Search

Neuromyelitis optica spectrum disorder (NMOSD) is a rare, chronic, autoimmune disease, characterized by astrocytopathic lesions in the central nervous system (Beekman et al., 2019; Fujihara et al., 2020). The main aim of NMOSD maintenance therapy is to reduce the frequency and severity of relapses and minimize future disability (Fujihara et al., 2020). Oral corticosteroids are used long-term to prevent relapses, but are associated with serious complications (Kessler et al., 2016; Kimbrough et al., 2012). In the SAkuraSky study, satralizumab reduced the risk of relapse in patients with NMOSD compared with placebo, with comparable rates of serious adverse events and infections between treatment arms (Yamamura et al., 2019). Here, we report on 16 patients who tapered their steroid dose during the openlabel extension (OLE) period of SAkuraSky.SAkuraSky was a phase 3, multicenter, randomized, double-blind (DB), placebo-controlled study of satralizumab in combination with immunosuppressive therapies (ISTs) in patients with NMOSD. Patients were randomized 1:1 to receive 120 mg subcutaneous satralizumab or placebo in addition to a stable dose of their baseline IST. After completing the DB period or experiencing relapse, patients could enter the OLE period. In the OLE, all patients received satralizumab, and IST doses could be tapered at the investigator's discretion. We assessed the different steroid tapering patterns and their impact on relapse and safety. Patients were considered to have tapered their steroids if their steroid dose at the clinical cut-off date (CCOD: February 18, 2020) was lower than on the first day of the OLE. Annualized relapse rate (ARR) was calculated as the number of relapses divided by the total number of patientyears at risk.Overall, 36 patients receiving oral corticosteroids entered the OLE, of whom 16 tapered their steroid dose. The mean age (range) at baseline was 44.9 (16-73) years, all 16 were female, 14 (88%) were Japanese, and 15 (94%) were AQP4-IgG seropositive. None were receiving any additional ISTs. Patients tapered their steroids from a median of 10 (range: 5-25) mg/day at OLE baseline to 2.75 (0-15) mg/day at the CCOD. Three patients discontinued steroids entirely, and all three remained relapse free. One patient who remained relapse free had temporary increases in steroid dose. Three relapses were observed in two patients who tapered steroids during the OLE; all three relapses required treatment. One of the relapses occurred shortly after a drop in steroid dose from 25 to 10 mg/day. The ARR for steroid-tapered patients was numerically lower in the OLE period than the satralizumab group in the DB period. The safety profile of satralizumab was in line with the overall SAkuraSky population. Two serious infections were observed in steroid-tapered patients in the OLE, both in the same patient: one event (hepatitis E) occurred before the patient began tapering their steroid dose; and one event (influenza) occurred while the patient was tapering.During the OLE of SAkuraSky, 16 patients tapered steroids and the ARR did not increase from the DB period. Patient numbers limit interpretation.

Background Leucine-rich glioma inactivated 1 (LGI1) is a component of the voltage-gated potassium channel complex. IgG antibodies against LGI1 are associated with immunotherapy-responsive encephalitis and epilepsies. LGI1-antibody concentrations are 10–100 times greater in serum than in cerebrospinal fluid (CSF). Oligoclonal IgG bands are rarely found in patients with LGI1-antibody encephalitis or epilepsy. These observations raise questions about the sources of the B cells that result in production of LGI1 antibodies and how the IgGs reach the brain. We aimed to investigate the migration and expansions of peripheral and central B cells to the production of LGI1-specific IgG. Methods We performed PCR amplification and next generation deep immune repertoire sequencing of immunoglobulin (Ig) heavy chain variable regions (VH) from CSF and subsorted peripheral blood B-cell populations from two patients with limbic encephalitis and faciobrachial dystonic seizures associated with LGI1 antibodies. Bioinformatics clustering of related IgM-VH or IgG-VH transcripts was used to determine whether active B-cell diversification could be observed, and whether intrathecal B-cell repertoires, if present, were related to peripheral B cells. Findings We identified clusters of related Ig-VH transcripts in the CSF of both patients. Within these clusters there was a range of somatic hypermutations along the IGHV germline segment-derived portion. In addition, we identified a large number of closely related Ig-VH clusters that were common to both CSF and peripheral blood, including a small number of dominating Ig-VH clusters that might represent the most active clonally related B-cell populations. Interpretation Our data suggest that some B-cell affinity maturation occurs inside the CNS compartment in LGI1-antibody encephalitis. Somatic hypermutation rates point to a CSF antigen-driven activation of clonally related B cells that shape the intrathecal immune repertoire. The target antigen or antigens of these clonally related B cells remain unknown; our work continues to determine the relative contribution of intrathecally activated and peripheral LGI1-specific B cells in this autoimmune CNS disease. Funding Wellcome Trust Intermediate Fellowship to SRI, Fulbright-MS Society, Epilepsy Research UK, BMA Vera Down Research Grant.

Background: Leucine-rich glioma inactivated 1 (LGI1) is a component of the voltage-gated potassium channel complex (VGKC). Antibodies against LGI1 are associated with limbic encephalitis and related central nervous system (CNS) disorders. In most patients, anti-LGI1 antibodies are found in the serum, while their occurrence in the cerebrospinal fluid (CSF) is much less common. Oligoclonal IgG (oligoclonal bands, OCB), a prominent indicator for intrathecal B cell activation, are rarely found in patients with LGI1 encephalitis. Overall, the immune pathology resulting in LGI1 autoimmunity is unclear. However, partially due to the absence of OCB, peripheral immune processes are believed to be the principal disease drivers. Objective: To investigate whether an active intrathecal B cell immune response is present in patients with LGI1 antibodies. Methods: We applied PCR amplification and next generation deep immune repertoire sequencing (DIRS) of B cell receptor immunoglobulin (Ig) heavy chain variable regions (VH) from CSF and peripheral blood B cells of 2 patients with LGI1 encephalitis. Extensive bioinformatics clustering of related IgM-VH or IgG-VH transcripts was employed to determine 1) whether active B cell immune mechanisms operate in LGI1 encephalitis, and 2) whether intrathecal B cell repertoires, if present, are related to peripheral B cells. Results: We identified numerous clusters of related Ig-VH transcripts in the CSF of both patients; within these clusters we found a range of somatic hypermutations (SHM) along the IGHV germline segmentderived portion suggesting that local antigen-driven activation of clonally related B cells was involved in shaping the intrathecal immune repertoire. Furthermore, we identified a large number of closely related Ig-VH clusters that were common to both, CSF and periphery. Interestingly, in each patient we identified a small number of dominating Ig-VH clusters that may represent the most active clonally related B cell populations. Conclusion: Our data suggest that in LGI1 encephalitis B cell affinity maturation occurs inside the CNS compartment. The target antigen(s) of the clonally related B cells described here remain unknown; ongoing work will determine whether intrathecally activated B cells and/or B cells that apparently target identical antigens in the periphery and CNS compartments are involved in LGI1 autoimmunity.