Faculty Bibliography

BACKGROUND/UNASSIGNED:Cognitive impairment occurs in approximately one-third of pediatric-onset multiple sclerosis (POMS) patients. The Symbol Digit Modalities Test (SDMT), a widely used cognitive screen in adults, has yet to be incorporated early into the standard care of POMS. OBJECTIVE/UNASSIGNED:To screen for cognitive impairment early in the course of POMS and analyze predictive factors. METHODS/UNASSIGNED:Of the 955 POMS or clinically isolated syndrome (CIS) patients prospectively assessed from March 2014 to July 2018, 500 POMS and 116 CIS patients met inclusion criteria (disease onset before the age of 18, one or more SDMTs, and 8 years or older at the time of testing). Those with relapse were analyzed separately from those who were relapse-free. RESULTS/UNASSIGNED: = 383, mean follow-up: 1.8 years), 14.1% had clinically meaningful decline predicted by older age of multiple sclerosis (MS) onset and male gender. Disease relapse or steroid use led to transient worsening on the SDMT. CONCLUSION/UNASSIGNED:Early in the disease, some POMS and CIS patients are at risk for cognitive impairment and subsequent decline.

Introduction: Treatment of pediatric MS is challenging as most disease-modifying therapies (DMT) lack efficacy data in children, and there are no comparative effectiveness studies to guide initial DMT choice.
Objective(s): We aim to assess the real-world effectiveness of initial treatment with newer compared to injectable DMTs on disease activity in MS and CIS treated before 18 years.
Method(s): This is a cohort study of children with MS/CIS followed at 12 clinics in the US Network of Pediatric MS Centers, who received initial therapy with newer (fingolimod, dimethyl fumarate, teriflunomide, natalizumab, rituximab, ocrelizumab) or injectable (interferon beta or glatiramer acetate) DMT. Propensity scores (PS) were computed with logistic regression to predict newer DMT use, including pre-identified confounders (sex, race, ethnicity, site, age at onset and first DMT, first event characteristics, height, weight, diagnosis, number of relapses in prior 6 months, new T2 hyperintense or gadolinium enhancing lesions in prior 6 months, baseline EDSS). Relapse rate was modeled with negative binomial regression for number of events on initial DMT, adjusted for PS quintile. Time to new/enlarging T2 and gadolinium enhancing lesions on MRI brain were modeled with midpoint survival analyses, adjusted for PS quintile.
Result(s): 741 children (66% female, 15% CIS) began initial therapy, 197 with a newer and 544 with an injectable DMT. Those started on newer DMT were older at onset (14.3 vs injectable 13.4 years), less likely to have a monofocal first event (37% vs injectable 55%), and more likely to have MS (87% vs injectable 79%). Number of relapses in the prior 6 months was slightly lower in those started on newer DMT (0.8 vs injectable 1.0), while first event severity and EDSS were similar between groups. Balance in confounders was acceptable within PS quintiles. In PS quintile adjusted analysis, those started on newer DMT had significantly lower relapse rate than those started on injectable DMT (rate ratio 0.45, 95% CI 0.29-0.70, p< 0.001). Those started on newer DMT also had lower rate of new/enlarging T2 (HR 0.52, 95% CI 0.37-0.73) and gadolinium enhancing lesions (HR 0.38, 95% CI 0.23-0.62) than those on injectable DMT.
Conclusion(s): Initial treatment of children with MS/CIS with newer DMTs led to better disease activity control compared to initial therapy with injectables, supporting greater effectiveness of newer therapies. Data on long-term safety of newer DMTs is required

Background: Walking impairments are one of the most impactful consequences of multiple sclerosis (MS). Recently, physical rehabilitation research has focused on developing synergistic protocols to enhance clinical benefit. Recent studies have shown that transcranial direct current stimulation (tDCS) and aerobic physical activity (PA) have converging activation pathways and when completed simultaneously, they may promote cortical neuroplasticity.
Objective(s): To harness cortical plasticity to improve gait for individuals with MS.
Aim(s): To investigate the effects of multiple sessions of PA with simultaneously administered tDCS on walking abilities.
Method(s): MS participants (EDSS: 1-6.5, Relapsing-Remitting or Secondary-Progressive subtype) with clinically significant gait deviations were recruited for a randomized controlled trial of 10 sessions of either active or sham tDCS paired with unloaded cycling for 20 minutes. Stimulation was administered over the primary motor cortex (2.5 mA-2.0 mA; anode over C3/cathode over FP2). Walking speed was assessed quantitatively by using a single inertial sensor placed on the lower back and perceived walking abilities were evaluated using the 12-Item MS Walking Scale (MSWS-12), a self-report questionnaire. Measurements were collected at baseline, the end of tDCS intervention, and 4-weeks post-intervention. Two-way repeated measures-ANOVA (Time, Treatment) was performed to investigate differences between active and sham conditions.
Result(s): Thirty-two participants were enrolled in the study, 22 underwent active treatment. No demographic differences were detected between active and sham groups (active:EDSS 4.3+/-1.2, age 55.5+/-10.3; sham:EDSS 4.5+/-1.5, age 49.7+/-13.9). Statistical analysis showed significant Treatment by Time interactions for gait speed and MSWS-12 score. Post-hoc analysis revealed that gait speed increased significantly after active treatment (Baseline vs. End Treatment, 0.98 vs. 1.16 m/s, p< 0.001; Baseline vs. Follow-up, 0.98 vs. 1.20 m/s, p< 0.001). Active group further reported significant improvement in self-report measure (Baseline vs. End Treatment, 58.04 vs. 49.73, p< 0.05). No significant difference was detected after sham stimulation.
Conclusion(s): Our results indicate that multiple sessions of tDCS administered simultaneously with PA induce cumulative and selfreport improvement in walking and benefits persisted until 4-week post-intervention

Objective: To use quantitative metrics from a large heterogenous population of MS PATHS (Partners Advancing Technology for Health Solutions) patients to derive an integrated view of MS disease course.
Background(s): A commonly used diagram to describe MS disease course shows how various measures change over time. The curves are derived hypothetically, and the best fit patterns, e.g. linear, accelerating, are uncertain. It is also unknown whether the diagrams reflect the current era of disease modifying therapies.
Method(s): In MS PATHS, 2 standardized MRI acquisition sequences (3D FLAIR and 3D T1 on Siemens 3T scanners) were incorporated into routine MS MRI protocols at all participating institutions. A software prototype (MSPie) was developed for automated calculation of brain parenchymal fraction (BPF), total T2 lesion volume (T2LV), and new T2 lesion counts (newT2). The Multiple Sclerosis Performance Test (MSPT) was used to complete neuroperformance tests and questionnaires, including Patient Determined Disease Steps (PDDS) and self-reported relapses. Serum was collected as part of an MS PATHS biomarker sub-study and analyzed by SIMOA kit assay to measure serum neurofilament light (sNfL). Cross-sectional data from patients with MRI metrics were analyzed using linear regression to calculate slopes, and tests for quadratic terms to test linearity, for each measure vs disease duration.
Result(s): 5215 unique patients (mean[sd] age=45.9[11.9]; disease duration=11.9[8.8] years) had MRI metrics. Over nearly 4 decades of MS, BPF showed a linear decrease (slope=-0.16%/year) while PDDS and T2LV showed a linear increase, with annual slopes of 0.076/year and 0.51ml/year, respectively. Linear terms (slopes) were highly significant (p< 10^-15); whereas quadratic terms were weak (p< 0.05). Markers of inflammatory activity, including newT2 and relapses, stayed constant/decreased over the course of MS, with annual slopes of -0.01 (p=0.174) and -0.01 (p< 10^-6), respectively. Log(sNfL) increased linearly (slope= 0.015/year, p< 10^-14).
Conclusion(s): Standardization of MRIs across an international network is feasible, enabling high quality MRI-based metrics and systematic learning from routine patient care. Although limited by the cross-sectional nature of the analyses, these results show strong linearity observed for various measures of disease progression, suggesting that MS neither stabilizes nor accelerates in later stages, unlike some hypothetical diagrams of disease evolution

Introduction: Cognitive impairment in pediatric onset multiple sclerosis (POMS) occurs in up to one third of cases.
Objective(s): To screen for cognitive impairment early in the disease course of POMS, measure predictive factors, and determine the effect of relapse on cognitive processing speed.
Aim(s): To identify cognitive processing speed impairment among POMS and pediatric clinically isolated (CIS) patients enrolled in the US Pediatric MS and Other Demyelinating Disease Registry. In March 2014, the Symbol Digit Modalities Test (SDMT) scores were analyzed from March 2014, when the SDMT was added to the clinical evaluation through July 2018, when the data set was locked.
Method(s): SDMT raw scores were converted to age-normative z-scores using validated age and sex adjusted means. Processing speed impairment was defined with z-score increments of-1.0. Clinically meaningful decline in longitudinal analyses was defined by a z-score decrease of 1.0 or more.
Result(s): For the POMS (n=500) and CIS (n=116) with at least one SDMT, the mean age at symptom onset was 13.5 years and the mean (+SD) disease duration at the initial SDMT assessment was 3.0 + 2.9 years. A total of 23.4% of MS and 16.4% of CIS patients had impaired processing speed at initial assessment. SDMT impairment was predicted by worse EDSS, longer disease duration, and lower level of mother's educational achievement. On longitudinal follow-up (n=383, mean follow-up: 1.8 years), 14.1% had clinically meaningful decline predicted by older age of MS onset, male gender, and longer test-retest interval. Disease relapse and steroid use were associated with transient SDMT worsening.
Conclusion(s): Early in the disease course, a subgroup of POMS patients are at risk for cognitive impairment and subsequent decline. Screening for cognitive slowing is critical for prompt identification of potential cognitive deficits and initiation of additional services

Introduction: Travel to clinic can be difficult due to barriers of time and cost and becomes even more burdensome for MS patients living with disabilities. Telemedicine platforms present a solution by providing supervised treatment and rehabilitation at home. Without barriers to access, patients may be more compliant and adherent to daily rehabilitation exercises. We have a large telerehabilitation research program in MS that pairs rehabilitation with transcranial direct current stimulation (tDCS), an emerging non-invasive brain stimulation technique used to improve outcomes. We provide real-time treatment administration and supervision via HIPAA compliant videoconference, termed remotely supervised tDCS or RS-tDCS.
Objective(s): To characterize the advantages of telemedicine for patients with MS in an urban setting.
Aim(s): To measure barriers to access for participants in our RS-tDCS telerehabilitation program, as well as compliance and adherence to a remotely supervised intervention.
Method(s): Participants with MS were recruited to complete a trial of cognitive remediation paired with RS-tDCS at-home. Participants were surveyed following completion of the intervention and asked to rate their difficulty in attending the clinic (on a 1-5 ordinal scale, 1 = no difficulty and 5 = nearly impossible difficulty) as well as the typical cost of attending clinic. Descriptive statistics and ordinal logistic regression models were used to evaluate the factors driving difficulty of travel.
Result(s): Participants (n=44) reported that round trip travel to the clinic requires an average of 2.3+/-2.3 hours of time and $27.04+/-38.13. Participants rated the difficulty associated with attending clinic as being moderate to significant (2.5+/-1.3). Regression analyses that included disease features produced better models and accounted for greater variance in difficulty attending the clinic, (p< 0.001, McFadden pseudo R2 = .515), as compared with socioeconomic variables alone (p< 0.001, McFadden pseudo R2 = .140). The RS-tDCS protocol was successful in providing treatment (95% compliance to treatment) and 93% of participants reported satisfaction with the treatment and remote protocols.
Conclusion(s): Participants with MS face considerable difficulty reaching the clinic, largely due to increasing neurologic disability. Telemedicine techniques such as RS-tDCS can increase treatment access, reduce physical and financial burden of travel and maintain high rates of treatment adherence