Rheumatology 2005;44:15421545

Advance Access publication 27 September 2005

doi:10.1093/rheumatology/kei080

B-cell depletion in the treatment of patients with

systemic lupus erythematosus: a longitudinal
analysis of 24 patients
M. J. Leandro, G. Cambridge, J. C. Edwards,
M. R. Ehrenstein and D. A. Isenberg

KEY WORDS: SLE, Rituximab, B-cell depletion.

The past 50 yr have seen a substantial improvement in the

mortality and morbidity gures for patients with systemic lupus
erythematosus (SLE). In the 1950s, the 4-yr survival for lupus was
thought to be approximately 50% [1], whereas most recently
published series [reviewed in 2] now estimate a 15-yr survival rate
of the order of 80%. This major improvement has been brought
about by the introduction of corticosteroids, other immunosuppressive drugs, such as azathioprine and cyclophosphamide, and
adjunctive therapies, including dialysis and renal transplantation.
Although the more recent gures are most encouraging, a cohort
of patients continues to suffer from very aggressive disease which,
even with full compliance, continues to cause worrying morbidity
and, sadly, mortality. In our own group of over 400 patients under
follow-up, 15% have died at an average age of 51 yr [3]. Part of
the ongoing problem is the severity of the disease in some patients,
while in other cases the side-effects of the currently available
immunosuppressives, notably the increased risk of infection,
continue to make a major contribution to this increased mortality.
There is therefore a continuing need to identify new,
more effective therapies for severely affected patients, with fewer

side-effects. As reviewed elsewhere, recent interest has focused on

the utility of removing C20-positive B lymphocytes in patients with
SLE with rituximab, a chimeric monoclonal antibody directed to
CD20 that is very effective in depleting normal and malignant
B lymphocytes in vivo [4, 5]. Some encouraging early results have
now been published [6, 7].
The notion that B-cell depletion might be useful as a form
of treatment for patients with an autoimmune rheumatic
disease, rheumatoid arthritis, was rst demonstrated in an open
study [8, 9]. A large double-blind multicentre controlled trial
has recently conrmed the value of B-cell depletion in rheumatoid
arthritis [10]. These early results were obtained by using a
combination of rituximab, a chimeric monoclonal antibody
recognizing the CD20 marker, together with corticosteroids (orally
or intravenously) and intravenous cyclophosphamide or oral
methotrexate.
In our rst report of six patients with SLE given rituximab,
intravenous cyclophosphamide and corticosteroids [6], we
described both clinical improvement and a lowering of doublestranded DNA (dsDNA) antibody levels and an increase in C3 in

Centre for Rheumatology, University College London, London, UK.

Submitted 26 April 2005; revised version accepted 19 July 2005.
Correspondence to: D. Isenberg, Centre for Rheumatology, University College London, 4th Floor, Arthur Stanley House, 4050 Tottenham Street,
London W1T 4NJ, UK. E-mail: d.isenberg@ucl.ac.uk
1542
The Author 2005. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Downloaded from http://rheumatology.oxfordjournals.org/ by guest on September 18, 2014

Objectives. To assess the clinical and basic serological consequences of B-cell depletion with rituximab in the treatment
of patients with systemic lupus erythematosus (SLE) who have failed conventional immunosuppression.
Methods. An open study of 24 patients with severe SLE followed for a minimum of 3 months is reported. In the majority
of patients (19 out of 24), 6 months follow-up data are described. Disease activity in these patients was assessed every
12 months using the British Isles Lupus Assessment Group (BILAG) system and estimates of anti-double-stranded DNA
antibodies and serum C3 levels. During the follow-up period, signicant side-effects were sought and the reduction in oral
prednisolone was recorded. It was our general practice to stop concomitant immunosuppression (e.g. azathioprine,
mycophenolate) when B-cell depletion was given (in most cases in the form of two 1 g intravenous infusions of rituximab
2 weeks apart accompanied by two 750 mg intravenous cyclophosphamide infusions and two methylprednisolone infusions of
250 mg each).
Results. Twenty-two patients were female and two male. At the time of B-cell depletion, the mean age was 28.9 yr (range
1749) and the mean disease duration was 7.9 yr (range 118). The global BILAG score (P < 0.00001), serum C3 (P < 0.0005)
and double-stranded DNA binding (P < 0.002) all improved from the time of B-cell depletion to 6 months after this treatment.
Only one patient failed to achieve B-lymphocyte depletion in the peripheral blood. The period of B-lymphocyte depletion ranged
from 3 to 8 months except for one patient who remains depleted at more than 4 yr. Analysis of the regular BILAG assessments
showed that improvements occurred in each of the eight organs or systems. The mean daily prednisolone dose fell from 13.8 mg
(S.D. 11.3) to 10 mg (S.D. 3.1).
Conclusion. In this open study of patients who had failed conventional immunosuppressive therapy, considerable utility in
the use of B-cell depletion has been demonstrated. Our data provide strong support for the performance of a full doubleblind control trial.

Treatment of SLE by B-cell depletion

some of the patients. We now report a long-term follow-up study
of up to 4 yr in 24 patients treated with B-cell depletion.

Patients, materials and methods

activity index [12]. This index distinguishes disease activity in

eight organs or systems using the principle of the physicians
intention to treat. A combination of clinical features and a small
number of blood test results allows the distinction of patients
from the most active, grade A, to the never active, grade E, in that
organ or system. A global BILAG numerical score can be
derived using the following notation: A 9 points, B 3 points,
C 1 point, and D/E 0 points. The patients routine full blood
count, serum creatinine, C3 (laser nephelometry), total immunoglobulin and dsDNA binding (Shield Diagnostics, Dundee, UK;
normal <50 units/ml) and urine protein to creatinine ratio were
tested in the hospital routine laboratories. Paired t-tests or
Wilcoxon matched-pairs signed-rank tests were used to compare
global BILAG scores and key laboratory test results prior to
and 3 months and 6 months after B-cell depletion.

Results
Twenty-four patients were treated (22 female, two male). At the
time of B-cell depletion, the mean age was 28.9 yr (range 1749)
and mean disease duration was 7.9 yr (range 118). Data at
baseline and 3 months include all patients. Data at 6 months
include 19 of the 24 patients. For patients 1, 18, 19 and 22, data
at 6 months are not available. Patient 1 was lost to follow up at
3 months. Patient 18 did not deplete and other treatment options
had to be considered. Patient 19 repopulated at 3 months and
was retreated. Patient 22 died at approximately 5 months with
pancarditis.
Only one patient (patient 18) failed to achieve B-lymphocyte
depletion in the peripheral blood (i.e. CD 19 count <0.005
109/l).
The period of B-lymphocyte depletion ranged from 3 to 8 months,

TABLE 1. Patients baseline characteristics

Sex

Age
(years)

Disease
duration
(years)

1
2
3
4

F
F
F
F

38
35
36
40

7
11
10
11

17

Afro-caribbean

6
7
8

F
F
F

20
49
21

12
10
4

White
Afro-caribbean
White

9
10
11
12
13
14
15

M
F
F
F
F
F
F

27
30
17
38
26
18
18

10
15
6
8
6
3
4

Afro-caribbean
Afro-caribbean
Asian
Afro-caribbean
White
White
White

16
17
18
19
20
21
22

M
F
F
F
F
F
F

23
21
35
30
40
28
24

7
1
9
12
8
7
1

Afro-caribbean
Afro-caribbean
White
White
Afro-caribbean
White
Afro-caribbean

23

34

18

Asian

24

32

13

Asian

Patient

Ethnic group
White
White
White
White

Main clinical manifestations

Fever, organic brain syndrome, arthritis
Arthritis, serositis, skin vasculitis
Lymphadenopathy, arthritis, serositis
Skin rash, arthritis, serositis, shrinking
lung syndrome, skin vasculitis, nephritis (IV)
Skin rash, arthritis, serositis,
skin vasculitis, nephritis (IV)
Skin rash, arthritis, nephritis (IV)
Arthritis, serositis, skin vasculitis, nephritis (IV)
Autoimmune thrombocytopaenia,
haemolytic anaemia, nephritis (V)
Arthritis, serositis, nephritis (IV)
Skin rash, arthritis, nephritis (IV)
Fever, skin rash, serositis, nephritis (IV)
Arthritis, nephritis (IV)
Fever, arthritis, nephritis (IV)
Autoimmune thrombocytopaenia
Fever, skin rash, CNS vasculitis,
arthritis, nephritis (IV), anaemia
Nephritis (IV), haemolytic anaemia
Nephritis (IV)
Arthritis, nephritis (IV)
Fever, skin rash, arthritis, skin vasculitis
Arthritis
Arthritis, nephritis (IV)
Fever, lymphadenopathy, organic brain
syndrome, arthritis, nephritis (IV), anaemia
Fever, skin rash, serositis,
skin vasculitis, nephritis (IV)
Fever, lymphadenopathy, skin rash,
arthritis, anaemia

Previous therapies
St,
St,
St,
St,

HXQ,
HXQ,
HXQ,
HXQ,

AZT, Cy MMF
MTX, CyA
AZT, CyA, MTX
AZT, Cy, MMF

St, HXQ, AZT, Cy, MMF

St, HXQ, AZT, Cy
St, HXQ, AZT, Cy, MTX, CyA
St, HXQ, AZT, MMF,
IVIG, Cy
St, AZT, Cy
St, HXQ, AZT, MMF, Cy
St, Cy, AZT, CyA, MMF
St, HXQ, Cy
St, HXQ, AZT, Cy
St, HXQ, AZT
St, AZT, Cy, MMF
St,
St,
St,
St,
St,
St,
St,

Cy, AZT
AZT, Cy, MMF
AZT, Cy, MMF
HXQ, AZT, Cy
HXQ, AZT, Cy, MTX
HXQ, AZT, Cy
HXQ, AZT, Cy

St, Cy, AZT

St, HXQ, AZT, Cy, MTX

HXQ, hydroxychloroquine; AZT, azathioprine; Cy, cyclophosphamide; CyA, cyclosporine A; MMF, mycophenolate mofetil; MTX, methotrexate;
St, steroids.

Downloaded from http://rheumatology.oxfordjournals.org/ by guest on September 18, 2014

All of the patients included in the study met four or more of

the revised classication criteria for the American College of
Rheumatology [11]. Most patients were treated on an urgent basis
because of clinical need as they had failed conventional immunosuppressive therapy. Local hospital ethical committee approval
and patients informed consent was obtained. Patients baseline
characteristics are summarized in Table 1.
Treatment protocols were based on rituximab. The rst
six patients (patients 16) were treated with two infusions of
500 mg rituximab (half dose) and two infusions of 750 mg
cyclophosphamide under oral steroid cover (2
prednisolone
60 mg for 5 days) given 2 weeks apart. Subsequently the patients
were treated with a similar protocol but with full-dose rituximab
(2
1000 mg) and intravenous methylprednisolone (2
250 mg)
instead of oral steroids. Two patients received rituximab monotherapy: one had an allergy to cyclophosphamide (patient 12) and
the other patient refused to have the drug (patient 9). Patients
were allowed to continue prednisolone and hydroxychloroquine.
Other immunosuppressives were stopped at baseline except for
two patients who continued their azathioprine (patient 15 started
at baseline on 100 mg; patient 16 continued on 150 mg). All except
three patients had previously been treated with intravenous
cyclophosphamide (Table 1).
The clinical assessment of these patients, most of whom were
followed every 12 months in our lupus clinic, was undertaken
using the British Isles Lupus Assessment Group (BILAG) disease

1543

M. J. Leandro et al.

1544

TABLE 2. Changes in detailed BILAG scores 6 months after B-cell depletion (n 19 patients)
Organ/system

A!B
A ! C/D
B!C
B!D

General

Mucocutaneous

CNS

Musculoskeletal

CVS/Respiratory

Vasculitis

Renal

Haematology

1
1
4

1
3

3
1

2
3

1
1
1

1
5
3

2
3
3
1

600

p<0.00001

10

400
P<0.08
300
200
100

P<0.0005)
600

anti-dsDNA (U/mL)

C3

0.75

C3

500

Protein/creatinine
ratio

0.5

0.25

Anti-dsDNA

500
P<0.002

400
300
200
100
0

6/12

6/12

FIG. 1. BILAG global scores, serum levels of C3 and anti-dsDNA antibodies, and urine protein:creatinine ratio at baseline and
6 months after B-cell depletion (n 19 patients; median  S.E.M.).
except for one patient who remains depleted at more than 4 yr
(patient 3).
The mean BILAG global score fell from a mean of 13.9 (S.D. 5.8)
at baseline to 6.8 (S.D. 4.7) at 3 months and 5.0 (S.D. 2.3) at
6 months. Table 2 shows more detailed data on sustained
improvement of at least two BILAG grades, i.e. A ! D, A ! C
and B ! D, and also B ! C in each of the eight organs or systems.
It is clear that B-cell depletion was able to improve some aspects
of lupus activity in every organ and system. In contrast, no lupus
patient developed any major sustained deterioration (i.e. new
A scores or D ! B scores) before B-cell repopulation of the
peripheral blood occurred. Manifestations such as fatigue,
arthralgia/arthritis, serositis, nephritis, thrombocytopenia and
haemolytic anaemia responded particularly well to the treatment.
It should be noted that relatively few patients with major CNS
disease were treated (n 3). No clear differences in outcome were
detected between patients treated with different protocols. The
mean daily dose of prednisolone before treatment was 13.8 mg
(S.D. 11.3) and for those completing 6 months of follow-up (n 21)
it had fallen to 10 mg (S.D. 3.1).
Mean total serum immunoglobulin levels decreased mildly
to moderately but remained within the normal range. Figure 1
shows the global BILAG score, clinical and laboratory parameters

before and 6 months after B-cell depletion. It is notable that

the fall in global BILAG score (P<0.00001), the fall in serum
anti-dsDNA antibody levels (P<0.002) and the rise in serum
C3 levels (P<0.0005) are highly statistically signicant. The
protein:creatinine ratio decreased following treatment but the
decrease did not reach statistical signicance (P<0.08).
One patient (patient 9) had a severe reaction to the second
rituximab infusion, with thrombocytopenia followed by pancytopenia, from which he recovered completely. One patient (patient
22) died approximately 5 months after B-cell depletion. She had
had very aggressive disease with major skin, renal, joint and
neuropsychiatric disease. Her BILAG score improved from 45
at baseline to 25 after 3 months but she later developed a fatal
pancarditis. B-cell repopulation of the peripheral blood was
already detectable before she developed sepsis. No other serious
adverse events, including serious or opportunistic infections,
have been observed. In particular, one patient (patient 3) remains
depleted more than 4 yr after treatment. She has not had any
serious or opportunistic infections.
The mean follow-up is now 23 months (range 351). At present,
13 patients (patients 2, 3, 6, 8, 9, 13, 14, 15, 16, 17, 21, 23 and 24)
remain well enough not to have had to restart other immunosuppressive therapy (follow-up ranging from 7 to 51 months).

Downloaded from http://rheumatology.oxfordjournals.org/ by guest on September 18, 2014

BILAG score

BILAG

Urinary protein/creatinine

15

Treatment of SLE by B-cell depletion

Seven patients have been retreated (eight treatments). One patient
failed to deplete on retreatment due to a documented specic
human anti-chimeric antibody (HACA) response (described
elsewhere [13]).

Discussion

levels of anti-DNA antibodies at baseline were noted, even though

a signicant decrease was not found in the group as a whole [15].
It should be noted that this was a phase I/II dose escalation
trial and patients were given three different doses of rituximab,
two below the recommended dose, and unaccompanied by cyclophosphamide or methylprednisolone. Their results also conrm
the safety of B-cell depletion as a therapeutic approach.
In summary, these encouraging long-term results on 24 patients
with lupus severely active despite years of immunosuppression
strongly encourage the view that removal of CD20-positive B cells
may be a very useful way of treating patients with active lupus. It is
now timely to undertake a large double-blind randomized control
trial of this treatment.
D.A.I. declares that in lieu of payment for a consultancy undertaken for Roche, they have paid a fee into a back arthritis charity
of which he is one of the trustees. The other authors have declared
no conict of interest.

References
1. Merrell M, Shulman LE. Determination of prognosis of chronic
disease: illustrated by systemic lupus erythematosus. J Chronic Dis
1955;1:1232.
2. Gladman DD, Urowitz MB. Prognosis mortality and morbidity in
systemic lupus erythematosus: In: Wallace DJ, Hahn BH, eds. Dubois
Lupus erythematosus, 6th edn. Philadelphia: Lippincott, Williams and
Wilkins, 2002:125574.
3. Moss K, Ioannou Y, Sultan SM, Haq I, Isenberg DA. Outcome of a
cohort of 300 patients with systemic lupus erythematosus attending
a dedicated clinic for over two decades. Ann Rheum Dis 2002;61:
40913.
4. Silverman GJ, Weisman S. Rituximab therapy and autoimmune
disorders. Arthritis Rheum 2003;48:148492.
5. Gorman C, Leandro M, Isenberg DA. Does B cell depletion have a
role to play in the treatment of systemic lupus erythematosus? Lupus
2004;13:3126.
6. Leandro MJ, Edwards JC, Cambridge G, Ehrenstein MR,
Isenberg DA. An open study of B lymphocyte depletion in systemic
lupus erythematosus. Arthritis Rheum 2002;46:26737.
7. Anolik JH, Campbell D, Felgar RE. The relationship of Fc IIIa
genotype to degree of B cell depletion by rituximab in the treatment
of systemic lupus erythematosus. Arthritis Rheum 2003;48:4559.
8. Edwards JC, Cambridge G. Sustained improvement in rheumatoid
arthritis following a protocol designed to deplete B cells.
Rheumatology 2001;40:20511.
9. Leandro MJ, Edward JC, Cambridge G. Clinical outcome in
22 patients with rheumatoid arthritis treated for B cell depletion.
Ann Rheum Dis 2002;61:8838.
10. Edwards JC, Szezepanski L, Scechinski J et al. Efcacy of B cell
targeted therapy with rituximab in patients with rheumatoid arthritis.
N Engl J Med 2004;350:257281.
11. Tan EM, Cohen AS, Fries JF et al. The 1982 revised criteria for
the classication of systemic lupus erythematosus. Arthritis Rheum
1982;25:12717.
12. 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;86:44758.
13. Tahir H, Rohrer J, Bhatia A, Wegener WA, Isenberg DA. Humanised
anti CD-20 monoclonal antibody in the treatment of severe resistant
systemic lupus erythematosus in a patient with antibodies against
rituximab. Rheumatology 2005, 44:5612.
14. Looney RJ, Anolik JH, Campbell DC et al. B cell depletion as a
novel treatment for systemic lupus erythematosus. Arthritis Rheum
2004;50:25809.
15. Anolik JH, Barnard J, Cappione A et al. Rituximab improves
peripheral B cell abnormalities in human systemic lupus erythematosus. Arthritis Rheum 2004;50:358090.

Downloaded from http://rheumatology.oxfordjournals.org/ by guest on September 18, 2014

This report signicantly extends both the clinical follow-up and

serological analysis following the use of B-cell depletion in a group
of patients with SLE whose disease has remained active in spite
of intensive immunosuppression. The results of this open study
conducted over a period of up to 4 yr provide further encouragement for the view that B-cell depletion may be a very useful
addition to the range of therapies available for patients with
very active lupus. In a group of 24 patients with active, refractory
SLE, all 23 patients who showed signicant depletion improved
following treatment. Encouragingly, we have seen signicant
clinical improvement in virtually all of the organs and systems
commonly affected by lupus. The possible exception to this is
patients with major CNS lupus, for whom we have insufcient
data as yet to make any tenable claim.
Whilst B-cell depletion using a combination of rituximab,
cyclophosphamide and prednisolone is unlikely to be a cure for
lupus, the fact that the majority of our patients were substantially
improved in spite of having failed years of conventional immunosuppression is notable. Our key observations are that: the
treatment regimens we have used almost always achieve a
substantial depletion of B cells in the peripheral blood (>99%);
the overwhelming majority of patients in whom this depletion is
achieved improved clinically, some going into remission; the
periods of B-cell depletion in the peripheral blood ranged from 3
to 11 months, except for one patient who remains depleted at more
than 4 yr; disease are does not occur until or a variable period of
time after the CD20-positive B cells return to the peripheral blood;
and the treatment was well tolerated in the majority of patients.
We did, however, note that one patient had a severe infusion
reaction and that another patient with very aggressive disease died
of pancarditis 5 months after being treated. B-cell repopulation
of the peripheral blood was already detectable before she died.
In this small, heterogeneous group of patients, we have not been
able to detect clear differences between the different protocols.
We would advise against using a smaller dose of rituximab
(2
500 mg) as this might predispose to the development of
HACAs. In our experience, the period of B-cell depletion in the
peripheral blood in patients with SLE is frequently shorter than
in patients with rheumatoid arthritis treated with the same
protocol or than what has been reported for patients with
lymphoma. It is likely that these differences are accounted for by
variation in the pharmacokinetics and effectiveness of the
mechanisms of depletion of CD20-positive B cells by rituximab.
The mean daily dose of prednisolone before treatment was
13.8 mg (S.D. 11.3) and for those completing 6 months of follow-up
(n 21) it had fallen to 10 mg (S.D. 3.1). Oral prednisolone was only
decreased slowly and progressively once clinical improvement was
clear. Patients who remained well after 6 months continued to
decrease the dose of prednisolone further.
The serological data are also of great interest. We have shown
in this report that antibodies to dsDNA and serum C3 levels are
signicantly improved following B-cell depletion. This improvement matches the global and individual BILAG organ/system
scores. Looney et al. [14] also reported a signicant improvement
in the global disease activity of 11 out of 17 of the patients
they treated (using the SLAM activity index) in the absence of a
signicant change in anti-dsDNA antibody or complement levels.
In the same trial these authors report a signicant decrease in antidsDNA antibody levels in four out of eight patients in whom
effective B-cell depletion was obtained and in whom elevated

1545