Objective To evaluate rituximab (RTX) in primary Sjögren's syndrome (pSS) with peripheral nervous system (PNS) involvement.
Methods Patients with pSS and PNS involvement who were included in the French AIR registry were analysed.
Results 17 patients (age 60 years (44–78 years); 14 were female) were analysed.
Neurological improvement was noted in 11 patients (65%) at 3 months. Rankin scale decreased from 3 (1–5) to 2 (1–5), 2 (1–5) and 2 (1–6) after 3, 6 and 9 months (p=0.02). European Sjögren's Syndrome Disease Activity Index decreased from 18 (10–44) to 11 (5–20), 11 (5–29) and 12 (5–30) after 3, 6 and 9 months (p<0.05).
RTX was effective in neurological involvement in 9/10 patients with vasculitis or cryoglobulinaemia (90%) (group 1) at 3 months and in 2/7 cases (29%) without cryoglobulinaemia and vasculitis (p=0.03). Rankin and European Sjögren's Syndrome Disease Activity Index scales decreased significantly in group 1.
Conclusion RTX seems effective in cryoglobulinaemia or vasculitis-related PNS involvement in pSS.
Statistics from Altmetric.com
Peripheral nervous system (PNS) involvement is one of the extraglandular features observed in primary Sjögren's syndrome (pSS).1 Various forms of neuropathy, including pure sensory polyneuropathy, mixed sensorimotor polyneuropathy and multineuritis, have been described. Vasculitis has been shown to be associated with better response to treatment and outcome.2
Recently, rituximab (RTX) was used in pSS, and some studies suggested its efficacy.2,–,9 No study has aimed to evaluate the effect of RTX on pSS with PNS involvement. We evaluated the safety and efficacy of RTX in pSS with PNS involvement, using the data of the French Autoimmunity and Rituximab (AIR) registry.
Patients and methods
Prospective data from patients with pSS and PNS involvement who were included in the AIR registry were analysed. All patients fulfilled the American–European Consensus Group criteria. PNS involvement was defined by the presence of clinical signs of peripheral neuropathy confirmed by the nerve conduction study. Three subtypes of neuropathy were defined: pure sensory polyneuropathy comprising sensory ataxic neuropathy, mixed sensorimotor polyneuropathy and multineuritis. Vasculitis was defined by the presence of vasculitis on neuromuscular biopsy or by the presence of neuropathy associated with purpura.
Laboratory data at inclusion were as follows: standard tests, serum γ-globulins, CH50, C3, C4, rheumatoid factor, protein immunofixation, anti-nuclear antibodies, serum cryoglobulin, anti-phospholipids antibodies, anti-neutrophil cytoplasmic antibodies, C-reactive protein, erythrocyte sedimentation rate, per cent CD19 cells, HIV serology and hepatitis C virus serology.
Neurological response to RTX was defined as partial or complete clinical and/or electrophysiological improvement at 3, 6 and 9 months. The modified Rankin scale, which consists in a clinical score of disability from 0 (asymptomatic) to 5 (severe), was retrospectively assessed.12 Disease activity was assessed retrospectively using the European Sjögren's Syndrome Disease Activity Index (ESSDAI), which assesses the activity level of 12 domains.13
Taking into account previous reports on the efficacy of RTX in pSS-associated cryoglobulinaemia and vasculitis,10 we divided the patients into two groups: patients with cryoglobulinaemia or vasculitis (group 1) and patients with neither cryoglobulinaemia nor vasculitis (group 2).
This study was approved by 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.
Data are presented as medians with ranges for continuous variables and as numbers with percentages for qualitative variables. Fisher's exact test was used to compare qualitative variables, and non-parametric Mann–Whitney U test or Wilcoxon test was used to compare continuous variables. p<0.05 was considered significant. Statistical comparison was realised from baseline to 3, 6 and 9 months and between different groups at the same times. Statistical analyses were realised using GraphPad Prism (GraphPad Software, San Diego, California, USA, 2007).
Among the 82 patients with pSS who were included in the AIR registry, 17 patients with PNS involvement (21%) (median age 60 years (48–78); 14 were female) received RTX. Ten patients in group 1 had either cryoglobulinaemia (n=9) or vasculitis without cryoglobulinaemia (n=1), and 7 patients were without both in group 2 (table 1). Sixteen of 17 patients received previous treatments for neuropathy: corticosteroids alone (n=3), Ig alone (n=2) or Ig with immunosuppressive agent (n=4), immunosuppressive agents with corticosteroids (n=6) or immunosuppressive agents alone (n=1).
Patients in group 1 (n=10) had higher ESSDAI (24 vs 12; p=0.002), more frequent skin involvement (7 cases vs none; p=0.009) and hypocomplementia (42% vs 89%; p=0.006) (table 1).
The RTX regimen was 1 g repeated 15 days later in eight patients (47%). RTX was administered for refractory neuropathy (n=8), systemic presentation (n=4), neurological involvement with skin (n=2), articular (n=2) or pulmonary involvements (n=1). Corticosteroids were associated with RTX in 15 cases (88%) at a median daily dose of 10 mg (5–80 mg). Other immunosuppressive agents were associated with RTX in five cases (29%) (methotrexate in two cases; mycophenolate mofetil, azathioprine and plasma exchange in one case each).
The median follow-up after the last RTX was 33 months (7–77 months).
Neurological improvement assessed by the doctor was noted in 11 patients (65%) at 3 months and persisted in nine cases (53%) at 6 months. Neurological response was more important in patients with sensorimotor neuropathy (p<0.05) and multineuritis (p=0.07) than in patients with sensory neuropathy. Rankin scale decreased from 3 (1–5) at baseline to 2 (1–5), 2 (1–5) and 2 (1–6) at 3, 6 and 9 months, respectively (p=0.02). Median ESSDAI decreased from 18 (10–44) at baseline to 11 (5–20), 11 (5–29) and 12 (5–30) after 3, 6 and 9 months, respectively (p<0.05).
Comparison of efficacy between two groups
RTX was effective in neurological involvement in 9/10 (90%) group 1 patients and in 2/7 (29%) group 2 patients (p=0.03) at 3 months. Among the two responders in group 2, neurological features were sensorimotor polyneuropathy (n=1) and sensory painful neuropathy (n=1). Response persisted at 6 months in 6/9 group 1 responders and in 2/2 group 2 responders. Rankin scale decreased significantly from 3.5 (1–5) at baseline to 2 (1–5) at 3 months in group 1 (p<0.05), but remained stable in group 2.
In group 1, ESSDAI decreased from 24 (17–44) at baseline to 14.5 (7–21) at 6 months (p=0.008), whereas no change was observed in group 2. In patients with vasculitis and/or cryoglobulinaemia, skin improvement was noted in all cases at 3 months. In these patients, joint improvement was noted in all three cases at 3 months, and only in 1/4 cases in patients without vasculitis nor cryoglobulinaemia. Cryoglobulinaemia disappeared in all cases, except in the patient who did not respond at 3 months.
Among responders to RTX, 7 (65%) experienced a relapse after 8 months (7–17 months) following RTX (skin in four cases, neurological in three cases and joint in two cases). Five of these seven relapsing patients had cryoglobulinaemia, which reappeared at relapse in 4/5 cases (80%). Six of them were retreated by RTX, and response was noted in 5/6 cases (83%) after retreatment with RTX. A patient with joint, neurological and skin relapse without cryoglobulinaemia did not respond to RTX retreatment.
Six (35%) patients experienced adverse events, including two mild arterial hypertension episodes during RTX. One patient experienced severe acute infusion reaction at the second RTX, which resulted in RTX discontinuation. A severe cutaneous infection occurred in one patient. A 74-year-old patient with acquired hypogammaglobulinaemia and mycophenolate mofetil developed cytomegalovirus infection 18 months after RTX.
In the present study, we analysed the safety and efficacy of RTX in the largest cohort of pSS-related neuropathy. Our results show the clinical and biological efficacy of RTX in pSS patients with cryoglobulinaemia-related and/or vasculitis-related neuropathy.
Our results confirm the previous findings concerning the efficacy of RTX in cryoglobulin-related neuropathy10 and case reports of pSS-related PNS involvement (table 2).5 7,–,10 14 15 In the present study, 8/9 patients with cryoglobulin-related neuropathy experienced a good response to RTX. This response to treatment was correlated with the disappearance of cryoglobulinaemia, and the only patient with persistent cryoglobulinaemia did not respond to RTX. In agreement with the better efficacy of immunosuppressive agents in vasculitis-related neuropathy,2 the presence of cryoglobulinaemia or vasculitis was a predictive factor of response to RTX. While patients with vasculitis had a more active disease, the response to RTX was noted in 90% of these patients at 3 months, correlated with a better Rankin scale and the disappearance of cryoglobulinaemia.
Regarding pSS-related non-vasculitis neuropathy, the use of immunosuppressive agents is disappointing. Three cases with favourable outcome were reported with RTX.7 14 15 In the present study, the neurological response was observed in 2/7 patients (27%) of group 2. Among the two patients who improved, one had sensorimotor polyneuropathy without any biopsy, and it has been shown that this neurological feature is usually associated with vasculitis. Interestingly, in group 2 patients, only 27% had anti-Sjögren's syndrome A antigen and 43% were men. Even if all these patients fulfilled the American–European Consensus Group criteria of pSS, these patients have a peculiar form of pSS, and this non-vasculitis peripheral neuropathy could be less driven by B cells.16 Thus, other studies are necessary in this subgroup of patients and, to date, RTX should not be recommended in these patients.
As much as 65% of responders experienced relapse and RTX retreatment. Limited data concerning maintenance treatment with RTX are actually available, but some studies on non-Hodgkin's lymphoma and vasculitis associated with anti-neutrophil cytoplasmic antibodies confirm the necessity of RTX maintenance treatment.17 18 Maintenance treatment may be discussed in patients with a good response to RTX.
The main limitation of the study is the open-label and observational design. Other immunosuppressive agents were associated with RTX in 29% of patients, and their additional effects could not be excluded. Neurological response is difficult to assess, but was associated with objectives markers such as Rankin scale, ESSDAI and immunological response.
In conclusion, AIR registry data suggest the efficacy of RTX in vasculitis-related and cryoglobulinaemia-related PNS involvement in pSS. Conversely, when the pathogenesis of neurological manifestations is not related to vasculitis, the efficacy of RTX could not be demonstrated. Additional studies could be warranted to confirm these findings and to assess the necessity of maintenance treatment.
The authors thank the French Society of Rheumatology (SFR), the Club Rhumatismes et Inflammation (CRI), the French National Society of Internal Medicine (SNFMI) and Isabelle Pane (bioinformatician and data manager of the AIR registry) for their implications in the AIR registry.
Competing interests PYH, CL, SR, JEG and XM were investigators of the TEARS phase II study evaluating the efficacy of RTX promoted by the university hospital of Brest and funded by Roche.
Ethics approval The study was approved by the Kremlin Bicêtre Ethic Committee.
Provenance and peer review Not commissioned; internally peer reviewed.
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.