B Cell Activation Biomarkers as Predictive Factors for the Response to Rituximab in Rheumatoid Arthritis

Transcription

1 ARTHRITIS & RHEUMATISM Vol. 63, No. 4, April 2011, pp DOI /art , American College of Rheumatology B Cell Activation Biomarkers as Predictive Factors for the Response to Rituximab in Rheumatoid Arthritis A Six-Month, National, Multicenter, Open-Label Study Jérémie Sellam, 1 Houria Hendel-Chavez, 2 Stéphanie Rouanet, 3 Karim Abbed, 2 Bernard Combe, 4 Xavier Le Loët, 5 Jacques Tebib, 6 Jean Sibilia, 7 Yassine Taoufik, 2 Maxime Dougados, 8 and Xavier Mariette 1 Objective. To examine whether serum B cell markers can predict response to rituximab, a B cell ClinicalTrials.gov identifier: NCT Supported by Roche France. 1 Jérémie Sellam, MD, PhD (current address: Hôpital Saint- Antoine, Assistance Publique Hôpitaux de Paris, and Université Pierre et Marie Curie Paris 6, Paris, France), Xavier Mariette, MD, PhD: Hôpital Bicêtre, Assistance Publique Hôpitaux de Paris, INSERM U1012, and Université Paris-Sud 11, Le Kremlin Bicêtre, France; 2 Houria Hendel-Chavez, PhD, Karim Abbed, MD, Yassine Taoufik, PhD: Hôpital Bicêtre, Assistance Publique Hôpitaux de Paris, and Université Paris-Sud 11, Le Kremlin Bicêtre, France; 3 Stéphanie Rouanet, PhD: Roche, Neuilly/Seine, France; 4 Bernard Combe, MD, PhD: Lapeyronie University Hospital, Montpellier I University, and UMR5535, Montpellier, France; 5 Xavier Le Loët, MD, PhD: Rouen University Hospital and INSERM U905, Rouen, France; 6 Jacques Tebib, MD, PhD: Centre Hospitalier Lyon-Sud, Pierre-Bénite, France; 7 Jean Sibilia, MD, PhD: Hôpitaux Universitaires de Strasbourg and Université de Strasbourg, Strasbourg, France; 8 Maxime Dougados, MD, PhD: Paris-Descartes University, UPRES-EA 4058, and Hôpital Cochin, Paris, France. Drs. Combe, Le Loët, Tebib, Sibilia, Dougados, and Mariette have received honoraria from Roche for serving on the scientific committee of the SMART study (less than $10,000 each). Dr. Sellam has received consulting fees, speaking fees, and/or honoraria from Roche (less than $10,000). Dr. Combe has received consulting fees, speaking fees, and/or honoraria from Roche, Wyeth-Pfizer, Schering- Plough, and UCB (less than $10,000 each). Dr. Le Loët has received consulting fees, speaking fees, and/or honoraria from Abbott, Pfizer, Roche, Sanofi-Aventis, and Schering-Plough (less than $10,000 each). Dr. Sibilia has received consulting fees, speaking fees, and/or honoraria from Roche (less than $10,000). Dr. Dougados has received research grants and honoraria from Roche to conduct studies and/or to participate at symposia organized by Roche (less than $10,000 each). Dr. Mariette has received consulting fees, speaking fees, and/or honoraria from Roche (less than $10,000). Address correspondence to Jérémie Sellam, MD, PhD, Service de Rhumatologie, Hôpital Saint-Antoine, 184 Rue du Faubourg Saint-Antoine, Paris, France ( jeremie.sellam@sat.aphp.fr); or to Xavier Mariette, MD, PhD, Service de Rhumatologie, Hôpital de Bicêtre, 78 Rue du Général Leclerc, Le Kremlin Bicêtre, France ( xavier.mariette@bct.ap-hop-paris.fr). Submitted for publication August 30, 2010; accepted in revised form December 30, depleting monoclonal antibody, in patients with refractory rheumatoid arthritis (RA). Methods. This rituximab re-treatment dose study (SMART [essai MAbthera sur la dose de Re- Traitement]) involved 208 patients with refractory RA. Serum markers of B cell activation (anti cyclic citrullinated peptide [anti-ccp] antibodies, rheumatoid factor [RF], serum IgG, IgA, and IgM levels, serum and free light chains, and serum BAFF) were assessed before the first rituximab cycle (1,000 mg on days 1 and 15). Univariate and multivariate analyses were performed to identify factors associated with a European League Against Rheumatism (EULAR) response at 24 weeks. Results. There were 149 responders (72%). Two baseline factors were associated with a EULAR response at 24 weeks in multivariate analysis: the presence of anti-ccp antibodies or RF (odds ratio 3.5 [95% confidence interval ]) and a serum IgG concentration above normal (odds ratio 2.11 [95% confidence interval ]), with synergy between them (odds ratio 6.0 [95% confidence interval ]). Conclusion. The presence of RF or anti-ccp antibodies and elevated IgG are 2 simple biomarkers that can be used routinely before therapy to predict response to rituximab in patients with refractory RA. Biologic therapies have profoundly changed the course of rheumatoid arthritis (RA), but factors that predict response, which could be used to optimize the use and selection of these costly agents that have potentially severe side effects, have not been identified. The anti-cd20 monoclonal antibody that targets B cells, rituximab, is an effective treatment for RA that is refractory to tumor necrosis factor (TNF) blockade (1,2). Because B cells play critical roles in RA (3 7) and 933

2 934 SELLAM ET AL represent the target of rituximab, serum biomarkers of B cell activation are potential predictive factors for rituximab efficacy (3 8). In the present study we investigated the relevance of the baseline concentrations of serum markers of B cell activation (anti cyclic citrullinated peptide [anti-ccp] antibodies, rheumatoid factor [RF], serum IgG, IgA, and IgM levels, serum and free light chains, and serum BAFF [or B lymphocyte stimulator] [9 12]) for predicting the response to rituximab among patients with RA. PATIENTS AND METHODS Patients. A total of 208 patients who had had RA for at least 6 months and who fulfilled the American College of Rheumatology 1987 revised classification criteria (13) were included in the SMART (essai MAbthera sur la dose de Re-Traitement) study. This study was a 2-year, national, multicenter, randomized, open-label study evaluating the efficacy and tolerability of 2 doses of rituximab for re-treatment after 1 course of rituximab (1,000 mg on days 1 and 15). A list of the SMART study investigators is provided in Appendix A. All patients had active disease, defined by a Disease Activity Score in 28 joints (DAS28) (14) using the C-reactive protein level (DAS28-CRP) of 3.2 associated with 6 of66 swollen joints and 6 of 68 tender joints, or a CRP level of 10 mg/liter (normal 5 mg/liter), or an erythrocyte sedimentation rate of 28 mm/hour (15). Each patient received a stable dose of methotrexate ( 10 mg/week for at least 4 weeks) and had experienced an inadequate response or an intolerance to TNF inhibitors; alternatively, patients could have had contraindications to the use of TNF inhibitors. All patients had discontinued etanercept for at least 4 weeks and had discontinued infliximab or adalimumab for at least 8 weeks. Steroids (prednisone or equivalent 10 mg/day) and nonsteroidal antiinflammatory drugs were permitted if their doses had remained stable for at least 4 weeks and 2 weeks, respectively, before enrollment. Study protocol. All patients received 1 course of rituximab (infusions of 1,000 mg on days 1 and 15). Treatment efficacy was evaluated 24 weeks after the first rituximab infusion, according to the European League Against Rheumatism (EULAR) response criteria (16 18). These response criteria were used to classify patients as good or moderate responders or as nonresponders and are based on the individual amount of change in the DAS28 and the level (low, moderate, or high) of DAS reached. For example, a good response corresponds to a DAS28 of 3.2 and a decrease in DAS28 from baseline of 1.2. Responders were randomly assigned to receive re-treatment with either one 1,000-mg dose or two 1,000-mg doses 2 weeks apart. This re-treatment was given every 6 months if the patient s DAS28-CRP was 3.2. In this ancillary study, we investigated whether the baseline values of the serum markers of B cell activation could be used to predict the response to rituximab 24 weeks before randomization. This study was approved by the local ethics committee (Groupe Hospitalier Pitié-Salpêtrière), and all patients provided informed consent. Serum B cell markers. Serum samples obtained before the first rituximab infusion were used to measure RF by nephelometry (BN prospec; Dade Behring) and anti-ccp IgG antibodies by enzyme-linked immunosorbent assay (ELISA) (DiaSorin). The cutoff was 15 IU/liter for RF and 25 IU/liter for anti-ccp antibodies. BAFF was measured by Quantikine ELISA (R&D Systems). Serum concentrations of IgG, IgA, IgM, and free light chains were assessed by nephelometry. The upper limit of normal (ULN) was gm/liter for IgG, 2.69 gm/liter for IgA, 2.09 gm/liter for IgM, 19.4 mg/liter for free light chains, and 26.3 mg/liter for free light chains. The lower limit of normal (LLN) was 6.82 gm/liter for IgG, 0.84 gm/liter for IgA, and 0.74 gm/liter for IgM. The cutoffs for Ig levels are based on testing performed previously in 3,500 French healthy donors. Statistical analysis. Statistical analysis was performed with SAS software, version 9.1. Tests were 2-sided, and a Type I error was set at Continuous data are shown as the mean SD or as the median and range. The relationship between EULAR response at 24 weeks and explanatory variables was analyzed by logistic regression. These variables were selected by univariate logistic regression among the characteristics of patients and B cell activation markers at baseline. Variables significant after univariate regression (P 0.15) were then entered in a stepwise multivariate model (entry level, P 0.15; level for staying in the model, P 0.10) adjusted to the DAS28-CRP. Results are expressed as odds ratios (ORs) with 95% confidence intervals (95% CIs). RESULTS Characteristics of the study population. The 208 RA patients (mean SD age years, mean SD disease duration 13 9 years, 176 women [85%]) included 194 patients with an inadequate response to anti-tnf and 14 for whom anti-tnf was contraindicated. Their baseline characteristics are shown in Table 1. Among the 194 patients who had been treated unsuccessfully with anti-tnf, anti-tnf had no effect in 158, and 32 could not tolerate anti-tnf. Data were not available on 4 patients. Among nonresponder patients, anti-tnf was administered for a median 12.2 months (range months). Anti-TNF was contraindicated in 14 patients because of personal previous neoplasia (n 7), personal or familial demyelinating disease (n 4), previous serious infections (n 2), or latent tuberculosis with side effects of the antituberculosis antibiotics (n 1). Parameters associated with response to rituximab. Of the 208 patients included, there were 149 responders 24 weeks after the cycle of rituximab (72%) (44 good responders [21%] and 105 partial responders [50%]) and 59 nonresponders (28%). The factors associated with a EULAR response in univariate analysis were the presence

3 B CELL ACTIVATION BIOMARKERS AND RESPONSE TO RITUXIMAB IN RA 935 Table 1. Baseline demographic and disease characteristics of the 208 patients* Women, no. (%) 176 (85) Disease duration, years 13 9 RF, no. (%) 136 (65) RF concentration, median (range) IU/liter 98 (17.0 2,430.0) Anti-CCP, no. (%) 159 (76) Anti-CCP concentration, median (range) 740 ( ,194.0) IU/liter DAS28-CRP HAQ DI score CRP, median (range) mg/liter 11.0 ( ) ESR, median (range) mm/hour 28.0 ( ) Radiographic erosions, no. (%) 184 (89) Prednisone 10 mg/day, no. (%) 165 (79) MTX dosage, mg/week Duration of MTX use, years Time since last anti-tnf treatment, 2.7 ( ) median (range) months * Except where indicated otherwise, values are the mean SD. RF rheumatoid factor; anti-ccp anti cyclic citrullinated peptide; DAS28-CRP Disease Activity Score in 28 joints using the C-reactive protein level; HAQ DI Health Assessment Questionnaire disability index; ESR erythrocyte sedimentation rate; MTX methotrexate; anti-tnf anti tumor necrosis factor. of RF or anti-ccp antibodies, an IgG concentration greater than the ULN and an IgA concentration greater than the ULN, the presence of radiographic erosions, a DAS28-CRP of 5.1, the absence of prescribed steroids, and a time of 6 months since the last TNF inhibitor treatment (Table 2). An immunoglobulin concentration below the LLN, serum IgM and free light chain concentrations above the ULN, or the value of the serum BAFF level was not associated with a particular clinical response, nor was the level of anti-ccp or RF in positive patients (data not shown). Multivariate analysis (Table 2) indicated that the presence of autoantibodies (RF or anti-ccp) was associated with clinical response (OR 3.5 [95% CI ], P ), as was a serum IgG concentration above the ULN (OR 2.11 [95% CI ], P 0.04). The presence of 1 of these 2 parameters increased the probability of a EULAR response (OR 2.5 [95% CI ], P 0.03), while the presence of both parameters further increased the probability of this response (OR 6.0 [95% CI ], P ) (Figure 1). When we compared EULAR moderate and good responders, the presence of autoantibodies (RF or anti- CCP) or a serum IgG level above normal was not associated with a particular profile of response (i.e., moderate or good response), with ORs of 1.2 (95% CI ) and 0.9 (95% CI ), respectively, for these 2 parameters (P values for both were not significant). Table 2. Clinical and biologic markers according to EULAR response* Characteristic EULAR nonresponders (n 59) EULAR responders (n 149) Univariate OR (95% CI) P Multivariate OR (95% CI) Radiographic erosions 2.0 ( ) 0.13 No (referent) 10 (42) 14 (58) Yes 49(27) 135 (73) DAS28-CRP 2.4 ( ) (referent) 22 (42) 30 (58) (24) 119 (76) Steroid prescription 0.4 ( ) 0.05 No (referent) 7 (16) 36 (84) Yes 52 (32) 112 (68) Time since last TNF inhibitor therapy 1.7 ( ) months 45 (32) 97 (68) 6 months (referent) 14 (21) 51 (79) RF or anti-ccp 3.2 ( ) ( ) Negative (referent) 19 (50) 19 (50) Positive 40 (23) 130 (77) IgG ULN 2.2 ( ) ( ) 0.04 No (referent) 45 (34) 88 (66) Yes 14 (19) 61 (81) IgA ULN 1.8 ( ) 0.07 No (referent) 41 (33) 83 (67) Yes 18 (21) 66 (79) * Values are the number (%) of responders or nonresponders with a given characteristic. EULAR European League Against Rheumatism; OR odds ratio; 95% CI 95% confidence interval; ULN upper limit of normal (see Table 1 for other definitions). Results of multivariate analysis were adjusted to the DAS28-CRP and the time since the last anti-tnf treatment. P

4 936 SELLAM ET AL Figure 1. Frequency of European League Against Rheumatism response according to the presence of autoantibodies and an IgG concentration above the upper limit of normal (ULN) (12.66 gm/liter). Results are presented as odds ratios (ORs) and 95% confidence intervals for patients having autoantibodies and/or an IgG concentration greater than the ULN compared with patients having no autoantibodies and an IgG concentration less than or equal to the ULN (referent). RTX rituximab; MTX methotrexate; RF rheumatoid factor; anti-ccp anti cyclic citrullinated peptide. DISCUSSION The management of RA has entered a new era with the arrival of biologic agents. However, we still lack the tools needed to predict response to these drugs, to allow us to tailor the treatment regimen. This apparent gap in knowledge may lead to the serial use of several immunosuppressive drugs, which may expose patients to an elevated risk of adverse side effects. This report describes 2 simple serum biomarkers (RF or anti-ccp antibodies and a serum IgG concentration) that may be used to predict the clinical response to a single course of rituximab in patients with refractory RA. Furthermore, these 2 markers may operate synergistically. Serum markers of B cell activation have already been documented in RA patients; they are elevated in established and in early RA, and some of them may be correlated with disease activity (10 12). Their elevated concentration as well as the presence of autoantibodies may indicate that some forms of RA are B cell driven and sensitive to B cell depletion therapy. Recent results presented only in abstract form suggest that the presence of RF or anti-ccp may also be a predictive factor for response to rituximab both in early RA and in established RA (19 23). However, serum concentrations of immunoglobulins, free light chains, and BAFF in these patients have never been measured. We found for the first time that a serum IgG concentration above the ULN is a factor predicting the response to rituximab; furthermore, patients with autoantibodies and an elevated serum IgG concentration had an 85% chance of responding to rituximab. To explain such a synergy between these 2 variables, we can speculate first that the presence of other, still-unknown autoantibodies may account for this increase in IgG, and that these autoantibodies may add to the presence of RF or anti-ccp to yield a B cell signature allowing better efficacy of rituximab. Second, high serum IgG concentrations may reflect a nonspecific polyclonal B cell activation, and the presence of anti-ccp or RF may reflect a more specific autoimmune B cell reactivity, and both of these conditions may synergize to yield a B cell activation state that also leads to better efficacy of rituximab. However, the serum BAFF concentration was not associated with positive response, indicating that this B cell activation is not driven by an increase in BAFF. Moreover, we have not found an association between levels of autoantibodies and EULAR response. This suggests that the presence of autoantibodies reflects a B cell mediated disease, but that high amounts of autoantibodies in RA patients with RF or anti-ccp do not correspond to more potent B cell activation. Of note, it is well recognized that the level of these autoantibodies is not correlated with disease activity. The main advantages of measuring concentrations of RF, anti-ccp, and serum IgG are that the assays are very easy and can all be performed on a standard serum sample, and the cost is low enough to make such assays feasible for all laboratories. An additional advantage to knowing the serum IgG concentration is that it has recently been shown in the French AutoImmunity and Rituximab registry that a baseline serum IgG concentration of 6 gm/liter is predictive of severe infection following rituximab treatment (24). Analysis of the blood B cell subpopulations may also help predict the response to rituximab, but this is technically more difficult since it must be carried out on a fresh blood sample, requires experienced technicians, and is much more expensive. Consequently, while flow cytometry analysis of B cells could also help predict the efficacy of rituximab (25), it is probably inappropriate for use in daily practice. In conclusion, the presence of RF and anti-ccp antibodies and an elevated serum IgG concentration can be used to predict the response to rituximab in patients with refractory RA, since these markers reflect a B cell driven disease. These markers should become an integral part of the risk/benefit analysis used to determine the best therapeutic approach to the management of RA.

5 B CELL ACTIVATION BIOMARKERS AND RESPONSE TO RITUXIMAB IN RA 937 ACKNOWLEDGMENTS We thank Dr. Owen Parkes for editing the English text. We also thank Dr. Rosemary Jourdan and Dr. Nadine Mackenzie (Roche, France) and Dr. Jamila Filipecki (previously at Roche, France) for their participation in the initial conception of the SMART study. AUTHOR CONTRIBUTIONS All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Sellam and Dr. Mariette had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. Sellam, Combe, Le Loët, Sibilia, Taoufik, Dougados, Mariette. Acquisition of data. Sellam, Hendel-Chavez, Abbed, Le Loët, Tebib, Sibilia, Taoufik, Dougados, Mariette. Analysis and interpretation of data. Sellam, Hendel-Chavez, Rouanet, Abbed, Combe, Sibilia, Taoufik, Dougados, Mariette. ROLE OF THE STUDY SPONSOR Roche France designed the SMART study but did not participate in the design, data collection, or interpretation of the results of this ancillary study, which was proposed by an independent scientific committee. Roche France supported the measurement of serum biomarkers and the statistical analysis. Roche France was not involved in the writing of the manuscript. Their agreement to submit the manuscript for publication was not required, their approval of the content of the submitted manuscript was not required, and publication of the manuscript was not contingent upon their approval. REFERENCES 1. Huizinga TW, Pincus T. In the clinic: rheumatoid arthritis. Ann Intern Med 2010;153:ITC Cohen SB, Emery P, Greenwald MW, Dougados M, Furie RA, Genovese MC, et al, for the REFLEX Trial Group. Rituximab for rheumatoid arthritis refractory to anti tumor necrosis factor therapy: results of a multicenter, randomized, double-blind, placebocontrolled, phase III trial evaluating primary efficacy and safety at twenty-four weeks. Arthritis Rheum 2006;54: Eisenberg R, Albert D. B-cell targeted therapies in rheumatoid arthritis and systemic lupus erythematosus. Nat Clin Pract Rheumatol 2006;2: Marston B, Palanichamy A, Anolik JH. B cells in the pathogenesis and treatment of rheumatoid arthritis. Curr Opin Rheumatol 2010;22: Martinez-Gamboa L, Brezinschek HP, Burmester GR, Dorner T. Immunopathologic role of B lymphocytes in rheumatoid arthritis: rationale of B cell-directed therapy. Autoimmun Rev 2006;5: O Neill SK, Glant TT, Finnegan A. The role of B cells in animal models of rheumatoid arthritis. Front Biosci 2007;12: Youinou P, Jamin C, Saraux A. B-cell: a logical target for treatment of rheumatoid arthritis. Clin Exp Rheumatol 2007;25: Dorner T, Kinnman N, Tak PP. Targeting B cells in immunemediated inflammatory disease: a comprehensive review of mechanisms of action and identification of biomarkers. Pharmacol Ther 2010;125: Aho K, Heliovaara M, Knekt P, Reunanen A, Aromaa A, Leino A, et al. Serum immunoglobulins and the risk of rheumatoid arthritis. Ann Rheum Dis 1997;56: Gottenberg JE, Aucouturier F, Goetz J, Sordet C, Jahn I, Busson M, et al. Serum immunoglobulin free light chain assessment in rheumatoid arthritis and primary Sjögren s syndrome. Ann Rheum Dis 2007;66: Gottenberg JE, Busson M, Cohen-Solal J, Lavie F, Abbed K, Kimberly RP, et al. Correlation of serum B lymphocyte stimulator and 2 microglobulin with autoantibody secretion and systemic involvement in primary Sjögren s syndrome. Ann Rheum Dis 2005;64: Gottenberg JE, Miceli-Richard C, Ducot B, Goupille P, Combe B, Mariette X. Markers of B-lymphocyte activation are elevated in patients with early rheumatoid arthritis and correlated with disease activity in the ESPOIR cohort. Arthritis Res Ther 2009;11: R 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: 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: Wells G, Becker JC, Teng J, Dougados M, Schiff M, Smolen J, et al. Validation of the 28-joint Disease Activity Score (DAS28) and European League Against Rheumatism response criteria based on C-reactive protein against disease progression in patients with rheumatoid arthritis, and comparison with the DAS28 based on erythrocyte sedimentation rate. Ann Rheum Dis 2009;68: 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: Van Gestel AM, Anderson JJ, van Riel PL, Boers M, Haagsma CJ, Rich B, et al. ACR and EULAR improvement criteria have comparable validity in rheumatoid arthritis trials. J Rheumatol 1999;26: Van Gestel AM, Haagsma CJ, van Riel PL. Validation of rheumatoid arthritis improvement criteria that include simplified joint counts. Arthritis Rheum 1998;41: Mariette X, Kivitz A, Isaacs JD, Stohl W, Tak PP, Jones RE, et al. Effectiveness of rituximab methotrexate in patients with early active rheumatoid arthritis and disease characteristics associated with poor outcomes [abstract]. Arthritis Rheum 2009;60 Suppl: S Wendler J, Wassenberg S, Tony HP, Bartz-Bazzanella P, Krause A, Rubbert-Roth A, et al. Predictive value of rheumatoid factor titre for treatment response to rituximab in the German rituximab treatment of active rheumatoid arthritis in daily practice trial [abstract]. Ann Rheum Dis 2010;69 Suppl:iii Isaacs JD, Olech E, Tak PP, Deodhar A, Keystone E, Emery P, et al. Autoantibody-positive rheumatoid arthritis patients have enhanced clinical response to rituximab when compared with seronegative patients [abstract]. Ann Rheum Dis 2009;68 Suppl:iii Van Vollenhoven RF, Chatzidionysiou K, Nasonov E, Lukina G, Hetland ML, Tarp U, et al. Six-month results from the Collaborative European REgistries for Rituximab in Rheumatoid Arthritis (CERERRA): efficacy of rituximab is highest in RF-positive patients and in those who failed at most one prior anti-tnf [abstract]. Arthritis Rheum 2009;60 Suppl:S Tak PP, Cohen SB, Emery P, Saadeh CK, De Vita S, Donhove JP, et al. Clinical response following the first treatment course with

6 938 SELLAM ET AL rituximab: effect of baseline autoantibody status (RF, anti-ccp) [abstract]. Ann Rheum Dis 2007;66 Suppl:ii Gottenberg JE, Ravaud P, Bardin T, Cacoub P, Cantagrel A, Combe B, et al, on behalf of the Investigators of the AutoImmunity and Rituximab registry and the French Society of Rheumatology. Risk factors of severe infections in patients with rheumatoid arthritis treated with rituximab in the AutoImmunity and Rituximab registry. Arthritis Rheum 2010;62: Vital EM, Dass S, Rawstron AC, Buch MH, Goeb V, Henshaw K, et al. Management of nonresponse to rituximab in rheumatoid arthritis: predictors and outcome of re-treatment. Arthritis Rheum 2010;62: APPENDIX A: SMART STUDY INVESTIGATORS The SMART study investigators, in addition to the authors of this article, are as follows (all in France): Dr. I. Azais (Poitiers), Dr. J. C. Balblanc (Belfort), Dr. F. Berenbaum (Paris), Dr. P. Bertin (Limoges), Dr. M.-C. Boissier (Bobigny), Dr. P. Bourgeois (Paris), Dr. A. Cantagrel (Toulouse), Dr. P. Carli (Toulon), Dr. P.-Y. Chouc (Marseille), Dr. M. Couret (Valence), Dr. L. Euller-Ziegler (Nice), Dr. P. Fardellone (Amiens), Dr. P. Fauquert (Berck/Mer), Dr. R.-M. Flipo (Lille), Dr. P. Gaudin (Echirolles), Dr. J.-L. Grauer (Aix en Provences), Dr. A. Heraud (Libourne), Dr. P. Hilliquin (Corbeil), Dr. S. Hoang (Vannes), Dr. E. Houvenagel (Lomme), Dr. D. Keita (Paris), Dr. K. Lassoued (Cahors), Dr. L. Le Dantec (Lievin), Dr. J.-M. Le Parc (Boulogne), Dr. L. Lequen (Pau), Dr. F. Lioté (Paris), Dr. C. Marcelli (Caen), Dr. O. Meyer (Paris), Dr. J.-L. Pellegrin (Pessac), Dr. A. Perdriger (Rennes), Dr. G. Rajzbaum (Paris), Dr. S. Redeker (Abbeville), Dr. J.-M. Ristori (Clermont- Ferrand), Dr. A. Saraux (Brest), Dr. G. Tanguy (La Roche sur Yon), Dr. T. Thomas (Saint-Priest-en-Jarez), Dr. L. Zabraniecki (Toulouse), Dr. C. Zarnitski (Montivilliers). DOI /art Erratum In the article by Montes et al in the March 2011 issue of Arthritis & Rheumatism (pages ), there was an error in the analyses comparing RF status in the different subgroups of patients with RA. The last full sentence on page 656 should have read, Within the group of anti-ccp antibody positive patients, the prevalence of RF positivity was higher in the anti-ccp /anti CEP-1 subgroup than in the anti-ccp / anti CEP-1 (85.6% versus 76.7%, respectively) (Figure 2), but this difference was not significant (P 0.09). The correct version of Figure 2 is shown below. We regret the error.