Therapeutic drug monitoring of infliximab compared to standard clinical treatment with infliximab: study protocol for a randomised, controlled, open, parallel-group, phase IV study (the NOR-DRUM study)

Infliximab (INX) and other tumor necrosis factor inhibitors (TNFi) have revolutionised the treatment of a range of prevalent chronic immune mediated inflammatory diseases and remission has become an achievable treatment goal. Unfortunately, more than half of the patients either do not respond sufficiently to therapy or lose efficacy over time [1‐8]. A failure to achieve or maintain disease control has a major impact on patients’ quality of life and puts the individual at risk of developing irreversible organ damage and disability. Loss of clinical efficacy may be due to subtherapeutic drug levels [9‐11]. Methods for measurement of serum drug concentrations of INX and other biological drugs have recently become available for use in clinical practice. In patients on standard doses of INX, significant interindividual variations in serum drug levels, ranging from undetectable to significantly above the presumed therapeutic range, have been revealed [9‐11]. One major reason for this interindividual variation is the development of anti-drug antibodies (ADAb) which occur in 10%–60% of patients on INX [9‐11]. The INX biosimilar CT-P13 has a similar immunogenicity profile to the innovator INX, and ADAb to INX are cross-reactive with CT-P13 [12‐14]. Low levels of ADAb may be transient, but high levels influence the pharmacokinetics of the drug and decrease serum concentrations [9‐11]. The presence of ADAb may also be associated with serious side effects of INX such as hypersensitivity reactions [9‐11].

In an effort to optimise efficacy, clinicians often intensify the INX treatment by increasing the dose or decreasing the interval between infusions [8, 15, 16]. Large cohort studies show that up to 50% of patients have had one or more dose escalations within the first year of treatment [8, 15, 16]. The utility of empiric dose escalation is probably minimal in patients with high drug levels or ADAb, explaining the conflicting results regarding its effectiveness [15, 16]. Furthermore, the economic consequences of dosage increase is considerable given the high cost of INX.

With this background, therapeutic drug monitoring (TDM), an individualised treatment strategy based on systematic assessments of serum drug concentrations, has been proposed as a clinical tool to optimise efficacy, patient safety and cost-effectiveness of TNFi [17]. A treatment strategy based on TDM may improve INX therapy by minimising undertreatment, by reducing overtreatment, by allowing early identification of ADAb development, by enabling the detection of treatment failures, by timely discontinuation of ineffective treatment and by prevention of hypersensitivity reactions. In addition, TDM may also aid in choosing the next drug in patients where therapy has failed.

Although the exact therapeutic window of INX has yet to be clearly defined, a trough level > 3 μg/mL during maintenance therapy and > 20 μg/mL during the induction period has been associated with improved clinical outcomes in several observational studies and post hoc analyses of clinical trials across different diseases [18‐26]. TDM has gained great interest within gastroenterology over recent years [27‐29] and has now also been put on the research agenda in international rheumatology [30]. The European League Against Rheumatism (EULAR) has recently established a Task Force named “Therapeutic Drug Monitoring of Biopharmaceuticals in Rheumatology,” aiming to further develop the research agenda within this field. Supported by observational data [18‐26] and clinical experience, TDM of INX treatment has already been implemented in clinical practice in some Norwegian and other European centres with available methodology and special interest in immunogenicity. Only two studies, however, have assessed the clinical effectiveness of TDM in a randomised controlled setting, both in inflammatory bowel disease (IBD) [24, 31]. In the TAXIT trial (Trough Level Adapted Infliximab Treatment study), dose optimisation increased the percentage of patients with Crohn’s disease (CD) in remission, but failed to show effectiveness of TDM during maintenance treatment [24]. TDM of INX treatment was also evaluated in the TAILORIX (A Randomised Controlled Trial Investigating Tailored Treatment With Infliximab for Active Luminal Crohn’s Disease) randomised controlled trial (RCT) [31]. In this trial, patients were randomised to either an algorithm based on clinical symptoms, biomarkers and TDM or to symptom-based management. No differences in the proportion of patients in steroid-free clinical and endoscopic remission (primary endpoint) were shown.

Data from RCTs to support guidelines and recommendations for implementation of TDM in standard care of patients on treatment with INX and other biological drugs are highly needed in an era of increasing use of therapeutic monoclonal antibodies. The main aim of the NORwegian DRUg Monitoring study (NOR-DRUM) is to assess the effectiveness of TDM, both in achieving remission in patients starting INX treatment (part A) as well as in maintaining disease control in patients on INX treatment (part B).

The clinical role of TDM of TNFi and other biopharmaceuticals used for immune mediated inflammatory disorders is still debated. Currently, implementation is based on clinical experience and observational studies, since only a few clinical trials have been conducted [24, 31]. To our knowledge, the NOR-DRUM study is the first RCT assessing the effectiveness and safety of TDM in INX treatment across a range of immune-mediated inflammatory disorders.

NOR-DRUM is investigator initiated and fully funded by a joint grant from the four Norwegian regional health authorities. The study is conducted as a shared effort by Norwegian rheumatologists, gastroenterologists and dermatologists, and 21 study centres across Norway participate in the data collection. Norway is particularly well suited to conduct this large scale trial on TDM. Due to a tender-based prescription system and to the early implementation of biosimilars, INX has, in recent years, been the preferred TNFi for Norwegian clinicians treating patients with immune-mediated inflammatory diseases. Additionally, the recent completion of the large NOR-SWITCH trial [32] has established a strong research collaboration between the different specialties using TNFi and good logistics for conducting clinical trials at the infusion units nationwide. Finally, the Department of Medical Biochemistry at Oslo University Hospital is a non-commercial, high-capability facility for antibody analyses and has developed assays for serum drug measurement of biological drugs and antidrug antibodies while performing these analyses on a self-cost basis.

The implementation of TDM depends on a validated therapeutic target range. The treatment algorithms in NOR-DRUM are based on an extensive literature review and expert opinions. They have been developed through a series of meetings in the project group consisting of national leading experts in this field (both clinicians experienced with TDM and laboratory physicians) and with additional input from international key experts in the scientific advisory board. The therapeutic level of INX is not definitely known for all the diseases, but there are strong indications that the lower limit is close to 3 μg/mL [20‐25]. According to the literature review and expert opinion, the upper limit has been set to 8 μg/mL. The borders of the proposed therapeutic range, the yellow zones in Figs. 4 and 5, allow for some clinical considerations regarding the INX dosing. In the induction phase, the limits of 20 μg/mL at infusion 2 and 15 μg/mL at infusion 3 are based on clinical observations and previous literature [26, 33]. There is still no consensus whether dose adjustments or interval changes are the most effective and cost-effective way to increase or decrease the INX dose. Initial pharmacokinetic modelling suggested that a higher trough level could be achieved using less INX over time by shortening the interval instead of increasing the dose [34]. More recent studies suggest that a dose of, for example, 10 mg/kg every eight weeks is probably equal to 5 mg/kg every four weeks [35], and halving the infusion intervals are not superior to increasing dose when it comes to both effect and drug costs [36]. The proposed algorithms allows for both options, but due to lower drug costs in recent years, patient convenience and high costs of running infusion units, the preferred option is dose increase by increasing each infusion dose and dose decrease by increasing the infusion interval.

The NOR-DRUM study has an open-label design with no blinding of the patient or the study personnel. The disease activity measures included in the primary outcome are composite measures of patient-reported outcomes (PROs,) objective assessments and investigators assessments. Blinded outcome assessment was not considered feasible in this study, as it would have required both blinding of the patient and study personnel. Blinding of the study participants has not been possible as the participants in the intervention group experience changes in the dosing intervals based on serum drug levels. Feasible data collection within the frame of daily clinical practice at a large number of study centres has been important in this trial. Blinding of study personnel who do the outcome assessments would have required two sets of personnel at each site, making the study unfeasible. The open label design is a recognised weakness of the NOR-DRUM trial. Dose adjustments may occur in both arms though it is based on the study algorithm in the intervention arm only. Patients in both arms are treated with the same biological agent with standardised frequency of follow-up and the same management of complications. Sensitivity analyses will be performed to compare results of subjective and objective outcome measures.

The study is not powered to demonstrate superiority within each diagnostic group as inclusion of the required number of patients is not achievable in Norway within a reasonable timeframe and funding resources. As such, the primary endpoint is designed to evaluate the occurrence of remission (NOR-DRUM A) or disease worsening (NOR-DRUM B) across diseases. The choice of these generic endpoints are based on well-established measures of disease activity and predefined cut-offs for disease state and change for each diagnosis. The feasibility of these outcomes in a clinical-trial setting was tested successfully in the NOR-SWITCH trial [32]. The study is powered to detect a difference in remission rate of 15% (NOR-DRUM A) and a decrease in proportion of patients with disease worsening of 12.5% (NOR-DRUM B), which are both considered clinically meaningful differences.

In addition to assessing the effectiveness of TDM (the primary objective), we will also address safety, i.e. occurrence of infusion reactions and infections, and the effect of TDM on cost-effectiveness. Important exploratory objectives are predictions of immunogenicity. i.e. by genetic testing.

This large national multicentre trial assessing the effectiveness, safety and cost-effectiveness of TDM in patients treated with INX is expected to provide valuable information that will hopefully contribute to a reduction in the burden of disease in a range of common chronic diseases with a potentially disabling disease course, as well as a reduction of the high expenses related to biological therapy. The impact of TDM of biological therapy is currently a topic of great interest to clinicians, both nationally and internationally. Being the first trial to assess the effect of TDM in patients with a wide range of immune mediated inflammatory diseases on treatment with a TNF inhibitor, we hope the NOR-DRUM study will contribute to evidence-based implementation of TDM in standard care of patients treated with INX and other biological drugs.

Protocol version 1.2; December 2017.

Recruitment of patients started 1 January 2017 and will continue until the last patient has been included in NOR-DRUM B (expected 1 January 2020). Last patient’s visit in NOR-DRUM A was in October 2019. Last patient’s visit in NOR-DRUM B is estimated to be January 2021.