “a bad case” and promptly report a serious adverse event, so that true rates of risk can be established. Were it not for this type of reporting, the multiple cases of HSTCL may never have come to light.
In conclusion, it is probably best to recognize that the HSTCL story is ongoing; we will know more next year than we do now. Clearly, recognition of this black hole in our knowledge base will create a degree of anxiety in all clinicians who care about their patients. We know infliximab works for IBD, but we still do not know the ideal strategy for the use of this agent. Mackey and her colleagues remind us that we do not know everything about potential serious adverse events or if there is anything that we can do to decrease that risk. With so many critical questions for which there are no data to provide a quality answer, how do we “look before we leap” and proceed with our patients? We hope that every patient who has received infliximab to date was made aware that in no way could all of the potential side effects of this new agent be known, but that a therapeutic decision was being made based upon the best clinical evidence of the day. We also realize that there are situations in which hesitating by not providing a new agent to a patient would in itself lead to a significant adverse outcome. Although time is likely the commodity in which we hold the fewest shares, we must take the time to talk to our patients and their families about the potential for known and unknown adverse drug events whether an intervention is new or as old as leeches. Providing them with this type of information will help empower our patients and their families to make reasonable risk–benefit decisions for themselves, and, just as is the case for the clinician, these decisions will undoubtedly be based upon past experience, severity of illness, and perceived level of risk.
We are entering increasingly exciting times for the management of the IBDs. New therapies are here and newer ones are coming. The science of testing these therapies in clinical trials will continue, as will the art of practicing patient care when it comes to deciding who will be exposed to a given agent. While we do so, we owe it to every patient with IBD to rapidly and fully report serious potential adverse events that develop when treating him or her with these emerging new agents. Only with such open reporting can we develop a meaningful risk–benefit analysis that can help reduce suffering and save lives.
The authors would like to thank Drs Jeffrey Hyams and Jim Markowitz for their helpful discussions and review of the manuscript.
1. Van Deventer SJ. Targeting TNF alpha as a key cytokine in the inflammatory process of Crohn's disease—the mechanisms of action of infliximab. Aliment Pharmacol Ther 1999; 13(Suppl 4):3–8.
2. Puntis J, McNeish AS, Allan RN. Long term prognosis of Crohn's disease with onset in childhood and adolescence. Gut 1984; 25:329–336.
3. Otley AR, Griffiths A, Hale S, et al
. Health-related quality of life in the first year following a diagnosis of pediatric inflammatory bowel disease. Inflamm Bowel Dis 2006; 12:684–691.
4. Sawczenko A, Ballinger AB, Croft NM, et al
. Adult height in patients with early onset of Crohn's disease. Gut 2003; 52:454–455.
5. Keane J, Gershon S, Wise RP, et al
. Tuberculosis associated with infliximab, a tumor necrosis factor alpha-neutralizing agent. N Engl J Med 2001; 245:1098–1104.
6. Infliximab [package insert]. Malvern, PA: Centocor; 2006.
7. Mackey AC, Green L, Liang L, et al
. Hepatosplenic T cell lymphoma associated with infliximab use in young patients treated for inflammatory bowel disease. J Pediatr Gastroenterol Nutr 2007; 44:265–267.
8. Sands B, Siegel C, Hass S, et al
. Crohn's disease patients' willingness to accept the risks of serious adverse events in exchange for clinical benefits. Gastroenterology 2006; 130(4):A656.
9. Adelman B, Sandrock A, Panzara MA. Natalizumab and progressive multifocal leukoencephalopathy. N Engl J Med 2005; 353:432–433.
10. Yousry TA, Major EO, Ryschkewitsch C. Evaluation of patients treated with natalizumab for progressive multifocal leukoencephalopathy. N Engl J Med 2006; 354:924–933.
11. Domm JA, Thompson M, Kuttesch JF, et al
. Allogeneic bone marrow transplantation for chemotherapy-refractory hepatosplenic gammadelta T-cell lymphoma: case report and review of the literature. J Pediatr Hematol Oncol 2005; 11:607–610.