Preemptive HLA Genotyping for the Safe Use of Infliximab-combination Therapy in Inflammatory Bowel Disease

Inflammatory bowel disease (IBD) affects over 250,000 individuals in Canada. It is comprised of ulcerative colitis (UC) and Crohn's disease (CD). The dysregulated inflammatory response targeting the gastrointestinal (GI) tract is the hallmark of IBD and can lead to significant physical and psychological morbidity amongst affected individuals. Hallmarks of the disease include hematochezia, diarrhea, and abdominal pain. Individuals with IBD are committed to long‐term immunosuppressive therapy to drive disease into remission; however, such treatments are associated with significant cost, as well as, a risk for significant drug‐related toxicities. Individuals who are resistant or lose response to traditional therapies may require hospitalization and intestinal resection or colectomy. This is also associated with significant costs to the health care system and to patients. The last decade has seen an expansion in the number of therapies, specifically monoclonal antibodies (biologics), available for the treatment of IBD, targeting and inhibiting different proteins involved in perpetuating the inappropriate inflammatory response. There is growing evidence to support the use of biologics early in the disease course, bypassing other less effective and older treatments. In Canada there are currently five biologic agents approved for the management of IBD: infliximab, adalimumab, golimumab, vedolizumab and ustekinumab. Infliximab, the first biologic approved for the management of IBD in Canada and the most widely used, is a chimeric human‐murine monoclonal antibody directed against the pro‐inflammatory cytokine, tumour necrosis factor‐α (TNF). The efficacy of infliximab in CD and UC has been demonstrated in landmark trials; ACCENT and ACT respectively. It is considered a standard of care for moderate to severe IBD in treatment algorithms. Unfortunately, up to 40% of patients who initially respond to a TNF antagonist such as infliximab will lose response by the one‐year mark. Additionally, up to 23% of individuals with IBD exposed to infliximab will have an immediate infusion reaction with flushing, urticaria, presyncope and dyspnea necessitating treatment cessation. A leading contributor to both loss of response and infusion reactions is the development of anti‐drug antibodies (ADAs). ADAs are a consequence of the "immunogenicity" of TNF antagonists. Immunogenicity refers to the immune response of the exposed individual against large molecule therapeutic proteins such as infliximab. The underlying mechanisms of immunogenicity in TNF antagonist‐exposed IBD patients are poorly defined. Clinically, ADAs are very relevant to IBD treatment as some ADAs can inhibit drug function or induce hypersensitivity in exposed patients. Studies have shown that the presence of ADAs correlates with a loss of response to infliximab as well as with a high risk of infusion reaction. Having the capacity to identify individuals at high risk of ADA formation and apply targeted combination therapy to those individuals and avoid combination therapy in others would be exceedingly valuable in clinical practice. Thus, we propose to assess the utility of preemptively screening patients with IBD who are being considered for infliximab therapy for HLADQA1*05A>G and applying co‐immunosuppression with an immunomodulator (methotrexate or azathioprine) to the variant carriers (AG or GG) compared to those received the current standard of care. We will assess the resultant impact on infliximab ADA formation in addition to highly relevant clinical outcomes such as infliximab loss of response, treatment discontinuation, and adverse drug events. Therapeutic drug monitoring, the ability to measure ADAs, in addition to serum drug concentrations, has revolutionized IBD treatment algorithms by providing objective evidence to inform clinical decision‐making. Unfortunately, the current tools are only able to identify ADAs once they have developed and thus, treatment adjustments are reactive as opposed to preemptive. Patients are often only screened for ADAs once loss of response or a hypersensitivity reaction have occured. One way clinicians attempt to reduce the risk of ADA formation is to empirically combine a second immune‐suppressing agent such as methotrexate or azathioprine (immunomodulators) with infliximab. The addition of an immunomodulator to infliximab‐based therapy (combination therapy) is associated with reduced ADA formation. The downside is that combination therapy may be associated with an increased risk of infection, malignancy and other side effects related to the immunomodulator (pancreatitis, myelotoxicity, hepatotoxicity). There is also concern over the use of dual immunosuppression in certain patient populations, including frail elderly or patients at high risk of infection or malignancy. Currently, there are no clinical tools that predict who will develop ADAs, lose response to or have a hypersensitivity reaction to infliximab. Additionally, there are few ways to predict the risk of adverse events in IBD patients treated with combination therapy. Recently, in an peer‐reviewed dataset, a group demonstrated that variation in the class 2 human leukocyte antigen (HLA) gene region (HLADQA1*05A>G, rs2097432) is linked to an increased risk of ADA formation against infliximab and to a lesser extent, its sister TNF‐antagonist, adalimumab18. In a separate, retrospective study, we have confirmed that variation in HLADQA1*05A>G (rs2097432) is independently‐associated with a significantly higher incidence of and faster progression to infliximab ADA formation. Moreover, we demonstrated that variant carriers had a higher risk of infliximab loss of response, treatment discontinuation as well as a faster progression to these outcomes (Wilson et.al. 2019 unpublished/Gastro, submitted). Interestingly, the addition of co‐immunosuppression (methotrexate or azathioprine) to infliximab therapy reduced the risk of antibody formation in variant carriers compared to that of an individual with a wild type genotype.