Objectives To evaluate the efficacy and safety of rituximab in patients with primary Sjögren's syndrome (pSS).
Methods The AutoImmune and Rituximab registry has included 86 patients with pSS treated with rituximab, prospectivey followed up every 6 months for 5 years.
Results Seventy-eight patients with pSS (11 men, 67 women), who already had at least one follow-up visit, were analysed. Median age was 59.8 years (29–83), median duration of disease was 11.9 years (3–32). Indications for treatment were systemic involvement for 74 patients and only severe glandular involvement in four patients. The median European Sjögren's Syndrome disease activity index (ESSDAI) was 11 (2–31). 17 patients were concomitantly treated with another immunosuppressant agent. Median follow-up was 34.9 months (6–81.4) (226 patient-years). Overall efficacy according to the treating physician was observed in 47 patients (60%) after the first cycle of rituximab. Median ESSDAI decreased from 11 (2–31) to 7.5 (0–26) (p<0.0001). Median dosage of corticosteroid decreased from 17.6 mg/day (3–60) to 10.8 mg/day (p=0.1). Forty-one patients were retreated with rituximab. Four infusion reactions and one delayed serum sickness-like disease resulted in rituximab discontinuation. Three serious infections (1.3/100 patient-years) and two cancer-related deaths occurred.
Conclusions In common practice, the use of rituximab in pSS is mostly restricted to patients with systemic involvement. This prospective study shows good efficacy and tolerance of rituximab in patients with pSS and systemic involvement.
- Sjøgren's Syndrome
- Autoimmune Diseases
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Primary Sjögren's syndrome (pSS) is a systemic autoimmune disease that is characterised in the majority of patients by dryness, fatigue and pain. Systemic manifestations occur in 20–40% of patients.1 ,2
B lymphocytes play a crucial role in pathogenesis of the disease by their capacity to secrete autoantibodies and cytokines or present autoantigens.3 ,4 Evidence-based therapy for pSS is largely limited to symptomatic treatments that improve dryness.3
Some efficacy of rituximab, an anti-CD20 monoclonal antibody, was suggested in case reports, retrospective or open studies and in two controlled trials.5 ,6 However, only very few of the reported patients had been treated with rituximab for systemic complications other than lymphoma. Rituximab is currently not labelled for pSS and no information is available concerning its use in common practice.
We therefore evaluated the indications, efficacy and safety of rituximab in pSS, using the prospective data of the AutoImmune and Rituximab (AIR) registry.
Patients and methods
AIR is an independent multicentre registry promoted by the French Society of Rheumatology and the Club Rhumatismes et Inflammation, which includes patients treated with rituximab for refractory autoimmune diseases, including rheumatoid arthritis (RA), pSS according to the European American consensus group criteria, systemic lupus erythematosus (SLE), myositis, vasculitis and other refractory autoimmune diseases. A total of 3662 patients has been included and prospectively followed up.7
Data were collected at baseline, 3 and 6-month follow-up and then every 6 months. This study was approved by the French authorities (Comité Consultatif sur le Traitement de l'information en matière de Recherche dans le domaine de la Santé and Commission Nationale de l'Informatique et des Libertés). Written informed consent was obtained from all patients.
The only parameter that was not prospectively collected by each investigator was the European Sjögren's syndrome disease activity index (ESSDAI),4 which did not exist when the electronic case report form (e-CRF) was created, and was retrospectively calculated before and after rituximab. The ESSDAI is a consensus clinical index set up by a European League Against Rheumatism task force and designed to measure systemic disease activity in patients with pSS through 12 different organ-specific domains of involvement.
Efficacy was assessed 6 months after the first cycle of rituximab according to the global opinion of the physician. Each systemic complication was evaluated by objective tests if possible. A summary of each chart (adverse events, efficacy) was validated by the treating physician. Serious adverse events were validated using chart copies by the two coordinators of the registry (XM and JEG).
Data were presented as medians (ranges) for continuous variables and numbers (percentages) for qualitative variables. The non-parametric Mann–Whitney U test was used for continuous variables and Fisher's exact test was used to compare qualitative variables. Statistical analyses were realised using GraphPad Prism software. A p value less than 0.05 was considered statistically significant.
Among the 86 patients with pSS enrolled in the AIR registry, 78, who already had at least one follow-up visit at 6 months, were analysed.
The median age was 59.8 years (29–83), median duration of pSS was 11.9 years (3–32). Twenty-four patients (30.8%) had no autoantibody, 54 (69.2%) were anti-SSA positive and 20 (25.6%) were anti-SSA and anti-SSB positive.
Previous treatments included corticosteroids in 66 patients, hydroxychloroquine in 44, methotrexate in 33, azathioprine in 19, mycophénolate mofetil in four, leflunomide in eight, intravenous immunoglobulins and plasmatic exchanges in three patients each. The main baseline characteristics of patients are detailed in table 1.
Indications for rituximab
Seventy-four patients (95%) were prescribed rituximab for systemic involvement (32 patients for one main systemic complication, 42 patients for multiple systemic complications) (table 2). Twenty-seven patients predominantly had articular involvement (only arthralgias in eight, synovitis in 19), six had pSS-related central nervous system (CNS) involvement with brain diffuse white matter MRI abnormalities. Central neurological features were progressive multiple sclerosis-like manifestations (n=4), transverse myelitis (n=1), anxiety and depression disorder (n=1) evidenced by neuropsychological testing. Twelve patients had peripheral nervous system (PNS) involvement defined by the presence of clinical signs of peripheral neuropathy confirmed by electroneuromyography. Eight patients had pSS-related vasculitis: five cryoglobulinaemias with purpura and articular, PNS, renal involvement and three biopsy-confirmed small vessel lymphoid vasculitis without any evidence of cryoglobulinaemia. Patients with neurological involvement or cryoglobulinaemia were previously reported (table 3). Nine patients had pSS-related pulmonary involvement: one bronchial involvement with altered respiratory function tests, eight interstitial lung diseases (lung CT abnormalities in all patients) including one with lymphocytic alveolitis evidenced by broncho-alveolar lavage). Six patients (8%) had renal involvement (one tubular acidosis and five interstitial biopsy-confirmed interstitial nephritis). Three patients had myositis (elevated creatine-phosphokinase approximately twofold over the upper limit of normal, myogen features at electroneuromyography and histological myositis). Two patients had autoimmune cytopaenias: one autoimmune anaemia (haemoglobin concentration of 8 g/dl with positive Coombs test) and one autoimmune thrombocytopaenia (platelet count: 30 giga/l). One patient had pSS-related autoimmune pancreatitis (serum lipase threefold over the lower limit of normal) despite corticosteroids and had normal serum levels of IgG4.
Four patients with only severe glandular involvement were prescribed rituximab: two with parotid swelling and one with lachrymal gland hypertrophy, previously refractory to corticosteroids, and one sclera vasculitis and corneal perforation.
Rituximab regimen and comedications
The rituximab regimen is indicated in table 1. All patients received standard premedication with antihistamine, paracetamol and 100 mg methylprednisolone. Sixty-seven patients were initially treated with two rituximab infusions of 1 g, 11 patients with four infusions of 375 mg/m2. Corticosteroids were associated with rituximab in 29 patients with a median dosage of 17.6 mg/day (5–60). Other concomitant immunosuppressant agents included methotrexate in seven patients, hydroxychloroquine in seven, azathioprine, mycophenolate mofetil and cyclophosphamide in one patient each. Sixty-one patients (78%) received rituximab without any concomitant immunosuppressant other than corticosteroids.
The median follow-up was 34.9 months (6–81.4) (226 patient-years). Overall efficacy 6 months after the first cycle after rituximab, assessed by the global opinion of the physician, was observed in 47 patients (60%).
No data were available regarding the evolution of dryness, pain and fatigue. Efficacy regarding each involvement is summarised in table 2. Efficacy was observed in 45/74 patients (61%) with systemic involvement. Efficacy was observed in two out of four patients (50%) with only glandular involvement (two parotid swellings). Parotid swelling was also improved in two out of four patients (50%) with parotid swelling associated with systemic involvement. Regarding articular involvement, the median tender and swollen joint count decreased from 9 to 0 (p=0.003) and 4 to 0 (p=0.004), respectively, among the 17 patients with available data. The two patients with autoimmune cytopaenia respectively normalised their haemoglobin level (from 8 to 11 g/dl) and their platelet count (from 30 to 153 giga/l). Among the patients with pulmonary involvement, the patient with bronchial involvement and six of the eight patients with interstitial disease responded to the first cycle of rituximab according to their physician. All the responder patients with interstitial lung disease were retreated (two to nine cycles). Median ESSDAI, which could be evaluated in 72 patients, decreased from 11 (2–31) before rituximab to 7.5 (0–26) 6 months after rituximab (p<0.0001). A decrease equal or greater than 20% and 30% of the initial ESSDAI was observed, respectively, in 29 (69%) and 23 (55%) of the 42 responders with systemic features and evaluated ESSDAI.
In the 29 patients initially treated with corticosteroids, the dose of corticosteroids 6 months after the first cycle of rituximab had increased in five, was stable in eight and decreased in 16 patients (discontinuation of corticosteroids in six and decrease in 10 patients) with the following involvements: articular, four; pulmonary, three; renal, one; PNS, two; CNS, one; vasculitis, two; myositis, one; autoimmune thrombocytopaenia, one; autoimmune pancreatitis, one. The median daily dose of prednisone decreased from 17.6 to 10.8 mg, 6 months after rituximab (p=0.1). No significant difference in efficacy was observed between patients without (65.2%) or with anti-SSA/SSB antibodies (54.5%, p=0.5), or in patients treated without (56.5%) or with concomitant immunosuppressant agents other than corticosteroids (62.5%, p=0.9).
Five (6.4%) serious infusion reactions occurred. Four of them were immediate infusion reactions (urticarial allergic reaction with shock (one patient, first infusion), Quincke oedema (two patients, first and second infusion), skin rash without fever or synovitis (one patient, second infusion of the third cycle) occurred. The latter was a delayed serum sickness-like disease (26 days after fourth infusion of 375 mg/m2) occurring in a patient treated with corticosteroids (10 mg/day) without any other immunosuppressant agent. Three serious infections occurred (1.3/100 patient-years): two bronchopulmonary infections (including one cytomegalovirus pneumopathy, 4 years after the initiation or rituximab and 6 months after the fifth cycle, associated with hypogammaglobulinaemia (4.3 g/l) at the time of infection; one severe Staphylococcus aureus pneumopathy treated in the intensive care unit, 4 years after the initiation of rituximab and 9 months after the second cycle, with normal levels of gammaglobulins at the time of infection; one multiple dental infection (4 months after the second infusion of the second cycle, with hypogammaglobulinaemia (5.6 g/l) at the time of infection).
Two cancers (0.9/100 patient-years) occurred: one squamous cell carcinoma of the vulva with hepatic and vertebral metastases (11 months after rituximab initiation, two cycles), one Paget's cancer of the nipple with multiple metastases (6 months after rituximab initiation, one cycle). Two deaths occurred and were related to these cancers.
Levels of gammaglobulins were assessed at the initiation of rituximab in 48 patients and 6 months after the first cycle in 22 patients. Hypogammaglobulinaemia (gammaglobulins < 6 g/l) was observed after the first cycle of rituximab in four out of 22 patients (two with myositis and two with cryoglobulinaemia). Before rituximab, one of these patients had hypogammaglobulinaemia, two had normal and one patient had increased levels of gammaglobulins, which fell from 20.3 to 5.3 g/l. Two patients with hypogammaglobulinaemia after rituximab developed a serious infection and were subsequently retreated with rituximab and concomitant intravenous immunoglobulins.
Forty-one patients were retreated with rituximab (two cycles: n=21; three cycles: n=8; four cycles: n=3; five to 12 cycles: n=9). The median number of rituximab cycles was 2.3 (1–12). The median duration between the first and second cycle was 11 months (5.4–29.4). The efficacy of rituximab according to the clinician was observed in 16 of 21 patients treated with two cycles, seven of eight patients treated with three cycles and 11 of 12 patients treated with four cycles or more. Eight patients (two CNS involvements, one with transverse myelitis and one with severe depressive and cognitive symptoms); two pulmonary; one renal; one articular; one autoimmune anaemia; one autoimmune pancreatitis) were persistent responders 30 months (16–58) after their first cycle of rituximab and were not retreated. After a median follow-up of 34.9 months (6–81.4), rituximab was discontinued in 41 patients, including 26 patients after the first cycle and 15 after more than one cycle (inefficacy: 35, serious adverse events: six patients, four after the first cycle, one after the second cycle and one after the fifth cycle).
Our study shows that the use of rituximab in pSS in common practice is mostly restricted to patients with systemic involvement. In patients with pSS and systemic involvement, rituximab had a good efficacy (60%) and satisfactory tolerance. The limitations of this study include its observational design, resulting in missing data and the definition of efficacy according to the opinion of the physician. It is regrettable that no patient data were collected on the evolution of dryness, fatigue and pain. This might also be related to the fact that rituximab was mainly prescribed for systemic complications.
The main messages of the present study therefore concern the possible indications for rituximab in pSS and its safety in this disease. Most of the patients treated with pSS previously reported were treated with rituximab either for lymphoma,17 ,18 or for glandular involvement.19 The lack of data on the efficacy of rituximab on systemic involvement other than lymphoma is related to the limited proportion of patients with severe systemic involvement compared to patients with dryness, pain and fatigue. Moreover, patients with severe systemic involvement were not included in the two controlled trials that compared rituximab with placebo. In contrast, the objectives of registries such as AIR are to collect data on efficacy and tolerance in real-life patients with refractory autoimmune diseases. In France, inclusions in AIR have thus contributed to allow the exceptional use of biological agents in SLE, pSS or vasculitis,16 ,20 ,21 thanks to ‘temporary therapeutic protocols’ provided by the French health agency. Therefore, the AIR registry allowed us to report the largest series of rituximab-treated patients with systemic complications of pSS (table 3).
Regarding tolerance, charts were systematically reviewed for serious adverse events, including infusion reactions, serious infections and cancers. The rate of serious infections was lower than observed in the same registry in RA or SLE.7 ,20 Four cases of serious immediate reactions to infusions occurred, and there was only one case of delayed serum sickness-like disease, an adverse event that was previously reported in eight patients with pSS (table 3).8 ,15 However, it is sometimes very difficult to differentiate serious immediate reactions from serum sickness disease. The main reason for discontinuing rituximab was inefficacy and not adverse events. Interestingly, even in patients with pSS and frequent hypergammaglobulinaemia, a decrease in gammaglobulin levels after rituximab can be observed and can be associated with the occurrence of serious infections. As in RA, it is therefore recommended to monitor levels of gammaglobulins before each cycle of rituximab.22 ,23
Efficacy after the first cycle was observed in 60% of patients. The majority of patients did not receive any other immunosuppressant agent than rituximab. The interpretation of the results has to take into account the observational design and the definition of efficacy according to the opinion of the physician. Of interest, in the majority of patients, it was possible to evaluate modifications in objective parameters such as the number of swollen joints, red or platelet blood count, proteinuria or glomerular filtration rate, for instance. Each clinician was asked to validate the data collected in the registry and to provide his assessment on the efficacy of rituximab. The importance of the clinician's assessment in heterogeneous and complex diseases was recently emphasised by the inclusion of the physician's global assessment in the SLE responder index.24 The decrease in the ESSDAI was another concordant appreciation of rituximab efficacy. A decrease of at least 20% in the ESSDAI was observed in two-thirds of the responders in whom the score had been assessed. A decrease in corticosteroid dosage was also observed. Forty-one patients (52.5%) received multiple cycles of rituximab, including 20 patients who were prescribed three cycles or more. Of course, the patients retreated with rituximab were those who responded to the first cycle and tolerated it well. This can explain their high rate of response to subsequent cycles. Interestingly, and in contrast to RA, eight patients had persistent response after a single cycle of rituximab. The main involvements that benefited from rituximab were joints, vasculitis, autoimmune cytopaenias, pulmonary and kidney involvement. The results were more disappointing in PNS and CNS involvements without any vasculitis, in which disease activity can be difficult to discriminate from damage. No efficacy was observed in the three patients with mild myositis, in contrast to positive results in other severe inflammatory myopathies.21 Finally, the present study also showed the efficacy of rituximab in severe parotid swelling, which is considered by many clinicians to be a systemic complication.
In conclusion, the present study showed good efficacy and tolerance of rituximab in patients with pSS and systemic involvement. Controlled trials are now required to confirm the therapeutic interest of rituximab in patients with pSS-related systemic complications.
The authors would like to thank the French Society of Rheumatology (SFR), the Club Rhumatismes et Inflammation, the French National Society of Internal Medicine (SNFMI) and Isabelle Pane (bioinformatician and data manager of the AIR registry) for their involvement in the AIR registry, Rosemary Jourdan (Roche), Jamila Filipecki and Nadine Mackenzie (formerly Roche). The authors also thank the 14 research study nurses (A Bourgeois, E Braychenko, F Carmet, MH Da Silva, S Delmas, D Guinement, R Lefebure, C Lehning, N Minot, FMA Ouattara, V Pinosa, M Reau, H Thibault and E Wallet), L Dongmo, V Martin, all working in Euraxipharma, the contract research organisation in charge of data collection in the AIR registry.
J-EG and GC contributed equally to this study.
Contributors JEG, GC and XM, designed the study, contributed to data collection, analysed the data and drafted the manuscript. PR is the methodologist of the registry. All the other authors contributed to data collection and analysed the data. JEG and CG contributed equally.
Funding Roche provided an unrestricted educational grant but was not involved in the design, protocol, data collection or statistical analysis of the study.
Competing interests None.
Ethics approval This study was approved by the Comité Consultatif sur le Traitement de l'information en matière de Recherche dans le domaine de la Santé and Commission Nationale de l'Informatique et des Libertés.
Patient consent Obtained.
Provenance and peer review Not commissioned; externally peer reviewed.