Faculty Bibliography

BACKGROUND:Tocilizumab (TCZ), a humanized monoclonal antibody against the interleukin-6 receptor, is approved for treatment of rheumatoid arthritis and several other immune-mediated disorders. Off-label use of the intravenous formulation of tocilizumab for Neuromyelitis Optica Spectrum Disorder (NMOSD) decreased relapse rates in two small case series. However, treatment protocol that requires frequent intravenous infusions may adversely affect adherence to therapy, especially in the more disabled patients, thereby reducing effectiveness. A subcutaneous formulation of tocilizumab was shown to be noninferior to the IV formulation for approved rheumatologic diseases. The effectiveness of subcutaneous TCZ for NMOSD is unknown. METHODS:We retrospectively reviewed clinical, radiological and serological data on all NMOSD patients who received subcutaneous TCZ in two tertiary referral centers between 2014-2019. RESULTS:Twelve NMOSD patients who received at least 6 months of subcutaneous TCZ were identified. Eleven were female; mean age was 46.9 ± 14.5 years and mean disease duration was 6.6 ± 4.6 years. Seven patients were seropositive for AQP-4 antibodies, two - for MOG-IgG antibodies, and three were doubly seronegative. During subcutaneous TCZ treatment, eight patients (66.6%) were relapse-free, one patient (8.3%) experienced 1 relapse, two patients (16.6%) - 2 relapses, and one patient (8.3%) - 3 relapses. The median relapse rate within 1 year after starting subcutaneous TCZ - 0 (interquartile range =1.75-0) - was significantly lower than in the year prior to treatment initiation (2, interquartile range = 4.0-0.25; p = 0.04). Overall, the annual relapse rate (ARR) decreased from a median of 2 (interquartile range = 5.75-1.29) prior to subcutaneous TCZ to 0 (interquartile range= = 1.0-0) on treatment (p = 0.0015). One TCZ-treated patient died following a severe myelitis attack. CONCLUSIONS:Effectiveness of subcutaneous TCZ in NMOSD appears to be similar to that reported for the IV formulation and has an advantage of at-home administration. Prospective, comparative studies of subcutaneous TCZ for NMOSD are warranted.

BACKGROUND:Aquaporin-4-IgG positive (AQP4-IgG+) Neuromyelitis Optica Spectrum Disorder (NMOSD) is an uncommon central nervous system autoimmune disorder. Disease outcomes in AQP4-IgG+NMOSD are typically measured by relapse rate and disability. Using the MSBase, a multi-centre international registry, we aimed to examine the impact immunosuppressive therapies and patient characteristics as predictors of disease outcome measures in AQP4-IgG+NMOSD. METHOD/METHODS:This MSBase cohort study of AQP4-IgG+NMOSD patients examined modifiers of relapse in a multivariable proportional hazards model and expanded disability status score (EDSS) using a mixed effects model. RESULTS:206 AQP4-IgG+ patients were included (median follow-up 3.7 years). Age (hazard ratio [HR] = 0.82 per decade, p = 0.001), brainstem onset (HR = 0.45, p = 0.009), azathioprine (HR = 0.46, p<0.001) and mycophenolate mofetil (HR = 0.09, p = 0.012) were associated with a reduced risk of relapse. A greater EDSS was associated with age (β = 0.45 (per decade), p<0.001) and disease duration (β = 0.07 per year, p<0.001). A slower increase in EDSS was associated with azathioprine (β = -0.48, p<0.001), mycophenolate mofetil (β = -0.69, p = 0.04) and rituximab (β = -0.35, p = 0.024). INTERPRETATION/CONCLUSIONS:This study has demonstrated that azathioprine and mycophenolate mofetil reduce the risk of relapses and disability progression is modified by azathioprine, mycophenolate mofetil and rituximab. Age and disease duration were the only patient characteristics that modified the risk of relapse and disability in our cohort.

OBJECTIVE:To determine the underlying etiology in a patient with progressive dementia with extrapyramidal signs and chronic inflammation referred to the National Institutes of Health Undiagnosed Diseases Program. METHODS:Extensive investigations included metabolic profile, autoantibody panel, infectious etiologies, genetic screening, whole exome sequencing and the phage-display assay, VirScan, for viral immune responses. An etiological diagnosis was established post-mortem. RESULTS:Using VirScan, enrichment of dengue viral antibodies were detected in cerebrospinal fluid as compared to serum. No virus was detected in serum or cerebrospinal fluid, but post-mortem analysis confirmed dengue virus in the brain by immunohistochemistry, in situ hybridization, quantitative polymerase chain reaction and sequencing. Dengue virus was also detectable by polymerase chain reaction and sequencing from brain biopsy tissue collected 33 months ante-mortem, confirming a chronic infection despite a robust immune response directed against the virus. Immunoprofiling and whole exome sequencing of the patient did not reveal any immunodeficiency and sequencing of the virus demonstrated wild-type dengue virus in the central nervous system. INTERPRETATION/CONCLUSIONS:Dengue virus is the most common arbovirus worldwide and represents a significant public health concern. Infections with dengue virus are usually self-limiting and chronic dengue infections have not been previously reported. Our findings suggest that dengue virus infections may persist in the central nervous system and should be considered in patients with progressive dementia with extrapyramidal features in endemic regions or with relevant travel history. Further, this work highlights the utility of comprehensive antibody profiling assays to aid in the diagnosis of encephalitis of unknown etiologies. This article is protected by copyright. All rights reserved.

OBJECTIVE:To use the large dataset from the Tysabri Outreach: Unified Commitment to Health (TOUCH) program to compare progressive multifocal leukoencephalopathy (PML) risk with natalizumab extended interval dosing (EID) vs standard interval dosing (SID) in patients with multiple sclerosis (MS). METHODS:This retrospective cohort study included anti-JC virus antibody-positive patients (n = 35,521) in the TOUCH database as of June 1, 2017. The effect of EID on PML risk was evaluated with 3 planned analyses using Kaplan-Meier methods stratified by prior immunosuppressant use. Risk of PML was analyzed by Cox regression adjusted for age, sex, prior immunosuppressants, time since natalizumab initiation, and cumulative number of infusions. RESULTS:< 0.001) for the primary and secondary analyses, respectively. Relative risk reductions were 94% and 88% in favor of EID for the primary and secondary analyses, respectively. The tertiary analysis included no cases of PML with EID. CONCLUSION/CONCLUSIONS:Natalizumab EID is associated with clinically and statistically significantly lower PML risk than SID. CLASSIFICATION OF EVIDENCE/METHODS:This study provides Class III evidence that for patients with MS, natalizumab EID is associated with a lower PML risk than SID.

Introduction: A 2017 analysis of the TOUCH database demonstrated significantly lower progressive multifocal leukoencephalopathy (PML) risk in anti-JC virus (JCV) seropositive multiple sclerosis patients on natalizumab extended interval dosing (EID) than on standard interval dosing (SID; 300 mg infusion every 4 weeks).The risk assessments did not consider anti-JCV index, a risk factor for the development of PML. A lower index in EID vs SID patients could bias the EID cohort toward lower PML risk.
Objective(s): To compare anti-JCV antibody index values in anti- JCV seropositive patients on natalizumab EID and SID in the 2017 TOUCH analysis of PML risk.
Method(s): De-identified TOUCH patient and Quest Diagnostics index data were matched. Only anti-JCV seropositive patients with available index were included. Analysis cohorts were as prespecified in the 2017 risk analysis: primary (EID or SID in the previous 18 months), secondary (any prolonged period of EID or SID), and tertiary (dosing history consisting primarily of EID or SID). For PML patients, the maximum index value >6 months prior to PML diagnosis was used. For non-PML patients, the overall maximum index value was used.
Result(s): Anti-JCV antibody index was available for 1555 (78.3%) EID and 9027 (68.8%) SID patients (primary analysis), 2494 (74.9%) EID and 10,453 (67.8%) SID patients (secondary analysis), and 647 (79.4%) EID and 15,835 (68.4%) SID patients (tertiary analysis). Patients with available index values had similar natalizumab exposure as the overall analysis populations (median total doses 34-53 vs 32-51 [EID] and 27-47 vs 26-46 [SID]) and identical average dosing intervals (median 34-41 days [EID] and 29-30 days [SID]). In each analysis cohort, the median anti-JCV index was higher in EID patients than in SID patients (primary 1.7 vs 1.3; secondary 1.5 vs 1.4; tertiary 1.6 vs 1.4). Greater proportions of EID than SID patients had index values >0.9 and >1.5 in the primary (index >0.9: 70.2% vs 61.1%; index >1.5: 55.0% vs 45.9%), secondary (index >0.9: 66.1% vs 63.5%; index >1.5: 51.0% vs 48.5%) and tertiary (index >0.9: 70.5% vs 63.2%; index >1.5: 53.9% vs 48.4%) analyses.
Conclusion(s): For each analysis of PML risk, EID patients had numerically higher anti-JCV index values than SID patients. Thus, the lower risk of PML with EID compared to SID is not due to lower anti-JCV index among EID patients. This supports the conclusion that natalizumab EID is associated with a lower risk of PML than SID

Objective/UNASSIGNED:To develop a resource of systematically collected, longitudinal clinical data and biospecimens for assisting in the investigation into neuromyelitis optica spectrum disorder (NMOSD) epidemiology, pathogenesis, and treatment. Methods/UNASSIGNED:To illustrate its research-enabling purpose, epidemiologic patterns and disease phenotypes were assessed among enrolled subjects, including age at disease onset, annualized relapse rate (ARR), and time between the first and second attacks. Results/UNASSIGNED:As of December 2017, the Collaborative International Research in Clinical and Longitudinal Experience Study (CIRCLES) had enrolled more than 1,000 participants, of whom 77.5% of the NMOSD cases and 71.7% of the controls continue in active follow-up. Consanguineous relatives of patients with NMOSD represented 43.6% of the control cohort. Of the 599 active cases with complete data, 84% were female, and 76% were anti-AQP4 seropositive. The majority were white/Caucasian (52.6%), whereas blacks/African Americans accounted for 23.5%, Hispanics/Latinos 12.4%, and Asians accounted for 9.0%. The median age at disease onset was 38.4 years, with a median ARR of 0.5. Seropositive cases were older at disease onset, more likely to be black/African American or Hispanic/Latino, and more likely to be female. Conclusions/UNASSIGNED:Collectively, the CIRCLES experience to date demonstrates this study to be a useful and readily accessible resource to facilitate accelerating solutions for patients with NMOSD.

Background: In many of the inflammatory neurologic diseases (IND) of the central nervous system, such as MS, total white blood cell count in the cerebrospinal fluid (CSF) is normal or only mildly elevated, yet ratios of lymphocyte subtypes differ from those seen in the non-inflammatory neurological disease (NIND).
Objective(s): 1. To determine whether lymphocyte ratios in CSF, as assessed by flow cytometry can be used to discriminate between IND and NIND; 2. To determine whether lymphocyte ratios can be used to differentiate between MS and other CNS neuro-inflammatory diseases (such as meningitis, MOG Antibody Syndrome, Susac's syndrome).
Method(s): We retrospectively reviewed the charts of 100 consecutive patients evaluated by NYU neurologists between 1/2013 - 3/2019 for whom lymphocyte subtyping in CSF was carried out. The following lymphocyte markers were assessed: CD19, CD20, kappa-light chain, lambda-light chain, CD20-large (B immunoblasts), CD20-large kappa, CD20-large lambda, CD38bright+/ CD19+/CD20- (plasma cells), CD38bright+/CD19+/CD20- Kappa, CD38bright+/CD19+/CD20- lambda, CD3, CD4, CD8, CD3-/CD7+ (NK-cells), and CD3-/CD4dim + (monocytes). Regression modelling was used to identify lymphocyte subsets that predicted IND vs NIND, as well as MS vs non-MS inflammatory disorder.
Result(s): 62 patients had IND (45 MS, 17 non-MS), 25 had NIND, and 13 did not receive a definitive diagnosis. Regression model that best separated IND from NIND, included CD19+ (higher in IND, p=0.019), CD4+ (higher in IND, p=0.015) and CD3 (nonsignificantly lower in IND, p=0.065). When used as individual predictors, CD19+ >=2% predicted IND with sensitivity of 44%, specificity 88%, positive predictive value of 90% and negative predictive value of 39%, while CD4+ >=55% predicted IND with sensitivity 49.1%, specificity 84.2%, positive predictive value 90% and negative predictive value 35.6%. No NIND patients and 10 IND patients (16%) had CD19+>=5%. Only 2 NIND patients (8%) and 18 IND patients (29%) had CD4+ count >60%. Lymphocyte counts were not useful for differentiating MS from non-MS subgroups except that CD3+ count was lower in the MS group.
Conclusion(s): Flow cytometry of CSF is a useful adjunct in the diagnosis of IND. Higher CD19+ and CD4+ counts were rarely observed in NIND and could be regarded as strong supportive evidence of IND. These results will need to be confirmed in a larger prospective cohort

Introduction: A model of individual response to therapy based on demographic and clinical information ('The MSBase model'; Brain. 2017, 140:2426) improves clinicians' ability to predict how individual patient will fare on specific disease-modifying therapy (DMT) during a 4-year follow up period. However, the range of predicted outcomes is often too wide to make it the arbiter of decision- making. MS severity score (MSSS), a decile rank of Extended Disability Severity Scale (EDSS) scores in patients from the reference dataset, was shown to be longitudinally stable in MS cohorts, but not for individual patients.
Objective(s): To test whether the addition of MSSS will improve the accuracy and robustness of the existing MSBase predictive model.
Method(s): All eligible patients from the global MSBase cohort who commenced a new DMT during the prospectively recorded follow-up were included. For each DMT, Andersen-Gill survival models were constructed for confirmed disability and relapse events. Principal component analysis was used to reduce dimensionality of the models, as previously described. MSSS was added to these models separate from the principal components.
Result(s): Among 13,780 MSBase enrollees (72% female; 86% with Relapsing MS; age 38.4 [10.6] years, disease duration 8.49 [7.73] years, EDSS 2.65 [1.80]), the most common DMTs were Interferon (IFN)b-1a sc (18%), IFNb-1b sc (12%) and Glatiramer acetate (11%). Higher MSSS was associated with an increased chance of 6-month confirmed recovery from disability (Hazard ratio (HR) 1.17-1.64 p< 0.003; no evidence of association only for alemtuzumab) and with relapse risk (HR 1.03-1.18 p< 0.04, and non-significant for 3 DMTs), but did not predict risk of confirmed disability progression for any DMT. The amount of variance that MSSS helped to explain in addition to the original MSBase model was modest, largest for disability regression (partial adjusted R2 0.01-0.05) and less substantial for risk of relapses (partial adjusted R2 - 0.001-0.026).
Conclusion(s): Mostly independent of DMT identity, higher MSSS was positively associated with a chance of confirmed disability regression, and relapse risk, but not disability worsening. Addition of MSSS to the MSBase model yielded improvements in its predictive accuracy, especially with regard to disability regression and relapse risk. MSSS adds to the prediction of recovery from disability and relapse activity but is not a suitable indicator of individual treatment response