Lupus

http://lup.sagepub.com/

Rituximab in systemic lupus erythematosus: an updated systematic review and meta-analysis B Duxbury, C Combescure and C Chizzolini Lupus 2013 22: 1489 originally published online 17 October 2013 DOI: 10.1177/0961203313509295 The online version of this article can be found at: http://lup.sagepub.com/content/22/14/1489

Published by: http://www.sagepublications.com

Additional services and information for Lupus can be found at: Email Alerts: http://lup.sagepub.com/cgi/alerts Subscriptions: http://lup.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav

>> Version of Record - Nov 15, 2013 OnlineFirst Version of Record - Oct 17, 2013 What is This?

Downloaded from lup.sagepub.com at Scientific library of Moscow State University on November 15, 2013

Lupus (2013) 22, 1489–1503 http://lup.sagepub.com

PAPER

Rituximab in systemic lupus erythematosus: an updated systematic review and meta-analysis B Duxbury1, C Combescure2 and C Chizzolini3 1

General Internal Medicine, University Hospital, Geneva, Switzerland; 2Division of Clinical Epidemiology, University Hospital, Geneva, Switzerland; and 3Immunology and Allergy, University Hospital and School of Medicine, Geneva, Switzerland

The wide spectrum of clinical manifestations and high relapse rate represent a therapeutic challenge in systemic lupus erythematosus (SLE). Observational studies suggested efficacy of rituximab (RTX), a B-cell-targeting antibody, to control the activity of SLE. Two randomized trials controlled by placebo did not prove the superiority of RTX when used in addition to conventional treatment in nonrenal (EXPLORER) and renal (LUNAR) lupus. A systematic review of studies exploring the efficacy of RTX in SLE patients was conducted. The pooled percentages of response were assessed. Thirty studies with 1243 patients were analyzed. In studies using the British Isles Lupus Assessment Group (BILAG), the complete response (CR) rate was 46.7% (95% CI 36.8%– 56.8%) and the partial response (PR) was 37.9% (95% CI 30.6%–45.8%). With the Systemic Lupus Erythematosus Disease Activity Index (SLEDAI), the CR was 56.6% (95% CI 32.4%– 78.1%) and the PR was 30.9% (95% CI 8.9%–46%). In renal lupus the CR was 36.1% (95% CI 25.2%–48.6%); PR was 37.4% (95% CI 28.5%–47.3%). In EXPLORER, CR was 12.4% and PR was 17.2%; in LUNAR CR was 26.4% and PR was 30.6%, in both cases not different from controls. Assessment and standardization of SLE response to treatment remain a challenge. The discrepancy in the perceived efficacy of RTX between controlled and observational studies reflects the heterogeneity of lupus and stringency in criteria of response. Further randomized trials focusing on selected SLE manifestations and using composite response indices are warranted. Lupus (2013) 22, 1489–1503. Key words: SLE; rituximab; SLEDAI; BILAG; complete response

Introduction Systemic lupus erythematosus (SLE) is a clinically and serologically extremely heterogeneous disease characterized by flares and remissions, potentially involving any organ and tissue with a broad range of clinical manifestations. SLE affects predominantly women of childbearing age and is more frequent and severe among Africans, Hispanics and Asians.1 Mortality is still high, with nearly 10% risk of death within 10 years of diagnosis.2 The aim of treatment is first to induce and then maintain remission, avoiding unnecessary toxicity. Corticosteroids, antimalarial drugs and Correspondence to: Carlo Chizzolini, Immunology and Allergy, University Hospital, Rue Gabrielle Perret-Gentil 4, 1211 Geneva 14, Switzerland. Email: [email protected] Received 26 May 2013; accepted 26 September 2013

immunosuppressive agents including, but not exclusively, cyclophosphamide (CYC), mycophenolate mofetil (MMF) or azathioprine (AZA) are the pillars of treatment. However, they are not always efficacious and the frequent relapses require additional treatment, ultimately leading to long-term organ damage due to drug toxicity.3,4 A strong rationale for targeting B-cells in SLE is supported by their central role in SLE pathogenesis, being precursors of plasma cells producing autoantibodies, precipitating inflammation by producing cytokines, activating T-cells by presentation of self-antigens, and regulating T-cells activity via co-stimulatory molecules.5 Rituximab (RTX) is a chimeric monoclonal antibody that kills B-cells by binding to the surface molecule CD20 present at all stages of B-cell maturation, with the exception of early B-cell progenitors and plasma cells. It was originally developed as a therapeutic agent for B-cell lymphoma in the

! The Author(s), 2013. Reprints and permissions: http://www.sagepub.co.uk/journalsPermissions.nav

Downloaded from lup.sagepub.com at Scientific library of Moscow State University on November 15, 2013

10.1177/0961203313509295

Rituximab in SLE B Duxbury et al.

1490

1990s then approved for use in rheumatoid arthritis. Since 2002, numbers of observational studies have indicated RTX efficacy in patients with refractory SLE, recently even in treatment-naı¨ ve patients. More than 1000 patients have been included in open-label trials. However, two recent randomized controlled trials failed to demonstrate RTX benefit compared to placebo when added to standard treatment. The aim of the present work is to review systematically published studies assessing RTX efficacy in SLE patients. In our meta-analysis we grouped studies according to the activity indices used, focusing on clinical outcomes. Major attention was put on assessing the differences between open-label and the two randomized controlled studies.

Methods

the only one using the Systemic Lupus Activity Measure (SLAM) to evaluate disease activity.6 Statistical analysis The percentages of response (partial, complete and global) were pooled by using the method of the inverse of the variance, with random effects,38 on the logit transformed proportions. The presence of heterogeneity was assessed by using the statistic I-squared test,39 which reflects the proportion of between-studies variance that is due to factors other than chance. The length of follow-up of studies was explored as a potential heterogeneity factor: a meta-regression with random effects was performed and the statistical significance of the slope coefficient was reported. All analyses were performed using Comprehensive Meta Analysis V.2 (Biostat, Englewood, NJ, USA) with the significance level set at 0.05.

Selection of trials We searched MEDLINE and EMBASE using the terms ‘‘rituximab’’ and ‘‘systemic lupus erythematosus’’ published from January 1, 2002 to October 31, 2012. Studies were included if they were randomized, controlled studies or case series of more than 10 adult patients centered on clinical outcomes. Papers were excluded if they were reviews or expert comments; if they were case series with fewer than 10 patients; if they were pediatric cases; if the main outcome was not clinical and if they were published only as abstract form. When several publications under the same group of patients were found, only the most recent and comprehensive paper was included, unless the publication was derived from another patient cohort. Among the 505 sorted publications, 30 studies met the inclusion criteria, with 28 retrospective case series, prospective case series, case-control studies and open-label randomized studies6–33 and two randomized controlled trials.34,35 For each paper, selected items were systematically searched for: number of patients included, follow-up, indication for treatment with RTX and dosage, concomitant immunosuppressive treatment, corticosteroid dosage and clinical and biological outcomes. The results were compiled according to the clinical score used to assess disease activity (British Isles Lupus Assessment Group (BILAG);36 Systemic Lupus Erythematosus Disease Activity Index (SLEDAI);37 and renal values) and outcome, whenever sufficient details were provided in the original publication. One study that qualified for inclusion in our analysis is not reported in the tables because it is

Results Open-label trials using BILAG to assess disease activity We found eight open-label trials that used BILAG to assess disease activity and response to treatment, with very small differences in defining complete and partial response, detailed in Table 1. Three prospective case series,11,13,30 four retrospective case series16,18,26,28 and one case-control trial.31 In six trials RTX was used in refractory SLE with severe organ involvement despite conventional treatment. One trial evaluated response to RTX in patients who flared after a previous course of RTX,28 and one included only patients with treatment-naı¨ ve SLE, in which RTX was used to induce remission.31 In total 188 patients were included. The dose of RTX was variable among studies, being 4  375 mg/m2 or 2  1 g in 31 and 139 patients, respectively. The follow-up period varied between three and 12 months. Six of eight trials documented the exact rate of clinical response.11,13,18,26,28,30 The pooled percentage of complete response (CR) (i.e. no BILAG score A or B) from these six papers was 46.7% (95% confidence interval (CI), 36.8%– 56.8%) and the pooled percentage of partial response (PR) was 37.9% (95% CI, 30.6%– 45.8%). The pooled percentage of global response was 82.8% (95% CI, 79%–89.6%) in the seven evaluable studies11,13,16,18,26,28,30 (Figure 1). In seven trials, RTX was used in conjunction with

Lupus Downloaded from lup.sagepub.com at Scientific library of Moscow State University on November 15, 2013

16

11

45

31

17

37

PCS

RCS

RCS

RCS

RCS

PCS

Jo´nsdo´ttir 200813

Reynolds 200916

Lu 2009c18

Catapano 201026

Turner-Stokes 2011b28

Vital 201130

14

N

PCS

Study design

Tanaka 200711

Studyref.

Downloaded from lup.sagepub.com at Scientific library of Moscow State University on November 15, 2013

Refractory SLE active disease (>1A or >2Ba) despite ttt

Flare following a previous treatment with RTX

Refractory SLE

Refractory SLE active SLE for 10.4 y (mean) despite ttt

Refractory SLE: active SLE despite 1 IS

Refractory SLE: active disease (>1 A or >2Ba) despite ttt

active disease (>1A or >2Ba) despite ttt

Refractory SLE

RTX indication

Oral CST only tapered

All

21g

All (16 pts)

4  375 mg/m2 (15 pts) or 2  1 g (16 pts)

21g

None

CST only (but one pt w/ MMF)

None

Oral CST only

IS continued

2  1 g (46 pts); 2  500 mg (4 pts)

2  750 mg (3 pts), 2  1 g (5 pts); 2  375 mg/m2 (2 pts) or 2  500 mg (1 pt)

4  375 mg/m2

4  500 mg (5 pts) or 2  1 g (10 pts)

RTX dose

Table 1 Studies using BILAGa score as outcome measure

2  100 mg MTP then oral CST and 2  10 mg/Kg CYC (three pts)

2  750 mg CYC and 2  100– 250 mg MTP

500 mg CYC (15 pts); 500– 1000 mg MTP (2 pts); IVIG (2 pts); 1  100 mg HCS

2  750 mg CYC and 2  100– 250 mg MTP

500–750 mg CYC (7 pts); 250–500 mg MTP (6 pts)

2  0.5 g/m2 CYC and 2  250 mg MTP then oral CST

None

IS added

12

6,5

60 mg/d 1 w then 30 mg/d 1 w

4

6

3

6

7

F/U

NM

10 mg/d at entrance

NM

30 mg/d at entrance; median reduction at 3 m: 7.5 mg/d

0.5 mg/kg/d 4 w then tapered

30 mg/d

CST dose (mean)

CR 54%; PR 32%

CR 41%; PR 24%

CR 55%; PR 32%

CR 42%; PR 47%

R 90%

CR 56%; PR 37.5%

CR 14%; PR 50%

Clinical outcome

CR: No A/B and no flare PR: 1A or >2B despite ttt

Active SLE induction only

RTX indication

Conventional treatment

Group 2: None (three controls for each study pt ¼ 24 pts) 2  1 g (169 pts) or placebo (88 pts) All: MTX, MMF, AZA, CST (if stable dose of 1 IS) High-dose CST or CYC excluded

CST tapered

IS continued

Group 1: 2  1 g (8 pts)

RTX dose

4  100 mg MTP then oral CST

NA

1  750 mg CYC and 2  100 mg MTP then 2 mg/kg AZA

IS added

0.5–0.75-1 mg/ kg) for 16 d then tapered

Cumulative CST dose at 6 m: 2834.6 mg

13.2 mg/d at entrance. Cumulative CST dose at 6 m: 1287.3 mg

CST dose (mean)

12

6

F/U

BILAG: No difference in major or partial clinical response in the two groups Group 1 (RTX): CR 12.4%; PR 17.2% Group 2 (Placebo): CR 15.9%; PR 12.5%

No difference in the two groups but CST sparing

Clinical outcome

CR: no A/B and no flare w 24 PR: no A/B and no flare w40 OR 1B and no flare w52 or 8) despite ttt Refractory SLE active disease (BILAG > 1A or > 2B) despite ttt Refractory SLE active SLE (SLEDAI >1) for > 6 m despite ttt Refractory SLE active disease despite  1 ttt

Refractory SLE active disease for 9 y (mean) despite ttt Refractory LN active WHO class III/IV/V despite ttt Refractory NPSLE

RTX indication

CST only

4  375 mg/m2

2  1 g (82 pts) or 4  375 mg/m2 (48 pts)

All (59 pts)

All

None

4  375 mg/m2

21g

4  40 mg MTP

CST only and MMF (6 pts)

Variable (12 pts)

2  500 mg MTP

2  0.5 g/m2 CYC and 2  250 mg MTP then oral CST 4  100 mg MTP

None

Oral CST only

2  375 mg/m2 (6 pts); 4  500 mg (2 pts); 1  375 mg/m2 (1 pt) or 2  1 g (1 pt) 4  375 mg/m2 or 2  1 g

None

2  80 mg MTP (2 pts)

All

All

4  375 mg/m2

IS added

2  0.5 mg–1 g

IS continued

RTX dose

Table 2 Studies using SLEDAI score as outcome measure

16 mg/d at entrance (dose adjusted during trial) 30 mg/d at entrance

18 mg/d at entrance

0.5 mg/kg/d 4 w then tapered

23 mg/d at entrance

15–40 mg/d

NM

33 mg/d at entrance

CST dose (mean)

6

6

3.5

6

12

6

3

7.9

F/U

MEX-SLEDAI: global score reduction of three points SELENASLEDAI: R 71% (no difference between pts receiving RTX and those w/ concomitant IS)

SLEDAI: R 63%

SLEDAI: CR 56% PR 25%

SLEDAI: R 83%

SLEDAI: CR 90%

MEX-SLEDAI: R 90%

SLEDAI: CR 54%; PR 15%

Clinical outcome

R: SELENASLEDAI reduction score >3

R: Improvement in SLEDAI score by >2 points

CR:SLEDAI 50%

Definition

NM

NM

NM

Significant decrease in anti-dsDNA Ig level

NM

NM

NM

NM

Immunological parameters

(continued)

Twelve infections; five serum sickness-like reactions; two infusion reactions

One infection; two serum-sickness like reactions

One serum sicknesslike reaction; one infusion reaction

None

None

None

One infection

One sepsis-like syndrome

Severe adverse events

Rituximab in SLE B Duxbury et al.

1494

46

10

PCS

PCS

RCS

GalarzaMaldonado 201027

Chen 201129

Ferna´ndezNebro 201232

Refractory thrombocytopenia Refractory SLE active disease despite  2 ttt

Refractory SLE

Refractory SLE active SLE for 5.6 y (mean) despite ttt

RTX indication

2  1 g (83 pts) or 4  375 mg/m2 (38 pts) Repeated course of RTX (69 pts)

4  100 mg

2  500 mg

Group 2: none (9 pts)

Group 1: 2  1 g (10 pts)

RTX dose

All

CST only tapered

CST only

None

IS continued

CST 127 pts; CYC 32 pts; MMF 15 pts; AZA 25 pts; MTX 18 pts; RTX alone 16 pts; RTX þ HCQ 16 pts

CST and 6  750 mg– 1 g CYC (9 pts) 2  500 mg MTP and 1 g/d MMF (LN) or 500 mg/d (non-LN) and 400 mg/d HCQ None

CST

IS added

0.5 mg/kg/d

NM

15–60 mg/d rapidly tapered

1 mg/kg/d tapered

1 mg/kg/d tapered

CST dose (mean)

6 (3)

6

24

‘‘

12

F/U

SELENASLEDAI: CR 19%; PR 44%

SLEDAI: R 60%

MEX-SLEDAI: CR 50%; PR 46%

MEX-SLEDAI: CR 89%

MEX-SLEDAI: CR 90%

Clinical outcome

R: SLEDAI Reduction score>3 CR: SELENASLEDAI