Rituximab for Rheumatoid Arthritis Refractory to Anti Tumor Necrosis Factor Therapy

Transcription

1 ARTHRITIS & RHEUMATISM Vol. 54, No. 9, September 2006, pp DOI /art , American College of Rheumatology Rituximab for Rheumatoid Arthritis Refractory to Anti Tumor Necrosis Factor Therapy Results of a Multicenter, Randomized, Double-Blind, Placebo-Controlled, Phase III Trial Evaluating Primary Efficacy and Safety at Twenty-Four Weeks Stanley B. Cohen, 1 Paul Emery, 2 Maria W. Greenwald, 3 Maxime Dougados, 4 Richard A. Furie, 5 Mark C. Genovese, 6 Edward C. Keystone, 7 James E. Loveless, 8 Gerd-Rüdiger Burmester, 9 Matthew W. Cravets, 10 Eva W. Hessey, 11 Timothy Shaw, 11 and Mark C. Totoritis, 10 for the REFLEX Trial Group Supported by Hoffmann-La Roche, Biogen Idec, and Genentech. Dr. Genovese s work was supported by the NIH (grant 5-M01- RR from the National Center for Research Resources). 1 Stanley B. Cohen, MD: Radiant Research, Dallas, Texas; 2 Paul Emery, MA, MD, FRCP: Leeds General Infirmary, Leeds, UK; 3 Maria W. Greenwald, MD: Desert Medical Advances, Palm Desert, California; 4 Maxime Dougados, MD: René Descartes University, Paris, France; 5 Richard A. Furie, MD: North Shore Long Island Jewish Health System, Lake Success, New York; 6 Mark C. Genovese, MD: Stanford University Medical Center, Stanford, California; 7 Edward C. Keystone, MD, FRCPC: University of Toronto, Toronto, Ontario, Canada; 8 James E. Loveless, MD: Intermountain Orthopedics, Boise, Idaho; 9 Gerd-Rüdiger Burmester, MD: Charité University Hospital, Berlin, Germany; 10 Matthew W. Cravets, MA, Mark C. Totoritis, MD: Biogen Idec, San Diego, California; 11 Eva W. Hessey, MSc, BSc, Timothy Shaw, BSc: Roche Products Ltd., Welwyn Garden City, UK. Dr. Cohen has received honoraria (less than $10,000 each) from Genentech, Biogen Idec, Scios, and Amgen. Dr. Emery has received honoraria (less than $10,000) from Genentech. Dr. Furie has received honoraria (less than $10,000 each) from Biogen Idec and Genentech. Dr. Keystone has received honoraria (less than $10,000 each) from Amgen, Wyeth, Centocor, Abbott, Schering-Plough, Bristol-Meyers Squibb, Genentech, Biogen Idec, and Roche. Dr. Genovese has received honoraria (less than $10,000 each) from Genentech, Biogen Idec, Amgen, Centocor, Abbott, and Bristol- Meyers Squibb. Mr. Cravets and Dr. Totoritis own stock in Biogen Idec. Mr. Shaw owns stock in Roche. Address correspondence and reprint requests to Stanley B. Cohen, MD, Radiant Research, 5939 Harry Hines Boulevard, Suite 441, Dallas, TX stanleycohen@radiantresearch.com. Submitted for publication December 17, 2005; accepted in revised form May 8, Objective. To determine the efficacy and safety of treatment with rituximab plus methotrexate (MTX) in patients with active rheumatoid arthritis (RA) who had an inadequate response to anti tumor necrosis factor (anti-tnf) therapies and to explore the pharmacokinetics and pharmacodynamics of rituximab in this population. Methods. We evaluated primary efficacy and safety at 24 weeks in patients enrolled in the Randomized Evaluation of Long-Term Efficacy of Rituximab in RA (REFLEX) Trial, a 2-year, multicenter, randomized, double-blind, placebo-controlled, phase III study of rituximab therapy. Patients with active RA and an inadequate response to 1 or more anti-tnf agents were randomized to receive intravenous rituximab (1 course, consisting of 2 infusions of 1,000 mg each) or placebo, both with background MTX. The primary efficacy end point was a response on the American College of Rheumatology 20% improvement criteria (ACR20) at 24 weeks. Secondary end points were responses on the ACR50 and ACR70 improvement criteria, the Disease Activity Score in 28 joints, and the European League against Rheumatism (EULAR) response criteria at 24 weeks. Additional end points included scores on the Functional Assessment of Chronic Illness Therapy Fatigue (FACIT-F), Health Assessment Questionnaire (HAQ) Disability Index (DI), and Short Form 36 (SF- 36) instruments, as well as Genant-modified Sharp radiographic scores at 24 weeks. Results. Patients assigned to placebo (n 209) and rituximab (n 311) had active, longstanding RA. At week 24, significantly more (P < ) rituximabtreated patients than placebo-treated patients demonstrated ACR20 (51% versus 18%), ACR50 (27% versus 5%), and ACR70 (12% versus 1%) responses and moderate-to-good EULAR responses (65% versus 22%). 2793

2 2794 COHEN ET AL All ACR response parameters were significantly improved in rituximab-treated patients, who also had clinically meaningful improvements in fatigue, disability, and health-related quality of life (demonstrated by FACIT-F, HAQ DI, and SF-36 scores, respectively) and showed a trend toward less progression in radiographic end points. Rituximab depleted peripheral CD20 B cells, but the mean immunoglobulin levels (IgG, IgM, and IgA) remained within normal ranges. Most adverse events occurred with the first rituximab infusion and were of mild-to-moderate severity. The rate of serious infections was 5.2 per 100 patient-years in the rituximab group and 3.7 per 100 patient-years in the placebo group. Conclusion. At 24 weeks, a single course of rituximab with concomitant MTX therapy provided significant and clinically meaningful improvements in disease activity in patients with active, longstanding RA who had an inadequate response to 1 or more anti-tnf therapies. Rheumatoid arthritis (RA) is a chronic systemic autoimmune disease characterized by symmetric inflammation of affected joints (1,2). The disease affects 1% of the population (3) and has been a significant cause of disability (4,5). Over the last 3 decades, there have been significant improvements in patient outcomes associated with the introduction of methotrexate (MTX) in the 1980s and the subsequent shift to its earlier and more intensive use. The development of therapies targeting tumor necrosis factor (TNF) and interleukin-1 has resulted in further improvement in outcomes related to the capacity of these treatments to retard radiographic progression of RA, resulting in substantial improvement in physical functioning and a reduction in disability, as reported in clinical trials of 6 24 months duration (6 9). Even with the advances in disease management, a population of patients with refractory RA exists. In the clinical trials that led to approval of anti-tnf therapies, 25 40% of patients failed to achieve a response on the American College of Rheumatology 20% improvement criteria (ACR20) (6 9). Feltelius et al (10), reporting on the Swedish Society of Rheumatology registry of patients with RA treated with etanercept between 1999 and 2003, found that 21% of RA patients initiating therapy with etanercept were no longer receiving this therapy at 24 months (10). A preliminary study from the Stockholm TNF antagonist followup registry found that only 44% of patients were still taking their original therapy at 5 years, and 25% were no longer taking any TNF antagonist at all (11). These patients currently have few treatment alternatives. As such, they represent a difficult-to-treat patient population for whom additional therapeutic options are needed. While the etiology of RA remains elusive, it is believed that unknown antigenic triggers, initiated within the background of genetic, environmental, and hormonal influences, initiate a self-perpetuating cascade of autoimmune inflammatory responses within the synovial compartment (2,12,13). Specific B and T lymphocytes, macrophages, monocytes, endothelial cells, and fibroblasts play putative roles in this process (2,14). Recent research indicates that B cells may act at multiple levels of the inflammatory cascade by disrupting antigen presentation by T cells as well as inducing an increased expression of proinflammatory cytokines and autoantibodies that, together, initiate and contribute to the inflammatory damage observed in RA (12,14 18). Rituximab is a genetically engineered chimeric monoclonal antibody that targets CD20 B cells (19). It has been approved by the Food and Drug Administration (FDA) for the treatment of relapsed or refractory, low-grade or follicular CD20 B cell non-hodgkin s lymphoma and for diffuse, large B cell CD20 non- Hodgkin s lymphoma in combination with approved chemotherapy. Clinical trials of rituximab have demonstrated significant efficacy and adequate safety in modifying the symptoms of RA (20,21) and have provided further evidence of the role of B cells in the disease pathogenesis. By binding to CD20, rituximab depletes subpopulations of peripheral B cells through several postulated mechanisms, including cell-mediated and complement-dependent cytotoxicity and promotion of apoptosis (19,22,23). CD20 is an attractive target for B cell depleting therapy because it is stable and highly expressed on B cells but not on stem cells or plasma cells (24). To further investigate the efficacy and safety of rituximab when administered in combination with MTX, we conducted a phase III trial in patients with active RA who had an inadequate response to 1 or more anti-tnf therapies. The results of this trial, as reported herein, support the recent FDA approval of rituximab treatment in combination with MTX for the reduction of signs and symptoms of moderately to severely active RA in adult patients who have had an inadequate response to 1 or more TNF antagonist therapies. PATIENTS AND METHODS Role of the funding sources. The study was cosponsored by Hoffmann-La Roche (Nutley, NJ), Biogen Idec (San

3 RITUXIMAB IN RA REFRACTORY TO ANTI-TNF THERAPY 2795 Figure 1. Design of the Randomized Evaluation of Long-Term Efficacy of Rituximab in RA (REFLEX) Trial, a 2-year, randomized, double-blind, placebo-controlled, phase III study conducted at multiple centers. Primary safety and efficacy were evaluated at 24 weeks. TNF tumor necrosis factor; DMARD disease-modifying antirheumatic drug; MTX methotrexate; iv intravenous. Diego, CA), and Genentech (South San Francisco, CA). Hoffmann-La Roche and Biogen Idec were responsible for data collection, and the statistical analyses were conducted by suitably qualified statisticians who were employees of either sponsor. The trial protocol was designed jointly by Hoffmann-La Roche, Biogen Idec, and the lead clinical investigators. All authors had access to and involvement in the interpretation of the data, as well as input into and control over the content of the manuscript, with Dr. Cohen as supervisor. Patients. Eligible adult patients were recruited from 114 rheumatology centers in the US, Europe, Canada, and Israel. Patients had RA for at least 6 months, according to the ACR 1987 revised criteria (25), and had active disease, which was defined as 8 swollen joints (of 66 joints assessed) and 8 tender joints (of 68 joints assessed), a C-reactive protein (CRP) level 1.5 mg/dl or an erythrocyte sedimentation rate (ESR) 28 mm/hour, and radiographic evidence of at least 1 joint with a definite erosion attributable to RA, as determined by a central reading site (a centralized organization independent of the sponsors that provided blinded radiographic assessments). Eligible patients had to be taking MTX (10 25 mg/week) for at least 12 weeks prior to screening, with the last 4 weeks at a stable dosage. Patients with a history of a rheumatic autoimmune disease other than RA (except secondary Sjögren s syndrome), significant systemic involvement secondary to RA (vasculitis, pulmonary fibrosis, or Felty s syndrome), or ACR functional class IV disease were excluded. It was recommended that each investigator review their patients current immunization status and, if required, provide appropriate vaccinations/boosters at least 4 weeks prior to enrollment in the trial. Enrolled patients had experienced an inadequate response to previous or current treatment with the anti-tnf agents infliximab ( 3 mg/kg; at least 4 infusions), adalimumab (40 mg every other week for 3 months), or etanercept (25 mg twice weekly for 3 months), or were intolerant to at least 1 administration of these agents. Patients discontinued etanercept for 4 weeks and infliximab or adalimumab for 8 weeks prior to randomization. The study was performed in accordance with the Declaration of Helsinki. All participating sites received approval from their governing institutional review board (or equivalent) and all patients provided written informed consent. Study protocol. The Randomized Evaluation of Long- Term Efficacy of Rituximab in RA (REFLEX) Trial is a 2-year, randomized, double-blind, placebo-controlled, phase III study conducted at multiple centers (see Appendix A for the principal investigators of the REFLEX Trial Group). Patients were assigned to 1 of 2 treatment groups: placebo or rituximab (1 course; 2 infusions of 1,000 mg each), both with background MTX (Figure 1). Primary efficacy and safety were evaluated at 24 weeks; after 24 weeks, patients entered a post-treatment period and were followed up every 2 months for 18 months, for an overall study duration of 24 months. This report includes the initial 24-week results of this ongoing 2-year study. Patients were randomized at a ratio of 3:2 to receive intravenous infusions of rituximab or placebo on days 1 and 15. Both groups continued to take stable dosages of MTX (10 25 mg/week orally or parenterally) and received folate ( 5 mg/ week or equivalent), intravenous methylprednisolone (100 mg 30 minutes before each infusion), and oral prednisone (60 mg on days 2 7; 30 mg on days 8 14) during the 2-week treatment period. Continued use of glucocorticoids ( 10 mg/day of prednisone or equivalent) and nonsteroidal antiinflammatory drugs was permitted if the dosage was stable for 4 weeks and

4 2796 COHEN ET AL 2 weeks, respectively, before screening. Use of any other disease-modifying antirheumatic drug (DMARD) was prohibited during the trial. Clinical and safety assessments (ACR disease activity measures and recording of adverse events [AEs]) were made at screening, on day 1, and every 4 weeks through week 24. After week 24, patients were followed up for 18 months posttreatment. Radiographs of the hands and feet taken at screening and at weeks 24, 56, and 104 were assessed using the Genant modification of the Sharp scoring method (26). Radiographic evaluations were performed by 2 independent readers (at the central reading site) who were blinded to the study treatment. Patients failing to respond adequately to treatment (defined as 20% improvement from baseline in both the swollen and tender joint counts) were eligible to receive rescue therapy (for placebo-treated patients, 2 infusions of 1,000 mg of rituximab preceded by 100 mg of intravenous methylprednisolone and for rituximab-treated patients, standard of care) between weeks 16 and 24; after week 24, patients who withdrew from the study were followed up for at least 48 weeks after withdrawal, with longer followup for patients whose B cells had not returned to pretreatment or normal levels. Patients starting rescue therapy or who withdrew for any reason were considered ACR nonresponders. Patients who demonstrated a response to treatment on or after week 16 and had active RA were eligible to receive additional courses of rituximab through a separate retreatment protocol. This was a blinded study, with the study sponsor, investigators, and patients unaware of the treatment assignment of each patient. The results of the total rheumatoid factor (RF) titer and B cell counts at each assessment after screening were also blinded. To prevent potential breaks in the blinding because of observed efficacy or changes in laboratory values, a dual assessor approach was used to evaluate efficacy and safety. An efficacy assessor was responsible for joint examinations as well as the physician s global assessment. A safety assessor was responsible for evaluating possible adverse events and abnormal laboratory findings. Additionally, the safety assessor remained blinded to the CD19 B cell counts for the duration of the study. Only the safety assessor had access to all clinical and laboratory (safety) data and was able to make any necessary changes to the patient s medical therapy, thereby minimizing the chance of unblinding of the efficacy assessor, who only had access to the efficacy data. In addition to routine laboratory tests, levels of CD19 B cells, CD3, CD4, and CD8 T cells, total RF, RF immunoglobulin isotypes, immunoglobulins (IgG, IgA and IgM), and human antichimeric antibodies (HACAs) were analyzed throughout the study. Because rituximab interferes with CD20 analysis by flow cytometry, CD19 was used as a surface marker, since it has a similar distribution on B cells. The Common Toxicity Criteria (CTC), version 2.0 (27), was used to grade and record AEs, which were monitored until they stabilized or returned to baseline status. The intensity, relationship to study treatment, action taken regarding study drug, and outcome to date were recorded. A serious AE was defined as any experience that was fatal, immediately lifethreatening, required hospitalization or prolongation of an existing hospitalization, resulted in persistent or significant disability/incapacity, was a congenital anomaly/birth defect, was medically significant, or required intervention to prevent 1 or more of the outcomes listed above. Clinical assessments. The primary end point was the proportion of patients with an ACR20 response at week 24, defined as at least a 20% improvement from baseline values in the swollen joint count and the tender joint count as well as in 3 of the 5 remaining disease activity measures: physician s global assessment of disease activity; patient s global assessment of disease activity, patient s assessment of pain, patient s assessment of physical function, and either the CRP level or the ESR (28). Secondary end points included ACR50 and ACR70 responses (50% and 70% improvement from baseline according to the ACR criteria, respectively), changes from baseline to week 24 in scores on the Disease Activity Score 28-joint assessment for swelling and tenderness (DAS28) (29), the EULAR response criteria (30), and the individual parameters of the ACR improvement criteria: swollen joint count, tender joint count, patient s and physician s global assessments of disease activity, patient s assessment of pain, patient s assessment of disability (using the Disability Index [DI] of the Health Assessment Questionnaire [HAQ]), the CRP level, and the ESR (28). Additional end points included changes from baseline to week 24 in the Functional Assessment of Chronic Illness Therapy Fatigue (FACIT-F) score, the Short Form 36 (SF-36) health survey score (assessing health-related quality of life), and the Genant-modified Sharp radiographic score (26). Regardless of their status in the study, all patients were to return for scheduled radiographs. Statistical analysis. Sample size calculations were based on the assumption of detecting a difference in the proportion of patients with an ACR20 response in the rituximab plus MTX (0.45) and the placebo plus MTX (0.30) groups. On the basis of these assumptions and using a conservative exact test (Fisher s) with a 2-sided 5% significance level, a sample size of 500 patients randomized to rituximab and placebo at a ratio of 3:2 provided the study with 91% power to detect a difference between the treatments. Efficacy analyses were performed on the intent-to-treat (ITT) population, defined as all randomized patients who received any part of an infusion of study medication, excluding some patients as described in the Results. Patients who withdrew prematurely from the study or who started rescue therapy were included in the ITT population as nonresponders. The primary analysis was a test of the difference in the proportions of patients in the placebo and rituximab groups with an ACR20 response at week 24, using a Cochran-Mantel- Haenszel analysis and stratifying by RF status (positive versus negative) and geographic region (US versus non-us) at baseline. Results were expressed as proportions, corresponding 95% confidence intervals of the difference between response rates, and P values. The null hypothesis assumed that the proportion of patients with an ACR20 response at week 24 for rituximab was the same as that for placebo. Rituximab treatment was judged effective compared with placebo if there was sufficient statistical evidence to reject the null hypothesis in favor of rituximab. The secondary end points of the ACR50 and ACR70 responses were analyzed using the same statistical methodology as described for the ACR20 response. Changes in the DAS28 were assessed using an analysis of variance (ANOVA)

5 RITUXIMAB IN RA REFRACTORY TO ANTI-TNF THERAPY 2797 model with baseline DAS28, RF status, geographic region, and treatment as terms in the model. Categorical DAS28 responders (EULAR response) were assessed using a Cochran- Mantel-Haenszel test stratified for RF status and geographic region. Changes from baseline in the individual parameters of the ACR improvement criteria (the swollen joint count, tender joint count, patient s and physician s global assessments of disease activity, disability according to the HAQ DI, patient s assessment of pain, the CRP level, and the ESR) were summarized by descriptive statistics. The difference between the rituximab group and the placebo group was assessed using ANOVA models with baseline parameter of interest, RF status, and geographic region as terms in the model. Changes in the ACR20, ACR50, ACR70, and DAS28 over weeks 8, 12, 16, 20, and 24 were assessed using descriptive, graphic, and ANOVA techniques, as appropriate. Figure 2. Disposition of patients from enrollment to week 24. With regard to the number of patients treated, 1 patient who was randomized to the rituximab group actually received placebo and was therefore included in the placebo subgroup for the safety analysis, resulting in 209 patients receiving placebo plus methotrexate (MTX) and 308 patients receiving rituximab plus MTX. RESULTS Patient characteristics. Of 751 patients screened, 520 were enrolled in the study: 209 were randomized to placebo and 311 to rituximab (Figure 2). Three patients were randomized but did not receive treatment. One patient randomized to rituximab received placebo in- Table 1. Baseline patient characteristics (safety population)* Characteristic Placebo plus MTX (n 209) Rituximab plus MTX (n 308) Age, years Sex, no. (%) of patients Female 169 (81) 251 (81) Male 40 (19) 57 (19) Disease duration, years Swollen joint count (66 joints assessed) Tender joint count (68 joints assessed) RF positive, no. (%) of patients 165 (79) 242 (79) RF titer, IU/liter C-reactive protein, mg/dl Erythrocyte sedimentation rate, mm/hour Disease Activity Score in 28 joints Health Assessment Questionnaire Disability Index score Total Genant-modified Sharp radiographic score No. of previous DMARDs (excluding MTX) Previous anti-tnf agents taken, no. (%) of patients Infliximab 169 (81) 219 (71) Adalimumab 38 (18) 71 (23) Etanercept 104 (50) 168 (55) No. of previous anti-tnf agents taken, mean SD anti-tnf agent, no. (%) of patients 125 (60) 186 (60) 2 anti-tnf agents, no. (%) of patients 65 (31) 94 (31) 3 anti-tnf agents, no. (%) of patients 19 (9) 28 (9) Inadequate efficacy of anti-tnf agents (%) Use of glucocorticoids at baseline, no. (%) of patients 127 (61) 200 (65) Weekly dose of MTX at baseline, mg * Except where indicated otherwise, values are the mean SD. MTX methotrexate; RF rheumatoid factor; DMARDs disease-modifying antirheumatic drugs; anti-tnf anti tumor necrosis factor. Defined according to the European League Against Rheumatism criteria. Previous DMARDs other than MTX or anti-tnf agents include leflunomide, azathioprine, cyclosporine, hydroxychloroquine, gold, anakinra, sulfasalazine, and penicillamine. Based on an intent-to-treat population of 201 patients taking placebo and 298 taking rituximab.

6 2798 COHEN ET AL Figure 3. Responses to treatment at week 24 in the intent-to-treat population. A, Percentages of patients achieving a response according to the American College of Rheumatology 20% improvement criteria (ACR20), 50% improvement criteria (ACR50), and 70% improvement criteria (ACR70) at week 24. The ACR20, ACR50, and ACR70 responses in rituximab-treated patients were statistically significant (P ) compared with those in the placebo-treated patients. B, Percentages of patients achieving a response according to the European League Against Rheumatism (EULAR) criteria at week 24. EULAR responses are defined as follows: moderate represents a Disease Activity Score 28-joint assessment for swelling and tenderness (DAS28) of 5.1 and an improvement from 0.6 to 1.2; good represents a DAS28 score of 3.2 and an improvement of 1.2; low disease activity represents a DAS28 score of 3.2; remission represents a DAS28 score of 2.6. The EULAR moderate and good responses in the rituximab group (65%) were statistically significant (P ) compared with the placebo group (22%). Numbers above the bars are the numerical percentages represented by the bars. stead; this patient was analyzed in the placebo group for safety, but the rituximab group for efficacy. Patients receiving a portion of at least 1 infusion of placebo or rituximab were included in the safety analysis (209 patients in the placebo group and 308 in the rituximab group). A total of 21 patients were excluded from the ITT population: those for whom treatment was unblinded because of breakage of the rituximab vial, those who never received treatment, those treated prior to randomization, and those enrolled at a center where the blinding of the efficacy assessor was potentially compromised. Thus, the ITT population consisted of 499 patients (201 in the placebo group and 298 in the rituximab group). Sensitivity analyses that included all patients demonstrated no change in the significance of the results. A total of 54% of the patients randomized to placebo and 82% of the patients randomized to rituximab completed the 24 weeks of study. Ninety-six patients in the placebo group who received study medication withdrew prematurely, with 80 of them receiving rituximab rescue therapy and 16 entering the safety followup phase predominantly because of inadequate response. Of the 309 treated patients in the rituximab group, 53 who received study medication withdrew early and entered the safety followup prior to week 24, 1 patient required rescue therapy, and 1 patient withdrew for unknown reasons. Both study groups were balanced for age, sex, disease activity, and previous and concomitant treatments for RA (Table 1). Patients had longstanding ( 12 years) and active disease, as determined by elevated levels of markers of inflammation, DAS28 and HAQ DI scores, Genant-modified Sharp radiographic scores, and numbers of swollen and tender joints. A majority of patients in each group (79%) were RF positive at baseline. Most patients in both groups (89%) had received DMARDs other than MTX (defined in Table 1), and all patients had received anti-tnf agents. In both treatment groups, over 90% of all patients had previously taken 1 or 2 anti-tnf agents, with 9% of patients receiving 3 anti-tnf agents, and 91% of the patients had demonstrated an inadequate response to anti-tnf therapy because of a lack of efficacy. A total of 15% of patients had received their last dose of anti-tnf therapy within 30 days of the screening visit; clinical responses of these patients did not differ from the overall patient population described below. Glucocorticoids, most commonly prednisone and triamcinolone, were taken by 61% of the placebo group and 65% of the rituximab group at baseline.

7 RITUXIMAB IN RA REFRACTORY TO ANTI-TNF THERAPY 2799 Figure 4. A C, Percentages of patients achieving a response according to the American College of Rheumatology 20% improvement criteria (ACR20), 50% improvement criteria (ACR50), and 70% improvement criteria (ACR70), respectively, and D, changes in scores on the Disease Activity Score 28-joint assessment for swelling and tenderness (DAS28) over the 24-week study period in the intent-to-treat population. P versus the placebo group, by 2-tailed test. Clinical efficacy. At 24 weeks, the proportion of patients with an ACR20 response was significantly higher for rituximab-treated patients than for placebotreated patients (51% versus 18%; P ) (Figure 3A). Logistic regression analysis showed that rituximabtreated patients were 5 times more likely to achieve an ACR20 response than placebo-treated patients. The proportion of patients achieving ACR50 and ACR70 responses at week 24 were also significantly higher for rituximab-treated patients than placebo-treated patients (27% versus 5% for ACR50 and 12% versus 1% for ACR70; P for each comparison). Significantly more rituximab-treated patients achieved good or moderate EULAR responses as compared with placebotreated patients (65% versus 22%; P ) (Figure 3B), with significant reductions in mean DAS28 scores from baseline ( 0.4 versus 1.9; P ). In addition, a greater proportion of patients receiving rituximab achieved EULAR low disease activity and remission responses compared with patients receiving placebo (Figure 3B). ACR20 response rates over time showed a statistically significant (P ) separation between rituximab and placebo treatment by week 8; ACR50 and ACR70 responses over time showed a statistically significant separation by week 12 and week 16 of treatment, respectively (Figures 4A C). DAS28 scores were statistically significant (P ) between treatment

8 2800 COHEN ET AL Table 2. Mean changes from baseline in the individual parameters of the ACR improvement criteria at week 24 (ITT population analysis)* Individual parameter of the ACR improvement criteria Placebo plus MTX (n 201) Rituximab plus MTX (n 298) P Swollen joint count (66 joints assessed) Tender joint count (68 joints assessed) Patient s global assessment of disease activity, mm (0 100-mm VAS) Physician s global assessment of disease activity, mm (0 100-mm VAS) Health Assessment Questionnaire Disability Index Patient s assessment of pain, mm (0 100-mm VAS) C-reactive protein, mg/dl Erythrocyte sedimentation rate, mm/hour * For all individual parameters of the American College of Rheumatology (ACR) improvement criteria, a negative change represents improvement. Differences between groups were assessed using analysis of covariance models, with the baseline parameter of interest, rheumatoid factor, and geographic region (US versus non-us) as covariates. Values are the mean SD change from baseline. ITT intent-to-treat; MTX methotrexate; VAS visual analog scale. groups at every time point from week 8 through week 24 (Figure 4D). The mean reductions from baseline in all parameters of the ACR improvement criteria (used to calculate the ACR responses) were significantly greater for patients receiving rituximab (Table 2). Notably, at week 24, the swollen and tender joint counts, patient s and physician s global assessments of disease activity, pain scores, HAQ DI scores, CRP levels, and ESRs were significantly decreased in rituximab-treated patients; placebo-treated patients reported little or no changes in the swollen or tender joint counts, HAQ DI scores, pain scores, CRP levels, or ESRs at week 24. The CRP level was essentially unchanged in placebo-treated patients, but decreased by a mean of 2.1 mg/dl in rituximabtreated patients (P ). Of the 298 rituximabtreated patients in the ITT population at week 24, 18 (6%) had a HAQ DI score of 0, 29 (10%) had a swollen joint count of 0, and 21 (7%) had a tender joint count of 0. Of the 201 placebo-treated patients in the ITT population at week 24, 1 (0.5%) had a HAQ DI score of 0, 2 (1%) had a swollen joint count of 0, and 3 (1.5%) a tender joint count of 0. A total of 80 of the 281 rituximab-treated patients (28%) and 18 of the 180 placebo-treated patients (10%) with elevated CRP levels at baseline had CRP levels in the normal range by week 24. ESR decreased by a mean of 18.5 mm/hour in rituximab-treated patients, compared with a decrease of 4.1 mm/hour in placebo-treated patients (P ). Patients receiving rituximab reported a significant reduction in fatigue, as measured by the FACIT-F scale, which decreased by a mean of 9.1 points, representing a 29.6% improvement from baseline (P ); patients receiving placebo reported a mean decrease of 0.5 points in the FACIT-F score. At week 24, mean SF-36 scores for the mental and physical health domains increased by 4.7 and 5.8, respectively, in rituximab-treated patients, and by 1.3 and 0.9, respectively, in placebo-treated patients; both differences were statistically significant (P ). Changes in radiographic end points at week 24 for observed cases (Table 3) showed a trend toward less progression of joint damage in rituximab-treated patients. Changes from baseline in the joint narrowing scores were significantly reduced in patients receiving Table 3. Mean changes from baseline in the radiographic end points at week 24 (observed case analysis)* End point Placebo plus MTX (n 177) Rituximab plus MTX (n 268) P Total Genant-modified Sharp radiographic score Total joint space narrowing score Total erosion score No worsening of erosions, no. (%) of patients 106 (60) 176 (66) * A total of 24 patients receiving placebo and 30 patients receiving rituximab were missing radiographs at week 24. Except where indicated otherwise, values are the mean SD change from baseline (screening). MTX methotrexate.

9 RITUXIMAB IN RA REFRACTORY TO ANTI-TNF THERAPY 2801 rituximab compared with patients receiving placebo (P 0.016). In rituximab-treated patients, the changes in total erosion scores and overall Genant-modified Sharp scores were half those in the placebo-treated patients at week 24. The difference in the proportion of patients without new erosions was not statistically significant between treatment groups, although it numerically favored rituximab. Analysis of the interaction between treatment and geographic region showed that while ACR20 response rates for rituximab patients were statistically significant (P ) for both US and non-us patients, the treatment effect was greater for non-us patients at the 10% level (P 0.08). In addition, fewer RF-negative patients than RF-positive patients achieved an ACR20 response at week 24, both in patients treated with placebo (12% versus 19%) and in patients treated with rituximab (41% versus 54%); differences in the ACR20 response rates between the placebo and rituximab groups were significant for both RF-positive patients (P ) and RF-negative patients (P ). There was no significant interaction between treatment and RF status (P 0.9), indicating that the effect of treatment on the ACR response was not dependent on baseline RF status. Pharmacodynamics and immunogenicity. Treatment with rituximab was associated with a rapid and complete depletion of CD19 peripheral B cells, with some recovery of cell counts beginning between weeks 16 and 20 (Figure 5A). Conversely, B cell counts in placebo-treated patients rose acutely and then remained stable throughout the study. These effects were likely due to the patients glucocorticoid regimens. Despite selective B cell depletion, mean immunoglobulin (IgG, IgM, and IgA) levels generally remained within normal limits in all patients throughout the 24 weeks of study. In the placebo group, 4 of 209 patients (1.9%) had at least 1 posttreatment IgM value below the lower limit of normal. In rituximab-treated patients, 17 of the 308 patients (5.5%) had IgM levels below the lower limit of normal, 1 patient ( 1%) had IgA values below the lower limit of normal, and 3 patients ( 1%) had IgG values below the lower limit of normal, all during the posttreatment period. Of the 17 patients with IgM levels below the lower limit of normal (40 or 50 mg/dl, depending on the laboratory), the mean reduction was 8.5 mg/dl below this level. Rituximab treatment was associated with a mean decrease in RF levels (55%), whereas treatment with placebo was associated with a mean increase in RF levels (37%), at week 24 in patients who were RF-positive at Figure 5. A, CD19 cell counts in the safety population analysis. Values are the median and interquartile range. B, Percentage change in total rheumatoid factor (RF) values in the safety population analysis for patients who were RF-positive at baseline. Values are the mean SEM. baseline (Figure 5B). Of the rituximab-treated patients with postdose samples for week 24 or their early withdrawal point, 13 patients (4.3%) were found to have positive HACA titers. There were no apparent associations between HACA titers and peripheral blood B cell

10 2802 COHEN ET AL Table 4. Summary of adverse events (safety population) No. (%) of patients taking placebo plus MTX (n 209) No. (%) of patients taking rituximab plus MTX (n 308) Adverse events Any adverse event 183 (88) 261 (85) Severe adverse event* 49 (23) 55 (18) Related adverse event 77 (37) 119 (39) Serious adverse event 21 (10) 23 (7) Adverse event leading to withdrawal 2 ( 1) 8 (3) Death 0 0 Adverse events reported at a 5% incidence 183 (88) 261 (85) Rheumatoid arthritis 87 (42) 65 (21) Headache 19 (9) 26 (8) Upper respiratory tract infection 14 (7) 24 (8) Nasopharyngitis 12 (6) 23 (7) Nausea 5 (2) 22 (7) Fatigue 12 (6) 21 (7) Hypertension 11 (5) 21 (7) Diarrhea 16 (8) 18 (6) Arthralgia 10 (5) 17 (6) Pyrexia 7 (3) 15 (5) Dizziness 8 (4) 14 (5) Bronchitis 12 (6) 13 (4) Cough 11 (5) 10 (3) Sinusitis 11 (5) 10 (3) Urinary tract infection 16 (8) 10 (3) * Grade 3 or 4 according to the Cancer Therapy Evaluation Program Common Toxicity Criteria, version 2.0. Related to any study drug (rituximab, methotrexate [MTX], or glucocorticoids). counts, specific adverse events, or ACR responses at week 24 for HACA-positive patients. Safety. Overall, 88% of placebo-treated patients reported an adverse event during the study period, as compared with 85% of rituximab-treated patients (Table 4). The majority of AEs were mild or moderate in severity (CTC grade 1 or 2), with 23% of patients in the placebo group and 18% in the rituximab group experiencing severe adverse events (CTC grade 3 or 4). Fewer than 40% of the AEs in both groups were considered to be related to the study treatment. Serious AEs were reported in more patients receiving placebo (10%) than in those receiving rituximab (7%). Two patients in the placebo group withdrew from the study because of AEs (1 because of gastric cancer and 1 because of thrombocytosis). Eight patients in the rituximab group withdrew because of AEs, 5 because of infusion-related reactions during the first infusion (anaphylactic reaction, urticaria, laryngeal edema, and cough and hoarseness) and 3 because of other events (cardiac tamponade, spontaneous abortion, and progressive RA). There were no deaths recorded during the 24-week double-blind portion of the study. With the exception of exacerbation of RA, which was reported twice as frequently in placebo-treated patients (42%) as in rituximab-treated patients (21%), urinary tract infection (8% in placebo patients, 3% in rituximab patients), and nausea (2% in placebo patients, 7% in rituximab patients), the frequencies of AEs reported at a 5% incidence were similar between groups (Table 4). The most common AEs were musculoskeletal and connective tissue disorders, which were seen more frequently in the placebo group (53%) than in the rituximab group (39%). This difference was predominantly due to the higher incidence of worsening RA symptoms in placebo-treated patients compared with rituximab-treated patients. Infusion-associated AEs. Any AEs occurring during or within 24 hours of infusion of rituximab occurred in a higher proportion of rituximab-treated patients during or following the first infusion (29%) as compared with placebo-treated patients (23%). The most common AE was headache, affecting 5% of patients in both treatment groups. A subset of these AEs occurring within 24 hours was thought to more plausibly represent acute infusion reactions (Table 5). Signs and symptoms of these acute infusion reactions (pruritus, urticaria/ rash, angioedema, fever, chills, rigors, sneezing, throat

11 RITUXIMAB IN RA REFRACTORY TO ANTI-TNF THERAPY 2803 Table 5. Acute infusion reactions from the first and second rituximab infusions* No (%) of patients taking placebo plus MTX (n 209) No (%) of patients taking rituximab plus MTX (n 308) First infusion Second infusion First infusion Second infusion Acute infusion reactions* 38 (18) 24 (11) 72 (23) 26 (8) Headache 10 (5) 2 ( 1) 15 (5) 3 ( 1) Hypertension 4 (2) 4 (2) 9 (3) 7 (2) Nausea 2 ( 1) 8 (3) 2 ( 1) Pruritus 2 ( 1) 7 (2) Urticaria 1 ( 1) 7 (2) Diarrhea 1 ( 1) 1 ( 1) 5 (2) 3 ( 1) Flushing 2 ( 1) 1 ( 1) 5 (2) 2 ( 1) Pyrexia 1 ( 1) 2 ( 1) 5 (2) Dizziness 4 (2) 2 ( 1) 4 (1) 2 ( 1) Hot flush 4 (1) 1 ( 1) Throat irritation 4 (1) 1 ( 1) Tachycardia 7 (3) 3 ( 1) 2 ( 1) Ear pruritus 3 ( 1) Oropharyngeal swelling 3 ( 1) Hypotension 1 ( 1) 2 ( 1) 2 ( 1) Asthma 2 ( 1) Vomiting 1 ( 1) 2 ( 1) Rash 1 ( 1) 1 ( 1) * Values represent acute infusion reactions occurring in at least 2 rituximab-treated patients. MTX methotrexate. irritation/tightness, cough, bronchospasm, with or without hypotension or hypertension) were experienced during or within 24 hours of the first infusion by a higher proportion of patients in the rituximab group (23%) than in the placebo group (18%). With the second infusion, acute infusion reactions occurred in fewer patients receiving rituximab (8%) than patient receiving placebo (11%). Two rituximab-treated patients experienced acute infusion reactions that were considered serious adverse events (anaphylaxis and hypertension, respectively), but most infusion reactions were mild or moderate in severity (CTC grade 1 or 2). For patients who experienced an infusion reaction, the recommended treatment included acetaminophen/ paracetamol plus intramuscular or slow intravenous administration of antihistamine (diphenhydramine HCl) and/or a bronchodilator, if indicated. In patients who experienced a severe infusion-related reaction, the infusion was immediately interrupted and symptomatic treatment was initiated. After all symptoms disappeared, the infusion was restarted at half the rate that had precipitated the reaction. Infection rates. The incidence of infections was slightly higher in patients treated with rituximab (41%) than in patients who received placebo (38%), but the overall infection rate per 100 patient-years was slightly higher in placebo-treated patients (154.6) than rituximab-treated patients (138.2). The most common infections in both groups were upper respiratory tract infections, nasopharyngitis, urinary tract infections, bronchitis, and sinusitis. Rates of serious infections (those considered serious AEs or requiring intravenous antibiotics) per 100 patient-years were 3.7 for placebo (3 serious infections) versus 5.2 for rituximab (7 serious infections). Of the 7 serious infections occurring in the rituximab-treated patients, 6 resolved without sequelae (gastroenteritis, pyelonephritis, cat-bite infection, influenza, fever of unknown etiology, and de novo hepatitis B), and 1 resolved with sequelae (gangrenous cellulitis, resulting in amputation of a toe). There were no reports of tuberculosis or opportunistic infections over the 24 weeks of this study. DISCUSSION In this 24-week clinical trial, we evaluated the efficacy and safety of rituximab in patients with active, longstanding RA who had an inadequate response to 1 or more anti-tnf agents and other DMARDs. At 24 weeks, patients receiving rituximab in combination with MTX demonstrated statistically significant improvement in primary and secondary outcomes compared with

12 2804 COHEN ET AL patients receiving placebo with MTX. The proportion of patients who met the primary end point by achieving an ACR20 response rate was significantly greater in rituximab-treated patients (P ). A greater proportion of patients in the rituximab group also demonstrated ACR50 and ACR70 responses at week 24 (P ) and moderate-to-good EULAR responses compared with patients in the placebo group (P ). Mean decreases from baseline in DAS28 scores were also significantly greater in rituximab-treated patients. Overall, patients receiving rituximab demonstrated statistically significant and clinically meaningful improvements in all individual measures used to calculate the ACR response. The efficacy of rituximab is also reflected in the lower dropout rate because of an insufficient therapeutic response among patients receiving rituximab. The results of this trial are clinically important, demonstrating that rituximab is a therapeutic option for RA patients with an inadequate response to 1 or more anti-tnf agents. Patients receiving rituximab also demonstrated clinically meaningful and statistically significant improvements in their levels of disability, health-related quality of life, and fatigue, as measured by improvements in patient-reported outcomes on the HAQ DI, SF-36, and FACIT-F, respectively. A significant reduction in fatigue and improvements in both the SF-36 physical health (chiefly body pain) and mental health component scores, were also observed in patients receiving rituximab as compared with patients receiving placebo. In addition, for the first time in a patient population who had an inadequate response to 1 or more anti-tnf agents, an exploratory analysis of radiographic data at 24 weeks demonstrated a trend toward less progression of structural joint damage. The mean increases from baseline in the total Genant-modified Sharp scores, erosion scores, and joint space narrowing scores at 24 weeks in rituximab-treated patients were half those in placebo-treated patients, with the joint space narrowing score reaching statistical significance at week 24. It is important to recognize that this portion of the trial was only 6 months in duration, allowed rescue therapy between weeks 16 and 24 (with 40% of the placebo group and 13% of the rituximab group entering the rescue protocol at or after week 16), and included intravenous and oral steroids for the first 2 weeks in both the placebo and rituximab groups. While glucocorticoids have been shown to reduce the rate of radiographic progression of RA (31), all patients in this trial received glucocorticoids. Therefore, the effects of glucocorticoids on radiographic progression could not be assessed. Additionally, all patients had previously taken anti-tnf agents. Glucocorticoid administration and previous therapy with anti-tnf agents may have affected the ability to detect significant differences in radiographic changes between treatment groups. The effect of rituximab on radiographic progression will be further evaluated in longer-term followup from this study as well as in an additional study. The responses seen in this study population clearly support the hypothesis that B cells are key contributors to the immunopathogenesis of RA through several potential mechanisms (12,32 35). The patients who were RF seronegative demonstrated a similar treatment effect as those who were seropositive, further suggesting that rituximab may act through several different pathways central to the immunopathogenesis of RA: those involving the production of RF and other autoantibodies, as well as pathways involving T cell activation and production of cytokines by macrophages or directly by B cells. As expected from its mechanism of action, treatment with rituximab led to a rapid and complete depletion of CD20 peripheral B cells. Because the CD20 marker is not expressed on stem cells and other precursor cells or plasma cells (24), it is not surprising that the mean levels of IgG, IgM, and IgA generally remained within the normal range throughout the 24 weeks. The prolonged depletion of CD20 peripheral B cells in rituximab-treated patients raises the question of whether these patients may have a greater risk of infection, although the data from this study do not appear to identify a significantly greater risk through 24 weeks of the study. Additionally, in the small number of patients with low immunoglobulin concentrations, the frequency of infections did not appear to be different from that in patients with normal immunoglobulin levels. Further, while the overall incidence and rate of serious infections were slightly higher in patients treated with rituximab, there was nothing atypical about the course of these infections, and there were no reported cases of tuberculosis or opportunistic infections. Infections were successfully treated with appropriate therapy, with no fatal outcome. Longer-term followup of rituximab-treated patients is required before a more definitive assessment of their infection risk can be made. The overall safety profile of rituximab observed in this trial using a single course of 2 infusions is consistent with the worldwide safety experience with rituximab, which includes more than 730,000 patient exposures, most of which were in patients with non-

13 RITUXIMAB IN RA REFRACTORY TO ANTI-TNF THERAPY 2805 Hodgkin s lymphoma. In this trial, the most common AE reported was exacerbation of RA, which was reported by twice as many patients in the placebo group, further validating the efficacy of rituximab in controlling the disease in this difficult-to-treat population. Acute infusion reactions were seen in a higher proportion of rituximab-treated patients. As expected, the frequency of these events was greater during the first infusion than during the second. This observation is consistent with other rituximab trials in patients with both lymphoma and RA, although overall, the incidence and severity of infusion-related AEs in this patient population is much lower than that reported in lymphoma patients, who may be prone to such events because of higher B cell burdens (36,37). With the exception of acute infusion reactions and, to a lesser extent, the rate of serious infections (5.2 per 100 patient-years for the rituximab group versus 3.7 per 100 patient-years in the placebo group), there were no obvious increases in AEs associated with rituximab therapy. In conclusion, our results demonstrate that a single course of two 1,000-mg infusions of rituximab given 2 weeks apart, in combination with glucocorticoids and background MTX, produced significant clinical and functional benefits at 24 weeks in patients with longstanding and active RA who had an inadequate response to 1 or more anti-tnf therapies. This suggests that rituximab may have an important place in the treatment of patients with RA who demonstrate an inadequate response to currently available TNF inhibitors. These results also expand the efficacy and safety profile of rituximab observed in earlier studies of patients with RA in whom previous DMARD therapy had failed (38 41). Ongoing studies will continue to examine the efficacy including the efficacy of repeated courses and longterm safety of rituximab in the treatment of patients with RA. ACKNOWLEDGMENTS The authors wish to thank Genentech for their support in the preparation of the manuscript. The authors wish to thank the principal investigators of the REFLEX Trial Group. The authors are also grateful to F. Magrini, P. Lehane, S. Leathers, J. Kalsi, A. Donohoe, A. Tanveer, J. Smith, K. Rowe, B. Mistry, and N. Saiedabadi (Roche Products Ltd.) and to T. Kheoh, K. Gilder, J-W Wan, J. Torrington, and M. Weaver (Biogen Idec). REFERENCES 1. Sangha O. Epidemiology of rheumatic diseases. Rheumatology (Oxford) 2000;39 Suppl 2: Smith JB, Haynes MK. Rheumatoid arthritis a molecular understanding. Ann Intern Med 2002;136: Lawrence RC, Helmick CG, Arnett FC, Deyo RA, Felson DT, Giannini EH, et al. Estimates of the prevalence of arthritis and selected musculoskeletal disorders in the United States. Arthritis Rheum 1998;41: Corbett M, Dalton S, Young A, Silman A, Shipley M. Factors predicting death, survival and functional outcome in a prospective study of early rheumatoid disease over fifteen years. Br J Rheumatol 1993;32: Pincus T, Callahan LF, Sale WG, Brooks AL, Payne LE, Vaughn WK. Severe functional declines, work disability, and increased mortality in seventy-five rheumatoid arthritis patients studied over nine years. Arthritis Rheum 1984;27: Keystone EC, Schiff MH, Kremer JM, Kafka S, Lovy M, DeVries T, et al. Once-weekly administration of 50 mg etanercept in patients with active rheumatoid arthritis: results of a multicenter, randomized, double-blind, placebo-controlled trial. Arthritis Rheum 2004;50: Lipsky PE, van der Heijde DM, St. Clair EW, Furst DE, Breedveld FC, Kalden JR, et al, and the Anti-Tumor Necrosis Factor Trial in Rheumatoid Arthritis with Concomitant Therapy Study Group. Infliximab and methotrexate in the treatment of rheumatoid arthritis. N Engl J Med 2000;343: Weinblatt ME, Keystone EC, Furst DE, Moreland LW, Weisman MH, Birbara CA, et al. Adalimumab, a fully human anti tumor necrosis factor monoclonal antibody, for the treatment of rheumatoid arthritis in patients taking concomitant methotrexate: the ARMADA trial. Arthritis Rheum 2003;48: Weinblatt ME, Kremer JM, Bankhurst AD, Bulpitt KJ, Fleischmann RM, Fox RI, et al. A trial of etanercept, a recombinant tumor necrosis factor receptor:fc fusion protein, in patients with rheumatoid arthritis receiving methotrexate. N Engl J Med 1999; 340: 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: Van Vollenhoven RF, Carli CC, Bratt J, Klareskog L. Six year report of the STURE Registry for biologicals in rheumatology: satisfactory overall results but plenty of room for improvement [abstract]. Arthritis Rheum 2005;52 Suppl 9:S Edwards JC, Cambridge G, Abrahams VM. Do self-perpetuating B lymphocytes drive human autoimmune disease? Immunology 1999;97: Panayi GS. The immunopathogenesis of rheumatoid arthritis. Br J Rheumatol 1993;32 Suppl 1: Dorner T, Burmester GR. The role of B cells in rheumatoid arthritis: mechanisms and therapeutic targets. Curr Opin Rheumatol 2003;15: Olsen NJ, Stein CM. New drugs for rheumatoid arthritis. N Engl J Med 2004;350: Metlay JP, Pure E, Steinman RM. Control of the immune response at the level of antigen-presenting cells: a comparison of the function of dendritic cells and B lymphocytes. Adv Immunol 1989;47: Pistoia V, Corcione A. Relationships between B cell cytokine production in secondary lymphoid follicles and apoptosis of germinal center B lymphocytes. Stem Cells 1995;13: Takemura S, Klimiuk PA, Braun A, Goronzy JJ, Weyand CM. T cell activation in rheumatoid synovium is B cell dependent. J Immunol 2001;167: 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: Edwards JC, Cambridge G. Sustained improvement in rheumatoid

14 2806 COHEN ET AL arthritis following a protocol designed to deplete B lymphocytes. Rheumatology (Oxford) 2001;40: 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: 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. Biochem Soc Trans 1997;25: Clynes RA, Towers TL, Presta LG, Ravetch JV. Inhibitory Fc receptors modulate in vivo cytotoxicity against tumor targets. Nat Med 2000;6: Sell S, Max EE. All about B cells. In: Sell S, Max EE, editors. Immunology, immunopathology, and immunity. 6th ed. Washington DC: ASM Press; p 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: Genant HK, Jiang Y, Peterfy C, Lu Y, Redei J, Countryman PJ. Assessment of rheumatoid arthritis using a modified scoring method on digitized and original radiographs. Arthritis Rheum 1998;41: Cancer Therapy Evaluation Program: Common Toxicity Criteria version 2.0 (CTC). Bethesda (MD): Department of Health and Human Services, National Institutes of Health, National Cancer Institute; Felson DT, 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: 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: 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: Kirwan JR, for the Arthritis and Rheumatism Council Low-Dose Glucocorticoid Study Group. The effect of glucocorticoids on joint destruction in rheumatoid arthritis. N Engl J Med 1995;333: Gause A, Berek C. Role of B cells in the pathogenesis of rheumatoid arthritis: potential implications for treatment. Bio- Drugs 2001;15: Reparon-Schuijt CC, van Esch WJ, van Kooten C, Ezendam NP, Levarht EW, Breedveld FC, et al. Presence of a population of CD20, CD38 B lymphocytes with defective proliferative responsiveness in the synovial compartment of patients with rheumatoid arthritis. Arthritis Rheum 2001;44: Shiokawa S, Matsumato N, Nishimura J. Clonal analysis of B cells in the synovial membrane of patients with rheumatoid arthritis. Scand J Rheumatol 2003;32: Zhang Z, Bridges SL Jr. Pathogenesis of rheumatoid arthritis. Role of B lymphocytes. Rheum Dis Clin North Am 2001;27: Kimbey E. Tolerability and safety of rituximab (MabThera). Cancer Treat Rev 2005;31: McLaughlin P, Hagemeister FB, Grillo-Lopez AJ. Rituximab in indolent lymphoma: the single-agent pivotal trial. Semin Oncol 1999;26 Suppl 14: Edwards JC, Szczepanski L, Szechinski J, Filipowicz-Sosnowska A, Close D, Stevens RM, et al. Efficacy and safety of rituximab, a B-cell targeted chimeric monoclonal antibody: a randomized, placebo-controlled trial in patients with rheumatoid arthritis [abstract]. Arthritis Rheum 2002;46 Suppl 9:S Emery P, Fleischmann R, Filipowicz-Sosnowska A, Schechtman J, Szczepanski L, Kavanaugh A, et al, for the DANCER Study Group. The efficacy and safety of rituximab in patients with active rheumatoid arthritis despite methotrexate treatment: results of a phase IIb randomized, double-blind, placebo-controlled, doseranging trial. Arthritis Rheum 2006;54: 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 9;S Pavelka K, Emery P, Filipowicz-Sosnowska A, Nahir M, Kaell A, Baldassare A, et al. Efficacy and safety following repeated courses of rituximab in patients with active rheumatoid arthritis [abstract]. Ann Rheum Dis 2005;64 Suppl III:435. APPENDIX A: PRINCIPAL INVESTIGATORS OF THE REFLEX TRIAL GROUP The principal investigators of the REFLEX Trial Group are as follows: Drs. J. Alloway, C. Appleboom, E. Arthur, A. Ashafzadeh, A. Baldassare, J. Bathon, G. Bayliss, M. Bell, S. Berney, M. Bingham, K. Blocka, E. Boling, S. Bombardieri, R. Brasington, B. Bresnihan, S. Bruno, G. Burmester, H. Busch, G. Butler, M. Carrabba, D. Caspi, A. Chubrick, A. Cividino, W. Clair, C. Codding, S. Cohen, J. Condemi, S. Cooper, J. Cush, S. De Vita, J. Devogelaer, R. Digiovanni, J. Donohue, A. Dooley, M. Dougados, D. Doyle, M. Ellman, P. Emery, M. Fairfax, G. Ferraccioli, J. Fiechtner, R. Fife, J. Forstot, P. Freeman, E. Fung, R. Furie, M. Genovese, R. Gorla, M. Greenwald, D. Halter, J. Isaacs, C. Jackson, A. Kaell, R. Katz, C. Kaufmann, A. Kavanaugh, E. Keystone, J. Kim, J. Kremer, T. Kvien, P. Langevitz, X. Le Loët, M. Liam, J. Loveless, C. Lue, M. Luggen, D. MacCarter, W. Maksymowych, E. Malaise, D. Malone, D. Mandel, X. Mariette, A. Martin, E. Matteson, B. Mazières, F. McKenna, P. Mease, K. Mikkelsen, K. Miller, Y. Molad, J. Molitor, M. Molloy, L. Moreland, F. Murphy, M. Nahir, S. Newell, D. Noritake, H. Nuesslein, R. Pappu, E. Peters, J. Pope, E. Quaidoo, T. Riise, I. Rosner, A. Rubinow, E. Ruderman, C. Saadeh, J. Sany, P. Saxe, F. Scott, T. Sheeran, Y. Sherrer, J. Sibilia, D. Smith, D. Stahl, J. Sundy, J. Taborn, P. Tak, H. P. Tony, J. Udell, A. Uknis, M. Veys, D. Wallace, S. Wassenberg, C. Wiesenhutter, W. Wilke, S. Wolfe, D. Yocum.