Background

Cognitive impairment is one of the most disabling symptoms of multiple sclerosis (MS), affecting about 50% of patients.

Objectives

We sought to define homogeneous cognitive phenotypes in a large cohort of MS patients by using a data-driven approach, and to assess their distinctive clinical and MRI features.

Methods

A cohort of 1212 MS patients and 196 healthy controls (HC) from 8 Italian MS centers underwent cognitive evaluation with Rao’s Brief Repeatable Battery and Stroop Color Word Test. A subgroup (172 MS patients and 50 HC) also underwent a 3T MRI examination, including 3D T1-weighted and dual-echo sequences. Latent-profile analysis was used on cognitive tests’ z-scores for identifying cognitive phenotypes. Linear regression and mixed effects models were used to define the clinical and MRI features of each phenotype.

Results

Five cognitive phenotypes were identified, characterized by “preserved-cognition” (19%), “mild verbal memory/semantic fluency” impairment (30%), “mild-multi-domain” impairment (19%), “severe-attention/executive” impairment with mild impairment of other domains (14%), and “severe-multi-domain” impairment (18%). “Preserved-cognition” patients had shorter disease duration and lower Expanded Disability Status Scale (EDSS) score than all other groups, and mildly impaired phenotypes included patients with shorter disease duration and less likely progressive disease compared to severely impaired groups. However, the “preserved-cognition” group also included patients with progressive disease and high EDSS scores, and severely impaired phenotypes were also represented in early MS stages. Comparing each phenotype to “preserved-cognition” group, distinctive MRI features emerged: “mild verbal memory/semantic fluency” patients had reduced hippocampal volume (p=0.02), “mild-multi-domain” reduced cortical gray matter volume (p=0.04), “severe-attention/executive” higher lesion volume (p=0.04) and severe-multi-domain” extensive brain damage (p<0.01 for lesion, brain, gray matter and thalamic volumes).

Conclusions

We identified five cognitive phenotypes of MS patients, with distinctive MRI substrates. By defining homogenous and clinically meaningful groups, this characterization may be useful for future research on cognitive impairment in MS, and for defining personalized management approaches and rehabilitative strategies in clinical practice.

Background

It is increasingly understood that grey matter (GM) atrophy is associated with disability progression and cognitive decline in patients with multiple sclerosis (MS). Previously, we demonstrated that treatment with cladribine tablets 3.5 mg/kg bodyweight (CT3.5; cumulative dose over 2 years) in the CLARITY study (NCT00213135) decreased brain atrophy compared with placebo, which was closely associated with a lower risk of disability progression [De Stefano et al. MSJ 2018].

Objectives

Post hoc evaluation of GM and white matter (WM) volume changes in patients with relapsing MS randomized to CT3.5 or placebo in the CLARITY study.

Methods

Images from pre-gadolinium T1-weighted magnetic resonance imaging scans of patients randomized to CT3.5 or placebo for 2 years in CLARITY were evaluated using SIENA-XL software. Images from 0–6 months (CT3.5, n=267; placebo, n=265) were analyzed independently of images from 6–24 months (CT3.5, n=184; placebo, n=186) to account for the potential effects of pseudoatrophy. Annualized mean changes in percentage of GM volume (PGMV) or WM volume (PWMV) between CT3.5 and placebo for 0–6 and 6–24 months were compared using a variance model.

Results

CT3.5 reduced GM and WM volume vs placebo in the first 6 months, consistent with pseudoatrophy [PGMV change: CT3.5 -0.53 vs placebo -0.25 (p=0.045); PWMV change: CT3.5 -0.49 vs placebo -0.34 (p=0.137)]. Brain volume loss from 6–24 months was reduced in patients randomized to CT3.5 with the difference between CT3.5 and placebo significant for GM [PGMV change: CT3.5 -0.90 vs placebo -1.27 (p=0.026); PWMV change: CT3.5-0.32 vs placebo -0.40 (p=0.52)].

Conclusions

Volume loss reduction in WM and, particularly, in GM was noted from 6-24 months in CT3.5-treated patients vs placebo, after a period of pseudoatrophy (0-6 months). Such findings suggest that CT3.5 significantly reduces brain atrophy predominantly in the GM, an effect that may contribute to lower risk of disability progression.

Background

Dihydroorotate dehydrogenase (DHODH) inhibition is an established mode of action for disease-modifying treatment of relapsing-remitting multiple sclerosis (RRMS). Vidofludimus calcium (IMU-838) is a selective and potent second-generation DHODH oral immunomodulator being developed for the treatment of several immune-mediated diseases including MS and COVID-19. The inhibition of the DHODH enzyme leads to metabolic stress in stimulated lymphocytes with subsequent reduction of pro-inflammatory cytokines and induction of apoptosis. Due to IMU-838’s pharmacological selectivity and lack of relevant off-target effects on kinases, increased rates of typical antiproliferative effects (neutropenia, alopecia and gastrointestinal disturbances) have not been observed in clinical trials. The serum half-life of approximately 30 hours allows quick on- and off-dosing.

Objectives

To report top­line efficacy, safety, and tolerability of IMU-838 in the relapsing MS EMPhASIS trial (NCT03846219), the first trial of IMU-838 in MS.

Methods

EMPhASIS is a phase 2 multicenter, double-blind, placebo-controlled, randomized, parallel-group trial to assess the efficacy and safety of two once-daily oral doses of IMU-838 (30 and 45 mg/day) in patients with RRMS. Inclusion criteria are age 18-55 years, active RRMS defined by the evidence of clinical and radiological disease activity, and Expanded Disability Status Scale (EDSS) 0-4. The primary endpoint is the cumulative number of combined unique active MRI lesions over 24 weeks. Secondary endpoints include efficacy, safety and tolerability parameters. The study also includes a subsequent optional, open-label treatment period to evaluate long-term safety and tolerability.

Results

210 subjects (65% women) were enrolled in 36 centers across four European countries. The mean age at baseline was 36.8 years (SD 8.8). 198 subjects (94%) completed the 24-week main treatment period, with the last follow-up visit in April 2020. Database lock will be in July 2020 and top-line data will be available shortly thereafter. Primary outcome and several secondary outcomes (including safety data) will be presented.

Conclusions

IMU-838 is an orally available, next-generation selective immunomodulator with a potentially more favorable profile than first-generation DHODH inhibitors. Top-line results of IMU-838 on primary and several secondary endpoints in patients with RRMS in the EMPhASIS trial will be presented.

Background

The MAGNIFY-MS study (NCT03364036) aims to determine the onset of action of cladribine tablets 3.5 mg/kg over 2 years (CT3.5) in patients with relapsing multiple sclerosis (RMS). Efficacy data from the pivotal trial CLARITY showed that outcomes in CT3.5-treated patients were superior to placebo with regard to number and relative reduction of standardized combined unique active (CUA) lesions over the 96-week trial. Carrying out early and frequent magnetic resonance imaging (MRI) will provide valuable insights into the onset of action of CT3.5.

Objectives

To report on the onset of action of CT3.5 by observing changes in counts of CUA MRI lesions during the first 6 months of the MAGNIFY-MS study.

Methods

MRI scans were performed at screening, baseline, and at months 1, 2, 3 and 6 following CT3.5 treatment on patients with highly active RMS. Differences in CUA lesions between post-baseline periods (period 1, months 1–6, period 2, months 2–6, and period 3, months 3–6) were compared to the baseline period. CUA lesion count was standardized to period length and number of MRIs in a period. A mixed effects linear model was used to account for within pooled centre correlation and adjusted for CUA lesion count during the baseline period, age, and baseline expanded disability status scale (EDSS; >3, ≤3). Type-I-error inflation due to multiple testing was controlled by a gatekeeping procedure.

Results

The full analysis set considered for primary analysis included 270 patients. Reductions in mean CUA count were observed from month 1 onwards compared to baseline; by -1.193 in period 1, -1.500 in period 2 and -1.692 in period 3 (all p<0.0001). In particular, the mean T1 Gd+ lesion counts were decreased from month 2 onwards compared to baseline; by -0.857 at month 2, -1.355 at month 3 and -1.449 at month 6 (all p<0.0001). Sensitivity analysis using negative binomial distribution showed that the treatment effect increased with time measured as lack of CUA in subsequent periods; by 61% in period 1, 77% in period 2, and 87% in period 3 (all p<0.0001). The proportion of patients without any CUA lesions increased in the first 6 months; by 52% in period 1 (p=0.0241), 66% in period 2 (p<0.001), and 81% in period 3 (p<0.001).

Conclusions

MRI was used to assess disease activity in a group of highly active RMS on CT3.5 treatment from one month onwards. Data show an early onset of action on CUA lesions that was significant from month 1 versus baseline, with a treatment effect that increased over the first 6 months.

Background

Brain-derived neurotrophic factor (BDNF) can promote neuronal growth and repair, playing a key role in synaptic plasticity, especially in the hippocampus. The BDNF Val66Met polymorphism was shown to strongly affect BDNF function, but its role in modulating gray matter damage in multiple sclerosis (MS) patients is still not clear.

Objectives

Considering BDNF relevance on hippocampal function, we aimed to explore the effect of BDNF Val66Met polymorphism on the atrophy of hippocampal subfields and its role in cognitive functioning in MS patients.

Methods

Using a 3T scanner, we obtained dual-echo and 3DT1-weighted sequences from 50 MS patients and 15 healthy controls (HC). MS patients also underwent genotype analysis of BDNF and an extensive neuropsychological evaluation. Hippocampal subfields were segmented by using Freesurfer 7.0.1 software. Multiple linear regression models adjusted for age, sex and disease duration were used for between-group comparisons and analysis of associations.

Results

The BDNF Val66Met polymorphism was found in 22 MS patients (44%). Compared to HC, MS patients had reduced volumes of: bilateral hippocampus-amygdala transition area (HATA); left cornus ammonis (CA)1, CA3 and granule cell layer of dentate gyrus (GCL-DG); and right fimbria and presubiculum. BDNF Val66Met polymorphism carriers compared to wild-type (Val66Val) MS patients had higher volume of left hippocampal CA1, CA3, CA4, GCL-DG, molecular layer of subiculum and HATA; and of right hippocampal tail, fissure and presubiculum. In MS patients, higher volume in left CA3 and in right presubiculum correlated with better performance in semantic fluency, while higher volume in left GCL-DG correlated with better visuo-spatial memory performance.

Conclusions

The BNDF Val66Met polymorphism has a protective role in MS patients against both hippocampal atrophy and cognitive deterioration. BDNF genotype may be a potential biomarker for predicting cognitive prognosis, and an interesting target to study for neuroprotective strategies.

Background

Understanding whether multiple sclerosis (MS) can have a favorable course is still challenging. However, a small group of patients who are not disabled after many years of disease can be identified. The mechanisms responsible for this ‘benign’ clinical course remain unclear, likely due to the lack of long-term studies.

Objectives

To assess brain damage in multiple sclerosis patients with no or minimal disability after a longstanding clinical course.

Methods

We compared 13 patients with long-term benign clinical course (LT-BMS, age >55 years, disease duration >30 years, Expanded Disability Status Scale [EDSS] <3.0) and 27 non-benign MS (non-BMS) patients (age >55 years, EDSS >3.0). MRI scans were retrospectively assessed (mean follow-up: 11 years, mean scan per patient: 3). Comparisons of brain volumes (BV) and total T2-lesion volume (LV) changes between the two groups were performed using a mixed effect model. Lesion probability maps (LPMs) of both groups were compared using a nonparametric permutation test.

Results

Patients with LT-BMS showed less over-time decrease in global BV (p=0.02) and grey matter (GM) volume (p<0.001) than non-BMS. Lower atrophy was seen in LT-BMS with no or mild cognitive impairment. By contrast, there was no over-time difference between patient groups in T2-LV accumulation and lesion frequency across brain.

Conclusions

Global brain and GM atrophy changes were mild in this unique patient group with long-standing and no or minimal physical and cognitive disability. These results support the relevant role of GM atrophy in characterizing MS patients who may have favorable long-term disease evolution.

Background

The cerebellum plays a relevant role in both motor and cognitive function due to the high number of cerebellar connections with the brain and spinal cord. Alterations in cerebellar functional connectivity may modulate the relationship between brain structural damage and clinical impairment in multiple sclerosis.

Objectives

To investigate whether resting-state functional connectivity changes of the sensorimotor cerebellum represent adaptive neuroplastic mechanisms to reduce the effects of structural damage on physical disability in patients with multiple sclerosis and no disability.

Methods

A total of 144 multiple sclerosis patients with a score of ≤1.5 on the Expanded Disability Status Scale and 98 healthy subjects were selected from the Italian Neuroimaging Network Initiative database and included in this study. Both patients and healthy subjects underwent multimodal 3T-MRI including functional MRI at rest. After parcellation of the cerebellum, the sensorimotor cerebellum (lobules I-V + VIII) was identified and used as a seed for resting-state functional connectivity analysis.

Results

In patients, brain areas with decreased and increased sensorimotor cerebellar functional connectivity were found to coexist with respect to healthy subjects. Areas of decreased cerebellar functional connectivity, i.e. the lingual gyrus, insula, and precentral and postcentral gyri, negatively correlated with T2 lesion load and white matter atrophy. Areas of increased cerebellar functional connectivity, i.e. the posterior cerebellum, nucleus accumbens, prefrontal cortex, cingulate/paracingulate gyri, and precuneus, positively correlated with T2 lesion load and cerebellar and thalamic atrophy. Areas of increased cerebellar functional connectivity with the cingulate gyrus and precuneus negatively correlated with global grey and white matter atrophy.

Conclusions

In patients with multiple sclerosis, the sensorimotor cerebellum extensively reorganizes its functional links with other brain regions. Areas of decreased cerebellar functional connectivity related to white matter damage are present even in the absence of clinical manifestations and may represent a preclinical condition. Areas of increased cerebellar functional connectivity related to both lesion burden and thalamic or cerebellar atrophy likely represent a compensative reorganization of brain circuits. Lastly, global atrophy may influence functional connectivity changes in posterior cortical areas.

Background

Clinical and imaging features of myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD) may overlap with those of aquaporin 4-neuromyelitis optica spectrum disorder (AQP4-NMOSD) and relapsing remitting multiple sclerosis (RRMS). There is an unmet need for MRI biomarkers which reflect biological mechanisms involved in MOGAD and can help in the differential diagnosis.

Objectives

We aim to identify imaging features able to differentiate between non-acute MOG-antibody disease, AQP4-NMOSD and RRMS.

Methods

In this ongoing retrospective, cross-sectional MAGNIMS study, we analyzed data collected from 8 centers. All subjects (n=352) had brain and cervical cord 3T MRI. Quantification of MRI biomarkers included brain white matter lesions (WMLs), cortical lesions (CL), brain parenchymal fraction (BPF), white matter fraction (WMF), cortical and deep grey matter fractions (GMF) and cross-sectional cervical cord area (CSA) at C1-C2. Linear regression models were used to compare MRI measures between groups, corrected for age, sex, and centre. Statistical significance was considered when p was <0.05.

Results

91 patients with MOGAD (50F, mean age: 41yrs [±15]), 85 with AQP4-NMOSD (68F, 49yrs [±14]), 90 with RRMS (56F, 41yrs [±11]) and 87 healthy controls (HCs) (54F, 36yrs [±11.6]) were collected. The most common phenotypes at onset were optic neuritis and transverse myelitis in MOGAD (93%) and AQP4-NMOSD (87%). WMLs were detected in 57% MOGAD, 79% AQP4-NMOSD, all RRMS (100%) patients, and in 15% HCs. The mean lesion load and number of lesions were higher in RRMS than both MOGAD (p=0.007, p<0.001) and AQP4-NMOSD (p=0.001, p<0.001). At least one CL was seen in 8% patients with MOGAD (total n=8), 10% patients with AQP4-NMOSD (n=7), and in 69% patients with RRMS (n=150). All patient groups showed lower BPF than HCs, with lower WMF in MOGAD and RRMS than HCs (all p<0.01). Between groups, deep GMF was lower in RRMS than MOGAD (p<0.001) and AQP4-NMOSD (p=0.001). CSA was reduced in all disease groups when compared to HCs (all p<0.01) and lower in AQP4-NMOSD than RRMS (p=0.01).

Conclusions

This ongoing study indicates that MOGAD and AQP4-NMOSD share similar MRI features, and no specific MRI biomarker can distinguish between them. Patients with AQP4-NMOSD showed greater spinal cord atrophy than RRMS, and RRMS patients had a higher number of cortical lesions, and greater deep GM atrophy than AQP4-NMOSD and MOGAD. The next step is to investigate whether lesion distribution differs between the two antibody-mediated disease.

Background

Cognitive deterioration affects a large proportion of multiple sclerosis (MS) patients, but it occurs also with healthy aging. The effects of aging on cognitive performance in MS have not been fully investigated yet.

Objectives

By evaluating a large multicentric cohort of healthy controls (HC) and MS patients, we compared the age-related decline of cognitive functions occurring in HC and MS patients.

Methods

Brief Repeatable Battery of Neuropsychological Tests (BRB-N) was evaluated in 301 healthy controls (HC) (150 females, age 18-76 years, mean education=14.9 years) and 664 MS patients (421 females; age 18-77 years; mean education=13.3 years; 536 relapsing-remitting and 128 progressive MS) recruited from 3 centers of the Italian Neuroimaging Network Initiative (INNI, www.inni-ms.org). BRB-N allowed to assess verbal memory (Selective Reminding Test [SRT]), visuospatial memory (10/36 Spatial Recall Test [SPART] and delayed-recall), information processing speed (Symbol Digit Modalities Test [SDMT], Paced Auditory Serial Addition Test [PASAT] 3” and 2”) and verbal fluency (Word List Generation [WLG]). Raw scores of each test were converted to Z-scores, based on HC’s cohort by running linear models to regress out the effects of sex, age, education and center. The residuals for both HC and MS patients were then divided by the HC’s error term. Linear models were built for investigating the association of standardized scores with age in MS patients.

Results

In HC, scores of all the BRB-N tests except PASAT 3” and WLG declined significantly with aging (p from <0.0001 to 0.003). Compared to HC, MS patients showed significant worse estimated mean performances already from the age of 20 years in all BRB-N tests (p from <0.0001 to 0.003), except for SPART, SPART delayed-recall and PASAT 3”, whose estimated mean scores significantly worsened later in age (p from 0.03 to 0.04). MS patients showed also a steeper age-related decline of SPART, SPART delayed-recall, SDMT, PASAT 3”, PASAT 2” performances compared to HC (p from <0.0001 to 0.008). No differential effect of age compared to HC was detected in MS patients for WLG and SRT.

Conclusions

Cognitive deficits already affect young adult MS patients and progress faster during patients’ lifespan compared to healthy aging. A different susceptibility to age-effect exists in the cognitive tests currently used to assess cognition in MS patients. The accumulation of MS-related damage combined with brain aging may have synergic detrimental effects on cognitive performances of MS patients.

Funding. This project has been supported by a research grant from the Fondazione Italiana Sclerosi Multipla (FISM2018/S/3), and financed or co-financed with the ‘5 per mille’ public funding.

Background

The MAGNIFY-MS study (NCT03364036) aims to determine the onset of action of cladribine tablets 3.5 mg/kg over 2 years (CT3.5) in patients with relapsing multiple sclerosis (RMS). Efficacy data from the pivotal trial CLARITY showed that outcomes in CT3.5-treated patients were superior to placebo with regard to number and relative reduction of standardized combined unique active (CUA) lesions over the 96-week trial. Carrying out early and frequent magnetic resonance imaging (MRI) will provide valuable insights into the onset of action of CT3.5.

Objectives

To report on the onset of action of CT3.5 by observing changes in counts of CUA MRI lesions during the first 6 months of the MAGNIFY-MS study.

Methods

MRI scans were performed at screening, baseline, and at months 1, 2, 3 and 6 following CT3.5 treatment on patients with highly active RMS. Differences in CUA lesions between post-baseline periods (period 1, months 1–6, period 2, months 2–6, and period 3, months 3–6) were compared to the baseline period. CUA lesion count was standardized to period length and number of MRIs in a period. A mixed effects linear model was used to account for within pooled centre correlation and adjusted for CUA lesion count during the baseline period, age, and baseline expanded disability status scale (EDSS; >3, ≤3). Type-I-error inflation due to multiple testing was controlled by a gatekeeping procedure.

Results

The full analysis set considered for primary analysis included 270 patients. Reductions in mean CUA count were observed from month 1 onwards compared to baseline; by -1.193 in period 1, -1.500 in period 2 and -1.692 in period 3 (all p<0.0001). In particular, the mean T1 Gd+ lesion counts were decreased from month 2 onwards compared to baseline; by -0.857 at month 2, -1.355 at month 3 and -1.449 at month 6 (all p<0.0001). Sensitivity analysis using negative binomial distribution showed that the treatment effect increased with time measured as lack of CUA in subsequent periods; by 61% in period 1, 77% in period 2, and 87% in period 3 (all p<0.0001). The proportion of patients without any CUA lesions increased in the first 6 months; by 52% in period 1 (p=0.0241), 66% in period 2 (p<0.001), and 81% in period 3 (p<0.001).

Conclusions

MRI was used to assess disease activity in a group of highly active RMS on CT3.5 treatment from one month onwards. Data show an early onset of action on CUA lesions that was significant from month 1 versus baseline, with a treatment effect that increased over the first 6 months.