The last decade has witnessed significant expansion in our understanding of inflammatory bowel disease (IBD) therapies and their pharmacological characteristics. This improved knowledge, along with the paradigm shift in IBD management — evolving from merely symptom control to blocking disease progression and preventing bowel damage — has resulted in the emergence of several treatment concepts aiming to maximize the therapeutic value of each agent. Therapeutic drug monitoring is one of these promising concepts. Its main premise is to personalize treatment in each patient based on the individual drug pharmacokinetic data.
In the case of biologic therapies, therapeutic drug monitoring involves measurement of the trough drug levels and anti-drug antibodies (ADAbs). Assuming the thresholds for the therapeutic trough levels of each drug at the different stages of treatment are well-established, like in organ transplantation, knowing the drug level and the status of the ADAbs can potentially have great value in objectively guiding treatment decisions. Unfortunately, while the concept sounds very attractive, the data supporting and guiding its use in patients with IBD are very limited. Nevertheless, despite the limited data, therapeutic drug monitoring continues to be widely utilized in IBD clinical practice and has further evolved into two debatable strategic approaches, “reactive therapeutic drug monitoring” in patients with apparent therapy failure and “proactive therapeutic drug monitoring” in patients with quiescent disease. In the next few paragraphs, we will try to make the case for reactive therapeutic drug monitoring.
To set the stage for discussion, it is always important to remember the overall treatment goal for IBD patients. A combined target of both clinical and endoscopic remission has been recently suggested as the optimal therapy goal for patients with ulcerative colitis and Crohn’s disease.1 Biomarkers, such as C-reactive protein (CRP) and fecal calprotectin, are advocated as adjunct targets. Drug levels have not been explored as targets for treatment, but are generally considered as tools to help in achieving the actual target of clinical and endoscopic remission. In patients with apparent therapy failure, reactive therapeutic drug monitoring can objectively assess the underlying mechanism of failure and guide the next step in management. Based on therapeutic drug monitoring, drug failure can be classified into:
1) Mechanistic failure, with therapeutic drug level and undetectable ADAbs.
2) Non-immune mediated pharmacokinetic failure, with subtherapeutic drug level and undetectable ADAbs.
3) immune-mediated pharmacokinetic failure, with subtherapeutic drug level and detectable ADAbs.
A different management strategy is proposed for each case scenario; switching to a drug out of class (different mechanism of action) for patients with mechanistic failure; dose escalation of the same drug for patients with non-immunemediated pharmacokinetic failure; and finally, for patients with immunemediated pharmacokinetic failure, an addition of an immunomodulator agent to optimize the index therapy in case of low ADAbs titer or switching to a different drug within the same class are proposed. In the recently published AGA clinical guideline on this issue, the evidence supporting reactive therapeutic drug monitoring was judged to be very low.2 However, the proposed alternative to the therapeutic drug monitoring-guided approach, empirically escalating the dose, can potentially delay the use of appropriate therapy in approximately 60 to 80 percent of patients who have underlying mechanistic drug failure or immune-mediated drug failure. Likewise, empirically switching therapy can result in a missed opportunity of optimizing a potentially effective drug in approximately 9 to 37 percent of patients with evidence of non-immune mediated pharmacokinetic failure.3-5
In patients with apparent therapy failure, reactive therapeutic drug monitoring can objectively assess the underlying mechanism of failure and guide the next step in management.
The main potential downside of the therapeutic drug monitoring-guided approach is the lack of well-defined thresholds for therapeutic trough levels and ADAb concentrations for our currently approved biological therapies in IBD, which may lead to misclassification and therapy changes in patients who may still benefit from dose escalation. For example, in the randomized control trial by Steenholdt et al., therapeutic drug monitoring-guided therapeutic approach was compared to empiric dose escalation in Crohn’s disease patients with secondary failure of response to infliximab. A threshold of 0.5 μg/ml was used to define therapeutic drug level.6 This low threshold may explain why a large proportion of patients (70 percent) in the therapeutic drug monitoring-guided arm were classified as having a mechanistic failure and were treated by switching to out of class agent as opposed to dose escalation which would be the appropriate action if a higher threshold was utilized. Nevertheless, the risk-benefit profile would still favor the therapeutic drug monitoring-guided approach in patients with active disease, as it can identify the less controversial clinical situations that have clear therapy plan, e.g. patients with very high ADAbs titer and undetectable level or patients with high trough level and no detectable ADAbs. Furthermore, our understanding of the appropriate thresholds continues to improve, reducing the risk of misclassifications and ensuring more appropriate actions.
Making the case for proactive therapeutic drug monitoring is more challenging. The proposed approach is to proactively monitor drug levels in patients with quiescent disease and make therapy changes based on these levels, i.e. targeting drug levels. In theory, this approach may reduce cost by identifying patients with high trough levels who may be candidates for dose reduction. It may also reduce chances of disease flare and increase the durability with low trough levels and low titer of ADAbs. The problem is that the evidence supporting this notion is very scarce and with many unanswered questions — we may cause more harm. For instance, the lack of well-established thresholds for the trough levels is more detrimental in proactive compared to reactive therapeutic drug monitoring as it may result in prematurely changing therapy in patients who were otherwise tolerating their medications and doing clinically well. It is also not clear whether we should apply therapeutic drug monitoring before each dose or periodically and if so, when is the best time to obtain therapeutic drug monitoring — during the induction phase, maintenance phase or both — and what are the thresholds for therapeutic trough levels at each stage? Finally, what is the cost burden associated with each approach? With all the uncertainties, routine proactive therapeutic drug monitoring cannot be generally recommended and well controlled prospective therapeutic drug monitoring studies are urgently needed.
In conclusion, while being proactive is always more attractive, until we are able to accurately define the appropriate trough and ADAbs thresholds for each biologic agent at the different stages of therapy and have higher quality data supporting its role, routine proactive TDM cannot be recommended. Conversely, reactive TDM can be utilized as a tool to assist in decision making in patients with apparent failure of therapy keeping in mind the gap in knowledge and the limitations of the current evidence.
Dr. Abdalla has no conflicts to disclose. Dr. Abdalla is a member of the AGA Trainee and Early Career Committee and the AGA International Committee.
Dr. Herfarth is a consultant for Merck, Celltrion, Lycera and Pfizer.
1. Peyrin-Biroulet, L., Sandborn, W., Sands, B.E. et al, Selecting Therapeutic Targets in Inflammatory Bowel Disease (STRIDE): Determining Therapeutic Goals for Treat-to-Target. Am J Gastroenterol. 2015;110(9): p. 1324-38.
2. Vande Casteele, N., Herfarth, H., Katz, J., Falck-Ytter, Y., Singh, S. American Gastroenterological Association Institute Technical Review on the Role of Therapeutic Drug Monitoring in the Management of Inflammatory Bowel Diseases. Gastroenterology. 2017; 153(3): p. 835-857.e6.
3. Paul, S., Del Tedesco, E., Marotte, H. et al, Therapeutic drug monitoring of infliximab and mucosal healing in inflammatory bowel disease: a prospective study. Inflamm Bowel Dis. 2013; 19(12): p. 2568-76.
4. Roblin, X., Marotte, H., Rinaudo, M. et al, Association between pharmacokinetics of adalimumab and mucosal healing in patients with inflammatory bowel diseases. Clin Gastroenterol Hepatol. 2014;12(1): p. 80-84.e2.
5. Yanai, H., Lichtenstein, L., Assa, A. et al, Levels of drug and antidrug antibodies are associated with outcome of interventions after loss of response to infliximab or adalimumab. Clin Gastroenterol Hepatol. 2015; 13(3): p.522-530.e2.