Long-term follow-up in lupus nephritis patients treated with rituximab clinical and histopathological response

Transcription

1 RHEUMATOLOGY Rheumatology 2013;52: doi: /rheumatology/kes348 Advance Access publication 3 January 2013 Original article Long-term follow-up in lupus nephritis patients treated with rituximab clinical and histopathological response Thórunn Jónsdóttir 1, Agneta Zickert 1, Birgitta Sundelin 2, Elisabet W. Henriksson 3, Ronald F. van Vollenhoven 4 and Iva Gunnarsson 1 Abstract Objective. To investigate the long-term clinical, histological and serological affects of B-cell-depleting therapy (BCDT) in patients with LN refractory to conventional treatment. Methods. Twenty-five patients, followed for a mean time of 36 (9 95 ), were included. Renal disease activity was evaluated with the BILAG index and renal response was determined according to the LN European consensus statement. Renal biopsies were performed for histological evaluation at baseline and follow-up. Results. Partial response (PR) or complete renal response (CR) was observed in 22 of 25 after a median of 12. Sixteen patients achieved CR after a median of 24. Six patients experienced a renal relapse. Proteinuria decreased significantly (P = ) from baseline to 36. A noteworthy histological improvement was seen in nearly all patients with a significant reduction in activity index (P = 0.01). Longer depletion time and low baseline values of IgM were indicative of achieving clinical remission during the first year after treatment (P = 0.03 and P = 0.04, respectively). Conclusion. In therapy-resistant LN, BCDT induced clinical and histological improvements in the majority of patients. Transition from PR to CR was mainly seen during the second year of follow-up. Patients with longer depletion time and low baseline levels of IgM were more likely to gain a faster remission, suggesting that the clinical benefit may be linked to suppression of autoreactive plasmablasts. Although formal evidence of BCDT in LN is lacking, our data may provide guidance to clinicians considering therapeutic options in patients with refractory LN. Key words: systemic lupus erythematosus, lupus nephritis, rituximab, anti-cd20, B-cell depletion. CLINICAL SCIENCE Introduction Nephritis remains a major cause of morbidity in patients with SLE and repeated renal flares are known to be a risk factor for development of end-stage renal disease [1]. Despite common knowledge and application of intensive treatment regimens involving CSs and CYC [2 6], and in 1 Department of Medicine, Rheumatology Unit, 2 Department of Oncology and Pathology, Pathology and Cytology, 3 Department of Neurobiology, Care Sciences and Society and 4 Unit for Clinical Therapy Research, Inflammatory Diseases (ClinTRID), Karolinska University Hospital, Karolinska Institutet, Stockholm, Sweden. Submitted 22 March 2012; revised version accepted 18 October Correspondence to: Thórunn Jónsdóttir, Department of Medicine, Rheumatology Unit, Landspitali University Hospital, 108, Reykjavík, Iceland. torunnjo@landspitali.is later years MMF [7 9], there is still significant morbidity and mortality associated with LN [10]. A significant proportion of patients still have refractory disease or are unable to tolerate conventional immunosuppressive agents [11]. Increased understanding of the immunopathogenesis of SLE and LN as well as the lack of treatment options in severe LN has resulted in the development of new immunomodulatory therapies. In later years, B-cell depletion with rituximab (RTX) has shown promising results in several open uncontrolled studies in patients with refractory LN [12 21]. However, controlled clinical trials with RTX in LN have so far been unsuccessful [22]. We have previously reported on the clinical and histopathological course of patients with LN treated with RTX in a short-term follow-up study [16]. Here we present! 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 Thórunn Jónsdóttir et al. longer-term clinical results and data regarding predictors of response to B-cell depletion with RTX in an expanded cohort of patients with LN refractory to conventional therapy. Patients and methods Patients and study design Twenty-five patients with active SLE and LN proven by a recent renal biopsy were treated with RTX in combination with i.v. CYC and glucocorticoids (GCs). All but two patients were earlier refractory to conventional therapy including CYC and/or MMF. The patients were followed according to a standardized protocol until a renal flare occurred or re-treatment with RTX was given for non-renal lupus manifestation. Minimum time of follow-up was 2 years. All patients fulfilled at least four of the ACR criteria for SLE [23]. Eighteen of the patients have previously been reported [24]. The study was approved by the local ethics committee (Karolinska Institutets Etikråd) and the Swedish Medical Product Agency. All patients were informed and consented to the treatment. The patient demographics are presented in Table 1. Treatment protocol The protocol has previously been reported in detail [16]. Briefly, i.v. RTX 375 mg/m 2 was given once weekly for 4 weeks. CYC 0.5 g i.v. was given in combination with the first and fourth RTX infusion. An exception was made in four patients, two of whom were treated with RTX in combination with MMF and two with RTX as monotherapy. Five patients were treated with two infusions of RTX at 1000 mg each, 2 weeks apart. One patient received only three doses of RTX and one dose of CYC due to an infusion reaction at the start of the third infusion. GCs were temporarily increased during the treatment weeks (0.5 mg/kg) but tapered rapidly thereafter. After RTX, and during the time of B-cell depletion, concomitant immunosuppression was stopped according to the treatment protocol used and no other immunosuppressive therapy than anti-malarials was allowed (except in one patient in whom prior MMF treatment was continued). All patients with the exception of three (nos 4, 5 and 6) were treated with angiotensin-converting enzyme blocker and/or angiotensin receptor blocker. After the reappearance of B cells in the circulation, 10 of 25 patients were put on maintenance therapy mainly with MMF (n = 8) or AZA (n = 2). Of the remaining 15 patients TABLE 1 Patient demographics at baseline No. Age, years Disease duration, years Previous and ongoing therapy Time of follow-up, Time of B cell return, Maintenance therapy Time to flare, Time to PR/CR, AML, AZA, MMF, MTX, CYC a 40 9 AML 40 b 6/ AZA, CYC, CSA a 95 2 MMF 24/ CYC, CSA a, MMF a MMF 60 6/ CYC, AML a, CSA, MTX 21 6 None 21 b 12/not AZA, CYC a 12 3 None 12 Not AM, MMF, AZA, MTX, CYC a 65 8 AML 6/ MTX, AZA, AML, CSA, CYC a 15 6 None 15 c 12/not CSA, CYC a 64 5 None 64 24/ CYC, AZA a 10 6 None 10 6/ MMF, CYC a 46 4 MMF 6/ None a MMF 24/ CYC, None a 41 2 AZA 6/not 13 d 58 4 CYC, MMF a 39 3 MMF 12/not AZA, MMF, CYC a 41 8 None 12/ CSA, CYC, AZA a, AML a AML 6/ MTX, CYC, MMF a 36 5 None 36/ CYC, AZA, MMF a 17 6 MMF 17 6/not 18 e AZA, CYC, MMF a 9 6 None NR Not AML, AZA, CYC, MMF a None 11 Not AZA, MMF a 12 Still depleted None 6/12 at CYC, AZA a, CSA a 71 ND MMF 24/not 22 d 31 9 CYC, AZA, AML a AZA 12/ AZA, CYC, AML a, MMF a 27 ND AML 6/24 24 e 18 1 AML, MMF a 18 6 MMF 18/not AML, MMF, CYC a 24 7 None 12/24 Patients nos 21 and 25 are male. Patients nos 22 and 24 are African and patients nos 3 and 20 are Asian. All other patients are Caucasian. NR: non-responder; ND: not determined; AML: anti-malarials. a Ongoing therapy. b Retreated for non-renal flare. c Leukeran for non-renal flare. d MMF instead of CYC in combination with RTX. e RTX monotherapy

3 Long-term follow-up in LN 11 were treated with low-dose GC, combined with antimalarials in four cases (Table 1). Follow-up evaluation All patients were followed using a standardized protocol every second to third month during the first year and every 6 thereafter. The global disease activity was evaluated using the SLEDAI [25]. Renal disease activity was assessed by the BILAG index [26]. Blood sampling included investigation of serum creatinine and serum albumin levels, serology and complement. Glomerular filtration rate (GFR) was assessed by urinary clearance of Iohexol according to clinical routine or assessed using the Modification of Diet in Renal Disease (MDRD) formula [27]. Circulating B cells in the peripheral blood were investigated by detection of CD19 + B cells and analysed by flow cytometry at baseline, month 1, month 2 and every other month thereafter. The limit for detection of CD19 + B cells was < /l (normal ). Renal evaluation and histopathology Renal evaluation included urinalysis (dip slide procedure), urine sediment and investigation of 24-h protein excretion. Repeated renal biopsies were performed related to the 6-month follow-up time point. All biopsies were graded according to the World Health Organization (WHO) classification [28] and activity and chronicity indices were recorded [29]. All biopsies were evaluated by light microscopy, IF and electron microscopy, including evaluation of at least two glomeruli. Renal response and relapse For evaluation of the renal response the LN European consensus statement was used [30]. In these statements the ACR response criteria for proliferative and membranous renal disease in SLE clinical trials [31] and the EULAR guidelines for clinical trials [32, 33] were taken into consideration along with opinions from a panel of experts on LN. According to these criteria, evaluation of renal function (GFR), urinary protein level (here assessed by 24-h urinary protein excretion), abnormalities in urinary sediment and adverse events were recorded in order to evaluate renal response, complete response (CR) and partial response (PR), no response and flares. According to these consensus statements a renal flare was defined as increased disease activity requiring more intensive therapy and indicated by an increase in proteinuria or serum creatinine, abnormal urinary sediment or a reduction in creatinine clearance due to active disease. We also regarded any new renal BILAG A or B as a renal flare. Serology and complement levels Analysis of serum IgG anti-dsdna antibodies was performed by a fluorescent enzyme immunoassay method (Pharmacia, Uppsala, Sweden). The cut-off for normal was <15 IU/ml but values below this level were recorded for statistical calculations. C3 and C4 were determined by nephelometry (normal range g/l for C3 and g/l for C4). Analysis of complement component C1q was performed by rocket electrophoresis [34] using rabbit anti-c1q as the antibody. Levels of C1q were expressed as the percentage of the levels of normal blood donors (normal %). Total serum Igs G, A and M were measured by ELISA (normal range g/l for IgG, g/l for IgA and g/l for IgM). Statistics For comparison of variables at baseline and follow-up, Student s t-test was used for normally distributed parameters and the non-parametric Mann Whitney test for non-normally distributed parameters. Correlations were calculated and significance determined by Fisher s r-to-z test. Kaplan Meier hazard plots were constructed for time to PR/CR and time to flare. For these analyses, Statview (SAS Institute, Cary, NC, USA) was used. P < 0.05 was considered significant. Results All but 2 of the 25 patients were female and all but 4 were of Caucasian origin (2 were Asian, 2 African). The mean age at treatment was 35 years (range years) and the mean time from SLE diagnosis was 8 years (range 1 22 years) (Table 1). At inclusion the mean SLEDAI score was 16 (range 6 36) and all patients had a renal BILAG A; the mean prednisolone dose was 17 mg/day (range 0 40 mg/day). At 6 of follow-up, the mean dose had decreased to 10 mg/day (range mg/day) (P = 0.02). The patients were followed for a mean of 36 (range 9 95 ). Eighteen of the patients reached 2 years and 14 patients reached 3 years of follow-up. Follow-up evaluation The mean creatinine at study start was 100 nmol/l (range nmol/l) and was reduced to a mean of 84 nmol/l (range nmol/l) at 6 (P = not significant). Serum albumin increased from a mean of 27 g/l (range g/l) to 33 g/l (range g/l) at 6 (P < ). The most prominent reduction of proteinuria was seen during the first 6, with a reduction from a mean of 3 g/24 h to 1.1 g/24 h (P = ). After 1 year an additional reduction was noted from a mean of 1.1 g/24 h at 12 to 0.5 g/24 h at 24 in the 17 patients available for follow-up. At 36 an additional improvement was noticed with a mean proteinuria of 0.1 g/ 24 h (P < 0.05) in 14 patients available for follow-up (Fig. 1). The mean estimated GFR remained stable during follow-up. All patients had complete peripheral blood B-cell depletion. The CD19 + B cells were detectable in the circulation after a mean time of 8 (range 2 26 ). Add-on of maintenance treatment was decided by the treating physician (Table 1). No relationship was found between the maintenance therapy and response or time to flare (data not shown)

4 Thórunn Jónsdóttir et al. FIG. 1Reduction in proteinuria Proteinuria, g/24h Histological assessment At baseline, 15 of the 25 patients had proliferative LN (WHO III or IV), 9 of 25 had membranous LN (WHO V) and 1 patient refused renal biopsy (Table 1). Nineteen patients underwent a re-biopsy after a mean time of 8 (range 3 12 ). Histological improvements with transformation in WHO class at re-biopsy was seen in a majority of the patients with proliferative nephritis. In all patients there was a significant reduction in activity index from 4.5 to 2.3 (P = 0.001) and an unchanged chronicity index (data not shown). For detailed information regarding biopsy findings, see Table 2. Renal responses P= P= P<0.05 P< Months Twenty-four-hour albumin excretion decreased from a mean of 3.0 g/day (range g/day) to 1.1 g/day (range g/day) at both 6 and 12 and was further reduced to 0.5 g/day (range g/day) at 24 (P < 0.05) and to 0.1 g/day (range g/day) at 36 (P < 0.05). A significant reduction was seen between 6 and 36 (P < 0.05) and between 12 and 36 (P < 0.05), respectively. A reduction was also seen between 12 and 24, although it was not significant (P = 0.057). When comparing the two different sets of response criteria we found full agreement between the PR/CR consensus criteria [30] and achieving BILAG C/D. As the two renal response criteria used here were fully comparable, we choose to clarify BILAG C as PR and BILAG D as CR. One patient was a non-responder with persistent BILAG A. Two patients were considered poor responders, with a BILAG B at best (Fig. 2). The 22 remaining patients achieved a PR at some time point during follow-up, after a median time of 12 (range 6 36 ). Sixteen patients achieved a CR after a median of 24 (range 6 36 ). The likelihood of achieving PR during the first year was >50%, while the majority attaining a CR achieved this during the period from 12 to 24. Figure 3A and B illustrates the time course for achieving PR and CR. Renal relapse Six renal flares occurred after a mean time of 29 (range ). Both patients considered poor responders had a worsening of disease within the first year after treatment (Fig. 2). Four patients reaching a PR or CR flared, one within the first year, one after 17 and the remaining two patients after 60 and 64, respectively (Fig. 2). The four early flares all had proliferative disease (WHO III/IV) on pre-treatment kidney biopsy, while the two late flares had membranous LN (WHO V). None of the patients experiencing a flare within the first year of follow-up had been treated with maintenance immunosuppressive therapy, while the patient flaring after 17 had been taking MMF for 6. One of the patients experiencing a late flare was treated with MMF as maintenance therapy, while the other was under treatment with low doses of oral GC only (5 7.5 mg daily). Eight of 15 patients without maintenance therapy continued to do well 1 year after RTX treatment (Table 1). Serology and complement levels Anti-dsDNA antibodies were detectable in all but one patient at study start, with a mean titre of 202 IU/ml (range IU/ml). At 6 the mean titre had decreased to 107 IU/ml (range IU/ml) (P = ). Complement levels C3, C4 and C1q increased from baseline to 6 of follow-up (P < 0.01) and were kept stable thereafter (data not shown). Total IgG and IgA levels were stable throughout the observation period. A significant reduction was seen in total IgM levels at 6 and 12 compared with baseline (P < 0.01), although values remained within the normal range (data not shown). Predictors of response In patients who achieved a CR within the first 12 a significantly longer time to repopulation of B cells in the circulation was noted (P = 0.03) (Fig. 4A) compared with patients who did not achieve this goal. Furthermore, in patients who achieved a CR, baseline titres of IgM were positively correlated to the time to response (r = 0.53, P = 0.04, 95% CI 0.14, 0.84) (Fig 4B). A trend between low baseline peripheral blood B cells (CD19 + cells) and early achievement of a PR (r = 0.5, P = 0.06) was noted, however, it did not reach statistical significance. There were no correlations regarding complement level or serological status in association with outcome or flares. No correlation was found between response and histopathological classification. Adverse events Adverse events noted during the first 6 of followup included urinary tract infections (n = 3), Salmonella infection (n = 1), reactivation of herpes zoster (n = 2), infusion reaction (n = 3, two mild and one moderate). Severe adverse events were noted in four patients. One neutropenic fever was diagnosed 1 week after the last CYC and RTX 850

5 Long-term follow-up in LN TABLE 2 Histological information on kidney biopsies at baseline and at re-biopsy No. WHO classification at baseline WHO classification at re-biopsy AI/CI at baseline AI/CI at re-biopsy Time from treatment to re-biopsy, 1 IIIb Ib 5/8 1/8 3 2 IVc Vb 5/7 3/7 5 3 Vb ND 4/6 ND 12 4 ND ND ND ND ND 5 IVb IVd 5/1 2/ IVc IIIb 7/3 4/ IIIb IIb 13/3 2/2 7 8 Vb Vb 2/0 2/0 7 9 IVb IIb 7/4 4/ IVb IIb 10/1 2/ Vb Va 2/1 3/ Vb IIa 2/6 1/ Vb Vb/IIb 3/5 2/ Vb/IIIb Vb 9/3 1/ IIIb Va ND 1/ IIIb Vasculitis 5/3 5/ IV/Vb IVa 3/0 1/ IV IVa/V 3/2 4/ IV c IIB/V 5/3 2/ Va VA 1/0 1/ IV IV ND ND 9 22 IV a ND 1/2 ND ND 23 III ND ND ND ND 24 V ND ND ND ND 25 IVa IB 3/4 1/6 10 ND: not determined; AI: activity index; CI: chronicity index. FIG. 2Flow chart with overview of patient s response to therapy. Total 25 patients with LN Clinical response PR or CR 22 patients Non- / poor responders 3 patients CR 16 patients PR 6 patients Flare Flare Flare Non-responder 3 patients 1 early (10 ) 2 late (60 vs 64 ) 1 patient (17 ) 2 patients (11/12 ) 1 patient 851

6 Thórunn Jónsdóttir et al. FIG. 3Kaplan Meier curves showing time to BILAG C (PR) and D (CR). Share achieving BILAG C Share achieving BILAG D infusions. Additionally two patients were diagnosed with severe neutropenia, one in close correlation with the first CYC infusion while the other was diagnosed 3 after the treatment period. Severe necrotizing fasciitis and septicaemia was diagnosed in one patient 5 after treatment. Discussion Time, Time, (A) Time to achieve a renal BILAG C (PR). After 6 40% of patients have reached a BILAG C (PR) and after 12 60%. (B) Time to achieve a renal BILAG D (CR). After 6 12% had reached a BILAG D (CR) and after 12 20%. After 24 50% of patients had achieved a renal BILAG D (CR). In long-term follow-up, treatment with the combination of RTX and CYC in patients with therapy-resistant LN was here shown to induce a beneficial renal response in the majority of patients. The findings thus confirm our previous results on short-term clinical and histological efficacy of RTX in severe proliferative LN [16], also shown by others [35 37]. In the current report, renal response mainly occurred during the first year following treatment, but many of the patients further achieved a CR after 2 years of follow-up. In parallel, repeat renal biopsies in the early phase showed a significant improvement, which confirms the clinical response seen. A faster response A B was seen in patients with lower baseline levels of IgM and longer depletion time was positively correlated with time to response. A trend towards an association between low absolute numbers of CD19 + cells at baseline and more rapid responses was noted. In the context of the present study it should be noted that the Lupus Nephritis Assessment with Rituximab (LUNAR) clinical trial, investigating the efficacy of RTX in patients with LN in a double-blind, randomized trial, failed to achieve its primary endpoint, although the serological improvement was statistically better in the RTX group [22]. However, our study differs significantly from the LUNAR trial in being an open observational study reflecting reallife clinical LN patients. Importantly in contrast to the patients in LUNAR, the population treated here had failed conventional treatment, thus having an increased likelihood to be poor responders. Despite this we were able to demonstrate a clinical as well as histopathological improvement, shown here by a significant decrease in renal activity indices and reduction of proteinuria. The time for development of PR varied to some extent, although the majority of patients achieved this within 12. In contrast, the time to CR was longer, in most cases being documented between 12 and 24 of follow-up. These data suggest that comparative trials of RTX, or other B-cell depleting agents, may have to be designed to include longer follow-up periods in order to increase the likelihood of capturing complete responders in the differentiation between active treatment and comparators. We reported earlier on the correlation of low baseline CD19 + cells and beneficial clinical outcome after treatment with RTX in lupus patients [38]. In the present report, this correlation did not reach statistical significance, although a trend was noted between low numbers of CD19 + cells at baseline and a better renal outcome. Interestingly, a delay in repopulation was found to increase the likelihood for achieving a CR within the first year. Furthermore, patients with lower values of IgM at baseline were found more likely to achieve a CR. It could be speculated that high baseline number of CD19 + cells and titres of IgM could reflect the presence of a greater number of autoreactive plasma cells in the lymphoid and/or target tissues. In accordance with these findings, it is believed that in patients with RA the disappearance of these cells is needed for a better clinical response [39, 40]. It has also been indicated that suboptimal depletion of B cells and B-cell-derived plasma cells in the tissue of RA patients may be associated with a poorer clinical response [41]. Many studies have reported on the beneficial effects of maintenance agents in LN regardless of the type of induction treatment [42]. It could thus be debated whether, or when, additional immunosuppressive therapy should be initiated routinely after RTX therapy in the prevention of renal flares. However, this study was not designed to address this issue. Despite inclusion of patients with previous poor response to immunosuppressives, the flare rate observed 852

7 Long-term follow-up in LN FIG. 4Time to BILAG D in association with depletion time and baseline IgM. A Months to return of CD19 + cells B Time to BILAG D, n =18 n No BILAG D within first 12 Mean 6 P= Baseline IgM, g/l was comparatively low, also taking into account that the patients included had relapsing disease at inclusion, thus being selected for having a more severe disease course. However, the low number and the variable time points make it difficult to draw any conclusions on factors predicting the development of renal flares in the current patient material. In a collaborative study we previously showed similar serological and urine outcomes comparing proliferative and membranous LN patients treated with RTX. In line with these findings, we found no difference in response rate regarding histopathological classification, although this has recently been described by others [20]. The main limitations of our study are its open observational nature, that it is non-blinded and that it has a limited number of subjects. It is strengthened by the facts that it presents real-life refractory patients and is supported by histological data before and after treatment. y n =4 Achieving BILAG D within first 12 Mean 10 r =0.53, P=0.04 (A) Patients achieving a renal BILAG D (CR) within the first 12 had a longer time to return of B cells in the circulation compared with those not achieving this goal (r = 0.53; 95% CI 0.25, 0.82; P = 0.03). (B) Patients with lower baseline values of IgM showed a significantly shorter time to achieve a renal BILAG D (CR) (r = 0.53; 95% CI 0.14, 0.84; P = 0.01). While formal evidence of the efficacy of RTX in LN is still lacking, our data may provide guidance to clinicians considering therapeutic options with RTX as induction treatment in patients with refractory SLE and renal disease. Rheumatology key messages. B-cell-depleting therapy with RTX and CYC may give long-term clinical and histological improvement in refractory LN.. In LN treated with RTX + CYC transition from PR to CR was mainly seen during the second year of follow-up.. Longer follow-up may be required to evaluate renal response in LN trials. Funding: This work was supported by a grant from King Gustaf V s 80th Birthday Fund and by funds from the Karolinska Institute, the Swedish Research Council and a fund for renal research and the Swedish Rheumatism Association. Disclosure statement: R.F.vV. has received consultancy fees and honoraria from Abbott, GSK, MSD, Pfizer, Roche and UCB Pharma, and grants and research support from Abbott, GSK, MSD, Pfizer, Roche and UCB Pharma. All other authors have declared no conflicts of interest. References 1 Cervera R, Khamashta MA, Font J et al. Morbidity and mortality in systemic lupus erythematosus during a 10-year period: a comparison of early and late manifestations in a cohort of 1,000 patients. Medicine 2003;82: Austin HA 3rd, Klippel JH, Balow JE et al. Therapy of lupus nephritis. Controlled trial of prednisone and cytotoxic drugs. N Engl J Med 1986;314: Boumpas DT, Austin HA 3rd, Vaughn EM et al. Controlled trial of pulse methylprednisolone versus two regimens of pulse cyclophosphamide in severe lupus nephritis. Lancet 1992;340: Gourley MF, Austin HA 3rd, Scott D et al. Methylprednisolone and cyclophosphamide, alone or in combination, in patients with lupus nephritis. A randomized, controlled trial. Ann Intern Med 1996;125: Illei GG, Austin HA, Crane M et al. Combination therapy with pulse cyclophosphamide plus pulse methylprednisolone improves long-term renal outcome without adding toxicity in patients with lupus nephritis. Ann Intern Med 2001;135: Houssiau FA, Vasconcelos C, D Cruz D et al. The 10-year follow-up data of the Euro-Lupus Nephritis Trial comparing low-dose and high-dose intravenous cyclophosphamide. Ann Rheum Dis 2010;69: Chan TM, Li FK, Tang CS et al. Efficacy of mycophenolate mofetil in patients with diffuse proliferative lupus nephritis

8 Thórunn Jónsdóttir et al. Hong Kong-Guangzhou Nephrology Study Group. N Engl J Med 2000;343: Chan TM, Tse KC, Tang CS, Mok MY, Li FK. Long-term study of mycophenolate mofetil as continuous induction and maintenance treatment for diffuse proliferative lupus nephritis. J Am Soc Nephrol 2005;16: Ginzler EM, Dooley MA, Aranow C et al. Mycophenolate mofetil or intravenous cyclophosphamide for lupus nephritis. N Engl J Med 2005;353: Croca SC, Rodrigues T, Isenberg DA. Assessment of a lupus nephritis cohort over a 30-year period. Rheumatology 2011;50: Houssiau FA, Ginzler EM. Current treatment of lupus nephritis. Lupus 2008;17: Leandro MJ, Cambridge G, Edwards JC, Ehrenstein MR, Isenberg DA. B-cell depletion in the treatment of patients with systemic lupus erythematosus: a longitudinal analysis of 24 patients. Rheumatology 2005;44: Sfikakis PP, Boletis JN, Lionaki S et al. Remission of proliferative lupus nephritis following B cell depletion therapy is preceded by down-regulation of the T cell costimulatory molecule CD40 ligand: an open-label trial. Arthritis Rheum 2005;52: van Vollenhoven RF, Gunnarsson I, Welin-Henriksson E et al. Biopsy-verified response of severe lupus nephritis to treatment with rituximab (anti-cd20 monoclonal antibody) plus cyclophosphamide after biopsy-documented failure to respond to cyclophosphamide alone. Scand J Rheumatol 2004;33: Jacobson SH, van Vollenhoven R, Gunnarsson I. Rituximab-induced long-term remission of membranous lupus nephritis. Nephrol Dial Transplant 2006;21: Gunnarsson I, Sundelin B, Jonsdottir T, Jacobson SH, Henriksson EW, van Vollenhoven RF. Histopathologic and clinical outcome of rituximab treatment in patients with cyclophosphamide-resistant proliferative lupus nephritis. Arthritis Rheum 2007;56: Vigna-Perez M, Hernandez-Castro B, Paredes- Saharopulos O et al. Clinical and immunological effects of rituximab in patients with lupus nephritis refractory to conventional therapy: a pilot study. Arthritis Res Ther 2006;8:R Smith KG, Jones RB, Burns SM, Jayne DR. Long-term comparison of rituximab treatment for refractory systemic lupus erythematosus and vasculitis: remission, relapse, and re-treatment. Arthritis Rheum 2006;54: Lindholm C, Borjesson-Asp K, Zendjanchi K, Sundqvist AC, Tarkowski A, Bokarewa M. Longterm clinical and immunological effects of anti-cd20 treatment in patients with refractory systemic lupus erythematosus. J Rheumatol 2008;35: Diaz-Lagares C, Croca S, Sangle S et al. Efficacy of rituximab in 164 patients with biopsy-proven lupus nephritis: pooled data from European cohorts. Autoimmun Rev 2012;11: Cervera R, Abarca-Costalago M, Abramovicz D et al. Systemic lupus erythematosus in Europe at the change of the millennium: lessons from the Euro-Lupus Project. Autoimmun Rev 2006;5: Rovin BH, Furie R, Latinis K et al. Efficacy and safety of rituximab in patients with active proliferative lupus nephritis: the Lupus Nephritis Assessment with Rituximab study. Arthritis Rheum 2012;64: Tan EM, Cohen AS, Fries JF et al. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982;25: Jonsdottir T, Gunnarsson I, Mourao AF, Lu TY, van Vollenhoven RF, Isenberg D. Clinical improvements in proliferative vs membranous lupus nephritis following B-cell depletion: pooled data from two cohorts. Rheumatology 2010;49: Bombardier C, Gladman DD, Urowitz MB, Caron D, Chang CH. Derivation of the SLEDAI. A disease activity index for lupus patients. The Committee on Prognosis Studies in SLE. Arthritis Rheum 1992;35: Hay EM, Bacon PA, Gordon C et al. The BILAG index: a reliable and valid instrument for measuring clinical disease activity in systemic lupus erythematosus. Q J Med 1993; 86: Leung YY, Lo KM, Tam LS, Szeto CC, Li EK, Kun EW. Estimation of glomerular filtration rate in patients with systemic lupus erythematosus. Lupus 2006;15: Churg J SL. Renal disease. Classification and atlas of glomerular disease. Tokyo: Igaku-Shoin, Austin HA 3rd, Muenz LR, Joyce KM et al. Prognostic factors in lupus nephritis. Contribution of renal histologic data. Am J Med 1983;75: Gordon C, Jayne D, Pusey C et al. European consensus statement on the terminology used in the management of lupus glomerulonephritis. Lupus 2009;18: The American College of Rheumatology response criteria for proliferative and membranous renal disease in systemic lupus erythematosus clinical trials. Arthritis Rheum 2006;54: Bertsias G, Ioannidis JP, Boletis J et al. EULAR recommendations for the management of systemic lupus erythematosus. Report of a Task Force of the EULAR Standing Committee for International Clinical Studies Including Therapeutics. Ann Rheum Dis 2008;67: Bertsias G, Gordon C, Boumpas DT. Clinical trials in systemic lupus erythematosus (SLE): lessons from the past as we proceed to the future the EULAR recommendations for the management of SLE and the use of end-points in clinical trials. Lupus 2008;17: Laurell AB, Johnson U, Martensson U, Sjoholm AG. Formation of complexes composed of C1r, C1s, and C1 inactivator in human serum on activation of C1. Acta Pathol Microbiol Scand 1978;86C: Boletis JN, Marinaki S, Skalioti C, Lionaki SS, Iniotaki A, Sfikakis PP. Rituximab and mycophenolate mofetil for relapsing proliferative lupus nephritis: a long-term prospective study. Nephrol Dial Transplant 2009;24: Pepper R, Griffith M, Kirwan C et al. Rituximab is an effective treatment for lupus nephritis and allows a reduction in maintenance steroids. Nephrol Dial Transplant 2009;24:

9 Long-term follow-up in LN 37 Catapano F, Chaudhry AN, Jones RB, Smith KG, Jayne DW. Long-term efficacy and safety of rituximab in refractory and relapsing systemic lupus erythematosus. Nephrol Dial Transplant 2010;25: Jonsdottir T, Gunnarsson I, Risselada A, Henriksson EW, Klareskog L, van Vollenhoven RF. Treatment of refractory SLE with rituximab plus cyclophosphamide: clinical effects, serological changes, and predictors of response. Ann Rheum Dis 2008;67: Edwards JC, Cambridge G. Sustained improvement in rheumatoid arthritis following a protocol designed to deplete B lymphocytes. Rheumatology 2001;40: Edwards JC, Cambridge G, Abrahams VM. Do self-perpetuating B lymphocytes drive human autoimmune disease? Immunology 1999;97: Thurlings RM, Vos K, Gerlag DM, Tak PP. Disease activity-guided rituximab therapy in rheumatoid arthritis: the effects of re-treatment in initial nonresponders versus initial responders. Arthritis Rheum 2008;58: Contreras G, Tozman E, Nahar N, Metz D. Maintenance therapies for proliferative lupus nephritis: mycophenolate mofetil, azathioprine and intravenous cyclophosphamide. Lupus 2005;14(Suppl 1):S