Management of nonresponse to rituximab in rheumatoid arthritisPredictors and outcome of re-treatment.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 62, No. 5, May 2010, pp 1273–1279 DOI 10.1002/art.27359 © 2010, American College of Rheumatology Management of Nonresponse to Rituximab in Rheumatoid Arthritis Predictors and Outcome of Re-treatment E. M. Vital,1 S. Dass,1 A. C. Rawstron,2 M. H. Buch,1 V. Goëb,1 K. Henshaw,1 F. Ponchel,1 and P. Emery1 compared. First-cycle nonresponders (n ⴝ 25) were treated with a second cycle of rituximab at least 6 months after the first cycle. HSFC was performed at baseline, immediately prior to the second infusion (week 2), 1 month after the second infusion (week 6), and then every 3 months for each cycle of rituximab. Complete B cell depletion was defined as being <0.0001 ⴛ 109 cells/liter. Results. At baseline, the number of preplasma cells was significantly higher in first-cycle nonresponders than in first-cycle responders (P ⴝ 0.003). Following the first infusion of the first cycle of rituximab, only 9% of first-cycle nonresponders (3 of 34) exhibited complete depletion of B-lineage cells, compared with 37% of first-cycle responders (22 of 59) (P ⴝ 0.007). Following the first infusion of the second cycle of rituximab, 38% of first-cycle nonresponders exhibited complete depletion. Twenty-six weeks after the second cycle, there was a significant improvement in the Disease Activity Score in 28 joints, with 72% of patients exhibiting a EULAR response. Conclusion. RA patients whose disease did not respond to an initial cycle of rituximab have higher circulating preplasma cell numbers at baseline and incomplete depletion. Our findings indicate that an additional cycle of rituximab administered prior to total B cell repopulation enhances B cell depletion and clinical responses. Objective. A proportion of patients with rheumatoid arthritis (RA) have disease that fails to respond to an initial cycle of rituximab. Using highly sensitive flow cytometry (HSFC), it has been shown that most patients who do not exhibit a response, as measured using the European League Against Rheumatism (EULAR) criteria, have persistent circulating B cell levels at week 2 after initial treatment with rituximab. This study was undertaken to examine whether an additional cycle of rituximab would improve B cell depletion and clinical response in patients whose disease did not respond to the initial cycle. Methods. Patients with RA (n ⴝ 158) were treated with a first cycle of rituximab (2 infusions of 1 gm each). Clinical responses were assessed using EULAR criteria, and patients were categorized as either first-cycle responders or first-cycle nonresponders. Baseline characteristics of first-cycle nonresponders (n ⴝ 38) and first-cycle responders (n ⴝ 65) with complete data were Dr. Vital is recipient of a Research Training Fellowship from the National Institute of Health Research. Dr. Goëb is recipient of a travel grant from the French Society of Rheumatology. Dr. Emery is an Arthritis Research Campaign Professor of Rheumatology. 1 E. M. Vital, MBChB, MRCP, S. Dass, MBBChir, MA, MRCP, M. H. Buch, MBChB, MRCP, PhD, V. Goëb, MD, PhD, K. Henshaw, FIBMS, F. Ponchel, PhD, P. Emery, MA, MD, FRCP: University of Leeds, Chapel Allerton Hospital and Leeds Teaching Hospitals National Health Service Trust, Leeds, UK; 2A. C. Rawstron, PhD: St. James’s Institute of Oncology, Leeds, UK. Drs. Vital, Dass, and Buch have received honoraria from Roche (less than $10,000 each). Dr. Emery has received consulting fees, speaking fees, and/or honoraria from UCB and Centocor (less than $10,000 each) and has served as a study investigator for Centocor. Address correspondence and reprint requests to P. Emery, MA, MD, FRCP, Section of Musculoskeletal Disease, Leeds Institute of Molecular Medicine, Chapel Allerton Hospital, Chapeltown Road, Leeds LS7 4SA, UK. E-mail: firstname.lastname@example.org. Submitted for publication July 6, 2009; accepted in revised form January 11, 2010. Rituximab is an effective therapy for resistant rheumatoid arthritis (RA); good responses may be achieved in patients whose disease has failed to respond to conventional disease-modifying antirheumatic drugs (DMARDs) and to anti–tumor necrosis factor (antiTNF) agents (1). Consequently, patients who do not 1273 1274 VITAL ET AL exhibit an adequate response to rituximab have frequently exhausted all other available therapy options, and treatment of these patients is challenging. Anti-TNF or newer biologic agents may be considered in this context, although there are limited data on the safety of these agents in patients in whom the number of B cells remains below normal. The findings of initial studies into the use of rituximab appeared to indicate complete depletion of circulating B cells in all patients (2), suggesting that the failure to exhibit a clinical response was due to B cell–independent pathways of inflammation or to the persistence of pathologic plasma cells. Recently, the use of more sensitive methods to measure circulating B cell depletion has offered an alternative explanation. According to highly sensitive flow cytometry (HSFC), 95% of patients who failed to exhibit a response, as measured using the European League Against Rheumatism (EULAR) criteria (3), also had incomplete B cell depletion (4). In addition, studies of synovial biopsies have indicated that, although B cell numbers are reduced after treatment, synovial B cell depletion is seldom complete and that the persistence of synovial B-lineage cells is associated with poorer responses (5–7). These observations suggest that the enhancement of the initial depth of B cell depletion could increase the rate of clinical response by reducing synovial B cell presence. However, it has been unclear what determines the initial depth of B cell depletion and therefore how depletion might be enhanced. Accordingly, we investigated the determinants of the depth of B cell depletion and of clinical response, including levels of peripheral blood B cell subsets, as measured using HSFC. We hypothesized that a second cycle of rituximab, administered before B cells were fully repopulated in patients who did not exhibit a EULAR response after the first cycle, would increase both B cell depletion and the clinical response rate. Herein, we report the findings of our study of clinical responses and the depth of B cell depletion in 25 patients with RA whose disease had not responded to the first cycle of rituximab and who were given a second cycle. PATIENTS AND METHODS Patients and study design. All patients were diagnosed according to the American College of Rheumatology (formerly, the American Rheumatism Association) criteria for RA (8) and were positive for either rheumatoid factor or anti– cyclic citrullinated peptide antibodies or both. All patients had exhibited an inadequate response or a toxic reaction to ⱖ2 nonbiologic DMARDs, including methotrexate. In addition, 62% of patients exhibited an inadequate response or a toxic reaction to ⱖ1 anti-TNF agent. There were 103 patients who received rituximab and had complete data. We have previously reported data regarding B cell depletion and response to the first cycle of rituximab for 60 of these patients (5). All patients received 2 infusions (1 gm each) of rituximab, each preceded by methylprednisolone (100 mg). Patients were maintained on concomitant methotrexate (n ⫽ 73) unless it was not tolerated or contraindicated, in which case leflunomide (n ⫽ 15), azathioprine (n ⫽ 2), or no DMARD (n ⫽ 13) was used. Disease activity was assessed by the Disease Activity Score in 28 joints using the C-reactive protein level (DAS28CRP) (9). Six months after treatment, EULAR response criteria were used to categorize patients into first-cycle responders and first-cycle nonresponders. HSFC was used to enumerate B cell subsets at baseline, immediately prior to the second infusion (week 2), 1 month after the second infusion (week 6), and then every 3 months after each cycle of rituximab. From the first-cycle nonresponders, 25 patients were re-treated with a second cycle, using the same rituximab dose, at a time when circulating B cell levels had not yet returned to baseline levels. DAS28-CRP scores were assessed 6 months after the second cycle and were compared with baseline scores to calculate the EULAR response. B cell HSFC. Peripheral blood B cell subset levels were measured using 6-color flow cytometry, counting 500,000 events to reliably enumerate B cells, as described previously (5). Briefly, B cell levels and subsets were enumerated following standard cell surface staining techniques using a sequential gating strategy. First, B cells were identified using phycoerythrin (PE)–Cy5.5–conjugated CD19 (Invitrogen) and PECy7–conjugated CD38 (BD Biosciences) expression and lightscatter characteristics. Fluorescein isothiocyanate–conjugated CD3 and PE-conjugated CD14 (BD Biosciences) were used to exclude contaminating events, and B cells were classified, according to expression of PE-Cy7–conjugated CD38 and allophycocyanin (APC)–conjugated CD27 (BD Biosciences), as naive (CD19⫹⫹CD27⫺), memory (CD19⫹⫹CD27⫹), and preplasma (CD19⫹/⫺CD27⫹⫹CD38⫹⫹). APC-Cy7– conjugated CD45 (BD Biosciences) was used to identify total leukocytes for calculation of absolute levels of B cell subsets. Statistical analysis. Baseline clinical variables and B cell subsets, as predictors of response, in first-cycle responders were compared with those in first-cycle nonresponders, using univariate logistic regression. In first-cycle nonresponders receiving re-treatment, changes in DAS28 scores and B cell counts over time were compared using paired t-test and Wilcoxon’s matched pairs signed rank test, respectively. P values less than 0.05 were considered significant. RESULTS Prediction of first-cycle nonresponse. Among first-cycle nonresponders, there was a trend toward lower DAS28 scores, and response rates were lower in patients who were not treated with a concomitant DMARD. Consistent with the findings of our previous study (4), complete depletion of B cells (measured using HSFC and defined as a total B cell count that is ⬍0.0001 ⫻ 109 cells/liter after both the first and the NONRESPONSE TO RITUXIMAB IN RA 1275 Figure 1. Baseline levels of B cell subsets in first-cycle nonresponders (NR) and in first-cycle responders (R). Data are shown as box plots. Each box represents the 25th to 75th percentiles. Lines inside the boxes represent the median. Bars outside the boxes represent the 10th and 90th percentiles. Circles and asterisks indicate outliers and extreme outliers, respectively. Note the different scales for each subset. second infusion of rituximab) was a predictor of EULAR response at 6 months. Among first-cycle nonresponders, 9% (3 of 34) had complete depletion, com- pared with 37% of first-cycle responders (22 of 59). Total B cell levels and levels of all circulating B cell subsets was higher at baseline in first-cycle nonre- Table 1. Baseline clinical variables and B cell subsets among responders and nonresponders, as determined using univariate logistic regression* Age DMARD use (n ⫽ 103), no. (%) Mono (n ⫽ 13) Combination (n ⫽ 90) Prior anti-TNF use (n ⫽ 103), no. (%) Yes (n ⫽ 63) No (n ⫽ 40) ESR, mm/hour† DAS28-CRP‡ RF, IU/liter§ B cell subset Naive cells¶ Memory cells¶ Preplasma cells¶ B cell depletion (n ⫽ 93), no. (%) Incomplete (n ⫽ 68) Complete (n ⫽ 25) Responders (n ⫽ 65) Nonresponders (n ⫽ 38) OR (95% CI) P 59.03 (55.97–62.08) 58.08 (53.48–62.67) 1.01 (0.98–1.04) 0.718 4/13 (30.8) 61/90 (67.8) 9/13 (69.2) 29/90 (32.2) 4.73 (1.35–16.65) 0.015 36/63 (57.1) 29/40 (72.5) 7.03 (6.53–7.52) 5.92 (5.63–6.20) 4.84 (4.34–5.34) 27/63 (42.9) 11/40 (27.5) 7.35 (6.55–8.15) 5.56 (5.17–5.96) 5.33 (4.68–5.98) 1.98 (0.84–4.65) 0.118 0.93 (0.76–1.13) 1.33 (0.92–1.93) 0.88 (0.70–1.09) 0.464 0.133 0.229 6.10 (5.79–6.42) 4.86 (4.57–5.15) 2.66 (2.45–2.88) 6.58 (6.17–7.00) 5.60 (5.26–5.94) 3.25 (2.88–3.62) 0.68 (0.44–1.04) 0.67 (0.44–1.03) 0.47 (0.28–0.78) 0.073 0.068 0.003 0.44 (0.25–0.80) 0.007 37/68 (54.4) 22/25 (88.0) 31/68 (45.6) 3/25 (12.0) * Values for continuous variables are the mean (95% confidence interval [95% CI]). OR ⫽ odds ratio; DMARD ⫽ disease-modifying antirheumatic drug; anti-TNF ⫽ anti–tumor necrosis factor; ESR ⫽ erythrocyte sedimentation rate; DAS28-CRP ⫽ Disease Activity Score in 28 joints using the C-reactive protein level; RF ⫽ rheumatoid factor. † Square root value; n ⫽ 63 responders, 36 nonresponders. ‡ n ⫽ 65 responders, 38 nonresponders. § Natural log of (value ⫹ 1); n ⫽ 62 responders, 38 nonresponders. ¶ Natural log of ([value ⫹ 0.0001] ⫻ 10,000). For naive and memory cells, n ⫽ 51 responders, 31 nonresponders; for preplasma cells, n ⫽ 54 responders, 32 nonresponders. 1276 Figure 2. Total B cell levels and levels of B cell subsets in 25 patients with rheumatoid arthritis that did not respond to cycle 1 of rituximab and who were re-treated with a second cycle. Data are shown as box plots. Each box represents the 25th to 75th percentiles. Lines inside the boxes represent the median. Bars outside the boxes represent the 10th and 90th percentiles. Circles and asterisks indicate outliers and extreme outliers, respectively. Note the different scales for each subset and time point (represented as weeks after each cycle). VITAL ET AL sponders than in first-cycle responders (Figure 1 and Table 1). The majority of persistent circulating B cells that were present for 3 months after the first cycle of rituximab were preplasma cells, as reported previously (5). Baseline clinical data, log-transformed B cell variables, and complete/incomplete B cell depletion were evaluated as potential predictors of response, using univariate analysis. DMARD therapy (P ⫽ 0.015), preplasma cell numbers (P ⫽ 0.003), and complete/ incomplete B cell depletion (P ⫽ 0.007) were significantly predictive of response to rituximab (Table 1). Re-treatment of nonresponders: enhanced B cell depletion. Total B cell counts after each cycle of rituximab are shown in Figure 2. Although only 9% of first-cycle nonresponders had complete B cell depletion after the first infusion of cycle 1, 38% had complete B cell depletion after the first infusion of cycle 2. Median B cell numbers were lower at all time points after the second cycle of rituximab compared with median numbers after the first cycle, and the numbers of memory and preplasma cell were significantly reduced 26 weeks after cycle 2, compared with the numbers of cells after cycle 1 (Figure 2). Re-treatment of nonresponders: enhanced clinical responses. As illustrated in Figure 2, rituximab re-treatment was administered to 25 first-cycle nonresponders prior to total peripheral B cell repopulation, with the total number of B cells at baseline for cycle 2 significantly lower than the number at baseline for cycle 1 (P ⬍ 0.0001). Six months after cycle 2, there was a significant improvement in DAS28 scores compared with scores either immediately before cycle 2 (P ⬍ 0.001) or at baseline (P ⬍ 0.001) (Figure 3). After re-treatment with rituximab, 18 patients (72%) exhibited a EULAR response, 8 patients (32%) exhibited a good EULAR response, and 4 patients (16%) had disease classified as being in remission, as compared with baseline. There was also significant improvement in each component of the DAS28 (Figure 3C). Across both cycles, a good or moderate EULAR response occurred in 50% of cases with complete B cell depletion and in 31% of cases with incomplete B cell depletion. Response occurred in 39% of cases receiving a concomitant DMARD and in 25% of cases not receiving a concomitant DMARD. Of the first-cycle nonresponders who exhibited response to the second rituximab cycle, 9 patients experienced a relapse. The median time from the beginning of cycle 2 to relapse was 11 months, which was the same time observed among first-cycle responders who experienced a relapse. No re-treated patient had NONRESPONSE TO RITUXIMAB IN RA 1277 Figure 3. Clinical responses as assessed 6 months after the second cycle of rituximab in 25 patients with rheumatoid arthritis that did not respond to the first cycle. A, European League Against Rheumatism response rates, which include patients who exhibited no response (Non), patients who exhibited a moderate or good response (Mod/Good), patients who exhibited a good response, and patients whose disease was classified as being in remission (Rem). B, Disease Activity Score in 28 joints using the C-reactive protein level (DAS28-CRP) measured at baseline (BL), 6 months after cycle 1 (C1), and 6 months after cycle 2 (C2). C, Components of the DAS28-CRP, including tender joint count (TJC), swollen joint count (SJC), CRP (mg/liter), and the visual analog scale assessment of global health (GH-VAS). ⴱ ⫽ P ⬍ 0.05 versus baseline. Values in B and C are the mean and 95% confidence interval. immunoglobulin titers below the lower limit of normal after cycle 1, but 1 patient developed low IgM after cycle 2. DISCUSSION The results of our previous study indicated that almost all patients who fail exhibit a clinical response to rituximab have persistent circulating B-lineage cells after the first infusion of rituximab and often after both infusions (4). This suggests that these patients may have disease that potentially would respond to B cell depletion if depletion were more effective. The present study extends these observations with 3 key findings. First, patients whose B cells are not adequately depleted and who do not exhibit a clinical response have significantly higher numbers of circulating preplasma cells at baseline. Second, in these patients, a second cycle of rituximab administered prior to total repopulation of B cells results in greater B cell depletion. Third, clinical responses are enhanced following a second cycle of rituximab. These findings have a number of implications for understanding of the role of B cells in RA, for using treatment to target them effectively, and for developing a rational approach to the management of resistant RA. Preplasma cells do not express CD20 (and hence cannot be directly depleted by rituximab) but are generated from CD20⫹ B cells following their participation in germinal center reactions (10). Their circulating halflife is short, and the finding that there are some patients with undetectable preplasma cells after treatment with rituximab demonstrates that these cells may be indirectly depleted. Conversely, their continuing presence in the circulation in the absence of naive and memory B cells suggests continuing B cell activity at other sites. This is consistent with the findings of previous studies that have shown increased numbers of synovial plasma cells in nonresponders at baseline (7) and the persistence of synovial B cells and plasma cells after treatment (5,7). Patients who do not exhibit complete B cell depletion and clinical response after treatment with rituximab may have a higher burden of B cell number, tissue infiltration, and activity. This may serve as a useful predictor of 1278 response and may identify a group of patients who require more intensive B cell–depleting therapy. The observation that B cell depletion could be enhanced by additional cycles of rituximab was not necessarily expected. A number of factors have been suggested as potential contributors to inadequate depletion after treatment with rituximab, and not all of them could necessarily be overcome by a higher dose. Fc␥ receptor polymorphisms affect the degree of B cell depletion in the treatment of lymphoma and systemic lupus erythematosus (11,12) but have not yet been investigated in RA. The B cell survival factor B lymphocyte stimulator and the complement inhibitory factor CD55 are expressed in lymphoid tissues and in the rheumatoid synovium and may inhibit B cell killing by rituximab (13,14). Some patients develop human antichimeric antibodies following initial exposure to rituximab (1), which may be expected to lead to even less effective depletion after subsequent treatments. The key finding from the current study is the demonstration that many patients who initially fail to exhibit a response to rituximab may exhibit a response to a second cycle. There was previously a paucity of data on how best to manage the disease in this challenging group of patients. One approach is to switch to a different biologic agent, such as an anti-TNF␣ agent, abatacept, or tocilizumab. However, data on the safety of using these agents in patients with reduced B cell numbers are limited (15). Previous experience suggests that blocking multiple immune or inflammatory pathways may have unpredictable consequences. In contrast, there are substantial data to indicate the safety of repeat cycles of rituximab (16). While re-treatment was administered early in the present study, incomplete initial B cell depletion, early repopulation, and the longer retreatment interval of subsequent cycles may offset concerns regarding safety. The EULAR consensus statement on rituximab suggests that re-treatment should be considered for residual disease activity, although this was not based on trial data (17). The present study provides the first evidence to support this strategy. Our results require confirmation in larger studies. However, as a result of these observations, retreatment of nonresponders to rituximab is now standard practice in our unit. HSFC is a convenient tool for assessing different therapeutic strategies in B cell– depleting therapies, and future research will use HSFC in order to assess different dosing regimens and patient populations. More rapid suppression of disease would be preferable, and future investigations will address this issue. Since baseline B cell levels and early B cell VITAL ET AL depletion have now been identified as convenient early biomarkers of response, in the future, re-treatment may be instituted more quickly after the first cycle, thereby reducing the time needed to achieve control of disease. Re-treating nonresponders prior to full B cell repopulation is one strategy to enhance depletion in nonresponders, but alternative strategies may include using a different B cell–depleting agent or a different initial dose of rituximab. In conclusion, these data suggest that RA patients whose disease did not respond to rituximab have higher circulating B cell levels at baseline, which can lead to inadequate depletion. However, this can be overcome in a majority of patients, using an additional cycle of rituximab. The hypothesis that patients exhibiting initial nonresponse to rituximab have B cell– independent disease is not supported by our current results. Rather, our findings indicate that the full potential of B cell–depleting therapies has not been realized by current approaches and that more patients than previously thought may benefit from this therapy if its use is refined. ACKNOWLEDGMENT We thank Dr. Elizabeth Hensor (University of Leeds, Chapel Allerton Hospital and Leeds Teaching Hospitals National Health Service Trust, Leeds, UK) for assistance with statistical analysis. 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. Emery had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. Vital, Dass, Ponchel, Emery. Acquisition of data. Vital, Dass, Rawstron, Buch, Goëb, Henshaw, Ponchel, Emery. Analysis and interpretation of data. Vital, Dass, Buch, Goëb, Ponchel, Emery. REFERENCES 1. 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;54:2793–806. 2. Breedveld F, Agarwal S, Yin M, Ren S, Li NF, Shaw TM, et al. Rituximab pharmacokinetics in patients with rheumatoid arthritis: B-cell levels do not correlate with clinical response. J Clin Pharmacol 2007;47:1119–28. 3. Van Gestel AM, Prevoo ML, van ’t Hof MA, van Rijswijk MH, van de Putte LB, van Riel PL. Development and validation of the NONRESPONSE TO RITUXIMAB IN RA 4. 5. 6. 7. 8. 9. 10. 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;39:34–40. Dass S, Rawstron AC, Vital EM, Henshaw K, McGonagle D, Emery P. Highly sensitive B cell analysis predicts response to rituximab therapy in rheumatoid arthritis. Arthritis Rheum 2008; 58:2993–9. Dass S, Burgoyne CH, Vital EM, Reece RJ, Rawstron AC, Ponchel F, Emery P. Reduction in synovial B cells after rituximab in RA predicts clinical response. Ann Rheum Dis 2007;66 Suppl II:ii90. Teng YK, Levarht EW, Hashemi M, Bajema IM, Toes RE, Huizinga TW, et al. Immunohistochemical analysis as a means to predict responsiveness to rituximab treatment. Arthritis Rheum 2007;56:3909–18. Thurlings RM, Vos K, Wijbrandts CA, Zwinderman AH, Gerlag DM, Tak PP. Synovial tissue response to rituximab: mechanism of action and identification of biomarkers of response. Ann Rheum Dis 2008;67:917–25. 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;31:315–24. 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;38:44–8. Radbruch A, Muehlinghaus G, Luger EO, Inamine A, Smith KG, 1279 11. 12. 13. 14. 15. 16. 17. Dorner T, et al. Competence and competition: the challenge of becoming a long-lived plasma cell. Nat Rev Immunol 2006;6: 741–50. Anolik JH, Campbell D, Felgar RE, Young F, Sanz I, Rosenblatt J, et al. The relationship of Fc␥RIIIa genotype to degree of B cell depletion by rituximab in the treatment of systemic lupus erythematosus. Arthritis Rheum 2003;48:455–9. Cartron G, Dacheux L, Salles G, Solal-Celigny P, Bardos P, Colombat P, et al. Therapeutic activity of humanized anti-CD20 monoclonal antibody and polymorphism in IgG Fc receptor Fc␥RIIIa gene. Blood 2002;99:754–8. Gong Q, Ou Q, Ye S, Lee WP, Cornelius J, Diehl L, et al. Importance of cellular microenvironment and circulatory dynamics in B cell immunotherapy. J Immunol 2005;174:817–26. Golay J, Zaffaroni L, Vaccari T, Lazzari M, Borleri GM, Bernasconi S, et al. Biologic response of B lymphoma cells to anti-CD20 monoclonal antibody rituximab in vitro: CD55 and CD59 regulate complement-mediated cell lysis. Blood 2000;95:3900–8. Genovese MC, Breedveld FC, Emery P, Cohen S, Keystone E, Matteson EL, et al. Safety of biological therapies following rituximab treatment in rheumatoid arthritis patients. Ann Rheum Dis 2009;68:1894–7. Keystone E, Fleischman RM, Emery P, Chubrick A, Dougados M, Baldassare AR, et al. Long-term efficacy and safety of a repeat treatment course of rituximab in rheumatoid arthritis patients with an inadequate response to one or more TNF inhibitors. Arthritis Rheum 2006;54 Suppl:S328. Smolen JS, Keystone EC, Emery P, Breedveld FC, Betteridge N, Burmester GR, et al. Consensus statement on the use of rituximab in patients with rheumatoid arthritis. Ann Rheum Dis 2007;66: 143–50.