Original article RHEUMATOLOGY

Transcription

1 RHEUMATOLOGY Rheumatology 2013;52: doi: /rheumatology/kes365 Advance Access publication 22 December 2012 Original article The effects of rituximab therapy on released interferon-c levels in the QuantiFERON assay among RA patients with different status of Mycobacterium tuberculosis infection Yi-Ming Chen 1,2,3, Hsin-Hua Chen 1,3, Kuo-Lung Lai 1,3, *, Wei-Ting Hung 1,3, *, Joung-Liang Lan 1,3,4 and Der-Yuan Chen 1,3,5 Abstract Objective. To investigate the influence of rituximab therapy on released IFN-g levels in the QuantiFERON- TB-Gold In-Tube (QFT-GIT) assay among RA patients of different Mycobacterium tuberculosis infection status. Methods. Change in levels of released IFN-g in the QFT-GIT assay was evaluated in RA patients who had received 1 year of rituximab treatment. A tuberculin skin test was performed using the Mantoux method and the QFT-GIT assay was performed by measuring IFN-g levels in whole blood treated with tuberculosis (TB)-specific antigens. Results. Among 56 patients, 43 patients were presumed to be without latent TB infection (LTBI), 7 patients had LTBI and 6 patients had anti-tnf-associated TB. During the 1-year period of rituximab therapy, no patient developed active TB or had QFT-GIT conversion. No significant change in released IFN-g levels on QFT-GIT assay after 1-year rituximab therapy was observed in patients with LTBI [3.39 (1.21) vs 2.47 (0.82) IU/ml] or in those with anti-tnf-associated TB [1.06 (0.22) vs 0.87 (0.39) IU/ml]. Rituximab did not inhibit TB antigen-stimulated IFN-g production ex vivo. The frequency of circulating CD19 + B cells was significantly decreased [8.56 (0.84) vs 1.52 (0.44)%, P < 0.001], paralleling the decrease in RF titre [318.5 (76.0) vs (32.1) IU/ml, P < 0.001] and DAS28 [6.49 (0.13) vs 4.59 (0.14), P < 0.001] after 1 year of rituximab therapy. However, there was no significant change in the frequency of CD3 + T cells after rituximab therapy. Conclusion. No occurrence of active TB or QFT-GIT conversion was observed in patients receiving 1 year of rituximab therapy. No significant effect of rituximab therapy on IFN-g release levels was observed in patients with LTBI and with anti-tnf-a-associated TB. Rituximab may be an alternative therapeutic agent for these patients. Key words: rituximab, interferon-g, QuantiFERON-TB-Gold In-Tube assay, tuberculosis, rheumatoid arthritis, TNF-a inhibitors. CLINICAL SCIENCE 1 Faculty of Medicine and 2 Institute of Clinical Medicine, National Yang-Ming University, Taipei, 3 Division of Allergy, Immunology and Rheumatology, Taichung Veterans General Hospital, Taichung, 4 Division of Immunology and Rheumatology, China Medical University Hospital, Taichung and 5 Institute of Biomedical Science, National Chung-Hsing University, Taichung, Taiwan. Submitted 9 July 2012; revised version accepted 25 October Correspondence to: Der-Yuan Chen, Taichung Veterans General Hospital, No. 160, Sec.3, Taichung-Kang Road, Taichung 40705, Taiwan. dychen@vghtc.gov.tw *Kuo-Lung Lai and Wei-Ting Hung contributed equally to this study. Introduction An increased prevalence of active tuberculosis (TB) has been reported in RA patients [1], and the risk of active TB is even higher for those receiving anti-tnf-a therapy [2 4]. Guidelines have recommended that effective TB screening should be routinely carried out and prophylactic therapy initiated before starting anti-tnf-a therapy if latent TB infection (LTBI) exists [5, 6]. However, active! The Author Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please journals.permissions@oup.com

2 Yi-Ming Chen et al. TB may occur in RA patients despite prophylactic therapy because of isoniazid resistance or inadequate chemoprophylaxis [7, 8]. It is clinically challenging to manage active RA patients with anti-tnf-associated TB. It is well known that Th1-mediated immunity plays an important role in the response to Mycobacterium tuberculosis infection [9, 10], and T cells as well as macrophages secrete pro-inflammatory cytokines such as IFN-g and TNF-a to defend against M. tuberculosis infection [11, 12]. Rituximab, a monoclonal antibody directed against B-cell marker CD20, has been shown to be effective in RA patients with an inadequate response to anti-tnf-a therapy [13, 14]. The safety profile of extensive use of rituximab in non-hodgkin s lymphoma suggests no increase in the risk of active TB [15], and no active TB was reported in clinical trials of rituximab in RA patients [14, 16]. Furthermore, some case reports have noted that rituximab has been used safely in RA patients with anti-tnf-associated TB [17]. Jung et al. [18] also reported a patient with active RA and a past history of pulmonary TB in whom rituximab monotherapy resulted in a significant reduction in RA activity without adversely affecting the pulmonary lesion. We therefore speculate that rituximab may not increase the same theoretical concerns as TNF-a inhibitors regarding LTBI reactivation in RA patients. In a survey conducted by the Emerging Infections Network (EIN), however, rituximab therapy was found to be associated with active TB in three cases [19]. There is emerging evidence that B cells might play an important role in the containment of M. tuberculosis. Ulrichs et al. [20] showed that B cells are present in active follicle-like centres resembling secondary lymphoid organs that are located on the outer portion of granulomas in TB-infected lung tissue. Maglione et al. [21] demonstrated that B cells have a role in host defence against M. tuberculosis infection, and B cell / mice have enhanced susceptibility to TB infection. Based on these observations, there still exists the possibility that rituximab therapy has the potential to increase active TB disease. However, there is currently no recommendation for LTBI screening before rituximab therapy [22]. The major aim of this longitudinal study was to observe the safety profile of rituximab therapy with regard to the risk of TB and to investigate whether the released IFN-g levels in QuantiFERON-TB-Gold In-Tube (QFT-GIT) assay were affected by 1 year of rituximab therapy in 56 active RA patients of different M. tuberculosis infection status. In addition, we investigated the ex vivo effects of rituximab on IFN-g response to TB-specific antigens. DAS (DAS28) [25], and active disease status was defined as DAS28 >3.2. Therapeutic response was evaluated 6 months after anti-tnf-a therapy, according to the European League Against Rheumatism (EULAR) response criteria [26]. Patients were categorized into nonresponders who had either DAS28 of <1.2 or a DAS28 at 6 months of 55.1 [26]. After anti-tnf-a therapy, rituximab therapy was initiated in 21 (37.5%) patients who were non-responders, 29 (51.8%) patients who had intolerance due to skin reactions and 6 (10.7%) patients who had anti-tnf-associated TB with completion of anti-tubercular therapy (Fig. 1). All enrolled subjects were prescribed fixed-dose treatment with rituximab (1000 mg twice with a 14-day interval) at 6-monthly intervals for a 1-year period. To prevent infusion reactions, methylprednisolone 100 mg was given intravenously before starting rituximab infusion. Their combined medications included oral MTX and prednisolone. The Clinical Research Ethics Committee at our hospital (Taichung Veterans General Hospital, Taiwan) approved this study (ML21738/S08092) and each participant s written consent was obtained according to the Declaration of Helsinki. Tuberculin skin test and QFT-GIT assay All patients were evaluated before anti-tnf-a therapy using a standardized interview, chest radiographs, tuberculin skin test (TST) and QFT-GIT assay. Blood was drawn for the QFT-GIT assay, followed by TST using the Mantoux method, which was performed by intradermal injection of 2 tuberculin units of PPD RT-23 (Staten FIG. 1Flow chart of enrolled RA patients before and after rituximab therapy. Methods Study population Fifty-six patients with active RA [23] who had received anti-tnf-a therapy were enrolled from Taichung Veterans General Hospital in Taiwan where TB was prevalent with a notification rate of 74.6/ in 2002 [24]. The disease activity of RA was assessed by the 28-joint Flow chart shows the distribution of results of TST and QFT-GIT in 56 patients with RA before and after rituximab therapy. Six patients developed active TB after anti-tnf-a therapy. INHP: isoniazid prophylaxis; patients were presumed to have LTBI, if they had concordant positive TST/QFT-GIT results, and those with concordant negative TST/QFT-GIT results were presumed to be without LTBI

3 Rituximab effect on released IFN-c levels in RA Serum Institute, Denmark). If the induration diameter was 55 mm, the result was defined as positive TST [27]. The QFT-GIT assay was performed according to the manufacturer s instructions (Cellestis Ltd, Victoria, Australia). The QFT-GIT results were defined as positive if the IFN-g level was IU/ml in TB-specific antigen (early secreted antigenic target-6, culture filtrate protein-10 and T7.7)-stimulated wells after subtracting the level of the nil well according to the manufacturer s recommendation. QFT-GIT conversion was defined as baseline IFN-g <0.35 IU/ml and follow-up IFN-g IU/ml. Because there is no gold standard for diagnosis of LTBI, we presumed subjects to have LTBI if they had concordant positive TST/QFT-GIT results, and those with concordant negative TST/QFT-GIT results were presumed to have no LTBI. For LTBI-positive patients receiving anti-tnf-a therapy, isoniazid prophylactic therapy was offered for 9 months before rituximab therapy was prescribed. All patients received examination using the QFT-GIT assay before (at least 4 weeks after discontinuation of TNF-a inhibitors) and at the 12th month of rituximab therapy to evaluate its effects on released IFN-g levels in the QFT-GIT assay. Ex vivo effects of rituximab on released IFN-g levels in the QFT-GIT assay To explore ex vivo effects of rituximab on IFN-g production in the QFT-GIT assay, we examined the changes in the released IFN-g levels from whole blood stimulated with TB-specific antigens. Whole-blood samples were obtained from eight RA patients with QFT-GIT-positive result and five patients with active TB. Briefly, 1 ml of whole blood was incubated with 1mg of TB-specific antigens (ESAT-6, CFP-10 and T7.7) and rituximab (Roche Ltd, Basel, Switzerland) at concentrations of 0.1, 1 or 10 mg/ml for 20 h at 37 C, a method modified from one previous report [28]. The levels of supernatant IFN-g were compared with those in samples incubated without rituximab. Determination of levels of RF-IgM, CRP and circulating frequencies of B cells and T cells Levels of RF-IgM and CRP were measured by nephelometry (Dade Behring Inc., Newark, DE, USA). A positive result for RF-IgM was defined as a concentration >15 IU/ml; The frequencies of circulating CD19 + B cells and CD3 + T cells were measured before and at the 12th month of rituximab therapy using flow cytometry analysis. Briefly, the frequencies of circulating B cells and T cells were enumerated following standard cell surface staining techniques using phycoerythrin cyanin 7 (PE-Cy7)-conjugated CD19 and phycoerythrin-texas Red Õ -X (ECD)-conjugated CD3 (Beckman Coulter Inc., CA, USA), respectively. Lymphocytes were gated on the basis of forward- and side-scatter properties and counts of at least events were analysed. The results were analysed using CXP software (Beckman Coulter). Statistical analysis Results are presented as the mean (S.D.) or mean (S.E.M.) unless specified otherwise. For comparison of IFN-g released levels, the frequencies of circulating B cells and T cells, DAS28 and serum levels of RF-IgM as well as CRP before and after 1-year rituximab therapy during follow-up, the Wilcoxon signed rank test was employed. P < 0.05 was considered statistically significant. Results Demographic data, clinical characteristics and results of TST and QFT-G assay As illustrated in Table 1, all RA patients had active disease [mean (S.D.) DAS (0.98) and range ] before starting rituximab therapy. Among the RA patients who did not develop active TB after anti-tnf-a therapy, 43 patients were presumed to be without LTBI, whereas 7 patients were presumed to have LTBI. There were no significant differences in age at entry, percentage of females, proportion of patients with BCG vaccination, positive rate of RF antibodies, DAS28, daily dosage of CSs or proportion of used MTX among RA patients without LTBI, with LTBI and with anti-tnf-a-associated TB. Among six RA patients with anti-tnf-associated TB before rituximab therapy, three (50%) had extrapulmonary involvement, including miliary (two) and pleura involvement (one). Change in IFN-g release levels in RA patients undergoing rituximab therapy During 1 year of rituximab therapy, no patient developed active TB disease or had QFT-GIT conversion. After 1 year of rituximab therapy, there were no significant changes in IFN-g release levels of whole blood treated with TB-specific antigens in RA patients without LTBI [mean (S.E.M.) 0.13 (0.03) vs 0.10 (0.03) IU/ml], in those with LTBI [3.39 (1.21) vs 2.47 (0.82) IU/ml] or in those with recent TB [1.06 (0.22) vs 0.87 (0.39) IU/ml] (Fig. 2A). There were no significant differences in the change of released IFN-g levels between patients with inadequate response and those with intolerance to anti-tnf-a therapy. Change in levels of circulating T cells as well as B cells, serum levels of RF and CRP, and disease activity in RA patients undergoing rituximab therapy After 1 year of rituximab therapy for all RA patients, the frequencies of circulating CD19 + B cells were significantly decreased [mean (SEM) 8.56 (0.84)% vs 1.52 (0.44)%, P < 0.001], paralleling the decrease in the titres of RF [318.5 (76.0) vs (32.1) IU/ml, P < 0.001], serum levels of CRP [2.81 (0.32) vs 0.82 (0.11) mg/dl, P < 0.001] and DAS28 [6.49 (0.13) vs 4.59 (0.14), P < 0.001]. As shown in Fig. 2, the frequency of circulating CD19 + B cells, RF titre and serum levels of CRP and DAS28 were significantly decreased after 1 year of rituximab therapy in 43 LTBI-negative patients, 7 LTBI-positive patients and 6 patients with anti-tnf-associated TB. However, there were no significant changes in the frequencies of 699

4 Yi-Ming Chen et al. TABLE 1 Demographic data and laboratory findings of RA patients of different M. tuberculosis infection status before starting rituximab therapy Without developing active TB LTBI ( ) (n = 43) LTBI (+) (n =7) Anti-TNF-associated TB [2 LTBI ( ) and 4 LTBI (+)] (n =6) Age at entry, years 52.5 (12.0) 51.6 (12.4) 53.5 (9.5) Female, n (%) 39 (90.7) 5 (71.4) 5 (83.3) Disease duration, years 7.4 (2.8) 6.5 (3.4) 7.9 (2.1) Radiographic stage (III + IV), n (%) 40 (93.0) 5 (71.4) 6 (100.0) BCG vaccination, n (%) 41 (95.3) 6 (85.7) 6 (100.0) RF-IgM positivity, n (%) 36 (83.7) 6 (85.7) 6 (100.0) RF-IgM titre, IU/ml 323 (636) 313 (307) 292 (212) CRP, mg/dl 2.7 (2.2) 3.0 (2.9) 3.5 (3.5) Baseline DAS (0.99) 6.90 (0.83) 6.33 (1.06) Daily steroid dose, mg 6.3 (1.4) 6.8 (1.2) 6.5 (1.7) Used MTX, n (%) 42 (97.7) 7 (100) 6 (100) TNF-a inhibitors, n (%) Etanercept 22 (51.2) 4 (57.1) 2 (33.3) Adalimumab 21 (48.8) 3 (42.9) 4 (66.7) Frequency of comorbidities, n (%) Diabetes mellitus 3 (7.0) 0 (0.0) 1 (16.7) Anaemia (<9.0 gm/dl) 8 (18.6) 1 (14.3) 1 (16.7) Values are mean (S.D.) or number (%) of patients. BCG: Bacillus Calmette Guérin. circulating CD3 + T cells after 1 year of rituximab therapy in RA patients of different M. tuberculosis infection status (Fig. 2B). Ex vivo effects of rituximab on released IFN-g levels in the QFT-GIT assay As illustrated in Fig. 3A and 3B, IFN-g production was not inhibited by rituximab at concentrations of mg/ml in RA patients with QFT-GIT-positive results or patients with active TB. There was no significant difference in the percentage of inhibition of TB antigen-stimulated IFN-g production between QFT-GIT-positive patients and active TB patients (Fig. 3C). Association of the degree of B-cell depletion with the decrease in RF titres, DAS28 and IFN-g levels in RA patients undergoing rituximab therapy The degree of B-cell depletion after rituximab therapy was positively correlated with the decrease in RF titres (r = 0.305, P < 0.05) and the decrease in DAS28 (r = 0.264, P = 0.055). However, there was no significant correlation between the degree of B-cell depletion and the change of released IFN-g levels in the QFT-GIT assay. Discussion Consistent with the safety profile of rituximab therapy in clinical trials and cases reports [14 18], our results showed that no patients developed active TB among the 56 RA patients who received 1 year of rituximab therapy, including six patients with anti-tnf-associated TB. Based on these observations, we hypothesized that rituximab treatment would not increase the risk of active TB disease. To test this hypothesis, we conducted the present study, which, to the best of our knowledge, is the first attempt to investigate the effects of rituximab therapy on the released IFN-g levels in the QFT-GIT assay in RA patients of different M. tuberculosis infection status. Our results showed no significant change in the released IFN-g levels or QFT-GIT conversion in RA patients with LTBI or in those with anti-tnf-associated TB. Moreover, rituximab did not inhibit TB antigen-stimulated IFN-g production ex vivo in the present study. Given that IFN-g has a protective role against M. tuberculosis infection [10, 11], our results provide evidence of the safety of rituximab treatment against TB infection and offers supporting evidence for the clinical use of rituximab in RA patients with anti-tnf-associated TB. However, the impact of rituximab therapy on TB risk must be evaluated against the background of increased TB prevalence due to RA itself [1] and regional differences in exposure to M. tuberculosis. It is interesting to note that our results showed no significant change in the proportions of circulating CD3 + T cells among RA patients after 1 year of rituximab. Given that T-cell-mediated immunity plays a critical role in the optimal containment of M. tuberculosis [9, 10], little change in circulating T cells from our RA patients may partly explain the lack of significant change in the released IFN-g levels in the QFT-GIT assay. These observations provide evidence supporting the safety profile of rituximab therapy in our RA patients with LTBI or anti-tnfassociated TB. IFN-g response might predict the development of active TB in LTBI-positive subjects [29], with higher rates of progression to active TB in interferon-g release 700

5 Rituximab effect on released IFN-c levels in RA FIG. 2Released IFN-g levels, circulating immune cells and RA disease activity before and after rituximab therapy. The change in IFN-g released levels in the QFT-GIT assay (A), the percentages of circulating T cells (B) as well as B cells (C), serum levels of RF (D), DAS28 (E) and CRP levels (F) before and after 1 year of rituximab therapy for RA patients without LTBI, with LTBI and with anti-tnf-associated TB (recent TB) and the completion of anti-tubercular therapy (post-c/t). Data are presented as mean (S.E.M.). *P < 0.05, **P < 0.001, vs values before rituximab therapy, determined by the Wilcoxon signed rank test. assay-positive patients [30]. Although Hatemi et al. [31] demonstrated that short-term [3.6 (0.2) months] anti-tnf-a treatment did not cause a significant change in IFN-g response to TB-specific antigens, the results of previous studies showed that TNF-a inhibitors could suppress IFN-g production [28, 32]. In the present study, no significant change in the released IFN-g levels was observed in RA patients who had received 1 year of rituximab therapy. Moreover, our results showed that rituximab did not inhibit TB antigen-stimulated IFN-g production. This discrepancy in the change of released IFN-g levels between TNF-a inhibitors and rituximab therapy may explain the different risk of developing active TB for RA patients receiving different biologics. However, the sample size of our enrolled RA patients undergoing rituximab therapy was too small and the follow-up duration was too short to draw a conclusion regarding active TB risk. There were some limitations in our study. The sample size of rituximab therapy for anti-tnf-associated TB was too small to draw a conclusion for the effects of rituximab 701

6 Yi-Ming Chen et al. FIG. 3M. tuberculosis antigen-stimulated IFN-g production in whole-blood samples treated with serial concentrations of rituximab. Whole-blood samples from eight RA patients with QFT-GIT-positive results (A) and five patients with active TB (B) were incubated with or without rituximab. IFN-g production was shown as a percentage compared with control (without rituximab). The data are presented as median and the 25th percentile (lower bar) to the 75th percentile (upper bar) for each group. (C) Comparison of M. tuberculosis antigen-stimulated IFN-g production in whole-blood samples treated with rituximab 1 mg/ml between QFT-GIT-positive patients and active TB patients. The data are presented as box-plot diagrams, with the box encompassing the range of levels from the 25th percentile (lower bar) to the 75th percentile (upper bar). The horizontal line within the box indicates the median value and the horizontal lines above and below the box represent the maximum and minimum values, respectively, for each group. therapy on QFT-GIT assays. However, there are no case series in the literature of rituximab therapy in RA patients with anti-tnf-associated TB. Although the follow-up period was short (12 months), the data from a national registry indicate that the majority of clinically relevant infections occur within 7 months of the first infusion of rituximab [33], and 1 year was a sufficient duration to observe LTBI reactivation in RA patients receiving anti-tnf-a therapy [8]. However, our results should be interpreted with caution until longer follow-up data are available. In addition, the patients enrolled in this study were not an anti-tnf-a-naïve RA population, and thus our results might not be directly applicable to early or anti-tnf-a-naïve RA patients. Therefore a long-term study enrolling a larger group of patients, including an additional early RA population or a control RA group using non-biologic anti-rheumatic drugs alone, is required to validate our findings. In conclusion, the present study demonstrated that 1 year of rituximab therapy in RA patients was relatively safe compared with anti-tnf-a therapy regarding the risk of TB infection. Our results also showed no significant effect of rituximab therapy on the levels of released IFN-g in the QFT-GIT assay in RA patients, especially in those with LTBI or with anti-tnf-associated TB. This finding is of clinical importance, since a significant suppression of the proportion of T cells and IFN-g production would predispose patients to an increased susceptibility to M. tuberculosis infection [9, 10, 34]. Although the sample size was too small to obtain a definitive conclusion, our data suggest that rituximab therapy may be an alternative therapeutic option for RA patients with anti-tnfassociated TB. However, emerging evidence shows that B cells might play a role in optimizing the host defence against M. tuberculosis infection [19 21], so it is still necessary to assess and monitor the development of active TB in RA patients receiving rituximab therapy, particularly in those who had previously received TNF-a inhibitors. Rheumatology key messages. No significant effect of rituximab therapy on IFN-g released levels was observed in RA patients.. Rituximab may be an alternative therapeutic agent for RA patients with LTBI or anti-tnf-associated TB. Funding: This study was supported by grant NSC B-075 A-001-MY3 from the National Science Council, Taiwan. Disclosure statement: The authors have declared no conflicts of interest. References 1 Carmona L, Hernandez-Garcia C, Vadillo C et al. Increased risk of tuberculosis in patients with rheumatoid arthritis. J Rheumatol 2003;30: Gomez-Reino JJ, Carmona L, Valverde VR et al. Treatment of rheumatoid arthritis with tumor necrosis factor inhibitors may predispose to significant increase in tuberculosis risk: a multicenter active-surveillance report. Arthritis Rheum 2003;48:

7 Rituximab effect on released IFN-c levels in RA 3 Gardam MA, Keystone EC, Menzies R et al. Anti-tumor necrosis factor agents and tuberculosis risk: mechanisms of action and clinical management. Lancet Infect Dis 2003; 3: Keane J. TNF-blocking agents and tuberculosis: new drugs illuminate an old topic. Rheumatology 2005;44: Furst DE, Cush J, Kaufmann S, Siegel J, Kurth R. Preliminary guidelines for diagnosing and treating tuberculosis in patients with rheumatoid arthritis in immunosuppressive trials or being treated with biological agents. Ann Rheum Dis 2002;61: Carmona L, Gomez-Reino JJ, Rodriguez-Valverde V et al. Effectiveness of recommendations to prevent reactivation of latent tuberculosis infection in patients treated with tumor necrosis factor antagonists. Arthritis Rheum 2005; 52: Sichletidis L, Settas L, Spyratos D, Chloros D, Patakas D. Tuberculosis in patients receiving anti-tnf agents despite chemoprophylaxis. Int J Tuberc Lung Dis 2006;10: Chen DY, Shen GH, Chen YM et al. Biphasic emergence of active tuberculosis in rheumatoid arthritis patients receiving TNFa inhibitors: the utility of IFNg assay. Ann Rheum Dis 2012;71: Boom WH. The role of T-cell subsets in Mycobacterium tuberculosis infection. Infect Agents Dis 1996;5: Barnes PF, Wizel B. Type 1 cytokines and the pathogenesis of tuberculosis. Am J Respir Crit Care Med 2000;161: Jurado JO, Pasquinelli V, Alvarez IB et al. IL-17 and IFN-g expression in lymphocytes from patients with active tuberculosis correlates with the severity of the disease. J Leukoc Biol 2012;91: Mohan VP, Scanga CA, Yu K et al. Effects of tumor necrosis factor alpha on host immune response in chronic persistent tuberculosis: possible role for limiting pathology. Infect Immun 2001;69: Cohen SB, Emery P, Greenwald MW et al. Rituximab for rheumatoid arthritis refractory to anti-tumour necrosis factor therapy. Arthritis Rheum 2006;54: Emery P, Fleischmann R, Filipowicz-Sosnowska A et al. The efficacy and safety of rituximab in patients with active rheumatoid arthritis despite methotrexate treatment. Results of a phase IIb randomised, double-blind, placebo-controlled, dose-ranging trial. Arthritis Rheum 2006;54: Mohrbacher A. B cell non-hodgkin s lymphoma: rituximab safety experience. Arthritis Res Ther 2005;7(Suppl 3): S Edwards JC, Szczepanski L, Szechinski J et al. Efficacy of B-cell-targeted therapy with rituximab in patients with rheumatoid arthritis. N Engl J Med 2004;350: Burr ML, Malaviya AP, Gaston JH, Carmichael AJ, Östör AJK. Rituximab in rheumatoid arthritis following anti-tnf-associated tuberculosis. Rheumatology 2008; 47: Jung N, Owczarczyk K, Hellmann M et al. Efficacy and safety of rituximab in a patient with active rheumatoid arthritis and chronic disseminated pulmonary aspergillosis and history of tuberculosis. Rheumatology 2008;47: Kevin L, Winthrop S, Yamashita S, Beekmann E, Polgreen PM. Mycobacterial and other serious infections in patients receiving anti-tumor necrosis factor and other newly approved biologic therapies: case finding through the emerging infections network. Clin Infect Dis 2008;46: Ulrichs T, Kosmiadi GA, Trusov V et al. Human tuberculous granulomas induce peripheral lymphoid follicle-like structures to orchestrate local host defence in the lung. J Pathol 2004;204: Maglione PJ, Xu J, Chan J. B cells moderate inflammatory progression and enhance bacterial containment upon pulmonary challenge with Mycobacterium tuberculosis. J Immunol 2007;178: Smolen JS, Keystone EC, Emery P et al. Consensus statement on the use of rituximab in patients with rheumatoid arthritis. Ann Rheum Dis 2007; 66: Arnett FC, Edworthy SM, Bloch DA et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988; 31: Hsieh PR, Liu YC, So J et al. Mycobacterium tuberculosis in Taiwan. J Infect 2006;52: Prevoo MLL, van t Hof MA, Kuper HH et al. 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 et al. 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: American Thoracic Society. Diagnostic standards and classification of tuberculosis in adults and children. Am J Respir Crit Care Med 2000;161: Ogata A, Mori M, Hashimoto S et al. Minimal influence of tocilizumab on IFN-g synthesis by tuberculosis antigens. Mod Rheumatol 2010;20: Andersen P, Doherty TM, Pai M, Weldingh K. The prognosis of latent tuberculosis: can disease be predicted? Trends Mol Med 2007;13: Diel R, Loddenkemper R, Meywald-Walter K et al. Predictive value of a whole-blood IFN-{gamma} assay for the development of active TB disease. Am J Respir Crit Care Med 2008;177: Hatemi G, Melikoglu M, Ozbakir F et al. Quantiferon-TB Gold in tube assay for the screening of tuberculosis before and during treatment with tumor necrosis factor alpha antagonists. Arthritis Res Ther 2012;14: R Hamdi H, Mariette X, Godot V et al. Inhibition of anti-tuberculosis T-lymphocyte function with tumor necrosis factor antagonsits. Arthritis Res Ther 2006;8: R

8 Yi-Ming Chen et al. 33 Tony HP, Burmester G, Schulze-Koops H et al. Safety and clinical outcomes of rituximab therapy in patients with different autoimmune diseases: experience from a national registry (GRAID). Arthritis Res Ther 2011;13:R Caccamo N, Guggino G, Joosten SA et al. Multifunctional CD4 (+) T cells correlate with active Mycobacterium tuberculosis infection. Eur J Immunol 2010;40: