B Lymphocyte Depletion with the Monoclonal Antibody Rituximab in ...

0021-972X/07/$15.00/0 Printed in U.S.A.

The Journal of Clinical Endocrinology & Metabolism 92(5):1769 –1772 Copyright © 2007 by The Endocrine Society doi: 10.1210/jc.2006-2388

BRIEF REPORT B Lymphocyte Depletion with the Monoclonal Antibody Rituximab in Graves’ Disease: A Controlled Pilot Study Daniel El Fassi, Claus H. Nielsen, Steen J. Bonnema, Hans C. Hasselbalch, and Laszlo Hegedu¨s Department of Endocrinology and Metabolism (D.E.F., S.J.B., L.H.), Odense University Hospital, 5000 Odense, Denmark; Institute for Inflammation Research (D.E.F., C.H.N.), Rigshospitalet National University Hospital, 2100 Copenhagen, Denmark; and Department of Hematology (H.C.H.), Odense University Hospital, 5000 Odense, Denmark Context: Graves’ disease (GD) is a common TSH receptor autoantibody (TRAb)-mediated disorder. Because B lymphocytes are important self-antigen presenting cells and precursors for antibody-secreting plasma cells, temporary B-lymphocyte depletion with the monoclonal antibody rituximab (RTX) might be of benefit in GD. Objective/Design: The objective of this prospective, controlled, nonrandomized study was to investigate the effect of RTX in GD. Setting/Patients: We studied 20 outpatients referred to a university clinic with newly diagnosed (four with relapse) untreated GD. Ten received RTX (⫹RTX), whereas 10 did not (⫺RTX). Intervention: The patients received methimazole (MMI) for a median of 102 d (⫹RTX) and 110 d (⫺RTX) before the study. Patients in the ⫹RTX group received 375 mg RTX/m2 iv on d 1, 8, 15, and 22, and all patients were withdrawn from methimazole (MMI) at d 22.

Results: Four patients in the ⫹RTX group remained in remission with a median follow-up of 705 d (range, 435–904 d), whereas all the patients in the ⫺RTX group had relapsed by d 393 (P ⬍ 0.05). All of the patients in remission had initial TRAb levels below 5 IU/liter (normal, ⬍0.7 IU/liter). However, none of the five patients in the ⫺RTX group with correspondingly low TRAb levels were in remission (P ⬍ 0.01). RTX treatment did not affect autoantibody levels to a greater extent than did MMI monotherapy. Two patients received glucocorticoids for joint pain after RTX therapy. Conclusions: RTX treatment may induce sustained remission in patients with GD with low TRAb levels. However, RTX did not affect autoantibody levels and seems ineffective in patients with high TRAb levels. At present, high cost, low efficacy, and potential side effects do not support use in uncomplicated GD. (J Clin Endocrinol Metab 92: 1769 –1772, 2007)

Main Outcome Measures: We measured time to relapse of hyperthyroidism and changes in autoantibody levels.

G

RAVES’ DISEASE (GD) affects approximately 2% of the adult population, with a strong female predominance (1). Hyperthyroidism causes symptoms in various organ systems and affects the quality of life, most pronounced in those with Graves’ ophthalmopathy (GO) (1, 2). Less than half of the patients achieve lasting remission after 12 to 18 months of antithyroid drug therapy (3, 4). For these patients, thyroid surgery and radioiodine therapy are the definitive treatment options, but both carry a high risk of permanent hypothyroidism. Rituximab (RTX) (F. Hoffmann-La Roche Ltd., Basel, Switzerland) is a chimeric monoclonal antibody against the surface marker CD20, which is expressed by immature and mature B lymphocytes (B cells) but not by stem cells or most plasma cells (5). RTX causes immediate peripheral B cell depletion lasting for 6 to 9 months in patients with malignant lymphomas and autoimmune diseases such as rheumatoid

arthritis (6). By 2005, RTX, which is generally considered safe, had been used in 540,000 patients, primarily for malignant lymphomas (7, 8). Treatment of GD with RTX may cause diminished or abolished autoantibody production by deletion of plasma cell precursors, disruption of ectopic germinal centers, and abrogation of B cell presentation of self-antigens to T cells (9). As for the latter mechanism, we have previously shown that B cells present the thyroid self-antigen thyroglobulin to T cells in humans (10, 11). Recent case reports by us (12) and by Salvi et al. (13) demonstrate a marked and sustained clinical effect of RTX in three patients with otherwise treatment-refractory GO. We present the first controlled study of RTX in any autoimmune endocrine disorder. Subjects and Methods Subjects

First Published Online February 6, 2007 Abbreviations: FT3, Free T3; FT4, free T4; GD, Graves’ disease; GO, Graves’ ophthalmopathy; MMI, methimazole; RTX, rituximab; TPOAb, thyroid peroxidase antibodies; TRAb, TSH receptor autoantibody. JCEM is published monthly by The Endocrine Society (http://www. endo-society.org), the foremost professional society serving the endocrine community.

Twenty Caucasian outpatients (18 females) between 18 and 65 yr of age, with newly diagnosed (four with recurrent disease) untreated GD referred to the Department of Endocrinology, Odense University Hospital, were included (Table 1). None of the patients had undergone thyroidectomy or treatment with radioiodine. None received immunosuppressive agents. The diagnosis of GD was based on elevated thyroid hormone levels, decreased TSH level, elevated TSH receptor autoanti-

1769

1770

J Clin Endocrinol Metab, May 2007, 92(5):1769 –1772

El Fassi et al. • B Lymphocyte Depletion in Graves’ Disease

TABLE 1. Characteristics of patients with GD treated with (⫹RTX) or without (⫺RTX) rituximab ⫹RTX (n ⫽ 10)

Sex (female/male) Age (yr) No. with recurrent GD No. with active GOa Initial thyroid size by ultrasound (ml) Initial FT3 (pmol/liter) Initial FT4 (pmol/liter) Initial TRAb (IU/liter) TRAb at time of RTX initiation (d 1) (IU/liter) TRAb at time of MMI withdrawal (d 22) (IU/liter) TPOAb at time of RTX initiation (d 1) (IU/ml) MMI treatment time until drug withdrawal (d 22) (d) Cumulative MMI dose (mg) Follow-up after MMI cessation (d)

10/0 42 (23– 64) 4 2 27 (13– 42) 37.1 (7.9 –59.1) 51.0 (15.0 to ⬎80) 11.4 (0.7–90.1) 8.3 (⬍0.7 to 94.4) 7.3 (⬍0.7 to 78.2) 280 (⬍60 to ⬎3000) 124 (48 –193) 1581 (255–1878) 740 (435–904)

⫺RTX (n ⫽ 10)

8/2 38 (18 –53) 0 0 22 (12– 40) 30.8 (11.3–75.0) 52.4 (24.0 to ⬎80) 12.2 (2.1–37.7) 5.7 (1.3–19.2) 3.5 (0.7–18) 168 (⬍60 to ⬎3000) 132 (60 –181) 1471 (1005–3460) 554 (99 –741)

P value

0.47 0.48 0.09 0.47 1.00 0.87 0.87 0.97 0.58 0.63 0.80 0.44 0.53 ⬍0.02

Data are given as number of individuals or median and range. Reference intervals are 4.3–7.4 pmol/liter for FT3, 9.9 –17.7 pmol/liter for FT4, less than 0.7 IU/liter for TRAb, and less than 60 IU/ml for TPOAb. a These two patients had NOSPECS classifications of 2c3b4a5060 and 2b3a405060; their clinical activity scores were 6 and 5, respectively (for details, see Ref. 12). bodies (TRAb) and/or GO, and ultrasonic and radionucleotide scans consistent with GD. Consecutive eligible patients were offered RTX treatment, and control patients were included among those who declined RTX therapy.

Study design Ten patients with GD received RTX (⫹RTX), and 10 did not (⫺RTX). All were rendered euthyroid with methimazole (MMI), which was given for approximately 4 months ending at the last RTX infusion (or the corresponding day in patients in the ⫺RTX group). RTX therapy was given as iv infusions of 375 mg RTX/m2 body surface area on d 1, 8, 15, and 22 after pretreatment with 1 g of acetaminophen orally and 2 mg of clemastine iv. Due to adverse events, two patients received supplementary antihistamine, and one received iv glucocorticoids during the first infusion. Seven patients received a full RTX course. Three patients received only two doses of RTX, two of these due to adverse effects and one due to the patient’s own wishes. All the patients were included in the analyses. The study was prospective and controlled, but neither randomized nor blind. Blood samples were collected from the patients at study entry, weekly for the first 6 wk after RTX initiation, and monthly for at least 12 months. The main study end point was relapse of hyperthyroidism. Relapse was defined as serum levels of free T4 (FT4) and/or free T3 (FT3) above the normal range. For ethical reasons, MMI therapy was resumed immediately upon relapse. The study was approved by the local ethics committee and is recorded on www.clinicaltrials.gov (NCT00150111). Informed consent was obtained from each subject.

Materials and methods Serum TSH was measured using a time-resolved fluorometric assay (AutoDELFIA hTSH Ultra Kit; Perkin-Elmer/Wallac, Turku, Finland); the reference range is 0.30 to 4.00 mU/liter. Serum FT4 and serum FT3 were determined using the AutoDELFIA FT4 and FT3 Kits (PerkinElmer/Wallac), respectively. For FT4, the reference range is 9.9 to 17.7 pmol/liter, and for FT3, it is 4.3 to 7.4 pmol/liter. Thyroid peroxidase antibodies (TPOAb) were measured by an ELISA (Varelisa; Pharmacia, Freiburg, Germany) with a reference range of less than 60 IU/ml. Serum TRAb were measured using a human radio receptor assay (DYNOtest; Brahms Diagnostica, Berlin, Germany) (14) with a reference range of less than 0.7 IU/liter. All assays had a coefficient of variation of less than 10%. Thyroid size was determined ultrasonically (15).

Statistical analysis Differences between the two main groups were tested using the Mann-Whitney test, and subgroups of patients were compared with

two-sided Fisher’s exact test. Based on Kaplan-Meier curves, the log rank (Mantel Cox) test was used to compare outcome between the treatment groups, and the hazard ratio for relapse was calculated. Spearman’s r was used to test for correlations. P values less than 0.05 were considered significant. The analyses were performed using the GraphPad Prism version 4.00 for Windows (GraphPad Software, San Diego, CA).

Results

All patients with GO and recurrent GD were in the ⫹RTX group, and the follow-up was longer in this group than in the ⫺RTX group. Aside from this, no significant differences were found between the two groups regarding any pertinent variables (Table 1). One year after MMI treatment was stopped, four patients in the ⫹RTX group remained euthyroid vs. one in the ⫺RTX group (Fig. 1A). The patients in the ⫹RTX group who remained euthyroid included two patients with recurrent GD and two with debut of GD. One of the two patients with GO remained euthyroid (12). RTX significantly affected the risk of relapse because the last patient in the ⫺RTX group had relapse on d 393, whereas the four patients in the ⫹RTX group, followed for 435, 666, 743, and 904 d, respectively, remain euthyroid (P ⬍ 0.05) (Fig. 1A). The ⫺RTX/⫹RTX hazard ratio for relapse is 2.30 (95% confidence interval, 1.08 –13.10). The duration of euthyroidism correlated inversely with TRAb levels on d 1 (Spearman’s r ⫽ ⫺0.79, P ⬍ 0.009 for patients in the ⫹RTX group; and Spearman’s r ⫽ ⫺0.83, P ⬍ 0.004 for patients in the ⫺RTX group). Four patients in the ⫹RTX group and five patients in the ⫺RTX group had initial TRAb levels less than 5 IU/liter (Fig. 1B), and among these patients, a significant effect of RTX on remission rate was observed at the end of follow-up (P ⬍ 0.01). RTX responders and nonresponders did not exhibit differences regarding median initial FT4 [45 pmol/liter (range, 15 to ⬎80) vs. 51 pmol/liter (range, 47 to ⬎80), respectively], thyroid size [29 ml (range, 13–36) vs. 23 ml (range, 13– 42), respectively], MMI treatment time [124 d (range, 52–193) vs. 120 d (range, 48 –167), respectively], or cumulated MMI dose [1168 mg (range, 255-1878) vs. 1584 mg (range, 645-1758),



1771

FIG. 1. The effect of RTX on thyroid status and TRAb. A, Kaplan-Meier plot showing the number of patients remaining euthyroid, without MMI, after receiving RTX treatment (⫹RTX, continuous line) or not (⫺RTX, broken line). The groups had received MMI for a median of 124 d and 132 d, respectively, before MMI withdrawal. The follow-up period of each subject in sustained remission is indicated by a vertical tick. B, The relationship between thyroid status at the end of follow-up and TRAb levels at d 1 is shown. Patients in the ⫹RTX group are represented by squares and patients in the ⫺RTX group by circles. Filled symbols indicate patients experiencing relapse. C, The relative TRAb levels during the observation period are shown as squares for patients in the ⫹RTX group and circles for patients in the ⫺RTX group. Two patients in the ⫹RTX group and one patient in the ⫺RTX group were excluded from the analysis due to TRAb levels less than 1.4 IU/liter (twice the detection limit) at d 1. The values were normalized to the levels observed at d 1 (100) and are shown as median and range. TRAb values less than 0.7 IU/liter were considered to be 0 IU/liter. All patients received MMI until d 22. MMI therapy was reinitiated in patients with relapse. Patients in the ⫹RTX group received RTX on d 1, 8, 15, and 22.

respectively]. One responder and two nonresponders received only two RTX doses. By d 22, when MMI treatment was stopped, TRAb levels had decreased to 81% (range, 60 –109 IU/liter) and 81% (range, 54 –114 IU/liter) of d-1 values in the ⫹RTX and ⫺RTX groups, respectively. The two groups also exhibited similar decreases in TRAb levels during the follow-up period (Fig. 1C). Moreover, the rate of decrease in TRAb levels of patients who received MMI for virtually the whole period, due to relapse of hyperthyroidism, did not differ between the two groups (data not shown). The decrease in TRAb levels of the patients in the ⫹RTX group who remained euthyroid did not differ significantly from those of patients with relapse at any time point (data not shown). TPOAb levels were not predictive of the response to RTX and declined at similar rates in the ⫹RTX and ⫺RTX groups (data not shown). Five patients experienced one or more RTX infusion-related adverse effects after the first infusion [hypotension (n ⫽ 4), nausea (n ⫽ 2), fever (n ⫽ 1), chills (n ⫽ 1), and tachycardia (n ⫽ 1)], which did not recur upon subsequent infusions. Four days after the second infusion, two patients (one of whom had no infusion reaction) developed a serum sicknesslike reaction with joint pain and fever. In these patients, RTX treatment was stopped after the second course. Also, one patient with no infusion reaction experienced worsening of gastrointestinal symptoms, subsequently diagnosed as ulcerative colitis, and axial arthritis. This patient, and one of the patients with a serum sickness-like reaction, received glucocorticoids for joint pain. Two patients in each group had a minor infection during the first year of follow-up.

Discussion

RTX acts by depleting B cells, the precursors of the autoantibody-producing plasma cells. Accordingly, it has been proposed that RTX treatment might be curative in autoantibody-mediated diseases such as GD (9, 16). In this study, the first controlled study of RTX therapy in any endocrine disorder, RTX decreased the relapse rate after short-term MMI therapy. Four of 10 RTX-treated patients remain euthyroid with a normal TSH vs. none of 10 patients in the ⫺RTX group. A longer follow-up period in the patients in the ⫹RTX group and the fact that all patients with GO and recurrent GD were included in the ⫹RTX group suggest that we may well have underestimated the effect of RTX (1). Only patients with low TRAb levels at RTX initiation remained euthyroid. In support of a genuine RTX-related effect, patients in the ⫺RTX group with similar low TRAb levels all relapsed. However, patients with correspondingly low TRAb levels might have obtained sustained remission without RTX given a longer MMI treatment (4). Importantly, RTX treatment did not mediate a greater decrease in autoantibody than treatment with MMI alone (Fig. 1C), which is supported by a preliminary report from an uncontrolled study of RTX in GO (17). Given a half-life of IgG of approximately 3 wk, de novo production of TRAb apparently occurred after initiation of RTX therapy. Indeed, RTX cannot be expected to affect the production of TRAb by CD20-negative plasma cells, of which at least some are longlived (18, 19). Moreover, although B cell depletion is profound in the peripheral blood (6), depletion may be less efficient in lymphoid (19) and thyroid tissue (13). In the patients with low levels of circulating TRAb, abrogated re-

1772


cruitment of new TRAb-producing plasma cells may, however, have been sufficient to lower the level of pathogenic autoantibodies below a critical level. In addition to their role as precursors for antibody-producing plasma cells, B cells are known to play a critical role as antigen-presenting cells in a number of autoimmune diseases (5, 19) and also efficiently take up and present thyroid autoantigens to T cells (10, 11). It is probable that the marked amelioration of eye symptoms by RTX in patients with severe GO (12, 13) is related to an abrogated autoantigen presentation by B cells. Half of the RTX-treated subjects experienced benign infusion reactions. Unexpectedly, two patients had to receive long-term glucocorticoid therapy due to late-onset articular pain. Adverse reactions, at least infusion reactions, might have been reduced by glucocorticoid pretreatment (20). In our opinion, the untoward effects, together with the high cost of treatment, raise questions about the use of RTX in GD despite an apparent efficacy in a subset of patients. Nonetheless, in view of the encouraging results obtained with RTX in previously treatment-refractory severe GO, RTX may still find a place in the treatment of GO (12, 13).


5. 6. 7. 8. 9. 10.

11.

12. 13.

Acknowledgments Received November 1, 2006. Accepted January 30, 2007. Address all correspondence and requests for reprints to: Laszlo Hegedu¨s, Department of Endocrinology and Metabolism, Odense University Hospital, Sdr. Boulevard 29, DK-5000 Odense, Denmark. E-mail: [email protected]. The Agnes and Knut Mørk Foundation, The Danish National Thyroid League, “Laegernes Forsikringsforening af 1891,” Codan Insurance A/S, The Novo Nordisk Foundation, and Roche Denmark A/S provided grant support. Roche Denmark A/S provided the study drug without charge. D.E.F. is the recipient of a research fellowship from The Institute of Clinical Research at The University of Southern Denmark. Disclosure Statement: D.E.F., S.J.B., and L.H. have nothing to declare. C.H.N. and H.C.H. have received lecture fees from Roche Denmark A/S.

18.

References

19.

1. Weetman AP 2000 Graves’ disease. N Engl J Med 343:1236 –1248 2. Watt T, Groenvold M, Rasmussen AK, Bonnema SJ, Hegedu¨s L, Bjorner JB, Feldt-Rasmussen U 2006 Quality of life in patients with benign thyroid disorders. A review. Eur J Endocrinol 154:501–510 3. Cooper DS 2005 Antithyroid drugs. N Engl J Med 352:905–917 4. Carella C, Mazziotti G, Sorvillo F, Piscopo M, Cioffi M, Pilla P, Nersita R,

20.

14. 15.

16. 17.

Iorio S, Amato G, Braverman LE, Roti E 2006 Serum thyrotropin receptor antibodies concentrations in patients with Graves’ disease before, at the end of methimazole treatment, and after drug withdrawal: evidence that the activity of thyrotropin receptor antibody and/or thyroid response modify during the observation period. Thyroid 16:295–302 Edwards JC, Cambridge G 2006 B-cell targeting in rheumatoid arthritis and other autoimmune diseases. Nat Rev Immunol 6:394 – 403 Leandro MJ, Cambridge G, Ehrenstein MR, Edwards JC 2006 Reconstitution of peripheral blood B cells after depletion with rituximab in patients with rheumatoid arthritis. Arthritis Rheum 54:613– 620 Boye J, Elter T, Engert A 2003 An overview of the current clinical use of the anti-CD20 monoclonal antibody rituximab. Ann Oncol 14:520 –535 Kimby E 2005 Tolerability and safety of rituximab (MabThera). Cancer Treat Rev 31:456 – 473 El Fassi D, Nielsen CH, Hasselbalch HC, Hegedu¨s L 2006 The rationale for B lymphocyte depletion in Graves’ disease. Monoclonal anti-CD20 antibody therapy as a novel treatment option. Eur J Endocrinol 154:623– 632 Nielsen CH, Leslie RG, Jepsen BS, Kazatchkine MD, Kaveri SV, Fischer E 2001 Natural autoantibodies and complement promote the uptake of a self antigen, human thyroglobulin, by B cells and the proliferation of thyroglobulin-reactive CD4(⫹) T cells in healthy individuals. Eur J Immunol 31:2660 – 2668 Nielsen CH, Hegedu¨s L, Leslie RG 2004 Autoantibodies in autoimmune thyroid disease promote immune complex formation with self antigens and increase B cell and CD4⫹ T cell proliferation in response to self antigens. Eur J Immunol 34:263–272 El Fassi D, Nielsen CH, Hasselbalch HC, Hegedu¨s L 2006 Treatment-resistant severe, active Graves’ ophthalmopathy successfully treated with B lymphocyte depletion. Thyroid 16:709 –710 Salvi M, Vannucchi G, Campi I, Rossi S, Bonara P, Sbrozzi F, Guastella C, Avignone S, Pirola G, Ratiglia R, Beck-Peccoz P 2006 Efficacy of rituximab treatment for thyroid-associated ophthalmopathy as a result of intraorbital B-cell depletion in one patient unresponsive to steroid immunosuppression. Eur J Endocrinol 154:511–517 Jensen E, Petersen PH, Blaabjerg O, Hansen PS, Hegedu¨s L 2005 Improved sensitivity of a thyrotropin receptor antibody assay. Clin Chem 51:2186 –2187 Hegedu¨s L, Perrild H, Poulsen LR, Andersen JR, Holm B, Schnohr P, Jensen G, Hansen JM 1983 The determination of thyroid volume by ultrasound and its relationship to body weight, age, and sex in normal subjects. J Clin Endocrinol Metab 56:260 –263 Edwards JC, Cambridge G, Abrahams VM 1999 Do self-perpetuating B lymphocytes drive human autoimmune disease? Immunology 97:188 –196 Vannucchi G, Campi I, Curro´ N, Pirola G, Bonara P, Rossi S, Guastella C, Ratiglia R, Beck-Peccoz P 2006 Rituximab treatment of Graves⬘ disease and thyroid-associated ophthalmopathy: effects on thyroid autoimmunity and thyroid morphology at ultrasound. Proc 31st Annual Meeting of the European Thyroid Association, Napoli, Italy, 2006, p 20 (Abstract) Slifka MK, Ahmed R 1998 Long-lived plasma cells: a mechanism for maintaining persistent antibody production. Curr Opin Immunol 10:252–258 Browning JL 2006 B cells move to centre stage: novel opportunities for autoimmune disease treatment. Nat Rev Drug Discov 5:564 –576 Emery P, Fleischmann R, Filipowicz-Sosnowska A, Schechtman J, Szczepanski L, Kavanaugh A, Racewicz AJ, Van Vollenhoven RF, Li NF, Agarwal S, Hessey EW, Shaw TM 2006 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, dose-ranging trial. Arthritis Rheum 54:1390 –1400

JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.