Dr. Maini has received consulting fees and honoraria (less than $10,000 each) from Roche and Centocor and has received consulting fees (more than $10,000) from Chugai Pharmaceuticals Ltd. Dr. Maini receives royalties from The Kennedy Institute of Rheumatology Trust for an anti–tumor necrosis factor patent. Dr. Maini is a coinventor of the coadministration of tocilizumab and methotrexate, for which a patent is pending. Dr. Maini previously owned stock in Johnson & Johnson, of which Centocor is a subsidiary. Dr. Maini is Non-Executive Director of Domantis and founder of Synovis Ltd. Dr. Taylor has received honoraria (less than $10,000) from Centocor. Dr. Emery has received consulting fees (less than $10,000) from Roche. Dr. Kishimoto holds a patent for tocilizumab.
Double-blind randomized controlled clinical trial of the interleukin-6 receptor antagonist, tocilizumab, in European patients with rheumatoid arthritis who had an incomplete response to methotrexate
Article first published online: 31 AUG 2006
Copyright © 2006 by the American College of Rheumatology
Arthritis & Rheumatism
Volume 54, Issue 9, pages 2817–2829, September 2006
How to Cite
Maini, R. N., Taylor, P. C., Szechinski, J., Pavelka, K., Bröll, J., Balint, G., Emery, P., Raemen, F., Petersen, J., Smolen, J., Thomson, D. and Kishimoto, T. (2006), Double-blind randomized controlled clinical trial of the interleukin-6 receptor antagonist, tocilizumab, in European patients with rheumatoid arthritis who had an incomplete response to methotrexate. Arthritis & Rheumatism, 54: 2817–2829. doi: 10.1002/art.22033
- Issue published online: 31 AUG 2006
- Article first published online: 31 AUG 2006
- Manuscript Accepted: 11 MAY 2006
- Manuscript Received: 5 OCT 2005
- Chugai Pharmaceuticals Ltd.
To establish the safety and efficacy of repeat infusions of tocilizumab (previously known as MRA), a humanized anti–interleukin-6 (IL-6) receptor antibody, alone and in combination with methotrexate (MTX), for the treatment of rheumatoid arthritis (RA).
The study group comprised 359 patients with active RA in whom the response to MTX was inadequate. During a stabilization period, these patients received their current dose of MTX for at least 4 weeks. Following stabilization, they were randomized to 1 of 7 treatment arms, as follows: tocilizumab at doses of 2 mg/kg, 4 mg/kg, or 8 mg/kg either as monotherapy or in combination with MTX, or MTX plus placebo.
A 20% response (improvement) according to the American College of Rheumatology criteria (ACR20 response) was achieved by 61% and 63% of patients receiving 4 mg/kg and 8 mg/kg of tocilizumab as monotherapy, respectively, and by 63% and 74% of patients receiving those doses of tocilizumab plus MTX, respectively, compared with 41% of patients receiving placebo plus MTX. Statistically significant ACR50 and ACR70 responses were observed in patients receiving combination therapy with either 4 mg/kg or 8 mg/kg of tocilizumab plus MTX (P < 0.05). A dose-related reduction in the Disease Activity Score in 28 joints was observed from week 4 onward, in all patients except those receiving monotherapy with 2 mg/kg of tocilizumab. In the majority of patients who received 8 mg/kg of tocilizumab, the C-reactive protein level/erythrocyte sedimentation rate normalized, while placebo plus MTX had little effect on these laboratory parameters. Tocilizumab was mostly well tolerated, with a safety profile similar to that of other biologic and immunosuppressive therapies. Alanine transaminase and aspartate transaminase levels followed a sawtooth pattern (rising and falling between infusions). There were moderate but reversible increases in the nonfasting total cholesterol and triglyceride levels and reversible reductions in the high-density lipoprotein cholesterol and neutrophil levels. There were 2 cases of sepsis, both of which occurred in patients who were receiving combination therapy with 8 mg/kg of tocilizumab plus MTX.
These results indicate that targeted blockade of IL-6 signaling is a highly efficacious and promising means of decreasing disease activity in RA.
Interleukin-6 (IL-6) is a proinflammatory cytokine that is abundantly expressed in patients with rheumatoid arthritis (RA) and is detectable in the joints and circulation of such patients during active phases of the disease (1–4). IL-6 binds to its soluble and membrane-bound receptors, and their interaction with gp130 transduces intracellular signals that mediate gene activation and a wide range of biologic activities (5). Preclinical and human studies have demonstrated IL-6 activities relevant to RA. These include the differentiation of B cells into immunoglobulin-secreting plasma cells, activation of T cells, induction of acute-phase proteins by hepatocytes, and production of platelets (6–9). IL-6 induces the recruitment of chemokines and leukocytes (10) and has been shown to induce proliferation of synovial fibroblasts (11). In addition, IL-6 has a profound effect on bone and can induce osteoclast differentiation and activation in vitro (12). IL-6 appears also to be involved in damage to cartilage, by decreasing aggrecan protein and type II collagen production by chondrocytes (13). The inflammatory destructive effects of IL-6 have been validated, with data in animals suggesting an important role of IL-6 in the induction and maintenance of chronic synovial inflammation (14–16). IL-6 gene–knockout mice have been demonstrated to be protected from arthritis (17). Because of the key role played by IL-6 in several phenomena typical of RA, IL-6 is a candidate target for therapeutic inhibition as a novel approach to the treatment of RA.
Monoclonal antibodies (infliximab and adalimumab) and a soluble tumor necrosis factor (TNF) receptor fusion protein (etanercept), which inhibit another proinflammatory cytokine TNF, have become established as effective options for the treatment of RA, especially in patients who experience an inadequate response to disease-modifying antirheumatic drugs (DMARDs), including methotrexate (MTX) (18). The efficacy of anti-TNF therapy is increased by concomitant treatment with MTX (19, 20). Nonetheless, 20–40% of patients have an inadequate response to such treatment, and there is an unmet need for new therapeutic interventions with new modes of action for the treatment of RA.
Tocilizumab (previously known as MRA) is one of the molecules in development that offers a new mechanism of action. Tocilizumab, a humanized anti-human IL-6 receptor antibody of the IgG1 subclass, was developed collaboratively by Osaka University and Chugai Pharmaceutical Company Ltd (Japan). Tocilizumab is humanized by the grafting of complementarity-determining regions of a mouse anti-human IL-6 receptor monoclonal antibody onto human IgG1, using recombinant DNA technology. Tocilizumab has been shown to compete for both the membrane-bound and soluble forms of human IL-6 receptor, thus inhibiting the binding of IL-6 to its receptors and its proinflammatory activity. Short-term clinical trials of tocilizumab in adult-onset RA and systemic-onset juvenile idiopathic arthritis have demonstrated acceptable safety and significant efficacy (21–24).
The objective of this multicenter randomized clinical trial (the Chugai Humanized Anti-Human Recombinant Interleukin-6 Monoclonal Antibody [CHARISMA] study) was to establish the safety and efficacy of repeat infusions of tocilizumab, alone and in combination with MTX, in patients with established RA in whom the response to MTX was inadequate and who had continuing disease activity. Tocilizumab was used either as monotherapy (by discontinuation of MTX) or concomitantly with MTX therapy and compared with placebo infusions in patients who continued to receive a fixed dose of MTX during this 20-week study.
PATIENTS AND METHODS
Role of the funding source.
Chugai Pharma Europe (a member of the Roche Group) provided funding for the trial and was responsible for supply of the study medication (active and placebo), for data processing and management, statistical analysis, and consulting via a clinical research organization (Covance), for ongoing medical safety reviews, and for reporting of serious adverse events. Dr. Maini was responsible for the trial design and manuscript preparation. Both Chugai and Roche Pharmaceuticals checked and approved the submitted manuscript.
Patient and treatment protocol.
In this randomized, double-blind, controlled trial, the study group comprised 359 patients fulfilling the American College of Rheumatology (ACR; formerly, the American Rheumatism Association) revised criteria for a diagnosis of RA (25); all patients had active disease, had experienced an inadequate response to MTX, and had been stabilized on their current dose of MTX for at least 4 weeks before randomization. Three groups of patients received intravenous infusions of tocilizumab (2 mg/kg, 4 mg/kg, and 8 mg/kg, respectively) every 4 weeks, plus MTX placebo once weekly. Another 3 groups of patients received intravenous infusions of tocilizumab (2 mg/kg, 4 mg/kg, and 8 mg/kg, respectively) every 4 weeks, plus MTX once weekly. A control group received placebo infusions every 4 weeks and MTX once weekly. Thus, all patients received a total of 4 infusions of tocilizumab or tocilizumab placebo once every 4 weeks, together with 10–25 mg of MTX or MTX placebo each week. End point efficacy evaluations were performed at week 16; followup evaluations for safety were performed at week 20, after which patients were no longer monitored. A total of 359 patients from 57 rheumatology centers in Europe were randomized to participate in this study. Members of the CHARISMA Study Group and the European countries involved are shown in Appendix A.
Randomization was performed centrally. When a patient was eligible for randomization into the study, an interactive voice response system was used to allocate treatment, by determining inclusion into the group that would minimize any imbalances between MTX dose level at baseline and between patients from a center. All patients and investigators were blinded to the study treatments.
Tocilizumab or placebo (an aqueous solution of sucrose and polysorbate 80, resembling the active treatment) was diluted in normal saline and administered by intravenous infusion over 1 hour. Pre-prepared capsules containing 10-mg, 12.5-mg, 15-mg, 17.5-mg, 20-mg, 22.5-mg, and 25-mg doses of MTX were blinded by over-capsulation using a lactose filler; the matching placebo capsules contained only filler.
The MTX dose remained unchanged for at least 4 weeks prior to baseline and throughout the study. Patients at all centers were supplied (separately, from the same source) 5 mg of folic acid weekly, for the duration of the study. Patients randomly allocated to receive tocilizumab monotherapy were switched to MTX placebo on day 0.
Inclusion criteria were a diagnosis of RA according to the ACR revised criteria (25), disease duration of at least 6 months, active disease (defined as ≥6 tender joints and ≥6 swollen joints, based on a 28-joint count), an erythrocyte sedimentation rate (ESR) ≥28 mm/hour, and/or a C-reactive protein (CRP) level ≥1.0 mg/dl. Patients must also have shown an inadequate response to MTX or a disease flare while receiving MTX (at a dosage of 10–25 mg weekly) during a minimum of 6 months of therapy. An inadequate response was defined as the presence of active disease, as described above, despite MTX therapy. If patients required concomitant treatment with nonsteroidal antiinflammatory drugs and/or oral steroids, the dose must have been stable for at least 4 weeks prior to study entry and during the course of the study (for steroids, ≤10 mg prednisolone or equivalent).
Exclusion criteria included leukopenia (white blood cell count <4.0 × 109/liter, absolute neutrophil count <2.0 × 109/liter) and/or thrombocytopenia (platelet count <150 × 109/liter), any hepatic dysfunction as shown by aspartate transaminase (AST) and alanine transaminase (ALT) levels >1.5-fold the upper limit of normal or significant renal impairment (serum creatinine level >1.5-fold the upper limit of normal). Patients who received DMARDs (excluding MTX) within 4 weeks prior to the start of the study were excluded, as were patients who received anti-TNF agents within 12 weeks or leflunomide within 6 months of infusion of study medication.
Ethical approval was obtained from the appropriate authorities in each country, and each patient gave informed consent. The first patient was enrolled on May 14, 2001, and the final patient exited the study on September 3, 2002.
The primary clinical end point was the percentage of patients who achieved improvement of 20% according to the ACR criteria (ACR20 response) at week 16; secondary end points included the ACR50 and ACR70 responses (26). For the analysis of efficacy (i.e., the ACR20 response at 16 weeks), all patients who withdrew before the week 16 assessment because of lack of efficacy were included in the analysis as nonresponders, as per protocol. For patients who withdrew for other reasons, the last recorded assessment of the ACR20 response was carried forward.
Other secondary end points included percent and absolute change from baseline in individual disease activity measures (swollen joint count, tender joint count, physician's global assessment of disease activity, patient's global assessment of disease activity, patient's pain score, ESR, CRP level, and Health Assessment Questionnaire [HAQ] score), as well as changes in the duration of early morning stiffness, and the Disease Activity Score in 28 joints (DAS28) (27). In this study, the DAS28 was calculated using the ESR, according to a previously described formula (28). Remission was defined according to the European League Against Rheumatism (EULAR) definition of a DAS28 score <2.6 (29).
Safety assessments included physical examinations, pre- and postdose electrocardiograms, and laboratory analyses of hematology, serum biochemistry, coagulation, immunologic parameters, and urine. Adverse events were recorded throughout the study. A treatment-emergent adverse event was defined as any adverse event that occurred after commencement of allocated treatment or an adverse event that occurred prior to the allocated treatment but worsened in severity after commencement of the allocated treatment. A treatment-related adverse event was defined as any adverse event whose relationship to the study drug was described as possible, probable, or definite, as documented in the case report forms.
The primary efficacy end point was the ACR20 response at week 16, using an intent-to-treat full-analysis set. All patients who were allocated to receive study medication, had received at least 1 dose, and had postrandomization efficacy data available were included in the full-analysis set. All patients who withdrew because of lack of efficacy before the week 16 assessment were included in the analysis as nonresponders. For patients who withdrew because of other reasons, the last recorded assessment of the ACR20 response was carried forward. Differences between treatment groups in the proportion of responders were assessed using logistic regression. A model was fitted for the treatment groups to which patients were assigned in a center and the stratification of weekly doses of MTX. Analyses similar to the primary efficacy analysis were performed for the secondary efficacy end points.
Treatment groups were formally compared at the week 4, week 8, week 12, week 16, and week 20 assessments, separately; changes from baseline to each assessment were analyzed using analysis of covariance, with an effect for treatment, center, and the respective baseline value fitted as a covariate. The P value for the overall test of difference between treatment groups from this model was determined. If results of the overall test of difference between treatment groups from this model were statistically significant, the linear and quadratic components of the dose-response relationship were assessed. P values for linear, quadratic, and average dose effect were determined for the groups receiving monotherapy and the groups receiving combination therapy.
Baseline characteristics and patient flow.
Three hundred fifty-nine patients were randomly allocated to the 7 treatment groups. Patient flow through the trial and randomization to each treatment arm are shown in Figure 1. Approximately equal numbers of patients were randomly allocated to each arm of the study. During the trial, 60 patients withdrew (34 patients withdrew due to adverse events and/or possible drug-related toxicity). Of the 359 patients randomized to receive study medication, all were included in the safety population, and 354 patients were included in the full-analysis set (5 patients were excluded from the full-analysis set because of a protocol violation). The baseline demographics of the treatment groups were similar, and no statistically significant differences were observed (Table 1).
|Characteristic||MRA monotherapy||Combination therapy, MRA plus MTX||MTX (n = 49)||All groups (n = 359)|
|2 mg/kg (n = 53)||4 mg/kg (n = 54)||8 mg/kg (n = 52)||2 mg/kg (n = 52)||4 mg/kg (n = 49)||8 mg/kg (n = 50)|
|No. of men/no. of women||9/44||13/41||14/38||7/45||12/37||11/39||11/38||77/282|
|Duration of RA, months||9.19||9.79||9.21||9.33||7.82||10.62||11.24||–|
|Tender joint count (28 joints)||15||15||15||15||13||15||16||15|
|Swollen joint count (28 joints)||11||11||12||11||11||11||12||11|
|C-reactive protein, mg/liter||26†||19||22‡||28§||31||24¶||32||27|
|Rheumatoid factor positive, no.||44||39||43||46||38||40||47||–|
|Patient's assessment of pain||58.9†||62.2||58.5‡||59.7§||60.2||57.0¶||63.8||–|
|Duration of MTX therapy, mean ± SD months||33.0 ± 25.5||34.1 ± 26.5||40.0 ± 30.5||38.8 ± 27.1||29.0 ± 17.7||40.4 ± 32.7||33.3 ± 27.8||35.5 ± 27.3|
|Patient's global assessment of disease activity||62.0†||63.6||60.7‡||62.3§||60.9||62.1¶||68.8||–|
|Physician's global assessment of disease activity||61.3†||62.1||59.5‡||59.2§||59.5||61.3¶||66.0||–|
|Duration of early morning stiffness, minutes||90.0†||110.0||120.0†||110.0§||90.0||120.0¶||90.0||–|
All patients had previously been treated with DMARDs, the most common of which was MTX. The mean duration of prior MTX treatment varied across the groups, with the longest duration recorded in the group receiving 8 mg/kg of tocilizumab plus MTX (40 months), and the shortest duration recorded in the group receiving 4 mg/kg of tocilizumab plus MTX (29 months). Most patients had previously been treated with 1–3 different DMARDs, although several patients had previously received more than 5 different DMARDs. Cyclophosphamides had been taken by 1.4% of patients, leflunomide by 6.4%, sulfasalazine by 53.5%, azathioprine by 6.7%, cyclosporine by 20.3%, etanercept by 2.8%, auranofin by 2.8%, chloroquine by 23.4%, gold by 3.1%, hydroxychloroquine by 12.0%, penicillamine by 7.2%, and sodium aurothiomalate by 21.2% of patients. Fifty patients had received TNF inhibitors (infliximab or etanercept) prior to study enrollment. A high proportion of patients in each treatment group was serum rheumatoid factor positive (overall mean 82.7%), and the proportion was highest in the group receiving placebo plus MTX (95.9%).
Primary efficacy end point.
The primary end point of the study, an ACR20 response at week 16, was achieved by 61% and 63% of the patients receiving monotherapy with 4 mg/kg and 8 mg/kg of tocilizumab, respectively, compared with 41% of patients receiving placebo plus MTX (P < 0.05). The responses of the group receiving tocilizumab at a dose of 2 mg/kg were not significantly different from those of the group assigned to receive placebo plus MTX (Figure 2). The ACR20 responses among patients receiving combination therapy with tocilizumab (2 mg/kg, 4 mg/kg, and 8 mg/kg) plus MTX exceeded those among patients receiving placebo plus MTX (64%, 63%, and 74%, respectively, versus 41%; P < 0.05, P < 0.05, and P < 0.001, respectively) (Figure 2).
Secondary efficacy end points.
The ACR50 and ACR70 response rates at week 16 were measured, as shown in Figure 2. Only combination therapy with 8 mg/kg of tocilizumab plus MTX was significantly better than placebo plus MTX in terms of both the ACR50 and the ACR70 (P < 0.05). There was a similar pattern of change in DAS28 scores, showing a clear dose response with tocilizumab and significant differences from week 4 onward, except in the group receiving monotherapy with 2 mg/kg of tocilizumab. The maximum reduction in the DAS28 was achieved with the 8-mg/kg doses. Among patients assigned to the 8-mg/kg dose of tocilizumab in combination with MTX, the mean DAS28 at 16 weeks was 2.9. All groups receiving tocilizumab, alone or in combination with MTX (except for the group assigned to monotherapy with 2 mg/kg of tocilizumab), showed a greater decrease in the DAS28 by week 16 than did the group receiving placebo plus MTX (for combination therapy with 8 mg/kg of tocilizumab and monotherapy with 8 mg/kg of tocilizumab, P < 0.001; for combination therapy with 4 mg/kg of tocilizumab, P < 0.05 versus placebo plus MTX) (Figure 3).
A further analysis was performed to calculate the percentage of patients who achieved remission according to the DAS, in the group receiving 8 mg/kg of tocilizumab plus MTX, the group receiving 8 mg/kg of tocilizumab as monotherapy, and the group receiving placebo plus MTX. The analysis showed that the rate of remission was 34% among those assigned to 8 mg/kg of tocilizumab plus MTX, 17% among those receiving 8 mg/kg of tocilizumab as monotherapy, and 8% among those receiving placebo plus MTX. Among patients responding to tocilizumab treatment, a response was observed by week 4, and this was maintained during the course of the study. The slope of the graph suggested that a peak response had not been reached by the end of the study, and that further improvement could be attained.
Each of the individual clinical components of the ACR response assessment (swollen joint count, tender joint count, physician's global assessment of disease activity, patient's global assessment of disease activity, patient's pain score, and HAQ score), as well as the duration of morning stiffness, showed an improvement at doses of tocilizumab ≥4 mg/kg. However, comparison with the placebo plus MTX control group at week 16 consistently showed that monotherapy with 2 mg/kg of tocilizumab performed less well than did placebo plus MTX. For the swollen joint counts, the mean reduction observed among patients receiving 8 mg/kg of tocilizumab, alone and in combination with MTX, was statistically significant compared with the reduction in the group assigned to placebo plus MTX (P = 0.01 and P < 0.001, respectively). Additionally, the mean reduction in the tender joint count in the group receiving combination therapy with 8 mg/kg of tocilizumab plus MTX was statistically significantly superior compared with the mean reduction in the group assigned to receive placebo plus MTX (P = 0.009).
The mean CRP level and the mean ESR showed a marked decrease over time in all of the patients receiving tocilizumab except those assigned to 2 mg/kg as monotherapy (both the ESR and the CRP level) and those receiving 4 mg/kg as monotherapy (the CRP level only), while placebo plus MTX had little effect on these laboratory parameters (Figure 4). Especially in the groups receiving 4 mg/kg of tocilizumab, as monotherapy or combination therapy, there was a sawtooth pattern, with a decrease in these parameters 2 weeks after dosing (maximum observed biologic effect) and an increase prior to the next infusion. The 8-mg/kg dose of tocilizumab produced a more sustained decrease. The mean CRP level normalized after the first infusions of 8 mg/kg of tocilizumab (as both monotherapy and combination therapy) and 4 mg/kg of tocilizumab as combination therapy, while the mean ESR normalized after the third infusions of tocilizumab at a dose of 8 mg/kg (as both monotherapy and combination therapy). The comparison between 8 mg/kg of tocilizumab, as monotherapy or combination therapy, and placebo plus MTX showed the most significant differences (P ≤ 0.001) at week 16. The mean ESR and the mean CRP level returned toward baseline values within 4–8 weeks after the last infusion of tocilizumab.
Other direct and indirect laboratory markers of inflammation that returned toward baseline levels after tocilizumab treatment included serum ferritin, serum amyloid A, C4 complement, fibrinogen, neutrophils, platelets, and hemoglobin.
In general, tocilizumab was well tolerated in this trial (see Table 2). Approximately half of the patients experienced adverse events after treatment, the majority of which were mild or moderate in intensity, and only one-fourth of which were judged to be treatment-related. The most frequent treatment-emergent adverse events were infections (mixed viral and bacterial), musculoskeletal disorders (all of which were compatible with RA), gastrointestinal disorders (a variety of minor disorders, the incidence of which was higher among those receiving tocilizumab compared with those receiving placebo plus MTX), skin rash, pruritus, and erythema.
|MRA monotherapy||Combination therapy, MRA plus MTX||MTX (n = 49)|
|2 mg/kg (n = 53)||4 mg/kg (n = 54)||8 mg/kg (n = 52)||2 mg/kg (n = 52)||4 mg/kg (n = 49)||8 mg/kg (n = 50)|
|Patients with ≥1 treatment-emergent AE||30||27||31||30||19||27||23|
|Patients with ≥1 treatment-related AE||13||14||15||12||9||14||9|
|Patients who discontinued due to an AE||4||5||5||3||6||6||4|
|Patients with ≥1 serious treatment-emergent AE||8||5||3||4||1||7||2|
|Patients with ≥1 serious treatment-related AE||5||2||1||2||0||4||0|
|Serious treatment-emergent infections†||4||0||0||0||0||3||0|
|Serious treatment-emergent anaphylactic reaction/shock‡||3||1||0||0||0||0||0|
There was no clear pattern of occurrence of adverse events by dose group, except for serious adverse events (Table 2). Thirty patients experienced 35 treatment-emergent serious adverse events. The incidence of treatment-emergent serious adverse events was highest in the group receiving 2 mg/kg of tocilizumab as monotherapy and lowest in the group receiving 4 mg/kg of tocilizumab plus MTX (Table 2). The most common treatment-emergent serious adverse events were aggravated RA (2 cases in the group receiving low-dose combination therapy and 1 case in the group receiving placebo plus MTX), 4 infections in 3 patients in the 2-mg/kg monotherapy group (1 patient with a limb abscess and osteomyelitis and 2 patients with respiratory infections), 3 infections in the 8-mg/kg plus MTX group (including 1 patient with infective arthritis and 2 with sepsis not otherwise specified), and anaphylactic shock/reaction and hypersensitivity (4 cases in the 2-mg/kg monotherapy group and 1 case in the 4-mg/kg monotherapy group). Of the 2 patients for whom sepsis was reported as a serious adverse event, septicemia developed in 1 patient following a Staphylococcus aureus–infected skin wound; in the other patient, septicemia was diagnosed, but there were no signs of focal infection and no organisms were isolated on blood culture, and pyelitis was suspected. That patient had a history of streptococcal septicemia due to a dental abscess (10 months previously).
Additionally, 2 cases of nonserious anaphylactic reactions and hypersensitivity were reported (1 in a patient receiving monotherapy with 4 mg/kg of tocilizumab, and 1 in a patient receiving combination therapy with 2 mg/kg of tocilizumab). Overall, anti-tocilizumab antibodies developed in 25 patients, all of whom were in the groups receiving either 2 mg/kg or 4 mg/kg of tocilizumab as monotherapy; anti-tocilizumab antibodies developed in none of the patients receiving treatment with 8 mg/kg of tocilizumab, as either monotherapy or combination therapy. No cases of reactivation of tuberculosis were seen, and no opportunistic infections were observed in this study.
Liver function test changes.
During the dosing period of this trial, the mean ALT level and, to a lesser extent, the mean AST level increased in all patients receiving tocilizumab, following a sawtooth pattern (rise and fall) between infusions (Figure 4). This increase was tocilizumab-related but was accentuated in patients receiving tocilizumab in combination with MTX (>3-fold the upper limit of normal in 2% of the patients receiving combination therapy). All of the mean values returned to near-baseline values within 8 weeks of the final infusion. The mean ALT level increased in a dose-related manner in the groups receiving tocilizumab monotherapy; the maximum increase from baseline at week 2 (45%) was observed in the group receiving 8 mg/kg of tocilizumab. Larger dose-related increases in the ALT level occurred in the combination-therapy groups; the maximum increase from baseline at week 2 (88%) was observed in the group receiving combination therapy with 8 mg/kg of tocilizumab plus MTX. No increase in the mean ALT level was observed in the group receiving placebo plus MTX. A similar pattern of changes in the mean AST level was seen, but these changes were of lesser magnitude.
Certain patients showed marked but reversible increases in the level of transaminases, which in some cases led to discontinuation of the study treatment. Five patients who had ALT levels >100 IU/liter were withdrawn from the study; all of these patients were in the groups receiving tocilizumab plus MTX. In 4 patients, the ALT level returned to a level within the normal range within 4 weeks of withdrawal. Overall, 127 patients treated with tocilizumab experienced an ALT level greater than the upper limit of normal during the study; 35 of these patients had mildly abnormal ALT values at baseline, compared with 12 patients who were treated with MTX alone. In 112 tocilizumab-treated patients, the ALT level returned to near-baseline values by the time of the followup visit. Eighteen patients had ALT values >100 IU/liter during the study, and all of these patients had been treated with tocilizumab.
Mean total bilirubin values increased gradually during the study period in tocilizumab-treated patients, but this increase was not exacerbated by the addition of MTX. Among the groups receiving tocilizumab monotherapy, the maximum increase from baseline in mean total bilirubin levels was 83% at week 16; this was observed in the group receiving 8 mg/kg of tocilizumab. Among patients receiving combination therapy, the maximum increase in bilirubin levels (59%) was observed at week 14 in the group receiving 4 mg/kg of tocilizumab plus MTX. Again, values had largely returned to normal by the followup visit, and the group receiving placebo plus MTX was unaffected. Twenty-one patients had bilirubin levels >21 μmoles/liter during the study, but these patients had high total bilirubin levels when they entered the study (mean 17 μmoles/liter, which is the upper end of the normal range) and demonstrated predominantly unconjugated hyperbilirubinemia. In 86% of this subgroup, total bilirubin levels returned to near-baseline values by the time of followup. No relationship between elevated ALT levels and elevated bilirubin levels was observed. No individual patient experienced significant elevations in both the ALT level and the bilirubin level simultaneously.
During the study period, moderate but reversible increases in the mean levels of nonfasting total cholesterol, high-density lipoprotein cholesterol, and triglycerides were seen in the groups assigned to receive treatment with tocilizumab. The levels increased initially and then stabilized and did not continue to increase during the treatment period (Figure 4). Importantly, the mean atherogenic index remained largely unchanged despite some rise and fall between infusions in the higher-dose groups. In both of the groups receiving 8 mg/kg of tocilizumab, the atherogenic index was reduced to below its initial level by the 20-week followup visit. Again, certain individuals experienced abnormally high lipid levels, but in the majority of these patients the levels were high at study entry and showed a relatively small increase during treatment.
There was a general dose-dependent reduction in the neutrophil count following treatment with tocilizumab, but the neutrophil count recovered when tocilizumab was withdrawn (Figure 4). Forty patients experienced neutropenia during treatment with tocilizumab. Reductions in the neutrophil counts of 47% and 43%, respectively, were seen at week 14 in patients receiving 8 mg/kg of tocilizumab as monotherapy and in those receiving 8 mg/kg of tocilizumab plus MTX. Of the 13 patients who experienced grade 2 neutropenia (between 1,000 and 1,500 neutrophils/mm3) or grade 3 neutropenia (between 1,000 and 500/mm3) during the dosing period excluding baseline, all had received higher doses of tocilizumab, with or without MTX. The addition of MTX did not seem to exacerbate the effect, based on the relative incidence of neutropenia in the groups receiving tocilizumab as monotherapy or combination therapy. Infections were not associated with reduced neutrophil counts. The lowest neutrophil count observed was 0.88 × 109/liter at week 8 (in patients receiving 8 mg/kg of tocilizumab), and no clear pattern in the occurrence of neutropenia was observed across study visits. Some patients experienced neutropenia after the first infusion of tocilizumab, and some patients experienced neutropenia after later infusions, and there was no evidence that continued dosing worsened the effect. The neutrophil count returned to the normal range by the followup visit in all except 1 patient, in whom the count was abnormally low at baseline.
The number of patients with low neutrophil counts was related to the dose of tocilizumab administered (0, 5, and 14 patients in the 2-mg/kg, 4-mg/kg, and 8-mg/kg tocilizumab groups, respectively). The proportion of patients who experienced infections was no different between those with low neutrophil counts (7 of 40 patients) and those with normal neutrophil counts (62 of 359 patients).
Previously, a monoclonal antibody to the IL-6 molecule was shown to induce improvements in patients with RA. However, the effects were short-lived, and neutralizing antibodies developed against this murine antibody (30). Clinical studies in Japan, in which tocilizumab was used as monotherapy (4 mg/kg or 8 mg/kg) in adults who had previously received DMARDs (including MTX at a dosage less than 8 mg/week) (22, 23), and a dose-ranging study in Europe in which patients received a single dose of either 0.1 mg/kg, 1 mg/kg, 5 mg/kg, or 10 mg/kg of tocilizumab (21) have provided proof of concept that IL-6 blockade by tocilizumab is efficacious in the treatment of RA. Based on these data and those from our study, it may be concluded that IL-6 is an important driver of inflammation in RA. The blockade by tocilizumab of the membrane-bound as well as the soluble IL-6 receptor thus appears to be a direct and manipulable method for achieving the target of reducing the pathologic effects of excess IL-6.
The results of this study clearly show that infusions of tocilizumab every 4 weeks, with or without background MTX therapy, can produce marked and dose-related improvement in RA disease activity as measured by ACR20, ACR50, and ACR70 responses and clinically meaningful changes in the DAS28. The study also showed that the 4-mg/kg and 8-mg/kg doses of tocilizumab were associated with high percentages of ACR50 and ACR70 responses after only 4 infusions (the difference between patients receiving placebo plus MTX and those receiving 8 mg/kg of tocilizumab in combination with MTX was statistically significant) (Figure 2). Assessment of efficacy over time showed continued improvement in all dose groups. This improvement was seen up to the end of the 16-week trial, and there was an indication that maximum efficacy may not have been achieved in this short-term study.
The mean DAS28 among patients receiving combination therapy with 8 mg/kg of tocilizumab plus MTX approached the EULAR criteria for remission (DAS28 <2.6). The primary and secondary efficacy end points for tocilizumab therapy at the highest dose demonstrate that the efficacy of this therapy is in the same range as that of the currently available anti-TNF therapies (18). Thirty-seven percent of the patients in this high-dose group also achieved an ACR70 response, and 34% achieved EULAR-defined remission based on the DAS28.
One feature of note in this study is that a relatively high proportion of patients in the group receiving placebo plus MTX also continued to respond. The ACR20 response rate of 41% in this study contrasts with a rate of 11% among placebo-treated patients in the Japanese study (23). Because the Japanese study compared a placebo infusion with treatment that commenced after a DMARD washout period, and patients in our study received a placebo infusion during continuation of MTX therapy, the high response rate associated with placebo likely indicates that our patients were not MTX nonresponders, but rather that they had not yet fully responded to MTX at the time of trial entry. In the current study, the response to placebo plus MTX was high, and a better control group might have been produced if the period of time during which controls received stable doses of MTX prior to trial entry had been longer than 4 weeks. The study design required that background MTX be withdrawn abruptly at the start of the study in the groups receiving tocilizumab monotherapy, thus placing these groups at a relative disadvantage in terms of efficacy in a population of incomplete MTX responders. This may have contributed to the negative comparison between the group receiving 2 mg/kg of tocilizumab and those receiving placebo plus MTX and to the apparent superiority of combination therapy over monotherapy at each dose level. However, the unexpectedly high rate of ACR20 responses in the group of patients who received MTX alone has no effect on the numbers of patients who achieved ACR50 or ACR70 responses while receiving tocilizumab plus MTX.
The durability of the response to tocilizumab was not investigated in this study; however, data from a phase II study performed in Japan demonstrated that with continued treatment, efficacy (as measured by ACR20, ACR50, and ACR70 responses) was maintained at least until week 72 (22). These data were obtained from an open-label extension study in which patients received tocilizumab at a dosage of 8 mg/kg every 4 weeks for an unrestricted period of time (22).
The most important safety signals in this study were related to changes in the results of liver function tests and serum lipid levels. A dose-related increase in the mean serum transaminase level was seen during tocilizumab treatment. This change appeared to be drug-related, due to the sawtooth pattern of rise and fall between infusions, most of which were observed after the first infusion of tocilizumab. More importantly, certain individuals experienced clinically significant increases in the transaminase level. The addition of MTX appeared to accentuate this effect. Mean bilirubin levels showed a gradual elevation over the study period. It is clear that the pattern of change for the mean bilirubin level was not the same as that for the mean transaminase level, and larger individual rises in the bilirubin level were not observed in the same patients as those experiencing rises in the levels of transaminases or lipids. Also, the total bilirubin elevations were mild (<3-fold the upper limit of normal). In a retrospective analysis of 8 patients with high levels of bilirubin, the increase appeared to be primarily in the unconjugated bilirubin fraction and possibly associated with underlying Gilbert's syndrome. Hemolysis may possibly have contributed to the bilirubin level, but there are no data supporting this hypothesis, which needs to be investigated in the future.
There was no evidence of frank clinical hepatitis or predictors of the occurrence of hepatitis in any of the patients who had elevated levels of ALT and bilirubin. Based on a review of individual patients at each time point, there was no temporal correlation between ALT, total bilirubin, and cholesterol and clinical symptoms consistent with hepatitis. No patient had any 2 of the 3 laboratory abnormalities at the same time. In the Japanese extension study, bilirubin levels did reach a plateau in patients treated for up to 72 weeks.
Treatment with tocilizumab also resulted in an increase in the total cholesterol level as well as an increase in the level of triglycerides and HDL cholesterol (in this study, low-density lipoprotein cholesterol was not measured). At the same time, the mean atherogenic index was unchanged. As with bilirubin, changes in the cholesterol level were not temporally related to changes in the ALT level; however, there was a temporal correlation between total cholesterol levels and the CRP levels (CRP levels were dramatically reduced). The degree of elevation of the cholesterol level may be related to the degree of suppression of inflammation (31). It has also been observed that use of the anti-TNF monoclonal antibody infliximab to treat RA leads to an elevated cholesterol level (32). Increases in plasma lipid levels and/or CRP levels are associated with an increased risk of cardiovascular events (33, 34), and a major cause of death in patients with RA is heart disease (35). The significance of the increases in lipid levels and the decreases in CRP levels observed in this study is unclear (especially because the lipid samples were nonfasting), and further studies are clearly indicated to investigate the clinical significance of these effects on the morbidity and mortality associated with ischemic heart disease. In addition, the mechanism of this lipid level elevation following DMARD treatment in general and tocilizumab in particular is unclear and should be further evaluated.
Favorable effects of tocilizumab were also demonstrated by other direct and indirect laboratory indicators of inflammatory disease (levels of serum ferritin, serum amyloid A, complement C4, fibrinogen, rheumatoid factor, and serum albumin, the neutrophil count, the platelet count, and the hemoglobin concentration), suggesting that suppression of inflammation by tocilizumab is not associated with only a few of the markers of inflammation.
A small number of patients in whom the neutrophil count was normal at the start of the study (unlike the majority of the study population) experienced neutropenia when treated with tocilizumab at a dose of 8 mg/kg. An increased incidence of infections, including life-threatening conditions, has been observed with other monoclonal antibodies targeting the components of the immune system. In this study, serious infections were seen in the group receiving combination therapy with 8 mg/kg of tocilizumab plus MTX, as might be expected with the combination of 2 potentially immunosuppressant drugs, although serious infections were also seen in the group receiving monotherapy with 2 mg/kg of tocilizumab. As the most clinically effective dose, 8 mg/kg of tocilizumab may be considered the optimal therapeutic dose. Although the evidence clearly is not conclusive, IL-6 is known to demarginate neutrophils (36), and inhibition of IL-6 function could be expected to marginate neutrophils, thus showing a decrease in the number of circulating neutrophils. Larger studies in which the period of exposure to the study drug is longer will be required to accurately assess the risk of serious infection associated with tocilizumab treatment.
Anaphylaxis and anaphylactoid reactions are also expected and serious adverse events associated with treatment with humanized monoclonal antibodies. In the case of tocilizumab, as with infliximab, such reactions occurred only at low doses and in the absence of MTX, against a background of “insufficient” immunosuppression, or high-dose antigen–induced immunologic tolerance may have occurred to prevent formation of anti-tocilizumab antibodies (19). Among patients in whom anti-tocilizumab antibodies developed, the number was highest in the groups receiving monotherapy, which suggests that MTX provides some protection against anti-tocilizumab antibody production. This trend is also seen when MTX is given in combination with infliximab (19).
The highest responses to tocilizumab monotherapy have been achieved using the 4-mg/kg (minimally efficacious) and 8-mg/kg (optimally efficacious) doses; therefore, those doses are proposed for use in future clinical studies. Monthly administration of 8 mg/kg of tocilizumab is proposed as the highest dosage, because in this study both efficacy and safety appeared to be dose-related, and increasing the dose beyond 8 mg/kg may increase the possibility that adverse events will occur. Combination therapy with tocilizumab plus MTX demonstrated superior efficacy compared with tocilizumab monotherapy.
In conclusion, the results of this study indicate that targeted blockade of IL-6 is a highly efficacious and promising means of decreasing disease activity in patients with RA, and that tocilizumab treatment (as monotherapy or in combination with MTX) is well tolerated in the majority of cases, having a safety pattern consistent with that of other biologic and immunosuppressive therapies. Large-scale phase III studies are currently ongoing in an attempt to validate the encouraging hypothesis generated in this and other studies, and to investigate the potentially beneficial radiographic outcomes related to treatment with tocilizumab in patients with RA.
We thank David Andrews for assistance with preparation of the figures and manuscript.
- 5Interleukin 6 and its receptor: ten years later. Int Rev Immunol 2004; 16: 249–84..
- 7B cell stimulating factor 2/interleukin 6 is a costimulant for human thymocytes and T lymphocytes. J Exp Med 1989; 557: 417–35., , , , , , et al.
- 9Human interleukin 6 is a direct promoter of maturation of megakaryocytes in vitro. Proc Natl Acad Sci U S A 1989; 8: 3546–9., , , , , .
- 13JAK/STAT but not ERK1/ERK2 pathway mediates interleukin (IL)-6/soluble IL-6R down-regulation of Type II collagen, aggrecan core, and link protein transcription in articular chondrocytes: association with a down-regulation of SOX9 expression. J Biol Chem 2003; 278: 2903–12., , , .
- 20TEMPO (Trial of Etanercept and Methotrexate with Radiographic Patient Outcomes) study investigators. Therapeutic effect of the combination of etanercept and methotrexate compared with each treatment alone in patients with rheumatoid arthritis: double-blind randomised controlled trial. Lancet 2004; 363: 675–81., , , , , , et al, and the
- 21Therapeutic benefit of blocking interleukin-6 activity with an anti–interleukin-6 receptor monoclonal antibody in rheumatoid arthritis: a randomized, double-blind, placebo-controlled, dose-escalation trial. Arthritis Rheum 2002; 46: 3143–50., , , , , , et al.
- 22Long-term safety and efficacy of anti–interleukin 6 receptor antibody (MRA) in patients with rheumatoid arthritis [abstract]. Arthritis Rheum 2003; 48 Suppl 9: S126., , , , , , et al.
THE CHARISMA STUDY GROUP
Members of the CHARISMA Study Group, in addition to the authors of this article, are as follows: Dr. S. Augustinova (Czech Republic), Dr. R. Bernstein (UK), Dr. M. C. Boisier (France), Dr. B. Bourke (UK), Dr. S. Bowman (UK), Dr. H. Brabcova (Czech Republic), Dr. F. C. Breedveld (The Netherlands), Dr. G. R. Burmester (Germany), Dr. E. Choy (UK), Dr. L. Czirjak (Hungary), Dr. J. P. Devogelaer (Belgium), Dr. C. Edwards (UK), Dr. A. Geusens (Belgium), Dr. H. Haentzschel (Germany), Dr. M. Hakala (Finland), Dr. P. J. Hedin (Sweden), Dr. K. Horslev-Petersen (Denmark), Dr. I. L. Johannesen (Denmark), Dr. J. Kalden (Germany), Dr. B. B. La Cour (Denmark), Dr. M. Leirisalo-Repo (Finland), Dr. T. Lorenzen (Denmark), Dr. M. Malaise (Belgium), Dr. K. Mikkelsen (Norway), Dr. N. D. Peters (Denmark), Dr. P. Prouse (UK), Dr. J. Sany (France), Dr. M. Schattenkircher (Germany), Dr. M. Schou (Denmark), Dr. P. Sheldon (UK), Dr. S. Sierakowski (Poland), Dr. K. Sirova (Czech Republic), Dr. A. K. L. So (Switzerland), Dr. H. Sorensen (Germany), Dr. Z. Szekanecz (Hungary), Dr. L. Szcezpanski (Poland), Dr. G. Thamsborg (Denmark), Dr. W. Tlustochowicz (Poland), Dr. H. P. Tony (Germany), Dr. D. Walker (UK), Dr. R. Westhovens (Belgium), Dr. J. Wojtulewski (UK), Dr. E. Veys (Belgium), Dr. P. Vitek (Czech Republic), Dr. R. Van Vollenhoven (Sweden), Dr. D. Zarowny (Poland).