Abstract

Background. Systemic lupus erythematosus is a relapsing autoimmune disease. Conventional therapy increases the risk of infection and malignancies; furthermore, a minority of patients suffer from refractory disease. B-cell depletion with the chimeric +AFw-anti-CD20 monoclonal antibody, rituximab, is an alternative therapy for relapsing and refractory systemic lupus erythematosus. We sought to assess the long-term efficacy and safety of rituximab in this patient subgroup.

Methods. Thirty-one sequential patients with relapsing or refractory systemic lupus erythematosus, 11 of whom had active lupus nephritis, received rituximab [either 375 mg/m2/week × 4 (n = 16) or 1000 mg × 2 (n = 15)]. The median follow-up was 30 months.

Results. Thirty of 31 (97%) patients had depleted peripheral B cells. Twenty-seven of 31 (87%) patients achieved remission (17 complete, 10 partial). Renal response occurred in 10/11 patients (4 complete, 6 partial) with active glomerulonephritis. Clinical improvement was reflected by reductions of disease activity, proteinuria and daily prednisolone dose. Eighteen of 27 (67%) patients relapsed after a median of 11 months. Relapses occurred on or after the return of circulating B cells in 10 but in the absence of B-cell return in 8. Re-treatment with rituximab was effective. Infusion reactions were common (18/31; 58%), and infections occurred in 8/31 (26%) patients.

Conclusions. Rituximab had a high rate of efficacy in relapsing or refractory systemic lupus erythematosus with or without renal involvement. Although relapse was common, it responded to re-treatment. The contribution of rituximab to infection risk was uncertain in view of the complex disease course and concomitant therapy of the patients studied.

Introduction

Systemic lupus erythematosus (SLE) is a relapsing autoimmune disease which affects multiple tissues and organs. The main immunological alterations are the loss of tolerance to self-antigens with polyclonal activation of B lymphocytes, auto-antibody production and altered T-cell function [1]. Nephritis is the most common severe manifestation of SLE, and is associated with the development of end-stage renal disease (ESRD) and increased mortality [2].

Current therapies, such as corticosteroids combined with cyclophosphamide or mycophenolate mofetil, increase the risk of infection and malignancy which also contribute to adverse outcomes [3]. Since 90% of lupus patients are young women, cyclophosphamide-induced ovarian failure [4] remains a potentially severe drawback. Responses to current therapies are often incomplete, and relapses are frequent as drug doses are reduced. A minority of patients suffer from refractory disease, due to treatment failure or intolerance. Newer, more effective therapies aimed at specific targets in the immune cascade with fewer side effects need to be explored.

B cells play a crucial role in the pathogenesis of SLE, through at least three broad mechanisms. Polyclonal B cells produce auto-antibodies, activating the complement system and immune responses. Activated B cells produce immunomodulatory cytokines such as tumour necrosis factor-α (TNF-α) and interleukin-6 (IL-6) which may cause inflammation and tissue damage. Furthermore, B cells acting as antigen-presenting cells activate CD4+-auto-reactive T cells, inducing production of pro-inflammatory signals from T cells.

Rituximab (RTX) is a chimeric (mouse/human) monoclonal antibody directed against the B cell-specific antigen CD20, an integral membrane protein believed to function in B-cell cycle initiation and differentiation. Indeed, rituximab selectively depletes CD20 mature and immature B cells through a variety of mechanisms including antibody-dependent cell-mediated cytotoxicity, apoptosis and complement-mediated cytotoxicity [5,6], while sparing plasma cells without significantly increasing susceptibility to infections [7]. Preliminary studies in relapsing and refractory SLE affecting adults and children have reported clinical efficacy (median response to treatment, 75% of cases) and safety of rituximab [8–19]. We report the long-term follow-up of 31 patients with refractory or relapsing SLE treated with rituximab.

Materials and methods

Thirty-one sequential patients receiving rituximab for refractory or relapsing SLE in the Vasculitis and Lupus Clinic at Addenbrooke’s Hospital, Cambridge, UK, between January 2002 and March 2008, were studied. All patients fulfilled at least four American College of Rheumatology diagnostic criteria. Disease activity was scored by the British Isles Lupus Assessment Group Index (BILAG). Demographic data, laboratory results, additional treatments and clinical course were extracted from two electronic databases and patients’ notes. Response to rituximab in 11 patients of this cohort was described in an earlier report [19].

Fifteen received four RTX infusions at a dose of 375 mg/m2/week (lymphoma regimen), after introduction a rheumatoid arthritis regimen have to be cancelled, the next 16 received two infusions at a dose of 1000 mg with a 2-week interval. Intravenous (IV) cyclophosphamide (500 mg) and IV methyl prednisolone (500–1000 mg) were administered prior to the first rituximab infusion in 15 and 2 patients, respectively. Two received intravenous immunoglobulin in association with rituximab. At first RTX course, all received prophylaxis against infusion reactions with IV chlorpheniramine (10 mg), IV hydrocortisone (100 mg) and paracetamol (1000 mg); at following courses, patients who experienced infusion reactions received 100 mg of IV methyl prednisolone before RTX infusion. The first 15 patients received further RTX only after RTX failure or at the time of flare; following a change of protocol, seven received protocol treatment of RTX 1000 mg every 6 months for 2 years.

B-cell depletion was defined as CD19 count <0.02 × 109 cells per litre and B-cell regeneration as CD19 count ≥0.02 × 109 cells per litre at any time.

‘Complete remission’ was defined as the absence of BILAG A-, B- and C-level disease activity, and ‘partial remission’ as the absence of BILAG A- and B-level disease activity. ‘Sustained remission’ was defined as remission following RTX therapy without relapse to the end of the observation period. ‘Complete renal response’ was defined as proteinuria <0.5 g/24 h, absent sediment, and stable or reduced serum creatinine; ‘partial renal response’ was defined as reduction by 50% of proteinuria, <30 red cells per high-power field (or <2 on urine dipstick) on sediment, and stable or reduced serum creatinine. ‘Relapse’ was defined as an increase in disease activity which required an increase in the prednisolone dose.

‘Infusion reactions’ were classified as ‘acute’ (occurring during the infusion) or ‘delayed’ (occurring at least after 24 h after the infusion), and ‘mild-to-moderate’ (short-lived reaction requiring oral symptomatic treatment only) or ‘severe’ (requiring hospitalization and intravenous drugs).

Statistical analysis was performed with SPSS version 15.0 software. The discrepancies of the numbers of patients are due to the fact that in all analyses, an individual patient’s data were excluded if more than two of the serial data points were missing. Non-parametric variables (BILAG score, prednisolone dose, proteinuria, albuminuria and dsDNA) were compared by Friedman’s test followed by post hoc Dunn’s multiple comparison test comparing subsequent time points with Time 0 if the Friedman's analysis was statistically significant. Parametric variables (creatinine, complement and immunoglobulins) were compared by a repeated-measures ANOVA followed by post hoc Dunnett’s multiple comparison test comparing subsequent time points with Time 0 if the initial ANOVA was statistically significant. Erythrocyte sedimentation rate (ESR) data were analysed by a repeated-measures ANOVA followed by a post hoc test for linear trend. The P-values in the text are from the post hoc analysis. Single missing data points in any series were dealt with by carrying forward the data value immediately preceding the missing value. Time to relapse and time to B-cell return between different groups of patients were evaluated by Kaplan–Meier survival curves and the log-rank test. The comparison of duration of B-cell depletion in those patients receiving more than one course of rituximab was performed using a Wilcoxon matched-pairs analysis. P-values <0.05 were considered significant. Plasma creatinine, C3 and C4 levels, and immunoglobulin G (IgG) and immunoglobulin M (IgM) levels are expressed as mean ± standard error.

Results

Thirty-one patients were identified (28 females and 3 males) [mean age, 40.2 ± 12.8 years; median number of organ systems involved, 3 (range, 3–8); median duration of SLE, 96 months (range, 6–396); and median follow-up, 30 months (range, 6–68)]. Eleven had renal involvement at the time of first RTX (biopsy-proven lupus nephritis, n = 9; clinical lupus nephritis, n = 1; and minimal change glomerulonephritis, n = 1). Three had ESRD (Patient 5, 11 and 26); of whom, one had been transplanted 4.5 years before RTX, and two were on maintenance dialysis. Three patients (Patient 2, 22 and 23) became pregnant during follow-up without any complication; all delivered healthy babies. All patients were previously treated with an immunosuppressive drug [median of two different drugs per patient (range, 1–6)], and 14 had received cyclophosphamide at a median cumulative dose of 10 g (range, 5–25). At the time of the first RTX course, 16 were taking immunosuppressive drugs, and the median prednisolone dose was 10 mg/day (range, 0–30) (Table 1).

Table 1

Characteristics of the patients at baseline

PatientSexAgeOrgan systems involved since diagnosisDisease duration, monthsNumber of previous IS therapiesIS therapies at the time of first rituximabPrednisolone dose, mg/day
59 S, J, A, O 108  10 
21 S, J, K (IV), L, BL 24  12.5 
45 S, K (bnp), C, O, H 168  10 
50 S, J, C, A 336  10 
36 S, J, HD, L, H, C, BL 144 AZA 50 mg, CYC 30 
47 S, J, C, O 180 CYC 10 
50 S, J, C 24  
44 S, J, A 108 CYC 10 
49 S, J, C, O 24 MMF 2 g, CyA 150 mg, CYC 
10 56 S, J, K (IV) 372 MMF 3 g, CYC 12.5 
11 25 S, J, KTX, H 96 RAPA 3 mg 
12 54 S, J, C, H 252 MMF 2 g 10 
13 37 J, C, O 48 CYC 10 
14 53 S, J, K (naMC), LI, C, O, BL MMF 2.5 g, PEX 30 
15 50 S, J, K (IV), L 144 AZA 150 mg, CYC 10 
16 42 S, J, L, BL 24 AZA 175 mg 12.5 
17 13 S, J, K (IV), L, O 14  30 IV MP 
18 35 S, J, C 24 IV Ig  
19 43 S, K (NAS), J, L, O 315  10 
20 27 S, J, O 36 FK 2.5 mg 15 
21 59 S, J, K (V), L 132 AZA 50 mg, CYC 20 
22 26 S, J, K (IV) MMF 2 g, CYC 10 
23 18 S, J, K (V), L, C, G, A, H 17 CYC 25 
24 48 S, J, L, C, H, BL 396 CYC 15 
25 36 S, J, K (V) 192 MMF 2 g, CYC 7.5 
26 44 S, J, HD, G 108 CYC 20 IV MP 
27 36 S, J, L, BL 12  10 
28 29 S, J, K (III), L, O 18 MMF 2 g, CYC 12.5 
29 61 S, J, L, O 84 MTX 15 mg 20 
30 22 S, J, L, C, O MMF 2 g, CYC, IV Ig 30 
31 32 S, J, K (aMC), L 168 MMF 1.5 g, FK 6 mg 17.5 
PatientSexAgeOrgan systems involved since diagnosisDisease duration, monthsNumber of previous IS therapiesIS therapies at the time of first rituximabPrednisolone dose, mg/day
59 S, J, A, O 108  10 
21 S, J, K (IV), L, BL 24  12.5 
45 S, K (bnp), C, O, H 168  10 
50 S, J, C, A 336  10 
36 S, J, HD, L, H, C, BL 144 AZA 50 mg, CYC 30 
47 S, J, C, O 180 CYC 10 
50 S, J, C 24  
44 S, J, A 108 CYC 10 
49 S, J, C, O 24 MMF 2 g, CyA 150 mg, CYC 
10 56 S, J, K (IV) 372 MMF 3 g, CYC 12.5 
11 25 S, J, KTX, H 96 RAPA 3 mg 
12 54 S, J, C, H 252 MMF 2 g 10 
13 37 J, C, O 48 CYC 10 
14 53 S, J, K (naMC), LI, C, O, BL MMF 2.5 g, PEX 30 
15 50 S, J, K (IV), L 144 AZA 150 mg, CYC 10 
16 42 S, J, L, BL 24 AZA 175 mg 12.5 
17 13 S, J, K (IV), L, O 14  30 IV MP 
18 35 S, J, C 24 IV Ig  
19 43 S, K (NAS), J, L, O 315  10 
20 27 S, J, O 36 FK 2.5 mg 15 
21 59 S, J, K (V), L 132 AZA 50 mg, CYC 20 
22 26 S, J, K (IV) MMF 2 g, CYC 10 
23 18 S, J, K (V), L, C, G, A, H 17 CYC 25 
24 48 S, J, L, C, H, BL 396 CYC 15 
25 36 S, J, K (V) 192 MMF 2 g, CYC 7.5 
26 44 S, J, HD, G 108 CYC 20 IV MP 
27 36 S, J, L, BL 12  10 
28 29 S, J, K (III), L, O 18 MMF 2 g, CYC 12.5 
29 61 S, J, L, O 84 MTX 15 mg 20 
30 22 S, J, L, C, O MMF 2 g, CYC, IV Ig 30 
31 32 S, J, K (aMC), L 168 MMF 1.5 g, FK 6 mg 17.5 

IS, immunosuppressive drugs; S, skin; J, joint; K, kidney; naMC, non-active minimal change; aMC, active minimal change; bnp, biopsy not performed; KTX, kidney transplant; HD, haemodialysis; NAS, nephroangiosclerosis; L, lung; C, central nervous system; O, ocular; H, heart; A, anti-phospholipid syndrome; BL, blood; LI, liver; G, gut; AZA, azathioprine; MMF, mycophenolate mofetil; CyA, cyclosporine; PEX, plasma exchange; IVIg, intravenous immunoglobulin; FK, tacrolimus; MTX, methotrexate; RAPA, rapamycin; CYC, one pulse of cyclophosphamide.

Table 1

Characteristics of the patients at baseline

PatientSexAgeOrgan systems involved since diagnosisDisease duration, monthsNumber of previous IS therapiesIS therapies at the time of first rituximabPrednisolone dose, mg/day
59 S, J, A, O 108  10 
21 S, J, K (IV), L, BL 24  12.5 
45 S, K (bnp), C, O, H 168  10 
50 S, J, C, A 336  10 
36 S, J, HD, L, H, C, BL 144 AZA 50 mg, CYC 30 
47 S, J, C, O 180 CYC 10 
50 S, J, C 24  
44 S, J, A 108 CYC 10 
49 S, J, C, O 24 MMF 2 g, CyA 150 mg, CYC 
10 56 S, J, K (IV) 372 MMF 3 g, CYC 12.5 
11 25 S, J, KTX, H 96 RAPA 3 mg 
12 54 S, J, C, H 252 MMF 2 g 10 
13 37 J, C, O 48 CYC 10 
14 53 S, J, K (naMC), LI, C, O, BL MMF 2.5 g, PEX 30 
15 50 S, J, K (IV), L 144 AZA 150 mg, CYC 10 
16 42 S, J, L, BL 24 AZA 175 mg 12.5 
17 13 S, J, K (IV), L, O 14  30 IV MP 
18 35 S, J, C 24 IV Ig  
19 43 S, K (NAS), J, L, O 315  10 
20 27 S, J, O 36 FK 2.5 mg 15 
21 59 S, J, K (V), L 132 AZA 50 mg, CYC 20 
22 26 S, J, K (IV) MMF 2 g, CYC 10 
23 18 S, J, K (V), L, C, G, A, H 17 CYC 25 
24 48 S, J, L, C, H, BL 396 CYC 15 
25 36 S, J, K (V) 192 MMF 2 g, CYC 7.5 
26 44 S, J, HD, G 108 CYC 20 IV MP 
27 36 S, J, L, BL 12  10 
28 29 S, J, K (III), L, O 18 MMF 2 g, CYC 12.5 
29 61 S, J, L, O 84 MTX 15 mg 20 
30 22 S, J, L, C, O MMF 2 g, CYC, IV Ig 30 
31 32 S, J, K (aMC), L 168 MMF 1.5 g, FK 6 mg 17.5 
PatientSexAgeOrgan systems involved since diagnosisDisease duration, monthsNumber of previous IS therapiesIS therapies at the time of first rituximabPrednisolone dose, mg/day
59 S, J, A, O 108  10 
21 S, J, K (IV), L, BL 24  12.5 
45 S, K (bnp), C, O, H 168  10 
50 S, J, C, A 336  10 
36 S, J, HD, L, H, C, BL 144 AZA 50 mg, CYC 30 
47 S, J, C, O 180 CYC 10 
50 S, J, C 24  
44 S, J, A 108 CYC 10 
49 S, J, C, O 24 MMF 2 g, CyA 150 mg, CYC 
10 56 S, J, K (IV) 372 MMF 3 g, CYC 12.5 
11 25 S, J, KTX, H 96 RAPA 3 mg 
12 54 S, J, C, H 252 MMF 2 g 10 
13 37 J, C, O 48 CYC 10 
14 53 S, J, K (naMC), LI, C, O, BL MMF 2.5 g, PEX 30 
15 50 S, J, K (IV), L 144 AZA 150 mg, CYC 10 
16 42 S, J, L, BL 24 AZA 175 mg 12.5 
17 13 S, J, K (IV), L, O 14  30 IV MP 
18 35 S, J, C 24 IV Ig  
19 43 S, K (NAS), J, L, O 315  10 
20 27 S, J, O 36 FK 2.5 mg 15 
21 59 S, J, K (V), L 132 AZA 50 mg, CYC 20 
22 26 S, J, K (IV) MMF 2 g, CYC 10 
23 18 S, J, K (V), L, C, G, A, H 17 CYC 25 
24 48 S, J, L, C, H, BL 396 CYC 15 
25 36 S, J, K (V) 192 MMF 2 g, CYC 7.5 
26 44 S, J, HD, G 108 CYC 20 IV MP 
27 36 S, J, L, BL 12  10 
28 29 S, J, K (III), L, O 18 MMF 2 g, CYC 12.5 
29 61 S, J, L, O 84 MTX 15 mg 20 
30 22 S, J, L, C, O MMF 2 g, CYC, IV Ig 30 
31 32 S, J, K (aMC), L 168 MMF 1.5 g, FK 6 mg 17.5 

IS, immunosuppressive drugs; S, skin; J, joint; K, kidney; naMC, non-active minimal change; aMC, active minimal change; bnp, biopsy not performed; KTX, kidney transplant; HD, haemodialysis; NAS, nephroangiosclerosis; L, lung; C, central nervous system; O, ocular; H, heart; A, anti-phospholipid syndrome; BL, blood; LI, liver; G, gut; AZA, azathioprine; MMF, mycophenolate mofetil; CyA, cyclosporine; PEX, plasma exchange; IVIg, intravenous immunoglobulin; FK, tacrolimus; MTX, methotrexate; RAPA, rapamycin; CYC, one pulse of cyclophosphamide.

All received one course of RTX. Fourteen received at least three courses of RTX (6, 5, 1, 1 and 1 patients received 3, 4, 5, 6 and 7 courses, respectively).

Efficacy

B-cell depletion

Following the first RTX course, 30/31 (97%) patients achieved peripheral blood B-cell depletion by 4 weeks after the first RTX dose (Figure 1). B-cell regeneration occurred in 17/30 (56%) patients after a median of 10 months (range, 2–16). In four who received only one course of RTX, B cells remained absent at 6, 12, 12 and 31 months. Nine who received repeated courses of RTX before B-cell reconstitution remained continuously B cell depleted.

Fig. 1

Changes in CD19 B-cell number after the first course of rituximab.

Twenty of 22 (90%) who received a second course of rituximab achieved B-cell depletion. B-cell regeneration occurred in 6/20 (30%) patients after a median of 6.5 months (range, 5–10). This was similar to the median duration of B-cell depletion after the first rituximab course for the same six patients (median time, 10.5 months; range, 4–12) (P = 0.16).

Fifteen were re-treated three times; all achieved B-cell depletion. B-cell regeneration occurred in 7 of 15 (45%) patients after a median of 6 months (range, 1–12).

Remission induction

Twenty-seven of 31 patients (87%) achieved remission, 17 full and 10 partial, after the first rituximab course with a median time to remission of 4 months (range, 1–9). Persisting features of lupus activity in the partial responders were proteinuria (n = 7), malaise and fatigue (n = 2), and pneumonitis (n = 1). In 18 patients with a 2-year follow-up, global BILAG fell from a median of 14.5 at entry to 3 at 12 months (P < 0.001) and 3.5 at 24 months (P < 0.001) (Figure 2A).

Fig. 2

Box plots, interquartile range, of BILAG (A) and daily prednisolone dose (B) changes following rituximab treatment in 18 and 22 patients with a 2-year follow-up (*P < 0.01, #P < 0.001 and ΔP < 0.05).

Renal response

Ten of 11 (91%) patients with active glomerulonephritis achieved a renal response. This was partial in six patients. There was a gradual reduction of proteinuria during follow-up (Figure 3).

Fig. 3

Box plots, interquartile range, of proteinuria changes before and following rituximab therapy in nine patients with active glomerulonephritis and a 2-year follow-up (*P < 0.01).

In nine patients with a 2-year follow-up, proteinuria fell from a median of 2.2 g/day at entry to 1.5 at 12 months (P > 0.05), 0.94 at 18 months (P < 0.01) and 0.5 at 24 months (P < 0.01) (Figure 3). There was a trend of serum albumin levels to increase (P = 0.20). Taking the group as a whole, renal function remained stable over 3 years of follow-up (plasma creatinine, 86.78 ± 17.24 μmol/L at baseline, 84 ± 11.75 μmol/L at 1 year, 78.33 ± 14.12 μmol/L at 2 years and 78 ± 16.67 μmol/L at 3 years; P = 0.9). In five patients with renal impairment at baseline (Patient 3, 10, 15, 17 and 21), estimated glomerular filtration rate increased in three and was stable in two during follow-up. All but one had a significant reduction of proteinuria. Haematuria, which was present at a median of 3 in eight patients at entry, disappeared after 12 months and remained persistently negative after 24 months in six. Of the six patients with proliferative glomerulonephritis, three (Patient 10, 15 and 28) achieved a complete and two a partial response. Three patients with membranous and one patient with minimal change glomerulonephritis responded partially to RTX.

Non-responders

Four patients failed to respond to the first course of rituximab. Three patients with long-standing SLE and persisting myalgia, arthralgia, and headache were re-treated with rituximab without improvement. One patient with class IV proliferative nephritis and a restrictive pulmonary defect associated with a recent history of pneumonia had persisting breathlessness and proteinuria after the first course but responded to a second rituximab course. Proteinuria fell slowly to 50% below baseline after 6 months.

Relapse and re-treatment

Eighteen of 27 responders (67%) relapsed after a median of 11 months (range, 4–24). Relapses occurred after B-cell return in 10. The median time between B-cell return and relapse was 6.5 months (range, 1–26). Relapse occurred in six patients without subsequent B-cell return and in two patients before B-cell return. Among 18 relapsing patients, there was no significant difference between those who were or were not taking immunosuppressive drugs in terms of time to relapse (P = 0.86) and time to B-cell return (P = 0.47). Similarly, there was no significant difference between the two different regimens of RTX in terms of time to relapse (P = 0.24).

Relapse was mild with constitutional symptoms, sweats, arthralgia and fatigue in two; moderate with skin and joint involvement (diffuse rash, alopecia, mouth ulcers and arthritis) in nine; and severe with renal, pulmonary or eye disease in seven patients.

Sixteen relapsing patients were re-treated with RTX; there was a complete remission in six, partial remission in nine and no response in one. The latter patient developed a nephritic flare with crescentic change on repeat renal biopsy following a respiratory tract infection and, subsequently, achieved complete remission after plasma exchange, cyclophosphamide and a third course of rituximab. The median time to second remission was 2 months (range, 1–5), and relapses following re-treatment occurred at a median of 12 months (range, 5–28).

At the time of reporting, nine patients were in sustained remission after a median of 32.5 months from their first RTX course; four of whom had received only one rituximab course, while five had received protocol RTX repeat treatment for remission maintenance.

Concomitant medications

In 25 patients with a 1-year follow-up, the median prednisolone dose decreased from 10 mg/day at entry to 8.3 mg/day at 6 months (P > 0.05) and 7.5 mg/day at 12 months (P < 0.05). In 22 patients with a 2-year follow-up, it decreased from 10 mg/day at entry to 7.5 mg/day at 12 months (P < 0.05) and 5.5 mg/day at 24 months (P < 0.001) (Figure 2B). Fifteen patients received one pulse of cyclophosphamide with the first rituximab course, and 16 patients had continued immunosuppressive drugs after rituximab, but neither influenced the duration of B-cell depletion, the initial response or the time to relapse. Five of 10 patients with a 2-year follow-up subsequently stopped immunosuppressive drugs, and five reduced the dose by 6 months. Five of nine patients with prolonged remission were taking immunosuppressive drugs. Six of 31 patients (19%) received hydroxychloroquine throughout, and two received IV Ig at the time of rituximab.

Serology

At baseline, anti-dsDNA levels were above the normal range of 50 IU/mL in 14 patients; of whom, 7 had active nephritis. Autoantibody levels following RTX fell from a median of 147 IU/mL at baseline to 67 IU/mL at 12 months (P > 0.05), 73 IU/mL at 18 months (P < 0.05) and 54 IU/mL at 24 months (P < 0.001) (Figure 4). Ten patients were positive for anti-extractable nuclear antigens (anti-ENA) at baseline, and all remained positive during follow-up.

Fig. 4

Box plots, interquartile range, of anti-dsDNA antibodies level changes following treatment with rituximab. The figures include the patients with a 2-year follow-up and anti-dsDNA levels above normal range (n = 12) at baseline (*P < 0.05 and #P < 0.001).

Before RTX, C3 and C4 levels were depressed in 9 and 13 patients, respectively.

In eight patients with a 1-year follow-up, C3 rose from 0.52 ± 0.07 at baseline to 0.68 ± 0.06 at 6 months (P > 0.05) and to 0.71 ± 0.09 at 12 months (P < 0.01). In seven patients with a 2-year follow-up data, they rose from 0.52 ± 0.09 at baseline to 0.73 ± 0.1 at 12 months (P < 0.01), 0.71 ± 0.11 at 18 months (P < 0.01) and 0.85 ± 0.12 g/L after 24 months (P < 0.01). In 11 patients with 1- and 2-year follow-up data, C4 levels rose from 0.08 ± 0.01 at baseline to 0.14 ± 0.01 at 12 months (P < 0.01), 0.13 ± 0.02 at 18 months (P < 0.01) and 0.15 ± 0.01 at 24 months (P < 0.01). The presence of hypocomplementaemia, anti-dsDNA or anti-ENA did not influence time to relapse. In 23 patients with a 2-year follow-up, ESR fell from a median of 21.5 mm/h at baseline to 17.5 mm/h after 24 months (P < 0.05).

Adverse events

Infusion reactions

Mild-to-moderate infusion reactions were common. Twenty episodes occurred in 14 patients (45%). Thirteen were immediate; the most common symptom was dyspnoea. Seven were delayed reactions, such as flu-like syndromes. Seven patients (23%) had severe infusion reactions (Table 2).

Table 2

Severe infusion reactions

PatientRTX courseReaction timeSymptomsIntervention with glucocorticoidsSubsequent RTX
2 weeks later Pericarditis P, 20 mg/day No 
48 h later Serum sickness reaction IV MP, 1500 mg No 
20 During infusion Throat swelling, rash and tachycardia None Yes (no complications) 
23 48 h later Breathlessness, polymyalgia None Yes (no complications) 
25 III During infusion Wide spread rash None No 
III During infusion Anaphylaxis P, 30 mg/day No 
1 week later Pyrexia and polyarthralgia None No 
PatientRTX courseReaction timeSymptomsIntervention with glucocorticoidsSubsequent RTX
2 weeks later Pericarditis P, 20 mg/day No 
48 h later Serum sickness reaction IV MP, 1500 mg No 
20 During infusion Throat swelling, rash and tachycardia None Yes (no complications) 
23 48 h later Breathlessness, polymyalgia None Yes (no complications) 
25 III During infusion Wide spread rash None No 
III During infusion Anaphylaxis P, 30 mg/day No 
1 week later Pyrexia and polyarthralgia None No 

Serum sickness reaction was characterized by polyarthritis and vasculitic rash; anaphylactic reactions were characterized by lip swelling, tachycardia, chest pain, headache and hypertension.

IV MP, intravenous methyl prednisolone; P, prednisolone.

Table 2

Severe infusion reactions

PatientRTX courseReaction timeSymptomsIntervention with glucocorticoidsSubsequent RTX
2 weeks later Pericarditis P, 20 mg/day No 
48 h later Serum sickness reaction IV MP, 1500 mg No 
20 During infusion Throat swelling, rash and tachycardia None Yes (no complications) 
23 48 h later Breathlessness, polymyalgia None Yes (no complications) 
25 III During infusion Wide spread rash None No 
III During infusion Anaphylaxis P, 30 mg/day No 
1 week later Pyrexia and polyarthralgia None No 
PatientRTX courseReaction timeSymptomsIntervention with glucocorticoidsSubsequent RTX
2 weeks later Pericarditis P, 20 mg/day No 
48 h later Serum sickness reaction IV MP, 1500 mg No 
20 During infusion Throat swelling, rash and tachycardia None Yes (no complications) 
23 48 h later Breathlessness, polymyalgia None Yes (no complications) 
25 III During infusion Wide spread rash None No 
III During infusion Anaphylaxis P, 30 mg/day No 
1 week later Pyrexia and polyarthralgia None No 

Serum sickness reaction was characterized by polyarthritis and vasculitic rash; anaphylactic reactions were characterized by lip swelling, tachycardia, chest pain, headache and hypertension.

IV MP, intravenous methyl prednisolone; P, prednisolone.

Six patients including the one who developed pericarditis appeared to have a transient worsening of their SLE symptoms commencing 2 weeks after the first RTX infusion, but in three, this was difficult to distinguish from concomitant infusion reactions.

Infections

There were 11 severe infections in eight patients (26%). Nine were chest infections; one was cutaneous herpes zoster and one, septicaemia. The median time from last RTX course to infection was 2 months (range, 1–15). Six patients received immunosuppressive drugs; two had moderate hypogammaglobulinaemia (IgG levels, 3.8 and 4.2 g/dL) prior to rituximab treatment. All the infections resolved with intravenous antibiotics.

Hypogammaglobulinaemia

IgG levels remained stable (Figure 5A).

Fig. 5

Changes in serum IgG (A) and IgM (B) levels following rituximab treatment. The dotted lines refer to the lower limit for IgG (7 g/dL) and IgM (0.4 g/dL) levels (*P < 0.01).

In 29 patients with a 1-year follow-up, IgM levels following RTX fell from 1.04 ± 0.15 g/dL at baseline to 0.8 ± 0.13 g/dL at 6 months (P < 0.05) and to 0.7 ± 0.09 g/dL at 12 months (P < 0.01). In 23 patients with a 2-year follow-up, they fell from 0.95 ± 0.17 g/dL at baseline to 0.64 ± 0.1 g/dL at 1 year (P < 0.01) and 0.62 ± 0.08 g/dL at 2 years (P < 0.01) (Figure 5B). IgG levels were below the normal range of 7 g/L in seven patients. In five, this was due to nephrotic syndrome; two had low IgG levels before commencing RTX.

Neutropaenia

One patient developed transient neutropaenia 3 weeks after the first RTX course (0.16 × 109/L) and again 4.5 months after the second RTX course (0.9 × 109/L) without complication.

Deaths

There were three deaths at 14, 46 and 48 months after the first course of RTX. The causes of deaths were active SLE and infection, sudden cardiac death, and severe sepsis associated with ESRD.

Discussion

Our report suggests that rituximab has the potential to be a long-term effective therapy for SLE patients with relapsing or refractory disease despite conventional therapy with immune suppression and prednisolone. The study is limited by a retrospective design and lack of a control group; however, all eligible patients were included, there was a prolonged period of evaluation and the cohort is representative of the spectrum of difficult-to-treat patients attending a tertiary referral centre.

Both rituximab regimens (375 mg/m2/week × 4 and 1000 mg × 2) were effective in achieving B-cell depletion, but failure to achieve depletion was associated with a poor clinical response. Although not a formal comparison, the 1000 mg × 2 regimen achieved the same level of clinical response as 375 mg/m2/week × 4 regimen. Clinical responses in 87% (32% partial) were reflected by falls in global BILAG and prednisolone requirement. The rate of response was similar to or better than that previously reported for rituximab in SLE [8–18]. Concomitant therapy with immune suppressive drugs and prednisolone may have contributed to the clinical response; however, all patients had previously failed such therapy, continued immune suppression did not prevent relapses after a response to rituximab and prednisolone doses could be significantly reduced during follow-up.

We observed clinical response in all of the organs and systems commonly affected by SLE.

Responses in lupus nephritis were similar to those with non-renal manifestations, although reduction in haematuria and proteinuria were delayed and did not reach a stable baseline until after 18 months.

Two-thirds of our patients relapsed, with a median time to relapse of 11 months. Although most relapses occurred after B-cell regeneration, eight relapsed without B-cell reconstitution; thus, conventional lymphocyte counts are not a reliable guide to rituximab re-treatment if relapse is to be avoided.

One-third of patients achieved a relapse-free response lasting at least 33 months. No baseline factors were found to correlate with this prolonged response; however, five from this group received protocol re-treatment with 1000 mg of RTX every 6 months for 2 years. The underlying mechanism by which RTX induces long-term remission remains unclear. Anolik et al. reported that rituximab improved abnormalities in B-cell homeostasis with a decreased proportion of auto-reactive memory B cells after treatment [20]. In view of the frequency of flare and the insensitivity of re-treatment based on the return of circulating B cells, a protocol of regular rituximab infusions to prevent relapse requires evaluation.

Serological abnormalities improved in parallel with clinical responses. Although changes in anti-dsDNA levels were varied between patients, on average they fell, and hypocomplementaemia improved, following rituximab. Positive baseline anti-ENA remained unchanged during follow-up, suggesting their production by longer lived plasma cells. This suggests that B-cell depletion, similar to conventional immunosuppression, has a greater effect on pathogenic antibodies, such as anti-dsDNA, than on autoantibodies thought to have less pathogenic potential, such as anti-ENA. Moreover, in contrast to a previous report [10], we found no associations of anti-ENA, anti-dsDNA or low complement levels with remission duration or relapse.

Rituximab was associated with infusion reactions in one-half of the patients; they were severe in seven. This contrasts with our experience in 74 vasculitis patients treated with RTX in our clinic in whom we have not observed any severe infusion reaction (unpublished observations). This suggests that SLE patients are more likely to develop severe infusion reactions; however, rituximab re-treatment after a prophylactic higher dose of steroid was possible, and there were no long-term sequelae to the reactions. It is unclear whether RTX contributed to the rate of serious infections in view of the concomitant therapies, prolonged disease and treatment course before rituximab. Rituximab has been associated with severe infection in other settings, and opportunistic infections have occurred in SLE patients after rituximab. The rate of infection we observed is comparable with that reported for trials with cyclophosphamide and prednisolone therapy [21,22]. The development of hypogammaglobulinaemia, or the presence of hypogammaglobulinaemia due to other causes prior to rituximab, is a potential barrier to rituximab therapy. Hypogammaglobulinaemia increases infection risk and may become a reason for rituximab withdrawal in a minority of patients. We did not measure human anti-chimeric antibodies (HACA) to rituximab in this study. HACA have been reported in up to 30% of SLE patients after rituximab [9,19] and may have caused the failure of one of our patients to deplete B cell after a second rituximab course. The introduction of humanized anti-B-cell antibodies may reduce this complication.

In conclusion, this long-term report on 31 patients with active lupus, including lupus nephritis, provides further evidence for the efficacy of rituximab in relapsing and refractory SLE. Our data support the use of RTX in ‘difficult-to-treat’ SLE patients refractory to conventional immune-suppressive drugs. Uncertainty remains as to the role of protocol re-treatment to prevent relapse and the contribution of rituximab to infection risk. Also, in a minority of patients, rituximab use is complicated by severe infusion reactions, HACA formation and hypogammaglobulinaemia. In the majority, however, rituximab has led to long-term stability of disease and reduced exposure to immunosuppressive drugs and prednisolone.

This work was supported by the NIHR Cambridge Biomedical Research Center. K.G.C.S was supported by the Wellcome Trust (083650/Z/07/Z) and F.C. by ERA–EDTA (short-term fellowship).

Conflict of interest statement. D.W.J. and K.G.C.S. have received research grant support and honoraria from Hoffman La Roche. F.C., A.N.C. and R.B.J. have not received any consultancy fee.

References

1
Nagy
G
Koncz
A
Perl
A
, et al. 
T- and B-cell abnormalities in systemic lupus erythematosus
Crit Rev Immunol
2005
, vol. 
25
 (pg. 
123
-
140
)
2
Balow
JE
Boumpas
DT
Austin
HA
RG
Lahita
Systemic Lupus Erythematosus and the Kidney
Systemic Lupus Erythematosus
1999
2nd edn.
New York, USA
Churchill Linvingstone
(pg. 
657
-
685
)
3
Abu-Shakra
M
Urowitz
M
Gladman
D
, et al. 
Mortality in systemic lupus erythematosus: results from a single center
J Rheumatol
1995
, vol. 
22
 (pg. 
1259
-
1264
)
4
Boumpas
Dt
Austin
HA
III
Vaughan
EM
, et al. 
Risk for sustained amenhorrea in patients with systemic lupus erythematosus receiving intermittent pulse cyclophosphamide
Ann Intern Med
1993
, vol. 
119
 (pg. 
366
-
369
)
5
Harjunpaa
A
Wiklund
T
Collan
J
, et al. 
Complement activation in circulation and central nervous system after rituximab treatment of B-cell Lymphoma
Leuk Lymphoma
2001
, vol. 
42
 (pg. 
731
-
738
)
6
Uchida
J
Hamaguchi
Y
Oliver
JA
, et al. 
The innate mononuclear phagocyte network depletes B lymphocytes trough Fc receptor-dependent mechanisms during anti-CD20 antibody immunotherapy
J Exp Med
2004
, vol. 
199
 (pg. 
1659
-
1669
)
7
Kimby
E
Tolerability and safety of rituximab
Cancer Treat Rev
2005
, vol. 
31
 (pg. 
456
-
473
)
8
Lindholm
C
Borjesson-ASP
K
Zendjanchi
K
, et al. 
Longterm clinical and Immunological Effects of anti-Cd20 treatment in patients with refractory systemic lupus erythematosus
J Rheumatol
2008
, vol. 
35
 (pg. 
826
-
833
)
9
Albert
D
Dunham
J
Khan
S
, et al. 
Variability in the biological response to anti-CD20 B-cell depletion in SLE
Ann Rheum Dis
2008
, vol. 
67
 (pg. 
1724
-
1731
)
10
Ng
KP
Cambridge
G
Leandro
MJ
, et al. 
B cell depletion therapy in systemic lupus erythematosus: long-term follow-up and predictors of response
Ann Rheum Dis
2007
, vol. 
66
 (pg. 
1259
-
1262
)
11
Jonsdottir
T
Gunnarsson
L
Risselada
A
, et al. 
Treatment of refractory SLE with rituximab plus cyclophosphamide: clinical effects, serological changes, and predictors of response
Ann Rheum Dis
2008
, vol. 
67
 (pg. 
330
-
334
)
12
Tanaka
Y
Yamamoto
K
Takeuchi
T
, et al. 
A multicenter phase I/II trial of rituximab for refractory systemic lupus erythematosus
Mod Rheumatol
2007
, vol. 
17
 (pg. 
191
-
197
)
13
Scheinberg
M
Hamerschlak
N
Kutner
JM
, et al. 
Rituximab in refractory autoimmune diseases: Brazilian experience with 29 patients (2002–2004)
Clin Exp Rheumatol
2006
, vol. 
24
 (pg. 
65
-
69
)
14
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
, vol. 
8
 pg. 
R83
 
15
Leandro
MJ
Cambridge
G
Edwards
JC
, et al. 
B-cell depletion in the treatment of patients with systemic lupus erythematosus: a longitudinal analysis of 24 patients
Rheumatology (Oxford)
2005
, vol. 
44
 (pg. 
1542
-
1545
)
16
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
, vol. 
52
 (pg. 
501
-
513
)
17
Gottenberg
JE
Guillevin
L
Lambotte
O
, et al. 
Tolerance and short term efficacy of rituximab in 43 patients with systemic autoimmune diseases
Ann Rheum Dis
2005
, vol. 
64
 (pg. 
913
-
920
)
18
Looney
RJ
Anolik
JH
Campbell
D
, et al. 
B cell depletion as a novel treatment for systemic lupus erythematosus: a phase I/II dose-escalation trial of rituximab
Arthritis Rheum
2004
, vol. 
50
 (pg. 
2580
-
2589
)
19
Smith
KGC
Jones
R
Burns
S
, et al. 
Long-term comparison of Rituximab treatment for refractory systemic lupus erythematosus and vasculitis
Arthritis and Rheum
2006
, vol. 
54
 (pg. 
2970
-
2982
)
20
Anolik
JH
Barnard
J
Cappione
A
, et al. 
Rituximab improves peripheral B-cell abnormalities in human systemic lupus erythematosus
Arthritis Rheum
2004
, vol. 
50
 (pg. 
3580
-
3590
)
21
Contreras
G
Pardo
V
Leclercq
B
, et al. 
Sequential therapies for proliferative lupus nephritis
N Engl J Med
2004
, vol. 
350
 (pg. 
971
-
980
)
22
Chan
TM
Tse
KC
Tang
CS
, et al. 
Long-term study of mycophenolate mofetil as continuous induction and maintenance treatment for diffuse proliferative lupus nephritis
J Am Soc Nephrol
2005
, vol. 
16
 (pg. 
1076
-
1084
)

Comments

0 Comments
Submit a comment
You have entered an invalid code
Thank you for submitting a comment on this article. Your comment will be reviewed and published at the journal's discretion. Please check for further notifications by email.