Rationale: A wide range of serum trough concentrations is observed in tocilizumab-treated rheumatoid arthritis (RA) patients, while 1 mg/L tocilizumab is sufficient to block systemic interleukin-6 receptor. A substantial proportion of patients has higher serum tocilizumab concentrations and is likely to be overexposed. We expect that patients can at least reduce the dose aiming for a concentration of 5 mg/L without reducing efficacy.

Objective: To evaluate the feasibility of the study after 20 weeks of follow-up, this includes the evaluation of the dose-reduction algorithm in tocilizumab-treated patients with RA.

Study design: Double-blind randomized controlled pilot study with a follow up of 20 weeks.

Study population: Consecutive RA patients that are treated with tocilizumab intravenously every four weeks for at least 24 weeks. Patients are screened for tocilizumab concentration after signing informed consent.

Intervention: Patients with a concentration below 5 mg/L will continue the dose. Those patients with a tocilizumab concentration above 5 mg/L are randomly assigned (2:1) to dose reduction or to continuation of the standard care tocilizumab dose. In the intervention group, the precise dose-reduction is calculated per patient in order to achieve a tocilizumab concentration of 5 mg/L (range 4-6 mg/L).

Main study parameters/endpoints:

The feasibility of the study logistics is evaluated according to the dropout rate and patients opinion about the study. Second, the proportion of patients achieving the targeted tocilizumab concentration after dose reduction is evaluated.

Nature and extent of the burden and risks associated with participation, benefit and group relatedness:

Dose-reduction will lead to lower drug costs and possibly to reduce the risk of adverse events. Since we lower the tocilizumab concentration in a proportion of the patients, risk of a exacerbation of the disease exists. In this case, patients will receive their original dose. Previous studies showed that disease activity is controlled adequately after returning to the standard dose. However, our algorithm is designed to reach concentrations of 5 mg/L (range 4-6 mg/L) and studies showed that 1 mg/L of tocilizumab is sufficient to maintain clinical effect. The expected burden of this study is low, since study visits are planned at the time of infusion and therefore do not take extra time. The additional burden consists of an extra blood sample taken every visit and the fingerprick that is performed once.

INTRODUCTION AND RATIONALE Rheumatoid arthritis (RA) is a chronic inflammatory disease with a symmetric polyarthritis. One of the cytokines playing a role in RA is interleukin-6 (IL-6)(1;2). Tocilizumab is a humanized monoclonal antibody targeting IL-6 receptor (IL-6R) and is proved to be effective in reducing inflammation and symptoms in RA. Tocilizumab intravenously (IV) is approved by the European Medical agency (EMA) in a standard dose of 8 mg/kg every 4 weeks for RA patients. The FDA approved a standard dose of 4 mg/kg every 4 weeks, with a possibility to increase the dose to 8 mg/kg. This raises the question whether a proportion of patients treated with 8 mg/kg are overexposed.

Previously, it has been showed that 8 mg/kg results in a wide range of serum tocilizumab concentrations, although serum concentrations above 1 mg/L are reported to be sufficient to inhibit IL-6 signaling and to normalize CRP levels (CRP ≤ 10 mg/L), suggesting a systemic IL-6R blockade (3;4). In a previous study, the median tocilizumab concentration was 10.6 mg/L after 24 weeks of treatment, suggesting that a substantial proportion of patients is likely to be overexposed to tocilizumab (4). A recent study showed that overexposure, defined by high serum drug concentration, can be reduced in adalimumab-treated RA patients by prolonging the dose-interval (5). The wide variation in tocilizumab concentrations and sufficient target blockade at a fixed serum trough concentration suggests that a similar approach to reduce overexposure is possible in tocilizumab treatment too. Moreover, Levi and colleagues showed improvements in the disease activity score using 28 joints (DAS28) were associated with a C50 (concentration at which the 50% of the effect is achieved) between 2-5 µg/ml. (8) We, therefore, expect that patients can reduce the dose aiming at a concentration of 5 mg/L without reducing efficacy. Around 70% of the patients had a tocilizumab concentration above 5 mg/L in a previous study and can potentially reduce the dose (3).

Reducing overexposure is necessary in light of high drug costs and, possibly, the risk of adverse events. Therefore, our aim is to study whether overexposure can be reduced without affecting disease activity in tocilizumab-treated RA patients. We have developed an algorithm to reduce the dose in order to achieve a tocilizumab concentration of 5 mg/L in each patient.

OBJECTIVES Primary objective: To evaluate the feasibility of reducing the dose of tocilizumab according to serum trough levels at baseline, after 20 weeks follow-up.

Secondary objective: To evaluate the dose-reduction algorithm in tocilizumab-treated RA patients with tocilizumab concentrations above 5 mg/L after 20 weeks of follow-up.

STUDY DESIGN Double-blind randomized controlled pilot study in RA patients treated with tocilizumab IV every four weeks for at least 24 weeks. Patients will receive their tocilizumab infusion at the outpatient clinic of the department of Rheumatology in Tel Aviv Sourasky Medical center. The treatment will be at the discretion of treating rheumatologist; patients can receive tocilizumab treatment as monotherapy or in combination with concomitant DMARD therapy or prednisone. After given informed consent, tocilizumab trough concentration will be determined and those patients with concentration above 5 mg/L will be randomly assigned (2:1 ratio) to dose reduction or to continuation of their tocilizumab dose. Patients with concentrations below 5 mg/L will continue the dose they received at the screening visit. All included patients will have a follow up of 20 weeks independent of the tocilizumab concentration, treatment arm or medication adjustments. Clinical data are collected at the baseline visit and at 12, and 20 weeks thereafter. Blood samples are collected every infusion, namely at baseline and at 4, 8, 12, 16 and 20 weeks thereafter.

The study visit will take place at the same day of the tocilizumab infusion. After performing the clinical/ laboratory assessments and infusion at 20 weeks of follow up, the allocation will be revealed. This allows the treating rheumatologist, in consultation with the patient, to decide whether to continue with the dose given during the study phase or to change the dose.

Intervention Those patients with tocilizumab concentration above 5 mg/L, that are assigned to the intervention group, will reduce the tocilizumab dose at the baseline visit. A dose-reduction algorithm is developed, using a population pharmacokinetic (PK) model, to calculate the dose needed for each patient to achieve a tocilizumab concentration of 5 mg/L (range 4-6 mg/L). As is shown in Table 1, the dose-reduction is dependent of the tocilizumab concentration measured at the screenings visit. Patients are followed for 20 weeks to evaluate the serum concentration after several infusions.

Table 1. Dose-reduction algorithm Tocilizumab concentration*, mg/L >30 ≥20 <30 ≥ 15 <20 ≥10 <15 ≥5 <10 Dose reduction‡, % 60 50 40 30 15

*Serum trough concentration measured at the screening visit. ‡Percentage of dose-reduction compared to prescribed dose at screening visit.

5. TREATMENT OF SUBJECTS 5.1 Investigational treatment Those patients that are allocated to the intervention group will reduce the dose at the baseline visit (week 0) and maintain that dose till 20 weeks of follow-up. The reduced dose is dependent of the tocilizumab serum concentration, measured at the screening visit, and is calculated according to the dose-reduction algorithm (as described in the "study design").

5.2 Escape medication Treatment can be changed in case of flare (defined in subheading "flare criteria" in study procedures). In this case, the original dose is given to the patient, without unblinding the patients and treating rheumatologist. If the flare persists after 4 weeks, treatment will be adjusted in consultation with the treating rheumatologist. This includes; increasing DMARDs or NSAIDs dosage, adding glucocorticoids or switching to another biologic/anti-rheumatic drug. If the patient or physician wants to reinstate the original dose, but the flare criteria are not met, this will be logged as protocol violation.

6. METHODS 6.1 Randomisation, blinding and treatment allocation Randomization allocation ratio will be 2:1 (therapeutic drug monitoring versus standard care). Randomization procedure: block randomization with variable block sizes (2 to 4 per block) generated by an online program. Variable blocks are used to ensure that the allocation of participants will not be predictable when the treating rheumatologist and participant are unblinded after the study. During the study, patients, physicians, nurses and researchers are blinded for drug concentrations and treatment allocation. The infusions for the study will be prepared by the, designated departmental nurse based on the randomization number. The physical appearance of the medication will be the same to enable blinding of participating patients and study staff. Allocation concealment: randomization list is an excel document managed by an independent co-worker of Reade and a backup independent co-worker.

Study procedures At screening patients will be checked for eligibility, informed consent will be signed and, afterwards, tocilizumab concentration will be determined. The assessments listed below will be performed at baseline (week 0), and 12 and 20 weeks thereafter. Blood samples are collected every infusion, namely at baseline and at 4, 8, 12, 16 and 20 weeks thereafter. In case of a (suspected) flare, an extra visit will be planned to perform the same assessments as during the regular visits Disease activity and functioning Swollen and tender joint count of 66/68 joints, patient pain visual analogue score(VAS) (0-100 scale), physician global VAS (0-100 scale), patient global disease activity VAS (0-100 scale). Functional status will be assessed with the simplified Health Assessment Questionnaire (sHAQ), which is included in the routine assessment of patient index data 3 (RAPID-3). DAS28-ESR, SDAI and CDAI will be calculated every visit.

Safety Patients will be asked for (serious) adverse events (AEs) during the past 3 months. This will be recorded by the examiner on an already designated questionnaire form. Nature, seriousness, duration, localization and consequences of the events will be recorded.

Laboratory investigations Erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), complete blood count (Hb, Ht, white cell count, platelets), creatinine, alanine aminotransferase (ALT) will be measured. Blood samples for serum (5 ml) are collected prior to each infusion and 15 minutes afterwards, in order to measure the tocilizumab concentration at trough and at peak moment, respectively. Tocilizumab concentrations (except screening samples) are measured in one batch at the end of the study.

Other variables Other variables recorded at baseline are: age, gender, duration of disease, immunoglobulin M (IgM) rheumatoid factor and anti-cyclic citrullinated peptide status, presence of erosive disease, length and weight, current medication, medication history regarding DMARD and NSAID therapy, prednisone and biological use. Patients perspective towards therapeutic drug monitoring (TDM) and current study procedures will be asked and noted during the first and final visit.

Flare criteria A flare is defined as an increase in DAS28 between visits of >1.2 (or >0.6 if current DAS28≥3.2) and/or DMARDs is increased, or a DMARD and/or triamcinolone treatment is added (both intramuscular or intra-articular). Clinical and laboratory assessment will be performed at the moment a patient is examined for a flare.

Medication:

Current medication use will be asked for each visit, with special attention to changes in DMARD or NSAID therapy and the use of glucocorticoids. It is aimed to leave all rheumatic co-medication unaltered during the study period. However, changes in use of paracetamol (acetaminophen), tramadol, NSAIDs, oral glucocorticoids, methylprednisone (Depo-medrol), triamcinolone (kenacort) and DMARDs are allowed during this study if a recent DAS28-score is reported. During each visit, patients are asked about the use of these medications. Changes in co-medication are left to the discretion of the treating rheumatologist, but treatment alterations without recent DAS28 score will be logged as protocol violations.

6.2 Follow-up of subjects withdrawn from treatment Follow up of patients withdrawn from treatment will be part of the study. Patients will be asked to complete 20 weeks follow-up.

6.3 Premature termination of the study The study will be prematurely terminated if safety of patients can no longer be guaranteed.

Inclusion Criteria:

• RA according to the ACR 1987 or 2010 criteria;
• Current use of tocilizumab IV, with a consistent interval of 4 weeks for at least 24 weeks.
• 18 years of age and older.

Exclusion Criteria:

• A potential subject will be excluded from participation in case of a scheduled surgery in the next 20 weeks or other preplanned reasons for treatment discontinuation.
• Children, pregnant women and individuals with a lack of judgement.