1. Top of page
  2. Abstract
  6. Acknowledgements


To examine the efficacy and safety of different rituximab doses plus methotrexate (MTX), with or without glucocorticoids, in patients with active rheumatoid arthritis (RA) resistant to disease-modifying antirheumatic drugs (DMARDs), including biologic agents.


A total of 465 patients were randomized into 9 treatment groups: 3 rituximab groups (placebo [n = 149], 500 mg [n = 124], or 1,000 mg [n = 192] on days 1 and 15) each also taking either placebo glucocorticoids, intravenous methylprednisolone premedication, or intravenous methylprednisolone premedication plus oral prednisone for 2 weeks. All patients received MTX (10–25 mg/week); no other DMARDs were permitted.


Significantly more patients who received 2 500-mg or 2 1,000-mg infusions of rituximab met the American College of Rheumatology 20% improvement criteria (achieved an ACR20 response) at week 24 (55% and 54%, respectively) compared with placebo (28%; P < 0.0001). ACR50 responses were achieved by 33%, 34%, and 13% of patients, respectively (P < 0.001), and ACR70 responses were achieved by 13%, 20%, and 5% of patients (P < 0.05). Changes in the Disease Activity Score in 28 joints (−1.79, −2.05, −0.67; P < 0.0001) and moderate to good responses on the European League Against Rheumatism criteria (P < 0.0001) reflected the ACR criteria responses. Glucocorticoids did not contribute significantly to the primary efficacy end point, ACR20 response at 24 weeks. Intravenous glucocorticoid premedication reduced the frequency and intensity of first infusion–associated events; oral glucocorticoids conferred no additional safety benefit. Rituximab was well tolerated; the type and severity of infections was similar to those for placebo.


Both rituximab doses were effective and well tolerated when added to MTX therapy in patients with active RA. The primary end point (ACR20 response) was independent of glucocorticoids, although intravenous glucocorticoid premedication improved tolerability during the first rituximab infusion.

Although cytokine-specific biologic agents have improved the treatment of rheumatoid arthritis (RA), some patients have an inadequate response to such therapies and, in clinical practice, most have only a partial response (1). The frequency of remission has remained <50% for patients with established RA, many of whom display “moderate” disease activity, according to their Disease Activity Score (DAS) (2). In order to achieve additional significant gains in RA therapy, new targets need to be identified. One such target is the B cell (3).

Rituximab is a monoclonal antibody that selectively depletes CD20+ B cells via 3 putative mechanisms: antibody-dependent cell-mediated cytotoxicity, complement-dependent cytotoxicity, and promotion of CD20+ B cell apoptosis (4–6). CD20 is not expressed on stem cells or plasma cells (7, 8); this is consistent with the observation that B cell recovery and Ig production were not compromised after a single course of rituximab (9).

Rituximab has shown efficacy in the treatment of CD20+ B cell non-Hodgkin's lymphoma (NHL) (10). More recently, a role for rituximab and B cell modulation has been investigated in rheumatoid factor (RF)–positive patients with active RA. In a phase IIa study, 2 1,000-mg infusions of rituximab (combined with methotrexate [MTX] and a 2-week treatment with glucocorticoids) significantly improved RA symptoms for up to 1 year compared with MTX alone (9). Moderate to good responses according to the European League Against Rheumatism (EULAR) criteria were also evident at 2 years in 33% of the rituximab plus MTX group (11).

To date, no studies have investigated the influence of the rituximab dosage or the role of glucocorticoids in the clinical efficacy and/or tolerability of rituximab regimens in RA. The objective of this study was therefore to examine these variables in patients with active RA who had an inadequate response to disease-modifying antirheumatic drugs (DMARDs) other than MTX or biologic agents, and who currently had an inadequate clinical response to MTX.


  1. Top of page
  2. Abstract
  6. Acknowledgements

Role of the funding source.

This study was funded by Roche, and Roche scientists contributed to the study design, collection of data, and data analysis. The manuscript was written by Dr. Emery. The agreement to submit the manuscript and the approval of the content were made jointly by the authors.

Study design.

This phase IIb, randomized, double-blind, double-dummy, placebo-controlled, international multifactorial trial of 9 different treatment regimens examined 2 dosage levels (and rituximab placebo) of both rituximab and glucocorticoids in a 3 × 3 configuration (Figure 1) with concomitant MTX in patients with active RA. Patients were evaluated at 4-week intervals between baseline and week 24.

thumbnail image

Figure 1. Study design. All groups received weekly methotrexate 10–25 mg parenterally or orally. a = 100 mg methylprednisolone administered intravenously (IV) prior to rituximab infusions on days 1 and 15. b = 100 mg methylprednisolone administered IV prior to rituximab infusions on days 1 and 15, and prednisone administered orally at 60 mg on days 2–7 and at 30 mg on days 8–14. c = primary end point: percentage of patients who met the American College of Rheumatology 20% improvement criteria at week 24.

Download figure to PowerPoint


Outpatients were eligible for study if they were between 18 and 80 years of age and had presented at least 6 months prior to randomization with moderate or severe RA (diagnosed according to the American College of Rheumatology [ACR; formerly, the American Rheumatism Association] revised criteria [22]) despite ongoing treatment with MTX at a dosage of 10–25 mg/week (orally or parenterally) for at least 12 weeks before randomization, with a stable dosage during the last 4 weeks. Active disease was defined as a swollen and tender joint count ≥8 and either an erythrocyte sedimentation rate ≥28 mm/hour or a C-reactive protein (CRP) serum level ≥1.5 mg/dl.

Patients must have failed prior treatment, manifesting as a lack or loss of response to treatment with at least 1 but not more than 5 DMARDs (other than MTX) and/or biologic response modifiers. Patients discontinued DMARDs (except MTX) and biologic therapy at least 4 weeks before randomization and discontinued infliximab, adalimumab, or leflunomide at least 8 weeks before randomization. Patients who were already taking oral glucocorticoids (≤10 mg/day of prednisone or equivalent) were included if the dosage had been stable for at least 4 weeks before trial entry, and patients who had received intraarticular or parenteral glucocorticoids were included, provided that the most recent administration was >4 weeks before screening. Concomitant treatment with nonsteroidal antiinflammatory drugs was permitted if the dosage had been stable for at least 2 weeks prior to entry. Concomitant treatment with any DMARD (other than MTX), anti–tumor necrosis factor α, or other biologic therapy was prohibited.

Patients were excluded if they had significant systemic involvement secondary to RA, evidence of significant other illnesses or laboratory abnormalities, a history of severe allergic or anaphylactic reactions to humanized or murine monoclonal antibodies, or previous treatment with rituximab or any lymphocyte-depleting therapies. In addition, patients were excluded if they had a history of recurrent significant infection.

The study was performed in accordance with the Declaration of Helsinki. All participating sites received approval from their governing institutional review board (or equivalent). See Appendix A for additional members of the Dose-Ranging Assessment: International Clinical Evaluation of Rituximab in Rheumatoid Arthritis (DANCER) Study Group and their locations. All patients provided written informed consent. The primary study population for efficacy analysis was composed of RF-positive patients. A limited number of RF-negative patients were also included for an initial exploratory analysis of the safety of rituximab in this group.

Study treatment.

Rituximab was administered by intravenous (IV) infusion in RF-positive patients: placebo, 500 mg or 1,000 mg on days 1 and 15 (total dose 0 mg, 1,000 mg, and 2,000 mg). Glucocorticoids were administered as placebo methylprednisolone, given IV 30–60 minutes before the infusion of rituximab (or rituximab placebo) on days 1 and 15, premedication methylprednisolone 100 mg, given IV on days 1 and 15 (250 mg prednisone equivalent), or premedication methylprednisolone 100 mg, given IV on days 1 and 15 plus 60 mg of oral prednisone on days 2–7 and 30 mg on days 8–14 (total glucocorticoid dose 820 mg prednisone equivalent) (Table 1). RF-negative patients received either placebo or rituximab (2 1,000-mg infusions), with or without glucocorticoids. All patients received a weekly regimen of MTX (10–25 mg orally or parenterally) with folate (≥5 mg/week).

Table 1. Randomization of trial medication and disposition of patients at 24 weeks*
Rituximab placebo, RF+/RF−Rituximab 2 500-mg doses, RF+/RF−Rituximab 2 1,000-mg doses, RF+/RF−
  • *

    RF = rheumatoid factor; IV = intravenous.

  • Rituximab or rituximab placebo was administered as 2 infusions: 1 on day 1 and 1 on day 15. All patients received 10–25 mg of methotrexate orally or parenterally on a weekly basis.

  • Methylprednisolone 100 mg IV on days 1 and 15.

  • §

    Methylprednisolone 100 mg IV on days 1 and 15 plus oral prednisone 60 mg on days 2–7 and 30 mg on days 8–14 (total oral prednisone dose 570 mg).

  • Patients received at least 1 dose of rituximab.

  • #

    In the rituximab placebo and rituximab groups receiving 2 1,000-mg infusions, 4 patients (3%) and 4 patients (2%), respectively, refused treatment.

Glucocorticoid placebo42/2141/043/22
IV glucocorticoid premedication42/041/042/20
IV glucocorticoid premedication + oral glucocorticoid§44/042/043/22
Total no. of patients randomized149124192
 Patients withdrawn, no. (%)52 (35)11 (9)27 (14)
  Insufficient response46 (31)8 (6)16 (8)
  Adverse events03 (2)6 (3)
  All other reasons#6 (4)05 (3)
 Completed 24 weeks, no. (%)97 (65)113 (91)165 (86)

Trial end points.

The primary end point of the trial was the proportion of RF-positive patients who met the ACR 20% improvement criteria (achieved an ACR20 response) (defined as a 20% improvement in the tender and swollen joint counts and a 20% improvement in 3 of the 5 other ACR improvement criteria) (13) at week 24. Secondary and exploratory analyses examined group differences in ACR50 and ACR70 responses, DAS 28-joint assessment for swelling and tenderness (DAS28) (14), EULAR responses (15), fatigue, as measured by the Functional Assessment of Chronic Illness Therapy–Fatigue (FACIT-F) subscale (16), and the effect on individual parameters of the ACR improvement criteria. The level of physical functioning and disability was assessed using the disability index (DI) of the Health Assessment Questionnaire (HAQ) (17).

Safety outcomes were expressed as the incidence of adverse events, graded according to National Cancer Institute (NCI) Common Toxicity Criteria (CTC), version 2.0 (18). CD19+ counts, measured by fluorescence-activated cell sorting, were used as a marker of the onset and duration of peripheral B cell depletion (7). Effects on Ig levels, protective antibody titers, and human antichimeric antibody (HACA) levels were also evaluated.

Statistical analysis.

The trial was planned to recruit 440 patients: 40 RF-positive patients per group (a total of 360 patients), who constituted the primary efficacy population, and 20 RF-negative patients in 4 of the 9 groups (a total of 80 patients), for exploratory analysis. Using a 2-sided 5% significance level, it was calculated that a sample size of 360 RF-positive patients would provide at least 80% power to detect differences between the proportions of patients who achieved an ACR20 response in the rituximab 2 1,000-mg and 2 500-mg infusion groups, as well as dose-response trends across the 3 pooled rituximab groups. Randomization was stratified by region (US versus non-US) and by RF status. Steroids were included in the main effects model of ACR20 response at 24 weeks to determine whether a significant effect of the steroid regimen had been achieved. The study was powered to test for an association between ACR20 response at week 24 and rituximab treatment while controlling for corticosteroid usage and region (US or non-US) for RF-positive patients, using a logistic regression model (main effects model). It was not powered to determine any individual effects of these controlling factors.

Primary analyses were based on intent-to-treat (ITT), with nonresponders imputed for all categorical end points (e.g., ACR and EULAR responses), and the last observation was carried forward for continuous variables (e.g., ACR improvement criteria and DAS28). All patients who dropped out of the study were considered nonresponders for purposes of categorical end points. A main-effect logistic regression model was used to evaluate ACR20 response at week 24 in the primary efficacy population across all study groups, with terms for rituximab, glucocorticoid usage, and region (US or non-US). This model also studied any effect of adding RF-negative patients to the analysis population. Contrast tests were conducted for rituximab 2 1,000-mg infusions versus placebo, and rituximab 2 500-mg infusions versus rituximab 2 1,000-mg infusions, in addition to a dose-response relationship in the ACR20 response across the upper- and lower-dose rituximab and placebo groups. No comparisons between the rituximab 2 500-mg infusions and placebo were originally planned.


  1. Top of page
  2. Abstract
  6. Acknowledgements

Patient demographics and characteristics.

A total of 465 patients were treated, 380 RF-positive patients and 85 RF-negative patients; this constituted the safety population (Table 1). Patients in each of the rituximab groups were generally well matched in terms of demographic parameters and region, baseline disease characteristics, and previous DMARD/biologic treatments (Table 2). Antimalarial agents had been taken by 38–68% of patients across treatment groups, sulfasalazine by 38–68% and immunosuppressive agents (cyclosporine, azathioprine, and/or leflunomide) by 24–44%. Overall, 29% of patients had previously taken biologic agents (infliximab, adalimumab, etanercept, or anakinra), with no significant differences between the randomized groups.

Table 2. Patient demographics and baseline disease characteristics*
 Rituximab placeboRituximab 2 500-mg infusionsRituximab 2 1,000-mg infusions
  • *

    RA = rheumatoid arthritis; DMARDs = disease-modifying antirheumatic drugs; MTX = methotrexate; TNFα = tumor necrosis factor α; ITT = intent to treat; HAQ = Health Assessment Questionnaire; DI = Disability Index; RF = rheumatoid factor; ESR = erythrocyte sedimentation rate; CRP = C-reactive protein; DAS28 = Disease Activity Score in 28 joints.

Safety population, no.149124192
  Female, %808380
  Age, mean years51.151.451.1
  Duration of RA, mean years9.311.110.8
  Previous DMARDs (other than MTX), mean no.
  Prior anti-TNFα treatment, %263328
  MTX dose, mean mg/week15.616.014.9
  US patients, %333334
Primary ITT population, no.122123122
 Disease characteristics   
  Swollen joint count, mean no.212222
  Tender joint count, mean no.353332
  Baseline HAQ DI score, mean1.71.81.7
  RF, mean IU/liter437421437
  ESR, mean mm/hour404541
  Serum CRP, mean mg/dl3.33.23.0
  DAS28, mean6.86.86.7

Of the 465 patients, 65% in the placebo group, 91% in the rituximab 2 500-mg infusions group, and 86% in the rituximab 2 1,000-mg infusions group completed the 24-week primary analysis period (Table 1). Fewer patients discontinued rituximab than placebo as a result of insufficient therapeutic response, but rituximab was associated with more withdrawals because of adverse events. Nine patients (all in rituximab groups) withdrew for safety reasons (mainly as a result of infusion-related events). One patient in the 2 500-mg rituximab infusions plus IV glucocorticoid group died due to a cerebrovascular accident.

The primary ITT efficacy population was composed of 367 RF-positive patients; the remaining RF-positive patients were excluded from this analysis for reasons related to accidental unblinding, drug dispensing errors, and unverifiable data.

Efficacy outcomes.

The proportion of patients in the primary ITT efficacy population who achieved an ACR20 response at 24 weeks, the primary end point of the trial, was significantly greater in each of the rituximab groups than in the placebo group (P < 0.0001) (Figure 2A). At week 24, 55% of the patients in the rituximab 2 500-mg infusions group and 54% of those in the 2 1,000-mg infusions group had achieved an ACR20 response, compared with 28% in the placebo group. There was no statistically significant difference in the odds of achieving an ACR20 response between the groups taking 2 1,000-mg infusions of rituximab versus 2 500-mg infusions of rituximab (odds ratio 0.93, P = 0.768). Analyses of the proportions of patients who achieved an ACR50 or ACR70 response showed a similar pattern. Compared with placebo, a significantly greater proportion of patients treated with either dosage of rituximab achieved an ACR50 response (P ≤ 0.001 versus placebo in each rituximab dosage group) and an ACR70 response (P = 0.029 for rituximab 2 500-mg infusions group and P ≤ 0.001 for rituximab 2 1,000-mg infusions group). Although ACR20 and ACR50 responses were similar for both rituximab dosages, ACR70 responses were numerically more frequent in the rituximab 2 1,000-mg infusions group than in the 2 500-mg infusions group (Figure 2A).

thumbnail image

Figure 2. Patients who experienced responses according to A, the American College of Rheumatology 20% (ACR20), 50% (ACR50), and 70% (ACR70) improvement criteria and B, the European League Against Rheumatism (EULAR) criteria at 24 weeks. Results are expressed using pooled data for glucocorticoid dose. Patients who withdrew or who had insufficient data from which to calculate an ACR score were classified as nonresponders. ∗ = P = 0.029; † = P ≤ 0.001; § = P < 0.0001, versus placebo.

Download figure to PowerPoint

Primary analysis within the main-effect model showed that glucocorticoid treatment had no significant effect on ACR20 response (P = 0.166). Dose-contrast tests between individual glucocorticoid treatments showed that the probability of achieving an ACR20 response in any of the main treatment groups was not significantly different for glucocorticoid placebo versus IV glucocorticoids (P = 0.62) or IV and oral glucocorticoids (P = 0.18). Patients who received glucocorticoids were, however, more likely to achieve an ACR20 response by week 4 than those who received no glucocorticoids. Consequently, since glucocorticoid treatment was not a prerequisite for the efficacy of rituximab, efficacy results are expressed using pooled glucocorticoid dose cohorts within each rituximab group.

The time course for achieving an ACR response showed early separation of rituximab from placebo. The proportion of patients who achieved an ACR20 response in the placebo group leveled off at ∼30% at week 4, while in the rituximab groups, it continued to increase, leveling off at ∼55% at 12 weeks. The same was essentially true for ACR50 and ACR70 responses, except in the rituximab 2 1,000-mg infusions group, in which ACR70 was seen in 10% of patients at week 12 and increased to 20% at week 24, suggesting a time-related effect on achievement of higher-level responses at this dose.

Changes in the DAS28 at week 24 reflected ACR response findings. Analysis of variance showed the adjusted mean change in the DAS28 from baseline to be significantly greater in patients treated with rituximab 2 500-mg and 2 1,000-mg infusions (−1.79 and −2.05, respectively) than in patients treated with placebo (−0.67; P < 0.0001). Similarly, EULAR responses in patients treated with rituximab at either dosage were significantly greater than those seen in patients who received placebo (Figure 2B). Compared with placebo, moderate or good EULAR responses were achieved in a significantly greater proportion of patients treated with rituximab 2 500-mg infusions (P < 0.0001) and rituximab 2 1,000-mg infusions (P < 0.0001), with a trend toward a higher proportion of patients achieving a good EULAR response with 2 1,000-mg infusions of rituximab.

CRP levels fell by a mean of 1.7 mg/dl between baseline and week 24 in patients treated with the active rituximab regimens. The corresponding mean change in CRP in the rituximab placebo group was 0.1 mg/dl. These differences in favor of the active rituximab groups were observed after week 4 and continued to week 24. In patients who were RF-positive at baseline, the proportion with RF titers below the level of detection by week 24 was 1.6% in the placebo group and 12% and 9% in the rituximab 2 500-mg infusions and 2 1,000-mg infusions groups, respectively. This reflected a mean increase in total RF titers in the placebo groups (ranging from 7.1% to 37.4% among placebo groups) compared with a mean decrease in the active rituximab groups (ranging from 11.5% to 47.9% among rituximab groups).

Patient-reported outcomes mirrored the improvements in clinical and laboratory findings in the primary ITT population. Changes in the mean DI scores of the HAQ between baseline and week 24 were −0.43 and −0.49 in the rituximab 2 500-mg and 2 1,000-mg infusions groups, respectively, compared with −0.16 in the rituximab placebo group. The proportions of patients with an improvement of −0.22 from baseline (minimum clinically important difference [MCID] [19]) were 63%, 67%, and 34%, respectively in these 3 groups. Furthermore, changes in fatigue, measured as the percentage improvement in the FACIT-F score between baseline and week 24, were 20% and 28% in the rituximab 2 500-mg and 2 1,000-mg infusions groups, respectively, compared with 4% in the placebo group. In both rituximab groups, the mean change from baseline in the FACIT-F score exceeded 4 points, which is the MCID (16). This was not the case in the placebo group.

The inclusion of data from RF-negative patients in the analyses did not alter the primary outcome. In the combined RF-positive and RF-negative population, 52% of the rituximab 2 1,000-mg doses group achieved an ACR20 response at week 24 compared with 32% of the placebo group. An exploratory analysis of RF-negative patients alone proved inconclusive, since it was confounded by an unusually high placebo response (52% versus 48% in the rituximab group) and a relatively small sample size (21 and 63 patients, respectively).

There was a significant difference at week 24 between the absolute proportions of US and non-US patients who achieved an ACR20 response across treatment groups, including placebo (P = 0.002), with an ACR20 response being more frequent in the non-US region. Between regions, the efficacy of rituximab relative to placebo (i.e., the difference between placebo and active drug responses) was similar, and consequently, there was no significant interaction between rituximab and region (P = 0.797). An exploratory analysis of the 2 regions showed that the only major baseline difference between the US and non-US populations was the proportion of patients previously exposed to biologic therapy (55% of US versus 23% of non-US patients). While there was no identified interaction between prior use of biologic therapies and rituximab treatment, ACR20 response rates were generally lower in patients with prior exposure to biologic therapies.

Safety outcomes.

Overall, 70% of patients treated with placebo reported at least 1 adverse event at any time during the study period, compared with 81% and 85% of patients in the rituximab 2 500-mg and 2 1,000-mg infusions groups, respectively. The majority (82%) of events in each group were mild to moderate in severity (NCI CTC grades 1–2). A similar proportion of patients reported severe (NCI CTC grade 3) events in each group (18% of patients receiving placebo and 17% and 18% of patients in the rituximab 2 500-mg and 2 1,000-mg infusions groups, respectively). The most frequently reported adverse events in each treatment group are shown in Table 3 and were mainly associated with the first infusion. Exacerbation of RA, reported as an adverse event, distinguished rituximab placebo from both doses of active rituximab, reflecting the efficacy of rituximab relative to placebo. There were no unexpected adverse events in the RF-negative subgroup.

Table 3. Most frequently reported adverse events in the safety population (n = 465)*
Adverse eventsRituximab placebo (n = 149)Rituximab 2 500-mg infusions (n = 124)Rituximab 2 1,000-mg infusions (n = 192)
  • *

    All adverse events affecting ≥5% of patients in any one group are shown. Values are the number (%). RA = rheumatoid arthritis; URI = upper respiratory tract infection.

  • Food allergy and exacerbation of RA in 1 patient each.

  • Cerebral infarction, fatal cerebrovascular accident unrelated to study treatment, convulsion, myocardial infarction, supraventricular tachycardia, drug hypersensitivity (occurring within 24 hours of rituximab infusion), failure of implant, lower limb fracture, and interstitial lung disease in 1 patient each.

  • §

    Epilepsy, serotonin syndrome, chest pain, generalized edema (occurring within 24 hours of rituximab infusion), exacerbation of RA, status asthmaticus, intestinal obstruction, metrorrhagia, and thromboangiitis obliterans in 1 patient each.

All events105 (70)100 (81)164 (85)
 Exacerbation of RA44 (30)21 (17)27 (14)
 Headache19 (13)14 (11)21 (11)
 Nausea13 (9)8 (6)19 (10)
 URI9 (6)10 (8)12 (6)
 Nasopharyngitis8 (5)7 (6)10 (5)
 Arthralgia5 (3)5 (4)11 (6)
 Diarrhea8 (5)7 (6)6 (3)
 Fatigue8 (5)5 (4)8 (4)
 Hypertension4 (3)5 (4)12 (6)
 Rigors3 (2)5 (4)13 (7)
 Dizziness6 (4)4 (3)10 (5)
Serious noninfection adverse events2 (1)9 (7)9 (5)§
Serious infections2 (1)04 (2)
Infusion-associated events.

Infusion-associated events (defined as any adverse event that occurred during or within 24 hours of rituximab infusion) occurred with the first infusion in 18% of patients who received placebo and in 31% and 38% of patients who received rituximab 2 500-mg and 2 1,000-mg infusions, respectively.

Acute infusion reactions.

Symptoms or signs suggesting an acute infusion reaction (pruritus, fever, urticaria/rash, chills, pyrexia, rigors, sneezing, angioneurotic edema, throat irritation, cough and bronchospasm, hypotension, or hypertension) were more commonly associated with the first infusion of rituximab compared with placebo. Overall, acute infusion reactions were reported at the first infusion in 17% of placebo-treated patients compared with 23% and 32% of patients who received 2 500-mg or 2 1,000-mg rituximab, respectively. However, premedication with IV glucocorticoid significantly reduced both the incidence and severity of acute infusion reactions. Without glucocorticoid premedication, these events were reported by 14% of the placebo group, 32% of the rituximab 2 500-mg infusions group, and 37% of the rituximab 2 1,000-mg infusions group, as compared with 19%, 19%, and 29% of these groups, respectively, following IV glucocorticoid premedication (Figure 3A).

thumbnail image

Figure 3. Effect of glucocorticoids on the frequency of acute infusion reactions after A, the first infusion and B, the second infusion. IV = intravenous; PO = by mouth.

Download figure to PowerPoint

The number of patients who experienced these symptoms following the second infusion of the treatment course was substantially lower in all groups, irrespective of glucocorticoid premedication (Figure 3B). The addition of 2 weeks of oral prednisone treatment did not appear to provide any further reduction of acute infusion reactions with the second infusion.

Two adverse events occurred within the first 24 hours of the first infusion: 1 case of drug hypersensitivity reaction occurred in the rituximab 2 500-mg infusions group, and 1 case of generalized edema occurred in the rituximab 2 1,000-mg infusions group. Both events occurred in patients who had not received IV glucocorticoid premedication. Five patients (2.6%) in the 2 1,000-mg infusions rituximab group were withdrawn from treatment because of acute infusion-associated events following the first infusion. Four of these patients had not received glucocorticoid premedication.

Infection-related adverse events.

Adverse events classed as infections and infestations were reported in 28% of the placebo group and in 35% of each of the active rituximab groups. The type and severity of infections were similar across the placebo and active rituximab groups (the most common being respiratory tract infections, urinary tract infections, and nasopharyngitis), so the pattern of infections did not distinguish between placebo- and rituximab-treated patients.

Six serious infections were reported. Two (1%)occurred in the placebo group (pneumonia and respiratory tract infection) and 4 (2%) in the rituximab 2 1,000-mg infusions group (2 cases of pyelonephritis, and 1 each of bronchitis and epiglottitis). These events resolved without sequelae. No serious infections occurred in the rituximab 2 500-mg infusions group. The rate of serious infections per 100 patient-years was 3.19 for the rituximab placebo group, 0 for the 2 500-mg infusions of rituximab group, and 4.74 for the 2 1,000-mg infusions of rituximab group.

Serious adverse events.

A total of 26 serious adverse events were reported, 4 (3%) in the placebo group (3%), 9 in the rituximab 2 500-mg infusions group (7%), and 13 in the rituximab 2 1,000-mg infusions group (7%), with no consistent pattern or signal. Of these, 20 were noninfection events in 2, 9, and 9 patients in the placebo, rituximab 2 500-mg, and rituximab 2 1,000-mg infusions groups (1%, 7%, and 5%, respectively) (Table 3). Five events were nervous system disorders, while others were distributed evenly across body systems in no discernible pattern. There was 1 fatal event, which occurred in the rituximab 2 500-mg infusions group. A 73-year-old woman died of a cerebral infarction 167 days after the first infusion of rituximab. This patient had a history of atrial fibrillation, hypertension, diabetes mellitus, and hyperlipidemia; at the time of enrollment into the trial, she was receiving concomitant celecoxib, MTX, folic acid, lisinopril, atorvastatin, and sertraline.


In the placebo group, peripheral B cell levels remained stable around baseline levels throughout the study period, after an initial transient increase during the 2-week glucocorticoid dosing period between rituximab placebo infusions. Treatment with rituximab was associated with nearly complete depletion of peripheral blood B cells, which persisted throughout the 24-week study period. Peripheral blood B cell recovery began after week 16 (Figure 4A).

thumbnail image

Figure 4. Levels of A, peripheral blood CD19+ B cells, B, IgG, C, IgA, and D, IgM during the 24-week treatment period.

Download figure to PowerPoint

Although peripheral blood B cell levels were depleted in rituximab-treated patients, mean and median Ig levels (IgG, IgA, and IgM) remained above the laboratory lower limit of normal at all times; the greatest change from baseline occurred in IgM (Figures 4B–D). Anti-tetanus titers remained stable.

Antibodies to the test agent, i.e., HACA (analyzed up to and including week 24), were detectable in 0.7%, 4.2%, and 2.7% of the placebo, rituximab 2 500- mg, and rituximab 2 1,000-mg infusions groups, respectively. There were no serious adverse events in the group of patients with HACAs. Moreover, 6 of the 10 rituximab-treated patients with positive HACAs at week 24 achieved ACR20 responses at that time.


  1. Top of page
  2. Abstract
  6. Acknowledgements

Results of the DANCER trial have confirmed that a single course of rituximab, given as 2 infusions 2 weeks apart, is highly effective over 24 weeks in the treatment of active RA in patients who have shown an incomplete response to MTX. Followup of these patients beyond the primary end point is ongoing, and radiologic outcomes (not measured in this study) will be addressed in future studies. Rituximab has a novel mode of action that results in the depletion of B cells, and it is therefore distinct from other biologic therapies for RA that target T cells and their related cytokines. The present study corroborates earlier studies that support a role for B cells in the inflammatory processes that underlie RA (9, 10, 20–24).

Both rituximab dosages were effective, with 54–55% of patients achieving an ACR20 response by week 24, and a high proportion achieving ACR50 or ACR70 responses. Importantly, clinical improvements in rituximab-treated patients were reflected in patient- and physician-reported outcome measures, as well as in changes in laboratory parameters. Moreover, the beneficial effects of rituximab were often evident early in the course of treatment.

In general, the efficacy results from the DANCER study broadly mirror those from a phase IIa study of rituximab in 161 RF-positive RA patients (9), although in the latter, a higher ACR20 response rate was observed. A possible explanation for this discrepancy is the prior use of biologic therapies at baseline (an indication of refractory disease). In the phase IIa study, only ∼10% of patients had been exposed to a biologic agent, whereas in the DANCER trial, the proportion was 30–40% in the active rituximab groups. A subanalysis of DANCER trial patients who received active rituximab with no prior exposure to biologic therapies determined an ACR20 response rate of 62% for this group, which is closer to previously reported rates (9). Previous use of biologic agents may also be a contributory factor to the lower absolute ACR response observed in US patients compared with non-US patients in this study. A greater proportion of US patients had received such treatment and may therefore have constituted a more refractory population. Nevertheless, rituximab was similarly effective relative to placebo both in patients who had received prior biologics and those naive to these therapies.

No dose-response relationship was established, based on the ACR20 criteria, for the 2 rituximab doses studied. There were, however, trends to indicate that the dose may influence the achievement of high-level response (i.e., ACR70 response and good EULAR response), and further evaluation of both rituximab doses is warranted.

The use of concomitant glucocorticoids with rituximab was based on the successful application of similar regimens in patients with NHL (10) and the initial trials in patients with RA (9, 20, 21). This trial has, however, established that glucocorticoids play no significant role in determining the primary end point of an ACR20 response at 24 weeks in RA patients treated with rituximab, although they may enhance the early response (up to 4 weeks) by virtue of their antiinflammatory properties. Consequently, the use of glucocorticoids does not appear to be a prerequisite for a clinical response to rituximab.

Both dosages of rituximab were generally well tolerated, with a low incidence of serious adverse events, including serious infections. The majority of adverse events were infusion associated, mild to moderate in intensity, and easily managed. Premedication with IV glucocorticoids did, however, reduce the incidence and severity of infusion-associated events following the first infusion of rituximab. The incidence of reactions following the second infusion was low in all groups, with 2 weeks of oral glucocorticoid treatment providing no additional safety benefit over IV glucocorticoid premedication alone. Overall, the incidence of infusion-related adverse events in RA patients appears to be lower than that observed in NHL patients (who may be prone to such events because of elevated levels of white blood cells or tumor lysis) (10, 25).

All patients had depleted B cell levels following rituximab treatment, with peripheral blood B cell levels falling to below the lower limit of normal. Mean and median Ig levels stayed within the normal range, but there was a downward trend specifically with IgM; however, there was no significant change in acquired immunity. Overall, the incidence of immunogenicity (HACAs) was low and had no apparent clinical relevance over the primary 24-week course of this trial, but patients continue to be monitored for this, together with all other pharmacodynamic parameters, including CD19 and Ig levels.

The DANCER study is a novel design for the study of RA: 9 treatment groups with a logistic regression main-effects model as its primary means of analysis. The complexity of the study was driven by a need to understand the effects of different components of a potential therapeutic regimen (i.e., rituximab dose and effects of glucocorticoids). Both objectives were met, but several factors must be taken into account when interpreting the results. First, extrapolation of these data to the larger RA population must be done with care, given the relatively small number of patients involved. In particular, specific comparison between any of the individual 9 groups is inappropriate; however, valuable guidance concerning both the use of glucocorticoids and the dosage of rituximab has been obtained. Second, it should be acknowledged that the present data represent just 24 weeks of followup; longer-term followup will be required to better characterize safety and, possibly, to identify potential differences between rituximab dosages. Last, with regard to the subset of RF-negative patients, small numbers of patients and a large placebo effect prevent the drawing of firm conclusions about the safety and efficacy of rituximab in this population.

In conclusion, both dosages of rituximab were highly effective and generally well tolerated when given with continuing MTX to patients with active RA. The study has also shown that concomitant glucocorticoids are not a prerequisite for the efficacy of rituximab in RA. However, premedication with IV glucocorticoids does appear to ameliorate acute infusion reactions. Both dosages of rituximab explored in this study warrant further differential exploration and longer-term followup.


  1. Top of page
  2. Abstract
  6. Acknowledgements

The authors wish to thank all the principal investigators involved in the study (see Appendix A). They are also grateful to the following at Roche Products: H. Tyrrell, S. Safa Leathers, P. Lehane, J. Kalsi, J. Gilbride, V. Mitchell, B. Mistry, and K. Rowe; and the following at Genentech: B. Wagner, A. Carroll, F. Maghsoudlou, and K. L. Sewell.


  1. Top of page
  2. Abstract
  6. Acknowledgements
  • 1
    Van Vollenhoven RF, Klareskog L. Clinical responses to tumor necrosis factor α antagonists do not show a bimodal distribution: data from the Stockholm tumor necrosis factor α followup registry. Arthritis Rheum 2003; 48: 15003.
  • 2
    Feltelius N, Fored CM, Blomqvist P, Bertilsson L, Geborek P, Jacobsson LT, et al. Results from a nationwide postmarketing cohort study of patients in Sweden treated with etanercept. Ann Rheum Dis 2005; 64: 24652.
  • 3
    Edwards JC, Cambridge G. Prospects for B cell-targeted therapy in autoimmune disease [review]. Rheumatology (Oxford) 2005; 44: 1516.
  • 4
    Anderson DR, Grillo-Lopez A, Varns C, Chambers KS, Hanna N. Targeted anti-cancer therapy using rituximab, a chimaeric anti-CD20 antibody (IDEC-C2B8) in the treatment of non-Hodgkin's B-cell lymphoma [review]. Biochem Soc Trans 1997; 25: 7058.
  • 5
    Clynes RA, Towers TL, Presta LG, Ravetch JV. Inhibitory Fc receptors modulate in vivo cytotoxicity against tumor targets. Nat Med 2000; 6: 4436.
  • 6
    Reff ME, Carner K, Chambers KS, Chinn PC, Leonard JE, Raab R, et al. Depletion of B cells in vivo by a chimeric mouse human monoclonal antibody to CD20. Blood 1994; 83: 43545.
  • 7
    Roitt I, Bostoff J, Male D. Immunology. 2nd ed. London: Gower Medical Publishing; 1989. p. 14.110.
  • 8
    Sell S, Max EE. Immunology, immunopathology and immunity. 6th ed. Washington, DC: ASM Press; 2001. p. 10149.
  • 9
    Edwards JC, Szczepanski L, Szechinski J, Filipowicz-Sosnowska A, Emery P, Close DR, et al. Efficacy of B-cell–targeted therapy with rituximab in patients with rheumatoid arthritis. N Engl J Med 2004; 350: 257281.
  • 10
    McLaughlin P, Grillo-Lopez AJ, Link BK, Levy R, Czuczman MS, Williams ME, et al. Rituximab chimeric anti-CD20 monoclonal antibody therapy for relapsed indolent lymphoma: half of patients respond to a four-dose treatment program. J Clin Oncol 1998; 16: 282533.
  • 11
    Emery P, Sheeran T, Lehane PB, Saiedabadi N, Shaw TM. Efficacy and safety of rituximab at 2 years following a single treatment in patients with active rheumatoid arthritis [abstract]. Arthritis Rheum 2004; 50 Suppl 1: S659.
  • 12
    Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988; 31: 31524.
  • 13
    Felson DJ, Anderson JJ, Boers M, Bombardier C, Furst D, Goldsmith C, et al. American College of Rheumatology preliminary definition of improvement in rheumatoid arthritis. Arthritis Rheum 1995; 38: 72735.
  • 14
    Prevoo ML, van 't Hof MA, Kuper HH, van Leeuwen MA, van de Putte LB, van Riel PL. Modified disease activity scores that include twenty-eight–joint counts: development and validation in a prospective longitudinal study of patients with rheumatoid arthritis. Arthritis Rheum 1995; 38: 448.
  • 15
    Van Gestel AM, Prevoo ML, van 't Hof MA, van Rijswijk MH, van de Putte LB, van Riel PL. Development and validation of the European League Against Rheumatism response criteria for rheumatoid arthritis: comparison with the preliminary American College of Rheumatology and the World Health Organization/International League Against Rheumatism criteria. Arthritis Rheum 1996; 39: 3440.
  • 16
    Cella D, Yount S, Sorensen M, Chartash E, Sengupta N, Grober J. Validation of the Functional Assessment of Chronic Illness Therapy Fatigue Scale relative to other instrumentation in patients with rheumatoid arthritis. J Rheumatol 2005; 32: 8119.
  • 17
    Fries JF, Spitz P, Kraines RG, Holman HR. Measurement of patient outcome in arthritis. Arthritis Rheum 1980; 23: 13745.
  • 18
    National Cancer Institute. Common Toxicity Criteria, version 2.0. URL:
  • 19
    Strand V, Cohen S, Crawford B, Smolen JS, Scott DL. Patient-reported outcomes better discriminate active treatment from placebo in randomized controlled trials in rheumatoid arthritis. Rheumatology (Oxford) 2004; 43: 6407.
  • 20
    Edwards JC, Cambridge G. Sustained improvement in rheumatoid arthritis following a protocol designed to deplete B lymphocytes. Rheumatology (Oxford) 2001; 40: 20511.
  • 21
    Leandro MJ, Edwards JC, Cambridge G. Clinical outcome in 22 patients with rheumatoid arthritis treated with B lymphocyte depletion. Ann Rheum Dis 2002; 61: 8838.
  • 22
    De Vita S, Zaja F, Sacco S, de Candia A, Fanin R, Ferraccioli G. Efficacy of selective B cell blockade in the treatment of rheumatoid arthritis: evidence for a pathogenic role of B cells. Arthritis Rheum 2002; 46: 202933.
  • 23
    Kneitz C, Wilhelm M, Tony HP. Improvement of refractory rheumatoid arthritis after depletion of B cells. Scand J Rheumatol 2004; 33: 826.
  • 24
    Kramm H, Hansen KE, Gowing HE, Bridges A. Successful therapy of rheumatoid arthritis with rituximab: renewed interest in the role of B cells in the pathogenesis of rheumatoid arthritis. J Clin Rheumatol 2004; 10: 2832.
  • 25
    Kimby E. Tolerability and safety of rituximab (MabThera) [review]. Cancer Treat Rev 2005; 31: 45673.


  1. Top of page
  2. Abstract
  6. Acknowledgements

In addition to the authors, other members of the DANCER study group are as follows: R. Day, MD, FRACP (Darlinghurst, New South Wales, Australia), G. Littlejohn, MBBS, FRACP, MPH (Clayton, Victoria, Australia), P. Nash, MD (Marrochydore, Queensland, Australia), R. Will, FRACP (Victoria Park, Western Australia), L. E. Andrade, MD (Sao Paolo, Brazil), S. C. Radominski, MD (Curitiba, Brazil), A. Samara, MD, PhD (Campinas, Brazil), H. El-Gabalawy, MD, FRCPC (Winnipeg, Manitoba, Canada), E. Keystone, MD, FRCPC (Toronto, Ontario, Canada), M. Khraishi, MD (St. John's, Newfoundland, Canada), A. Klinkhoff, MD (Vancouver, British Columbia, Canada), L. Martin, MD (Calgary, Alberta, Canada), J. Pope, MD, MPH, FRCP (London, Ontario, Canada), K. Pavelka, MD (Prague, Czech Republic), P. Hannonen, MB, FRACP (Jyva“skyla”, Finland), M. Hakala, MD (Heinola, Finland), M. Leirisalo-Repo, MD (Helsinki, Finland), M. Fleck, MD (Regensburg, Germany), U. Mueller-Ladner, MD (Regensburg, Germany), A. Rubbert-Roth, MD (Cologne, Germany), H. D. Stahl, MD (Leipzig, Germany), H. P. Tony, MD (Wuerzburg, Germany), S. Wassenberg, MD (Ratingen, Germany), M. Carrabba, MD (Milan, Italy), S. De Vita, MD (Udine, Italy), B. Seriolo, MD (Genoa, Italy), F. Trotta, MD (Ferrara, Italy), L. Barile, MD (Mexico City, Mexico), R. Burgos-Vargas, MD (Mexico City, Mexico), D. Galarza-Delgado, MD (Monterrey, Mexico), J. Miranda Limo'n (Azcapotzalco, Mexico), C. Ramos Remus, MD (Jalisco, Mexico), P. Gow, MD (Auckland, New Zealand), J. Singh, MD (Auckland, New Zealand), J. Szechin'ski, MD (Wroclaw, Poland), J. Alvaro-Gracia, MD (Madrid, Spain), S. Bustabad, MD (Tenerife, Spain), J. Gomez-Reino, MD, PhD (Santiago de Compostela, Spain), A. Laffon, MD (Madrid, Spain), J. L. Marenco, MD (Seville, Spain), E. Martin-Mola, MD, PhD (Madrid, Spain), B. Morck, MD (Gothenburg, Sweden), P. Dawes, MB, MRCP, FRCP (Stoke-on-Trent, UK), B. Hazelman, MD (Cambridge, UK), R. Hughes, MA, MD, FRCP (Chertsey, UK), R. W. Jubb, MD (Birmingham, UK), J. Alloway, MD (Greenville, NC), A. Baldassare, MD (St. Louis, MO), V. Bocoun, MD (Wausau, WI), P. Chatpar, MD (Plainview, NY), M. Davis, DO (Wausau, WI), A. Deodhar, MD, MRCP (Portland, OR), A. H. Dikranian, MD (San Diego, CA), J. Donohue, MD (Boston, MA), M. Ellman, MD (Chicago, IL), J. Fiechtner, MD, MPH (Lansing, MI), R. Fife, MD (Indianapolis, IN), J. Forstot, MD (Boca Raton, FL), P. Freeman, MD (Orlando, FL), D. Furst, MD (Los Angeles, CA), N. Gaylis, MD (Aventura, FL), A. Goldman, MD (Glendale, WI), M. Greenwald, MD (Palm Desert, CA), M. Hagansee, MD (New Orleans, LA), D. Halter, MD (Houston, TX), M. Hooper, MD (Beachwood, OH), C. Jackson, MD (Salt Lake City, UT), A. Kaell, MD (Smithtown, NY), A. Kivitz, MD (Duncansville, PA), G. C. Liang, MD (Chicago, IL), J. Looney, MD (Rochester, NY), J. Loveless, MD (Boise, ID), C. Lue, MD (Little Rock, AR), M. Luggen, MD (Cincinnati, OH), A. R. Mabaquiao, MD (San Diego, CA), D. Mandel, MD (Chardon, OH), A. Martin, MD (Tulsa, OK), P. Mease, MD (Seattle, WA), L. Moreland, MD (Birmingham, AL), G. J. Morgan, MD (Lebanon, NH), N. Neal, MD (Long Beach, CA), K. O'Rourke, MD (Winston- Salem, NC), K. Oelke, MD (Glendale, WI), W. Rigby, MD (Lebanon, NH), A. Rosen, MD (Largo, FL), M. Sayers, DO (Colorado Springs, CO), M. Sedrish, MD (Slidell, LA), Y. Sherrer, MD (Fort Lauderdale, FL), D. Smith, MD (Indianapolis, IN), S. Solomon, MD (Voorhees, NJ), S. Stern, MD (Louisville, KY), P. Sutej, MD (Winston-Salem, NC), N. Sweiss, MD (Chicago, IL), R. Tierney, MD (Minneapolis, MN), W. Washington, MD (Wausau, WI), M. Weitz, MD (South Miami, FL), C. Wiesenhutter, MD (Coeur D'Alene, ID), S. Wolfe, DO (Dayton, OH), E. Zanetakis, MD (Tulsa, OK).