2nd MAGNIMS virtual meeting

Map Unavailable

Date(s) - 17/12/2020
3:00 pm - 6:00 pm

2nd Virtual Meeting

17th December 2020

Download agenda by clicking below

Download file attached


17th December 2020

15.00 pm-18.40





Welcome and introduction (M. Rocca, C. Enzinger)

N=93 participants

MAGNIMS web-based project tracker (Jaume Sastre-Garriga)

  • Website www.magnims.eu well frequented, e.g. 76 last 7 days before submission deadline fellowships
  • NEW: MAGNIMS Project Overview; currently outside, but will be embedded in website, see screenshots; each project can be tracked – info on title, centre, contact for person who uploaded it
  • Stream/Forum; managed by general administrators, but each project has an own administrator; can invite followers
  • Well received, congrats to Sergio Vergara and Jaume Sastre-Garriga
  • Next step, load up projects!
  • Upon discussion, consensus to provide a template – what is needed

Update on MAGNIMS data-sharing (Hugo Vrenken)

  • Framework agreement, all core centers have signed and it is in effect
  • Affiliate centers: texts accepted by 9 affiliate centers, legal issues prevent joining of Paris at this stage
  • Hugo thanks Jaume and Mara for great efforts and in turn Mara thanks Hugo for great success so far

Download file attached


  • Presentation of first round 2020 ECTRIMS/MAGNIMS fellowships (M. Rocca) – background information: 1st course spring, 2nd one closed, 5 fellowships received (before X-mas upon solving of bureaucratic issue will be send to reviewers)
    • Antonio Carotenuto (Host: Milan), title of the project: Network Centrality in Multiple Sclerosis (participating centers: Milan, Amsterdam, Rome, Naples) – Supervisor MAR; 3D T1, FLAIR/PD T2, rsfMRI; 870 subjects from Milan, Naples, Rome; 234 HCs; most RRMS; data collection closed 30 Dec 2020; SPM + Degree centrality Map (REST toolbox); clear plan for study conduction.

Download file attached

    • Monica Margoni (Host: Milan), title of the project: Exploring in-vivo brain myelin content in patients with multiple sclerosis through the T1/T2 ratio mapping: a multicentre study (participating centers: Milan, Amsterdam, London, Oxford, Barcelona): Supervisor MF; T1-/T2w radio-values in different clinical phenotypes, compared to HC, both in NAWM and GM. 7 centres (2x Milan): 449 patients; 1y FU 231, 64 HC; 3D T1; created a new pipeline (based on Ganzetti et al, Front Hum Neurosci 2014); Pitfalls: eye and temporal masks were not evaluable in the subject´s space + temporal mask needed to be manually edited; histograms for each centre – different with high-intercentre but low intra-centre variability; next steps defined. Some more progressive patients will be requested from participating centres and info on cognition, ambulation and hand performance. Invitation to others to contribute if they wish.

Download file attached

  • Presentation of 2019 MAGNIMS/ECTRIMS Fellowships
    • Lukas Haider (Host: London), title of the project: Dirty” Virchow-Robin Spaces in Controls and Multiple Sclerosis – Histological and Clinical Validation of a Novel MRI Feature – unable to join the meeting due to clinical work. Update: The first part of the project (dirty VRS in NMO/RRMS/MOG/HC) has been written up and hopefully will be circulated soon. The second part (histological evaluation of dVRS) will be presented at the next meeting. FB: delayed due to histopathology, received a shipping from Oxford
    • Rosa Cortese (Host: Siena; Home London Ciccarelli), title of the project: Towards a better understanding of MOG-Antibody-associated disease and new MAGNIMS MRI criteria. To identify differences in MRI lesion distribution MOGAD, AQP4-NMOSD and RRMS; 2. Brain and spinal cord atrophy. Retrospective: 5 subjects per phenotype Minimum per site, 16 centres; 665 scans (162/172/189/152HC). Demographic differences as expected. Percentage with WML: RRMS 100%, MOGAD 68%, AQP4-NMOSD 82%. Lesion distribution all groups deep WM; cervical cord involvement 0% HC, highest NMOSD; cervical cord atrophy C1/2 lower in all groups vs. HC; discriminators for different diseases sought; only cervical cord lesions distinguish between MOGAD and NMOSD. Discrimination between RRMS and the other diseases possible, the both are more similar than different. Next steps: clinical features (recovery, etc.), ON + thoracic + lumbar SC lesions. Limitations discussed: definition of black holes, cervical cord images not acquired regularly; mixed bag of acute and chronic cases (e.g., swelling SC); length SC lesions; application of DIS and DIT criteria; why cortical lesions on DIR in HCs (n=2). Correct LPMs for lesion numbers (MS has more)

Download file attached

New affiliated center proposal

  • Manuela Vaneckova and Dana Horakov, University of Prague (affiliated to Siena): presentation of the group; 15 minutes

Download file attached

  • Introduction by Nicola de Stefano: have attended to three meetings already
    • Neuroradiologist, MRI unit; 2 3T Scanners (Siemens Skyra for volumetry, Philips Ingenia for whole body); in 29 min dedicated qMRI protocol incl. 3D FLAIR, 3D MPRAGE; >7000 MS patients in database, 3500 in active FU
    • Started with AVONEX-Steroid-AZA study 1999-2003 n=181, yearly MRI until know; 2005-2009 Early IFN ß-1a in high risk CIS n=220; all followed-up
    • Philips 1.5T 3430 patients, 20.053 MRI scans; 2016- 3T Skyra 2617 patients
    • Uher et al, MSJ Predictors of disability – CC atrophy; Uher cut-offs for brain-atrophy; Uher T semiautomated pipelines for measurement of brain and SC atrophy
    • Volumetry measurement (MorphoBox) and atrophy (FreeSurfer) in daily reporting
    • New projects: spinal cord project n=639; early MS NFL n=137, HC n=102
    • Quantitative characterization T1 z-scores periventricular gradient, more pronounced in progressive MS
    • Focused MRI on early CC and SC atrophy; personalized T1-/T2 mapping; NFL correlations (cooperation with Kuhle)
  • Discussion: what MAGNIMS projects would you be interested in? Atrophy, T1-mapping; we have huge data; what experience do you have with advanced imaging? Now T1/T2 mapping; atrophy, no diffusion no DTI; all patients from one centre; MRI and clinical data in one database; all appreciate large number of scans

New projects

  • 3D-FLAIR Brain Age Modelling for application to clinical MS imaging (Frederik Barkhof; Olga Ciccarelli, Tarek Yousry, London). Presented by Olivia Goodkin. Brain Age Modelling to predict chronological age (Cole et al). Mostly needed with volumetric T-1, not used with 3D FLAIR. Hypothesis: will capture adequate age-related information, not inferior to T1. Reference dataset 3D T1 and FLAIR MS and HC, MAGNIMS and open access data sources. Anonymised data requested for both, UCL Ethics. No minimum number of datasets; younger age groups needed

Download file attached

  • Data-driven analysis of (neurodegeneration) subtypes in multiple sclerosis (Frederik Barkhof; Viktor Wottschel; Olga Ciccarelli, Amsterdam, London). FB on behalf of Viktor: FU project from event-based modelling work (Eshaghi et al. Brain); Subtype and Stage Inference (SuStain), preliminary findings: 2 inherent models of atrophy: group 1 first brainstem and deep GM atrophy, group 2 early ventricle enlargement. Mild effects on EDSS and disease duration, age. Viktor is leaving end of year, but still working in academic institution in Amsterdam

Download file attached

  • COSMOS: aCcuracy Of central vein Sign in predicting Multiple sclerosis in nOn-typical preSentations (Nikos Evangelou, Jens Wuerfel, Massimo Filippi) – modified proposal needs to be send to SC


  • DECISIve – DiagnosE using the Central veInSIgn. A prospective diagnostic superiority studycomparing T2* MRI and lumbar puncture in patients presenting with possible Multiple Sclerosis: a multicentre study (Nikos Evangelou, Jens Wuerfel, Massimo Filippi) – modified proposal needs to be send to SC

Updates on planned and on-going projects

  • Cortical remyelination in multiple sclerosis: a magnetization transfer ratio multicenter study (Bruno Stankoff, Paris). Added 15 patients from Graz cohort and 36 from Paris; Milan data did not allow analyses, needed to be excluded. Final long. 142 MS Patients (110 RRMS, 32 PMS); new method to minimize WM misclassification; Request to obtain clinical data 5 yrs after baseline – ok for Graz, Siena will check.

Download file attached

  • A multi-centric study of the combined sensitivity and specificity of cortical lesions and central vein sign for MS diagnosis and differential diagnosis (Cristina Granziera, Jens Wuerfel, Ludwig Kappos): 18-80 yrs, def. diagnosis CIS, MS, NMOSD and all differentials; 3T including T1w, T2w, FLAIR; Milan, Bar, London, Verona, Lausanne: Data + DTA ok, Graz pondering to obtain additional data. N=670; 326 MS, 143 HC, rest mimics. Aiming at cortical lesion and CVS; cortical lesions: MPRAGE; MP2RAGE; DIR; PSIR; CVS: T2* GRE; T2* 3D-EPI; SWI. Discussion: Synthetic DIR?

Download file attached

  • Prognostic value of longitudinal network dynamics – Investigating their impact on emerging functional impairment (Sergiu Groppa, Mainz): CIS and RRMS disease duration <5yrs; 2 retrospective MRI at least 1yr apart; 5yr FU; n=251 multi-centre; 28.7% with disease progression over 5yrs. DTA almost cleared. Asks for additional data and centers to contribute. Presentation on website

Download file attached

  • Restriction spectrum imaging as supplementary early demyelination detection technique (Daniel Rinker, Hanne Harbo, Oslo). Hanne Harbo introduces, Einar Hogestol (post Doc) presents (Slides on website): updated proposal, RSI sequence change. Mainz, Barcelona, London test scans, passed initial processing. Slowed down due to COVID. Multi-shell data optimization. Will speed up during next months

Download file attached

  • Assessing treatment response to oral drugs for MS in real world setting (Serena Ruggieri, Claudio Gasperini, Rome). Pooled n=1020. Most switch from IFN or GA. 1-yr predictors (n=780): 16% CDP, 18% relapse, 26% lack of efficacy, currently data collection completed. Data on orals (particularly fingolimod and teriflunomide) are needed, email will be resend

Download file attached

  • Comparison of the 2017 and 2010 revisions of the McDonald criteria in patients with CIS suggestive of MS: a multicentre study (Paolo Preziosa, Massimo Filippi, Alex Rovira). 958 subjects, 9 centres; final cohort 785; demographics in line with different studies on this topic. Mean FU in non-converters 54 mths. Criteria performance: DIS 2017 more sensitive, accuracy similar. DIT the same. DIS plus DIT more sensitive, less specific, +OCBs improves performance; survival probability estimates of not developing MS: good discrimination over long term FU; median time to MS diagnosis 58 months when awaiting 2nd attack; reaching EDSS >=3.0 predictable by fulfilling of criteria, but predictive power not different between them. Next steps: 3 instead of 1 PV lesion, performance according to time of onset. Starting to draft the paper. Discussion: good to see that DIS and DIT demonstrate different performance, test effect of age strata, more fine-grained analyses would be good; also test for different presentations.

Download file attached

  • Spinal cord atrophy in MS – comparison of different approaches to enhance SC multicentre analysis (Carsten Lukas, Bochum). In progress, not presented, as per request of chairs to limit number of presentations. Additional analyses performed, now completed. Paper will be circulated early 2021.


  • Whole-cord and voxel-based assessment of cervical cord atrophy in MS patients with different clinical phenotypes: a multi-centre assessment (Paola Valsasina, Massimo Filippi, Mara Rocca, Milan) Presented by Paola Valsasina. 54 HC and 110 MS. Regional analyses of cervical cord atrophy in MS predominantly posterior regions. In all MS patients progression of atrophy over 1yr, also posterior regions, no cord atrophy progression in CIS, in PPMS limited to C1/C2 very focused manner. Other aim of the whole project: long term 5yr FU data collected from Barc, London, Oxford; info requested Bochum, Amst, Lugano, Mannheim and Naples

Download file attached

  • Periventricular white matter abnormalities and cortical thinning (Christian Enzinger, Graz). 808 scans from 7 sites; significant correlation of PVLL% and CMT in all MS patients; includes also progressive phenotypes (strongest effects RRMS); analyses finished, figures completed; paper draft needs to be finished before circulation, plan to do this by end of January 2021 at the latest


  • Study of the visual pathway in patients presenting with a CIS (JaumeSastre-Garriga, Barcelona). In progress, not presented


  • Structural connectivity in MS phenotypes (Sara Llufriu, HC Barcelona). Milan provided more patients with SPMS, longitudinal data being processed, regional analyses will be performed. Inclusion closed, but some data needed

Download file attached

  • Non-invasive perfusion patterns in MS phenotypes (Ilaria Boscolo, Massimiliano Calabrese, Verona). Complete analyses for 43 new subjects (5 discarded). 106 subjects from 3 centres Verona , London, Bochum analysed. Few visible cortical lesions. CBF results: trend across groups confirmed reduction towards this gradient HC > RRMS > PPMS, but different for centres. Group maps, but no statistics, demonstrate reduced perfusion. Study goal 70/80 RRMS (currently 70), 30/40 PPMS (currently just 9), 30 HC (currently 27)

Download file attached

Publication status

Review papers

  • Structural and functional connectivity (London workshop June 2017) (Declan Chard). Accepted, Nature Reviews Neurology
  • Quantitative MRI towards clinical application in multiple sclerosis (Basel workshop December 2019) (Cristina Granziera, LK). Accepted, Brain
  • Big Data and Deep Learning: New Avenues for MRI in MS? (Amsterdam workshop 2018) (HV). Major revisions, Neurology
  • MAGNIMS Guidelines (Graz workshop April 2019) (AR). One merged paper submitted to the Lancet Neurology, Joint with NAIMS and CMSC
  • Harmonization on MRI data in multicentre studies (Siena workshop November 2018) (NDS). No update
  • Atypical MS (Graz workshop April 2019) (CE) – interest by NRN; needs to be revisited and updated regarding literature; manuscript skeleton with assignment of tasks for paragraphs will follow by the end of this year

Project papers

  • Characterization of MS-related grey matter atrophy in relevant networks of the human brain: a data-driven, multivariate analysis using source-based morphometry (Mara Rocca, Massimo Filippi, Milan) – Accepted, Neurology
  • Predicting clinical conversion of CIS patients to MS using PRONTO (Deborah Pareto) – Under Review in Neuroimage Clinical (from September)
  • Manual and automated tissue segmentation confirm the impact of thalamus atrophy on cognition in Multiple Sclerosis: a multicenter study (Burggraaff) – revision Submitted Neuroimage Clinical
  • Development and evaluation of a manual segmentation protocol for deep grey matter in multiple sclerosis: towards accelerated semi-automated references (de Sitter/Burggraaff). Under revision, Neuroimage Clinical
  • Large-scale Assessment of Brain Volume Changes in Healthy Individuals (Marco Battaglini, Nicola De Stefano, Siena). No update
  • Quantification of cervical cord cross-sectional area: which software? which vertebral level? spinal cord or brain MRI? A multi-center comparison on a healthy volunteer – C. Lukas. In preparation

Recently Published papers

  • The role of pontine lesion location in differentiating multiple sclerosis from vascular risk factor related small vessel disease (Ruth Geraldes/Jackie Palace, Oxford) – Published, MSJ

Next MAGNIMS meetings

  • Meeting pipeline postponed – 36th meeting 25th-26thMarch 2021 Rome (CG) – rescheduled September (October) 2021
  • SC TC Meeting January 2021 – date and time to be fixed

FINALLY: Just a kind reminder to all presenters to send Jaume slides for website upload. Many thanks!


Thanks to Jackie Palace and team for organising the conference!