Dr. Pennec is deceased.
Improvement of Sjögren's syndrome after two infusions of rituximab (anti-CD20)
Article first published online: 28 FEB 2007
Copyright © 2007 by the American College of Rheumatology
Arthritis Care & Research
Volume 57, Issue 2, pages 310–317, 15 March 2007
How to Cite
Devauchelle-Pensec, V., Pennec, Y., Morvan, J., Pers, J.-O., Daridon, C., Jousse-Joulin, S., Roudaut, A., Jamin, C., Renaudineau, Y., Roué, I. Q., Cochener, B., Youinou, P. and Saraux, A. (2007), Improvement of Sjögren's syndrome after two infusions of rituximab (anti-CD20). Arthritis & Rheumatism, 57: 310–317. doi: 10.1002/art.22536
- Issue published online: 28 FEB 2007
- Article first published online: 28 FEB 2007
- Manuscript Accepted: 7 JUN 2006
- Manuscript Received: 11 JAN 2006
- Brest University Medical School Hospital
- 2003 Clinical Research Hospital Program (PHRC 2003)
- Primary Sjögren's syndrome;
- B cells
There is evidence to support a dominant role for B cells in the pathophysiology of primary Sjögren's syndrome (SS). Therefore, we evaluated the safety and efficacy of anti-CD20 monoclonal antibody.
Sixteen patients who met the new American-European Consensus Group criteria for primary SS and scored >50 on at least 2 of 4 visual analog scales (VAS; 100 mm) evaluating global disease, pain, fatigue, and global dryness received infusions of low-dose rituximab (375 mg/m2) at weeks 0 and 1 without steroid premedication.
Slow rituximab infusions (100 mg/hour) were well tolerated, with only 1 patient experiencing serum sickness–like disease. There was a dramatic reduction in B cells of the blood and salivary gland (SG). At week 12, VAS scores with respect to fatigue and dryness (P < 0.05), tender point count (P < 0.035), and quality of life as evaluated by the Short Form 36 questionnaire (SF-36; P < 0.001) were significantly improved. At week 36, significant improvements were noted in the 4 VAS scores (P < 0.05), tender joint count (P = 0.017), tender point count (P = 0.027), and SF-36 (P < 0.03). Pulmonary manifestations were ameliorated in 1 patient. Patients with improvements on at least 3 of the 4 VAS scores at any visit (n = 11) had a shorter disease duration than the other patients (n = 5; mean ± SD duration 3.8 ± 5.4 versus 30.1 ± 29.5 years; P = 0.02).
Low-dose rituximab infusions were well tolerated without the benefit of steroids. Infusions induced a rapid depletion of B cells in the blood and SG and could improve primary SS. Controlled studies are needed.
Primary Sjögren's syndrome (SS) is an autoimmune disorder affecting 0.2–3% of the general population (1). Its hallmark is chronic inflammation of the salivary and lacrimal glands, where lymphocytic infiltrates result in tissue destruction and cause loss of secretory function. Systemic manifestations are common, and B cell lymphoma occurs in nearly 5% of these patients (2–4). Pharmacologic treatments have the capacity to ameliorate the sicca symptoms, often transiently, but not to modify the course of the disease.
Although the pathogenesis of primary SS remains unclear, T cells as well as B cells have been shown to be involved. There is evidence for a critical role of B cells (5–9): antibodies to self antigens are produced systemically, and accumulation of memory-type B cells in the inflamed parotid glands leads to disturbed distribution of the B cell subsets in the peripheral blood.
CD20 is a 33–36-kd transmembrane phosphoprotein that is expressed on the surface of pre–B cells and mature B cells (10), but is absent from stem cells, normal antibody-secreting plasma cells, and plasmablasts. Rituximab is a chimeric anti-CD20 monoclonal antibody (mAb) composed of the constant region of human IgG1κ immunoglobulin and of murine variable regions. Rituximab was first developed for the treatment of B cell lymphoma (11–15) and subsequently has been applied to various diseases (16–22). Edwards et al claimed remarkable efficacy of rituximab in rheumatoid arthritis (RA) in an open-label study (23) and then confirmed this in a controlled study (24). Open-label trials of systemic lupus erythematosus (SLE) demonstrated that rituximab induced significant B cell depletion and clinical remission of lupus nephritis (25–27). In early studies, the treatment regimen was similar to that used for lymphoma, i.e., full-dose rituximab for 4 weeks in combination with high-dose corticosteroid therapy. However, studies aimed at determining the optimal dose in autoimmune disorders have never been conducted. Furthermore, the concomitant use of high-dose corticosteroid therapy may induce a bias in the evaluation of clinical and biologic effects and has not been validated in patients with autoimmune disorders.
A recent open-label study of primary SS demonstrated that full-dose rituximab was effective (28). However, that study can be criticized for 2 main reasons. First, a high dose of corticosteroids was associated with rituximab. Second, the study group included heterogeneous patients with either mucosa-associated lymphoid tissue (MALT)–type lymphoma or recent-onset primary SS without lymphoma. Therefore, the documented efficacy of rituximab on lymphoma acted as a confounding factor.
In a recent study of patients with RA, 2 infusions of 500 mg or 1,000 mg of rituximab produced similar results (29). Importantly, neither high nor low doses of corticosteroids influenced tolerance. Moreover, routine corticosteroid use is not recommended in combination with other chimeric mAb, such as infliximab.
These considerations led us to conduct an open-label pilot study of rituximab in patients with primary SS receiving rituximab without corticosteroids in a dosage of 2 weekly infusions of 375 mg/m2 each. Our objective was to determine whether rituximab was safe in primary SS, whether low-dose rituximab depleted circulating CD20 B cells, and whether this treatment induced sustained clinical, laboratory, and histologic improvements.
PATIENTS AND METHODS
Sixteen patients were enrolled in 2 departments at the University Hospital in Brest (France) between April and December 2004. Patients were eligible if they fulfilled the new American-European Consensus Group criteria for primary SS (30) and had active disease as assessed by values >50 mm on 2 of 4 visual analog scales (VAS; 0–100 mm) for global disease activity (including extraglandular manifestations), pain, sicca symptom, and fatigue, respectively, over the last week. Additional inclusion criteria were as follows: age 18–50 years, stable doses of nonsteroidal antiinflammatory drugs, and no prescription of immunosuppressive agents for the last 4 weeks. Patients were excluded if they had secondary SS; if they had received cytotoxic drugs within the last 4 months; if they had severe renal or hematologic failure, a history of cancer, hepatitis B or C, human immunodeficiency virus, tuberculosis, severe diabetes, or any other chronic disease or evidence of infection; or if they were unable to understand the protocol. The study design was approved by our institutional review board and by the Brest Ethics Committee. All patients gave their written informed consent.
All patients received 375 mg/m2 of rituximab at weeks 0 and 1 by intravenous infusion. No concomitant corticosteroid or immunosuppressive treatment was administered. The initial infusion rate of 50 mg/hour was increased by 50 mg/hour every 30 minutes to a maximum of 400 mg/hour if the drug was well tolerated. Symptoms of hypersensitivity or infusion-related reactions were sought every 30 minutes. Stable symptomatic treatment of oral and ocular discomfort was provided during the followup. This protocol was modified after the first 2 patients experienced side effects due to the infusion rate of 200 mg/hour. At that point, patients received an increasing rate of rituximab infusions up to a maximum rate of 100 mg/hour.
The primary end points were the evaluation of the safety and biologic effects of rituximab infusion at low a dose, without corticosteroid premedication. We studied B cell depletion in the peripheral blood and labial salivary gland (LSG). From baseline to week 36, we studied the variations of serum IgG, IgA, and IgM; T cell counts; rheumatoid factor (RF); antinuclear antibody (ANA); C-reactive protein (CRP) level; and erythrocyte sedimentation rate (ESR).
Secondary end points were exploratory in nature. These included the variations from baseline to week 36 of 4 clinical VAS scores: global disease activity, pain, sicca syndrome, and fatigue. We also evaluated the variations of 0–100-mm VAS scores for dry mouth, dry eyes, dry trachea, dry vagina, and dry skin; tender and swollen joint counts; tender points; unstimulated salivary flow rate; Schirmer's test; and Van Bijsterveld scores.
All patients underwent standardized evaluations including VAS score (31), physical examination, and routine laboratory tests at baseline and at weeks 1, 2, 4, 8, 12, 16, 20, 24, 28, 32, and 36. Tender points (n = 18), tender and swollen joint counts (interphalangeal, metacarpophalangeal, wrist, elbow, shoulder, hip, knee, ankle, and metatarsophalangeal joints; n = 42), and quality of life (evaluated using the Short Form 36 [SF-36] &lsqbr;32&rsqbr;) were recorded at weeks 0, 12, 24, and 36. Xerostomia was evaluated using the unstimulated salivary flow rate (5 minutes; samples were weighed and 1 gram corresponded to 1 ml) at 8:00 AM on an empty stomach. Ocular dryness was evaluated by an ophthalmologist (BC) using Schirmer's test (33) and the Van Bijsterveld score (34). These tests were carried out at weeks 0, 12, and 36. LSG biopsies were performed at weeks 0 and 12. Only 1 patient was taking sialagogue at inclusion and continued this treatment during the study, except the morning of visits.
All patients underwent laboratory tests at each visit. Standard tests included blood and urine parameters; ESR, CRP level, and blood cell counts; renal and liver function tests; creatine phosphokinase; serum levels of IgG, IgA, and IgM by nephelometry; and quantification of serum gamma globulins by serum protein electrophoresis. Immunologic tests were repeated at weeks 0, 12, and 36. These included latex tests and isotype-specific in-house enzyme-linked immunosorbent assays (ELISAs) for IgM-RF, IgG-RF, and IgA-RF; tests for anti–citrullinated peptide antibody using commercial kits, ANAs detected by an in-house indirect immunofluorescence test with HEp-2 cells as the substrate, and titers of anti–double-stranded DNA (anti-dsDNA) determined by in-house ELISA and by indirect immunofluorescence with Crithidia luciliae as the substrate; and IgA-, IgG-, and IgM-containing circulating immune complexes (CICs) using a method previously described (35).
B and T lymphocyte subpopulations were analyzed on an Epics-XL flow cytometer (Beckman Coulter, Hialeh, FL) with EDTA-treated blood. All mAb were from Beckman Coulter, unless specified otherwise. After red cell analysis using the Multi-Q-Prep system (Beckman Coulter), B cells were identified based on CD19 expression. In all patients, CD19+ cells were assessed at baseline and once a month throughout the study. The absolute number of cells was calculated from the size of the lymphocyte subpopulation and the leukocyte count. T cells were identified using fluorescein isothiocyanate (FITC) anti-CD3, phycoerythrin (PE) anti-CD4, and PE–cyanin 5 anti-CD8 monoclonal antibodies, and natural killer (NK) cells were identified using CD56 PE.
Salivary gland analysis.
Two LSG biopsy samples were obtained from all patients at inclusion and week 12. These samples were divided into 2 parts, 1 for anatomopathologic evaluation and 1 for immunofluorescence analysis.
For anatomopathologic evaluation, the tissues were examined by a blinded trained anatomopathologist (IQR). To obtain continuous data before and after treatment, we combined the Chisholm and Mason score (grades 1–4) (36) and the focus score, which describe the number of foci (number of infiltrates) >50 mononuclear cells/4 mm2 (grades 1–12) (37).
For immunofluorescence analysis, fresh LSG biopsy specimens from all patients before (week 0) and 3 months after (week 12) rituximab infusion were incubated overnight at 4°C in phosphate buffered saline (PBS) containing 15% sucrose (Sigma, Saint Louis, MO). Tissue sections were mounted onto poly-L-lysine–coated slides and were incubated at room temperature for 40 minutes with 1:50 goat anti-CD20 antibody (Neomarker, Fremont, CA) and 1:50 anti-CD3 mAb. After extensive washing in PBS, the slides were incubated for 40 minutes at room temperature with 1:200 FITC-conjugated donkey anti-mouse antibody (Jackson ImmunoResearch, West Grove, PA) and with 1:200 lissamine rhodamine–conjugated donkey anti-goat antibody (Jackson ImmunoResearch) in PBS supplemented with 2% donkey serum. The sections were fixed with cold paraformaldehyde at 4%. After several rinses in PBS, sections were coverslipped with Vectashield (Vector, Burlingame, CA) before analysis using a confocal microscope (Leica, Westlar, Germany). For each experiment, controls included the removal of primary antibody, resulting in negative staining.
SPSS software, version 12.0 for Windows (SPSS, Chicago, IL) was used. Data are reported as the mean ± SD. Visits (week 12 versus week 0 and week 36 versus week 0) were compared using Wilcoxon's matched pairs signed rank test for paired data. P values less than 0.05 were considered significant.
Sixteen patients with active disease were included in the study. There were 2 men and 14 women with a mean ± SD age of 54.8 ± 12.5 years. The mean ± SD disease duration was 13.3 ± 10.3 years. All patients had positive ANA, 13 of 16 had anti-SSA antibodies, and 7 of 16 had anti-SSB antibodies. All patients received the 2 infusions. Only 1 patient dropped out of the study at week 24 due to lymphoma. The other 15 patients completed the followup visits through week 36.
During their initial rituximab treatment, the first 2 patients developed mild to moderate infusion-related reactions after the rituximab infusion rate was raised to 200 mg/hour. The symptoms consisted of fever, headache, fatigue, flush, and pruritus. These symptoms resolved without treatment after the infusion rate was decreased to 100 mg/hour. The subsequent infusions were administered at rates no higher than 100 mg/hour for these 2 patients and for all following patients. These infusions were then well tolerated. Only very moderate hypersensitivity reactions were noted in another 8 patients, who reported transient headache or fatigue during the first infusion. These adverse events were considered negligible and rapidly resolved during the infusions without any specific medication and/or reduction of the infusion rate. No adverse events were noted during the second infusion.
Arthritis developed in 4 patients after rituximab treatment; of these 4 patients, 2 had experienced arthritis within the 6 months before study inclusion. Ten days after the last infusion, 1 of these 2 patients displayed ankle arthritis with joint swelling and urticaria of the legs. Such symptoms were considered common by the patient, resolved spontaneously, and never reappeared during the followup. The second of the 2 patients had ankle arthralgia at inclusion. This patient developed synovitis and purpura 2 months after infusions, and finally developed a lymphoma. Two other patients experienced ankle arthralgia 7 days after the last infusion. In 1 of these patients, the symptoms with joint swelling required hospital admission, but they resolved with corticosteroid treatment (20 mg/orally for 2 days). No infections or other adverse events were detected. There were no changes in hepatic, renal, or hematologic tests.
All patients completed therapy, but 1 patient did not attend followup visits through week 36. This patient had monoclonal gammopathy with type I cryoglobulinemia, arthritis, and purpura, but no evidence of lymphoma before inclusion. The patient had already experienced severe renal failure 3 months before inclusion, which resolved after a high dose of corticosteroid. Serologic tests and LSG biopsy samples were unable to prove lymphoma 3 months before inclusion. It is important to note that rituximab therapy failed to prevent progression to renal failure. Two months after the last infusion, the patient was hospitalized. Then lymphoplasmacytic lymphoma was confirmed by the presence of polyadenopathy, serum monoclonal IgM, and monoclonal proliferation in LSG but not in the bone marrow. CIC levels in sera were unchanged. Hospitalization was warranted due to a flare, i.e., due to an unsatisfactory efficacy rather than a complication of the treatment. This event was a serious adverse event. Our interpretation was that this patient had a lymphoplasmacytic lymphoma before inclusion. This lymphoma did not respond to a low dose of rituximab alone. The patient was, therefore, treated by fludarabine, mitoxantrone, and 4 infusions of rituximab. These infusions were well tolerated, and the patient has been in remission since then.
Efficacy on subjective parameters.
At week 12, i.e., at the first evaluation time point, subjective variables including fatigue and dryness were significantly improved (P < 0.05). Although there were trends toward improvements with respect to ocular and oral dryness evaluated independently, these differences did not reach significance. Seven of 16 patients at week 24, 7 of 15 at week 32, and 9 of 15 at week 36 improved their 4 VAS scores (global, pain, fatigue, and dryness). Those who improved at least 3 of the 4 VAS scores at any visit (n = 11) had a shorter disease duration than the remainder of the patients (n = 5; mean ± SD disease duration 3.8 ± 5.4 versus 30.1 ± 29.5 years; P = 0.02). Interestingly, at week 36, significant improvements were noted for 7 of 9 VAS scores: global disease, fatigue, dryness, pain, dry mouth, dry skin, and dry trachea (Figure 1).
Between week 0 and week 12, the tender point count changed significantly (P = 0.035). At week 36, significant improvements were noted in both the tender point count (P = 0.027) and the tender joint count (P = 0.017).
Quality of life evaluated using the generic SF-36 questionnaire (mental and physical components) was significantly ameliorated (P < 0.001 at weeks 12 and 24; P = 0.03 at week 36). At week 24, 7 patients had a 50% improvement in the SF-36 score.
Efficacy on objective parameters.
There were no significant changes (Table 1) in the unstimulated salivary flow rate; the salivary gland focus score; the ophthalmologic evaluation; the titers of ANA, anti-SSA, and anti-SSB; complement levels; or CICs and levels of serum IgG and IgA (data not shown). The IgA-RF titers were significantly decreased at week 36 (P = 0.04). IgM levels were lower at week 12 than at week 0 (P < 0.01).
|Week 0||Week 12||P Weeks 0–12||Week 24||P Weeks 0–24||Week 36||P Weeks 0–36|
|Global disease VAS, mm||71.9 ± 13.4||64.3 ± 23.2||0.53||56.4 ± 26.6||0.033†||55 ± 30||0.03†|
|Pain VAS, mm||60.4 ± 19.2||48.8 ± 28.3||0.19||40.6 ± 25.5||0.02†||33 ± 28||0.006†|
|Fatigue VAS, mm||77.3 ± 14.5||56.9 ± 28.4||0.001†||54.6 ± 30.2||0.005†||58 ± 31||0.006†|
|Dryness VAS, mm||8.4 ± 11.9||57.6 ± 29.3||0.036†||52.4 ± 32.5||0.005†||53 ± 30||0.006†|
|Tender point count||4.5 ± 6||2.0 ± 5.2||0.035†||1.0 ± 1.5||0.049†||2.0 ± 4.7||0.027†|
|Tender joint count||6.0 ± 10.9||4.8 ± 11.7||0.175||1.8 ± 4.8||0.16||2.7 ± 9.8||0.017†|
|Swollen joint count||1.0 ± 0.9||0.3 ± 0.6||0.94||0 ± 0||0.16||0 ± 0||0.15|
|Salivary flow rate, ml/minute||0.1 ± 0.1||0.1 ± 0.1||0.35||0.18 ± 0.26||0.35||0.12 ± 0.1||0.86|
|Schirmer test, mm||8.4 ± 8.3||8.1 ± 8.86||0.95||7.1 ± 7.5||0.63||9.0 ± 7.8||0.79|
|Chisholm score ≥3, no./total no.||9/15||5/15||–||NA||–||NA||–|
|Focus score||1.6 ± 1.7||1.4 ± 1.8||0.9||NA||–||NA||–|
|Anti-SSA||80 ± 41||76 ± 34||0.4||75 ± 34||0.06||70.5 ± 35.0||0.25|
|ESR, mm/hour||30.1 ± 26.0||26 ± 18||0.12||27.7 ± 24.1||0.4||29 ± 25||0.6|
|Latex test||14.8 ± 21.7||22.9 ± 43.8||0.5||14.6 ± 22.6||0.1||11.7 ± 18.7||0.1|
|IgA-RF, IU||0.3 ± 0.3||0.3 ± 0.3||1||0.3 ± 0.3||0.5||0.2 ± 0.2||0.04†|
|IgA, mg/liter||2.7 ± 1.2||2.83 ± 1.0||1||2.6 ± 1.0||0.6||2.9 ± 1.2||0.7|
|IgG, mg/liter||20.1 ± 13.2||20.2 ± 8.8||0.33||20.7 ± 12.4||0.2||24.4 ± 17.0||0.2|
|IgM, mg/liter||1.41 ± 0.57||1.07 ± 0.55||0.011†||1.13 ± 0.8||0.1||1.07 ± 0.7||0.2|
Efficacy on pulmonary manifestations.
One patient had a severe persistent dry cough with evidence of bronchiolitis and interstitial pneumonia by chest radiography and computed tomography. Pulmonary function test results demonstrated a severe restrictive pattern, with a diffusing capacity for carbon monoxide (DLCO) <40% due to decreased permeability of alveolocapillary membrane (DLCO/alveolar volume decreased by 0.8 mmol/min/kd/liter [mean ± SD normal 1.5 ± 0.2]) and of alveolocapillary membrane exchange surface (alveolar volume and total lung capacity [TLC] decreased by 78%).
The forced expiratory flow (FEF) at the 25% and 75% of the pulmonary volume revealed a moderate obstruction of small airways. After rituximab therapy, the cough resolved rapidly. At week 12, the computed tomography abnormalities were markedly amended (Figure 2) and the pulmonary test results were slightly rectified with a TLC at 85%.
B cell depletion and regeneration.
B cells (CD19), T cells (CD3+), T helper cells (CD3+, CD4+) cytotoxic T cells (CD3+, CD8+), and NK cells (CD56+) were numerated in peripheral blood at weeks 0, 1, 2, 12, 16, 20, 24, 28, 32, and 36. B cells diminished rapidly within the first week. After 12 weeks, depletion of peripheral B cells was complete in most of the patients. The counts of NK cells, T helper cells, and cytotoxic T cells remained unchanged (Figure 3). The rate of B cell regeneration varied across the patient group (Figure 4). At week 20, B cells were detectable in most patients, but regeneration was usually not complete until week 36. One patient experienced early regeneration, starting at week 8. This patient was unresponsive to rituximab and received a diagnosis of lymphoma during the study. Immunohistochemistry of LSG biopsy samples obtained at week 12 showed complete B cell depletion in all patients. However, only 6 patients had a significant B cell infiltration of their LSGs at week 0, as illustrated in Figure 5.
The 2 main goals of this open-label study were to determine the safety and the biologic effects of rituximab-induced B cell inhibition in 16 patients with active primary SS. In previous studies using rituximab to treat autoimmune diseases such as RA or SLE, patients received rituximab in combination with other immunosuppressants and high-dose corticosteroids (22–26). In contrast, rituximab was used alone in the present study, and all patients received 2 infusions of 375 mg/m2 each. This treatment induced rapid and complete B cell depletion in the peripheral blood and LSG, suggesting that a full effect may be obtained with this dose. Rituximab used alone at low infusion rates of 100 mg/hour was well tolerated, although fatigue and headache were common during the first infusion. Adverse events did not occur during the second infusion. Our first 2 patients experienced infusion-related reactions when the infusion rate was increased to 200 mg/hour, but these side effects improved when the rate was decreased to 100 mg/hour. For all subsequent patients, the infusion rate was slowed down to 100 mg/hour. Thereafter, infusions were well tolerated except for slight transient headache or fatigue. None of our patients had severe side effects. In most studies of autoimmune diseases, the incidence of adverse events was low compared with that noted in patients with lymphoma. Patients with autoimmune disease are probably at a lower risk for cytokine release syndrome caused by B cell lysis than those with lymphoma. One of our patients required hospital admission and low-dose corticosteroid treatment for arthralgia and purpura occurring 7 days after the last infusion, in keeping with earlier reports. This patient had extraglandular involvement with pulmonary manifestations. Another patient developed a lymphoma with renal failure. However, because this patient previously had similar symptoms, this was not regarded as being related to rituximab. Rituximab could induce renal abnormalities due to immune complex glomerulonephritis in a serum sickness reaction. There were no changes in the level of CICs in this patient, but renal biopsy was not performed. In the literature, rituximab has been successfully used in patients with lymphoma with renal failure (38) and in patients with lupus glomerulonephritis (39).
Unfortunately, validated criteria for evaluating treatment efficacy in patients with primary SS are not available. It is difficult to select a single criterion, because the predominant clinical manifestation of primary SS may be glandular (dryness) or extraglandular (fatigue, joint pain, nonerosive synovitis, secondary fibromyalgia, myalgia, lung involvement, cryoglobulinemia, or other). VAS scales have been validated for pain, fatigue, or sicca syndrome evaluation (40–42) and have been used in previous studies of SS (31) so that they were regarded as end points. Nevertheless, we also evaluated all validated objective tests for sicca syndrome and extra glandular manifestations.
We also demonstrated that anti-CD20 can deplete B cells in LSGs. All of our patients with B cell infiltration at inclusion had a complete depletion of B cells after treatment with rituximab. This is the first study of anti-CD20 efficacy in the salivary gland. The effect described by Pijpe et al was noted in 1 patient with MALT lymphoma (43). However, no increases in LSG function were noticed, but longer followup or treatment at an early stage are probably necessary.
Several clinical parameters improved significantly in our study, including fatigue and dryness (P < 0.05) at week 12. Efficacy was greatest at week 36, when significant improvements occurred for most of the VAS scores. However, objective parameters did not change. Many factors may have contributed to this finding. First, some patients did not have objective sicca syndrome at inclusion. Second, our patients had longstanding and severe disease: all patients had a salivary flow rate <0.5 ml/minute, and 10 of 16 patients had a Schirmer test result <6 mm. Conceivably, treatment may lack efficacy in longstanding disease. Third, all test results showed wide variability. Nevertheless, a lack of objective efficacy on dryness remains possible.
In contrast, we found improvements in tender point and tender joint counts from baseline to weeks 12 and 36. These satisfactory effects may have contributed to the improvement in SF-36 score. Although tender point and tender joint counts are not the hallmarks of primary SS, 7 patients were recruited from the rheumatology department and most patients with primary SS who are referred to rheumatologists have either tender points or tender joints.
In our study, various autoantibody titers were evaluated. Only IgA-RF decreased with rituximab therapy. In patients with RA, B cell depletion using rituximab in combination with cyclophosphamide or prednisolone induced a decrease in titers of RF and anti–cyclic citrullinated peptide antibodies. In addition, clinical relapse seemed to be more closely associated with autoantibody titer elevation than with B cell count elevation (31). In patients with SLE, rituximab induced significant B cell depletion, but changes in anti-dsDNA antibody titers and complement varied across studies and across patients (25, 27). In our study, anti-dsDNA antibody and IgM-RF titers were unchanged after B cell depletion, in keeping with the long lifespan of antigen-specific plasma cells that do not express anti-CD20. Cell turnover is not required to ensure prolonged antibody production (44).
Interestingly, rituximab therapy improved systemic manifestations of primary SS in some of our patients. One patient had a severe cough and computed tomography evidence of interstitial pneumonitis, which resolved completely after rituximab therapy. The cough disappeared and the controlled computed tomography was normal at week 12. Another patient had renal tubulopathy but normal creatinine levels. After the rituximab infusions, this patient remained free of joint manifestations until week 36. A third patient had monoclonal proliferation within the LSGs and cryoglobulinemia but no evidence of lymphoma at baseline. Renal failure and lymphoma developed despite rituximab therapy.
Our results clearly demonstrate that low-dose rituximab induced acute and complete CD20 cell depletion in the peripheral blood and LSGs. Low-dose rituximab infusions were well tolerated, although corticosteroid therapy was not administered. Our results suggest that rituximab could improve both subjective glandular manifestations and extraglandular signs. Controlled studies are needed and should include patients with active disease of less than 10 years' duration and/or extraglandular manifestations and must compare low- and high-dose rituximab versus placebo at week 24. This study may help to determine the sample size for future trials. Quality of life evaluated using the generic SF-36 questionnaire may be a valuable criterion for evaluating this pleomorphic disease.
Dr. Saraux had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study design. Devauchelle-Pensec, Renaudineau, Youinou, Saraux.
Acquisition of data. Devauchelle-Pensec, Morvan, Pers, Daridon, Jousse-Joulin, Roudaut, Jamin, Renaudineau, Quintin Roué, Cochener, Saraux.
Analysis and interpretation of data. Devauchelle-Pensec, Pers, Daridon, Jousse-Joulin, Jamin, Renaudineau, Youinou, Saraux.
Manuscript preparation. Devauchelle-Pensec, Saraux.
Statistical analysis. Saraux.
- 29The efficacy and safety of rituximab in patients with active rheumatoid arthritis despite methotrexate treatment: results of a phase IIb, double blind, placebo-controlled, dose-ranging trial. Arthritis Rheum 2006; 54: 1390–1400., , , , , , et al.