ARTICLE OPEN ACCESS

Multimodal longitudinal study of structural braininvolvement in amyotrophic lateral sclerosisHannelore K van der Burgh MSc Henk-Jan Westeneng MD Renee Walhout MD PhD

Kevin van Veenhuijzen MD Harold HG Tan MD Jil M Meier PhD Leonhard A Bakker MSc

Jeroen Hendrikse MD PhD Michael A van Es MD PhD Jan H Veldink MD PhD

Martijn P van den Heuvel PhD and Leonard H van den Berg MD PhD

Neurologyreg 202094e2592-e2604 doi101212WNL0000000000009498

Correspondence

Dr van den Berg

LHvandenBerg

umcutrechtnl

AbstractObjectiveTo understand the progressive nature of amyotrophic lateral sclerosis (ALS) by investigatingdifferential brain patterns of gray and white matter involvement in clinically or geneticallydefined subgroups of patients using cross-sectional longitudinal and multimodal MRI

MethodsWe assessed cortical thickness subcortical volumes and white matter connectivity from T1-weighted and diffusion-weighted MRI in 292 patients with ALS (follow-up n = 150) and 156controls (follow-up n = 72) Linear mixed-effects models were used to assess changes instructural brain measurements over time in patients compared to controls

ResultsPatients with a C9orf72 mutation (n = 24) showed widespread gray and white matter in-volvement at baseline and extensive loss of white matter integrity in the connectome over timeIn C9orf72-negative patients we detected cortical thinning of motor and frontotemporalregions and loss of white matter integrity of connections linked to the motor cortex Patientswith spinal onset displayed widespread white matter involvement at baseline and gray matteratrophy over time whereas patients with bulbar onset started out with prominent gray matterinvolvement Patients with unaffected cognition or behavior displayed predominantly motorsystem involvement while widespread cerebral changes including frontotemporal regions withprogressive white matter involvement over time were associated with impaired behavior orcognition Progressive loss of gray and white matter integrity typically occurred in patients withshorter disease durations (lt13 months) independent of progression rate

ConclusionsHeterogeneity of phenotype and C9orf72 genotype relates to distinct patterns of cerebraldegeneration We demonstrate that imaging studies have the potential to monitor diseaseprogression and early intervention may be required to limit cerebral degeneration

These authors contributed equally to this work

From the Department of Neurology (HKvdB H-JW RW KvV HHGT JMM LAB MAvE JHV LHvdB) Center of Excellence for Rehabilitation Medicine (LAB) andDepartment of Radiology (JH) UMC Utrecht Brain Center University Medical Center Utrecht De Hoogstraat Rehabilitation (LAB) Utrecht and Department of Complex TraitGenetics (MPvdH) Center for Neurogenomics and Cognitive Research VU University Amsterdam the Netherlands

Go to NeurologyorgN for full disclosures Funding information and disclosures deemed relevant by the authors if any are provided at the end of the article

The Article Processing Charge was funded by UMC Utrecht

This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 40 (CC BY-NC-ND) which permits downloadingand sharing the work provided it is properly cited The work cannot be changed in any way or used commercially without permission from the journal

e2592 Copyright copy 2020 The Author(s) Published by Wolters Kluwer Health Inc on behalf of the American Academy of Neurology

Amyotrophic lateral sclerosis (ALS) is clinically and geneti-cally a heterogeneous disease1 Up to 15 of patients developfrontotemporal dementia and another 35 have some degreeof cognitive or behavioral impairment23 A repeat expansionin C9orf72 appears to be the most common mutation inpatients with ALS45 The pathologic mechanisms underlyingthe process of neurodegeneration in ALS have not yet beenresolved Knowing more about how neurodegenerationdevelops over time and about its associated biomarkers inpatients with ALS and subgroups based on phenotype orgenotype may help to identify therapy that slows or stopsdisease progression

Neuroimaging might be valuable in tracing disease pro-gression in ALS Previous cross-sectional imaging studies haveconsistently reported abnormalities within the motor area6

frontotemporal regions78 and basal ganglia910 It has alsobecome clear that patients with a C9orf72 repeat expansionhave a distinct neuroimaging phenotype of widespread grayand white matter involvement1112 Longitudinal imagingstudies may provide an anatomical characterization of diseaseprogression related to clinical phenotype andmight guide ourunderstanding of underlying neurodegenerative processesMoreover whereas many imaging studies applied a singleimaging technique a multimodal approach will be valuable forestablishing the relative sensitivity of various imaging tech-niques in different stages of ALS and thus provide greaterinsight into when gray and white matter become involvedduring disease development

In the present study we investigated longitudinal structuralbrain changes in a large cohort of patients with ALS applyingamultimodal imaging approach assessing both gray and whitematter of the brain

MethodsParticipantsPatients were recruited from Prospective ALS Study TheNetherlands (PAN) an ongoing population-based casendashcontrol study performed in the Netherlands since January2006 shortly after diagnosis13 Clinical characteristics in-cluding functional cognitive and behavioral status diseaseduration and site of onset were recorded Functional statuswas assessed using the revised ALS Functional RatingScalendashrevised (ALSFRS-R)14 Cognition was assessed using 3neuropsychological tests verbal fluency index (VFI) Edin-burgh Cognitive and Behavioral ALS Screen (ECAS) and

frontal assessment battery (FAB) Abnormality in at least one ofthese tests was regarded as cognitive impairment BecauseECAS was developed after the study started different meth-odologies to assess cognitive or behavioral impairment wereapplied VFI scores were transformed to z scores A patient hadabnormal cognition if the score was below the 5th percentile15

The ALS-specific scores of the ECAS1617 were used to assesscognitive status a patient was labeledwith abnormal cognition ifthe residual score (ie observed score ndash expected score) was inthe lowest 5 of the Dutch normative population derived usingthe percentile rank method on residuals16 FAB scores lt13points indicated abnormal cognition18 Behavior was assessedusing the Amyotrophic Lateral SclerosisndashFrontotemporal De-mentia Questionnaire (ALS-FTD-Q)19 and the behavioralscreen of the ECAS (asking the caregiver to report changes inbehavior)17 Behavioral impairment was defined as apathy orscoring at least 2 other items on the ECAS behavioral screen3 oran ALS-FTD-Q score ge22 points19

Disease duration in months was calculated from symptomonset to scan date Disease progression rate was calculated by(48 ndash ALSFRS-R score)disease duration (in months)20

C9orf72 FUS TARDBP and SOD1mutations were assessedas described previously21 Follow-up of patients was con-ducted approximately every 4ndash6 months At each visitpatients were examined by a clinician specialized in motorneuron diseases and their functional cognitive and behav-ioral status assessed

Controls also came from the aforementioned PAN study13

Follow-up scans were conducted for this reference group afterapproximately 1 year to account for physiologic changes overtime (ldquoagingrdquo) Patients with a history of epilepsy stroke orany overt structural brain abnormalities were excluded

Image acquisition and preprocessingAnatomical T1- and diffusion-weighted images of the brainwere acquired using a 3T Philips (Best the Netherlands)Achieva Medical Scanner as described previously2223

Cortical and subcortical brain regions were analyzed using T1images and FreeSurfer (V530 surfernmrmghharvardedu)In total 92 distinct brain regions (68 cortical 14 subcorticaland 6 ventricular regions 4 cerebellar volumes) were par-cellated and segmented (editing cortical parcellations wasnot necessary)24 according to the Desikan-Killiany atlas25

Longitudinal analysis was performed by creating an unbiasedwithin-subject template space and image using robust

GlossaryALS = amyotrophic lateral sclerosis ALS-FTD-Q = Amyotrophic Lateral SclerosisndashFrontotemporal Dementia QuestionnaireALSFRS-R = ALS Functional Rating Scalendashrevised DWI = diffusion-weighted imaging ECAS = Edinburgh Cognitive andBehavioral ALS Screen FAB = frontal assessment battery FDR = false discovery rate FA = fractional anisotropy LME = linearmixed-effects PAN = Prospective ALS Study The Netherlands VFI = verbal fluency index

NeurologyorgN Neurology | Volume 94 Number 24 | June 16 2020 e2593

inverse consistent registration26 Participants with a singlescan were also preprocessed within the longitudinal pre-processing stream of FreeSurfer

Using diffusion-weighted imaging (DWI) white matter tractsforming the structural brain network were reconstructedbased on fractional anisotropy (FA)2223 Nodes were taken as83 segmented brain regions (cortical and subcortical regionsand brainstem) and the interlinking connections representedthe white matter pathways For each participant an individualundirected weighted brain network was reconstructed22 Toavoid potential spurious fibers we only considered con-nections that were present in 50 of all participants27

Statistical analysesStatistical analyses were performed using the R softwarepackage version 343 (R Foundation for Statistical Comput-ing R-projectorg) Connectome-based analyses were imple-mented using MATLAB Release 2017b (The MathWorksInc Natick MA mathworkscom) Measures of corticalthickness subcortical volumes and structural connectivitywere statistically compared between patients with ALS andcontrols Furthermore we investigated patients with C9orf72separately as they have a distinct imaging phenotype andperformed sensitivity analyses to examine possible influenceof other pathogenic mutations Age and sex were used ascovariates in all analyses For subcortical volumes we alsoincluded total intracranial volume as covariate

Cross-sectional analyses were conducted on the baselinescans using linear models and permutation tests (10000random permutations) Longitudinal analyses were per-formed based on all available scans applying linear mixed-effects (LME) models to assess change of structural measuresover time while accounting for random between-subjectvariation using random slopes and random intercepts28

Participants with single assessments were included to takeinto account potential attrition bias and different time inter-vals between scans were incorporated LME models are well-suited to analyzing longitudinal data with irregular and dif-ferent follow-up intervals as was the case for patients andcontrols2829 To investigate disease-specific effects in patientswith ALS compared to controls an interaction term (follow-up interval times group) was included Here follow-up intervaldenoted the time from baseline scan up to the moment of thefollow-up scan and group refers to patient or control Strati-fied analyses for site of onset (bulbar vs spinal) cognitive orbehavioral impairment and disease duration were performedto investigate brain involvement in relation to clinicalphenotype

All gray matter analyses were corrected for multiple testingusing false discovery rate (FDR) p Values lt005 after FDRcorrection were considered statistically significant Signifi-cance of connectome findings was assessed by means ofNetwork-Based Statistics30 using the property of connectomedisease effects to be more plausible when occurring in

connected components rather than in isolation Edges wereinitially labeled as affected if the linear model indicated a pvalue lt005 The size of the largest connected component ofthese affected connections was tested for significance bycomparing it to a null distribution obtained by 1000 randompermutations of group assignments

Fair comparisons between cognitively or behaviorally im-paired and unaffected patients require equal sample sizes Wetherefore obtained 50 random subsamples of unaffectedpatients to equal those of the impaired subgroups and per-formed 400 and 40 permutations per subsample for the grayand white matter analyses respectively which we sub-sequently pooled To achieve result stability and account forsubsampling analysis we carried out twice as many permu-tations (in total) Sensitivity analyses were performed by ex-cluding all participants with abnormal ECAS scores to assesswhether results were driven by participants with cognitiveimpairment Finally we compared baseline scans of patientswithout follow-up data to those of patients with longitudinaldata to assess attrition bias

Standard protocol approvals registrationsand patient consentsThe Medical Ethical Committee of the University MedicalCenter Utrecht approved the study Written informed con-sent was obtained from all participants

Data availabilityThe data that support the findings of this study are availablevia the corresponding author on reasonable request Thereporting of this study conforms to the Strengthening theReporting of Observational Studies in Epidemiology (STROBE)statement31

ResultsDemographicsIn total 292 patients with ALS and 156 controls partici-pated in the study (table 1) Twenty-four of the patientstested positive for the C9orf72 repeat expansion DWI datawere available for 251 patients with ALS (21 C9orf72-pos-itive patients) and 135 controls Follow-up data (up to 6visits tables e-1 and e-2 doi105061dryad8931zcrkv)were available for 133 C9orf72-negative patients with ALS(ie patients without the C9orf72 repeat expansion) and 17C9orf72-positive patients with ALS with a median follow-up time between the first and second visit of 523 monthsThis relatively short follow-up time was due to the pro-gressive nature of the disease Genetic data were availablefor 211 patients and 7 patients had a pathogenic mutationother than the C9orf72 repeat expansion (table e-3 doi105061dryad8931zcrkv) Of the total control group 72controls had 2 or 3 follow-up visits (median follow-up time115 months) Participants dropped out at follow-up be-cause they were unable to undergo second scan (n = 73)

e2594 Neurology | Volume 94 Number 24 | June 16 2020 NeurologyorgN

were deceased at follow-up (n = 58) or second scan was notscheduled yet according to study protocol at time of analysis(n = 102)

To investigate brain involvement in relation to clinical phe-notype C9orf72-negative patients were stratified for site ofonset (table e-4 doi105061dryad8931zcrkv) Patients

were categorized into 4 groups based on quartiles of diseaseduration at the firstMRI scan (ie gray matter le92 92ndash133133ndash212 and gt212 months white matter le10 10ndash1414ndash21 and gt21 months table e-5 doi105061dryad8931zcrkv) Cognitive and behavioral status was available fora subset of patients (tables e-6 to e-8 doi105061dryad8931zcrkv) Significant correlations between the MRI

Table 1 Demographic and clinical characteristics of patients with amyotrophic lateral sclerosis (ALS) and controls

ALS C9orf72-negative ALS C9orf72-positive Controls

Baseline Follow-up Baseline Follow-up Baseline Follow-up

Number 268 133 24 17 156 72

Follow-up interval mo 53 (42ndash62) 51 (37ndash64) 115 (92ndash178)

Male n () 178 (664) 89 (669) 16 (667) 12 (706) 102 (654) 48 (667)

Age at baseline y 634 (546ndash685) 613 (519ndash667) 567 (521ndash643) 543 (516ndash631) 612 (497ndash666) 617 (539ndash680)

Age at onset y 615 (531ndash669) 597 (498ndash648) 561 (506ndash633) 539 (472ndash618)

Handedness right n () 224 (836) 116 (872) 18 (750) 14 (824) 130 (833) 66 (917)

Type familial n ()a 15 (56) 9 (68) 14 (583) 8 (471)

Site of onset n ()

Bulbar 74 (276) 30 (226) 5 (208) 3 (176)

Spinal 192 (716) 102 (767) 19 (292) 14 (824)

Thoracic 2 (07) 1 (08)

FTD n () 2 (07) 1 (42)

Cognition n ( abnormal) 208 (168) 117 (137)

VFI z scores n ( abnormal) 144 (111) 86 (140)

ECAS ALS specific n ( abnormal) 118 (93) 54 (130)

FAB n ( abnormal) 182 (82) 111 (54)

Behavior n ( abnormal) 149 (342) 84 (298)

ALS-FTD-Q n ( abnormal) 61 (475) 36 (472)

ECAS behavior n ( abnormal) 83 (349) 41 (293)

El Escorial criteria n ()

Definite 37 (138) 16 (120) 2 (83) 2 (118)

Probable 114 (425) 54 (406) 5 (208) 2 (118)

Probable laboratory-supported 75 (280) 39 (293) 3 (125) 3 (176)

Possible 42 (157) 24 (181) 14 (583) 10 (588)

Progression rate 051 (027ndash079) 043 (021ndash065)b 064 (037ndash083) 066 (042ndash081)

Disease duration mo 133 (92ndash212) 144 (100ndash217) 115 (82ndash153) 114 (80ndash159)

ALSFRS-R baseline 41 (36ndash43) 42 (38ndash44)b 41 (38ndash44) 41 (38ndash43)

Abbreviations ALS-FTD-Q = Amyotrophic Lateral SclerosisndashFrontotemporal Dementia Questionnaire ALSFRS-R = ALS Functional Rating Scalendashrevised ECAS =Edinburgh Cognitive and Behavioral ALS Screen FAB = frontal assessment battery FTD = frontotemporal dementia VFI = verbal fluency indexValues aremedian interquartile range (IQR) unless otherwise stated Values in follow-up columns are calculated based on baseline data Disease progressionrate was calculated as follows (48-ALSFRS-R)disease duration (in months) The label of familial ALS was used when there were 2 or more cases of ALS or FTDin the family of the patienta Missing values n = 3b Significantly different compared to baseline with p lt 005

NeurologyorgN Neurology | Volume 94 Number 24 | June 16 2020 e2595

measurements and total ALSFRS-R score are shown (ap-pendix e-1 and figure e-1 doi105061dryad8931zcrkv)

C9orf72-negative ALSComparing 268 C9orf72-negative patients with ALS with 156controls at baseline significantly thinner cortex was found inthe primary motor frontal and temporal regions (figure 1A)The left accumbens nucleus and right thalamus bilateralhippocampi and left amygdala were also significantly smallerin patients (table e-9 doi105061dryad8931zcrkv) Theinferior lateral and third and fourth ventricles were signifi-cantly larger in patients with ALS compared to controls grayand white matter volumes of the cerebellum were not sig-nificantly different (table e-10 doi105061dryad8931zcrkv) Baseline connectome analysis revealed the largestconnected component of reduced structural connectivity(comprising 140 connections) in the motor network (p lt0001 figure 1C)

At follow-up additional cortical thinning was found in thebilateral precentral gyri right paracentral gyrus and frontaland temporal regions (p lt 005 figure 1B) Volumes of the leftcaudate right thalamus and accumbens nucleus right cere-bellar cortex and bilateral hippocampi decreased significantlyover time Volumes of the lateral third and fourth ventriclesincreased significantly over time Connectome analysesdemonstrated a small component (27 connections) with FAreductions over time and none with FA increase (p = 0016figure 1D)

C9orf72-positive ALSAt baseline more extensive cortical thinning in C9orf72-positive patients compared to controls than in the analysiswith C9orf72-negative ALS was observed (figure 1E) Re-duction in volume of the bilateral thalami caudate putamenpallidum accumbens nuclei hippocampi and amygdala andenlargement of the ventricles were found (table e-9 doi105061dryad8931zcrkv) Connectome analysis revealeda component of reduced connectivity (86 connections) mostnotably in connections between the bilateral precentral andparacentral gyri basal ganglia and temporal lobe (p = 0005figure 1G)

At follow-up no additional cortical effects were observed(figure 1F) enlargement of the (inferior) lateral and thirdventricles could be detected and progressive white matterinvolvement was shown by FA reductions in a large compo-nent (71 connections p = 0048 figure 1H)

Bulbar- vs spinal-onset ALSCompared to controls at baseline C9orf72-negative patientswith ALS with bulbar onset were found to have more wide-spread cortical involvement than patients with spinal onset(figure 2 A and E) both onset groups showed thinning of theprimary motor regions and the right entorhinal cortex butbulbar-onset patients also showed thinning of frontotemporaland parietal regions Regarding subcortical effects compared

Figure 1 Brain involvement at baseline and additionalchanges over time in C9orf72-negative and-positive patients with amyotrophic lateral scle-rosis (ALS)

(A B) Comparison of cortical thickness between C9orf72-negative patientswith ALS (n = 268) and controls (n = 156) Regions with a significantly lowercortical thickness at baseline (A) and additional cortical changes over time(B) in patients with ALS are marked in blue (p lt 005 false discovery ratendashcorrected) The cortical thickness of each region of interest (ROI) (separatelyfor left and right hemisphere) was taken for the analyses Each resulting pvalue was mapped onto the surface of the entire ROI to indicate significantregional differences between groups (C D) Connectome analysis with af-fected connections in patients with ALS (n = 230) compared to controls (n =135) at baseline (140 connections p lt 0001 [C]) and additional changes overtime centralized in the motor network (D) as expressed by a decrease infractional anisotropy marked in blue (27 connections p = 0016) (E F)Comparison of cortical thickness between C9orf72-positive patientswith ALS(n = 24) and controls (n = 156) at baseline shows widespread cortical in-volvement (E) and no additional cortical thinning over time (F) (G H) Con-nectome analysis reveals widespread connectivity effects in patients withALS at baseline (n = 24 86 connections p = 0005 [G]) and extensive addi-tional white matter effects over time (71 connections p = 0048 [H])

e2596 Neurology | Volume 94 Number 24 | June 16 2020 NeurologyorgN

to controls bulbar-onset patients showed larger inferior lat-eral ventricles and reduced volumes of the bilateral hippo-campi thalami accumbens nuclei and right putamen (tablee-11 doi105061dryad8931zcrkv)

Spinal-onset patients showed larger inferior lateral ventriclesand reduced volumes of the bilateral hippocampi compared tocontrols Bulbar- and spinal-onset patients showed stronglyoverlapping effects of the most severely affected connectionsmainly involving bilateral precentral gyri and paracentrallobules (figure 2 C and G) but the spinal-onset group dis-played a more widespread pattern of affected connections(bulbar onset 48 connections p = 0020 spinal onset 152connections p lt 0001)

No additional changes could be detected in bulbar-onsetpatients (figure 2 B and D) In contrast patients witha spinal onset showed progressive cortical thinning in theprimary motor and frontotemporal regions additional de-crease in volumes of the bilateral hippocampi left caudateright thalamus and accumbens nucleus and enlargementof the ventricles as well as a nonsignificant componentof decreased connectivity over time mainly involvingfrontotemporal regions (22 connections p = 0065 figure 2F and H)

Cognition and behaviorIn patients with unaffected cognition or behavior cerebralchanges were mainly characterized by thinner precentral gyri(figures 3A and 4A) The inferior lateral ventricles were en-larged in both groups and smaller bilateral hippocampi werefound for patients with unaffected cognition (tables e-12 ande-13 doi105061dryad8931zcrkv) The largest connectedcomponent in cognitively or behaviorally unaffected patientswas centralized around the motor system (cognition 44connections p = 0026 figure 3C behavior 51 connections p= 0021 figure 4C)

At follow-up there was no additional gray or white matterinvolvement for cognitively or behaviorally unaffectedpatients only patients with unaffected behavior showeda smaller right accumbens nucleus and both groups displayedenlarged ventricles

At baseline cognitively impaired patients had thinner bilateralprecentral gyri and frontotemporal regions (figure 3E) andshowed smaller volumes of the bilateral amygdala hippo-campi caudate left accumbens nucleus and right putamenand enlargement of the inferior lateral and third ventricles(table e-12 doi105061dryad8931zcrkv) The largest con-nected component included frontal and temporal con-nections (46 connections p = 0023 figure 3G)

At follow-up there was no additional gray matter thinning incognitively impaired patients They showed decreasing bi-lateral hippocampi and left caudate volumes and enlarging

Figure 2 Analysis of cortical thickness and connectivityeffects stratified for site of onset

(AndashD) Bulbar onset At baseline frontotemporal involvement and connec-tivity effects involving the primary motor regions are shown (p lt 005 forcortical thickness [A] 48 connections p = 0020 [C]) with no additionalcortical changes (B) and only small changes in connectivity over time (11connections p = 0063 [D]) (EndashH) Spinal onset At baseline cortical thinningis confined to the motor cortex (E) whereas additional cortical changes arefound in frontotemporal regions during follow-up (F) Connectome analysisshows a large component of decreased connectivity strength (152 con-nections p lt 0001 [G]) again including small changes in the primary motorand frontotemporal regions over time (22 connections p = 0065 [H])Regions unique for bulbar onset were the bilateral fusiform medial orbi-tofrontal gyri pars orbitalis superior frontal and temporal gyri and tem-poral lobes left entorhinal gyrus right inferior temporal gyrus insulamiddle temporal gyrus pars opercularis pars triangularis rostral anteriorcingulate and the bilateral amygdala Regions unique for spinal onset wereall situated in the left hemisphere middle and inferior temporal gyri insulapars triangularis and the paracentral gyrus

NeurologyorgN Neurology | Volume 94 Number 24 | June 16 2020 e2597

Figure 3 Analysis of cortical thickness and connectivityeffects in patients with normal and abnormalcognition

Patients with normal cognition show predominantly involvement of themotor system at baseline (A and C) and minor cerebral change over time (Band D) In contrast cerebral changes were widespread for gray and whitematter and included extramotor regions in patients with abnormal cogni-tion (E and G) Over time these patients only showed loss of white matterintegrity (47 connections p = 0043 [F and H])

Figure 4 Analysis of cortical thickness and connectivityeffects in patients with normal and abnormalbehavior

Patients with normal behavior show predominantly involvement of themotor system at baseline (A and C) and minor cerebral change over time (Band D) In contrast patients with abnormal behavior demonstrated wide-spread cerebral changes comprising nearly the entire temporal and frontallobes (E and G) Over time behaviorally impaired patients showed addi-tional loss of white matter integrity (22 connections p = 0021 [F and H])

e2598 Neurology | Volume 94 Number 24 | June 16 2020 NeurologyorgN

ventricles The significant largest connected component ofreduced structural connectivity included frontotemporalregions (47 connections p = 0043 figure 3H)

Patients with impaired behavior showed thinner bilateralprecentral gyri frontal parietal and temporal regions andreduced volumes of the bilateral hippocampi amygdalaaccumbens nuclei right thalamus putamen and caudate andenlarged inferior lateral and third ventricles (table e-13 doi105061dryad8931zcrkv) The significant largest connectedcomponent included motor temporal frontal and parietalconnections (110 connections p lt 0001 figure 4G)

At follow-up these patients showed no additional thinningbut ventricular volumes increased The significant largestconnected component included frontal connections (22connections p = 0021 figure 4H)

Effect of disease durationPatients were compared between quartile groups of diseasedurations The first quartile group had a higher progressionrate at baseline compared to the others (ie 078 compared to062 052 and 026 respectively)

At baseline patients with a disease duration of less than 9months from symptom onset (first quartile group) showedsignificant thinning of the rostral anterior cingulate cortex(figure 5A) the second quartile group showed cortical thin-ning in the bilateral precentral gyri and right entorhinal cortex(figure 5C) the third quartile group showed cortical thinningcongruent with the second quartile group and thinning of thebilateral pars orbitalis pars opercularis right pars triangularisand rostral middle frontal gyrus (figure 5E) the fourthquartile group showed the most extensive cortical thinningincluding primary motor regions and frontotemporal regions(figure 5G) Regarding subcortical structures there was sig-nificant enlargement of the inferior lateral ventricles for allquartiles compared to controls (table e-14 doi105061dryad8931zcrkv) The first quartile also showed reducedvolumes of the left hippocampus and amygdala Subcorticaleffects were found in the second quartile (bilateral hippo-campi right pallidum amygdala and thalamus and thirdventricle) and the fourth quartile (right hippocampus andfourth ventricle) Baseline connectome analyses of the quar-tile groups showed consistent involvement of the motornetwork (precentral paracentral and brainstem) and limitedextramotor involvement (first quartile group 65 connectionsp = 0002 second quartile group 68 connections p = 0012third quartile group 98 connections p = 0003 and fourthquartile group 121 connections p lt 0001 figure 6 A C Eand G)

Patients with a disease duration of less than 9 months fromsymptom onset showed significant additional cortical thin-ning over time including primary motor and frontotemporalregions (figure 5B) The second quartile group displayedsignificant additional changes restricted to the bilateral

precentral gyri and right temporal pole (figure 5D) The thirdand fourth quartile groups showed no significant additionallongitudinal cortical changes (figure 5 F and H)

At follow-up the first quartile group showed an additionaldecrease in subcortical volumes of the bilateral thalami andhippocampi left caudate and right accumbens nucleus and

Figure 5 Analysis of cortical thickness stratified for diseaseduration

(A C E and G) Baseline (B D F and H) Additional effects over time Patientswith amyotrophic lateral sclerosis (ALS) with a short disease duration showcortical thinning limited to the rostral anterior cingulate cortex at baseline(A) and additional cortical changes over time in motor and frontotemporalregions (B) Patients with a longer disease duration show cortical in-volvement of similar frontotemporal regions at baseline and no progressivecortical changes during follow-up

NeurologyorgN Neurology | Volume 94 Number 24 | June 16 2020 e2599

enlargement of the ventricles The second and third quartilegroups only showed ventricular enlargement over time Noadditional effects were detected in the fourth quartile groupPatients with the shortest disease duration were the only

disease duration subgroup that showed significant additionalconnectivity effects over time (first quartile 23 connections plt 0001 second quartile p = 0070 third quartile p = 0056fourth quartile p = 0436 figure 6 B D F and H)

Because patients with an early MRI scan have a faster diseaseprogression we further explored the effect of disease progressionon cerebral changes The progressive loss of gray and whitematter integrity in patients with shorter disease durations (lt13months) is independent of progression rate (appendix e-2 tablee-15 and figures e-5 and e-6 doi105061dryad8931zcrkv)

Sensitivity analysesSensitivity analyses excluding participants with an abnormalECAS score or a pathogenic mutation other than C9orf72showed similar results to those described above In most sub-groups patients without follow-up data displayed more cerebralinvolvement than patients with longitudinal data with consid-erable overlap of the involved regions (figure e-2 doi105061dryad8931zcrkv) Disease effects measured in mean diffusivityare similar to FA frontal connections are present more fre-quently in the largest connected components (appendix e-3 andfigure e-3 doi105061dryad8931zcrkv) Figure e-4 (doi105061dryad8931zcrkv) displays direct comparisons betweensubgroups (appendix e-4 doi105061dryad8931zcrkv)

DiscussionWe investigated whether structural MRI could serve asa biomarker of disease progression in clinical trials We ap-plied multiple modalities of brain imaging cross-sectionally atbaseline and longitudinally to a large cohort of patients withALS and controls Longitudinal effects in the total group ofpatients were detected in terms of progressive cortical atrophyin primary motor and frontotemporal regions shrinkingsubcortical volumes ventricle enlargement and loss of whitematter integrity in a connected component around the motorcortex We were also able to identify differential patterns ofgray and white matter involvement in subgroups of patientswith specific clinical characteristics disease duration or ge-notype The heterogeneity of our findings among subgroupsof patients may have important implications for cross-sectional and follow-up MRI studies in ALS and the timingof neuroimaging in the disease course for clinical trials andfollow-up studies

We performed a comprehensive study using cross-sectionallongitudinal and multimodal information In the LMEmodels we could use all 292 baseline scans of patients withALS as well as the available follow-up scans from 150 patientsThereby we account for between-subject variation and at-trition bias and thus improve accuracy of the longitudinalanalyses as was reported previously28 A limitation of allimaging studies in ALS including the present study may beselection bias towards a cohort of patients able to undergoMRI scanning In our study this limitation applies in partic-ular to the longitudinal part these patients less frequently had

Figure 6 Analysis of connectivity stratified for diseaseduration

(A C E and G) Baseline (B D F and H) Additional effects over time Patientswith a long disease duration show extensive connectivity effects at baseline(121 connections p lt 0001 [G]) and no significant additional effects overtime (H) Patients with a short disease duration show effects centeredaround the motor network at baseline (65 connections p = 0002 [A]) andadditional connectivity effects over time (23 connections p lt 0001 [B])

e2600 Neurology | Volume 94 Number 24 | June 16 2020 NeurologyorgN

a bulbar onset and a slower disease progression than thosewith only 1 MRI scan However by closely monitoring thispossible bias during inclusion of patients we were able for thefirst time to analyze MRI scans of a relatively large number ofpatients with fast-progressing disease and patients at an earlystage of the disease At baseline more extensive cerebral in-volvement was observed in patients with a single scan com-pared to patients with follow-up scans (figure e-2 doi105061dryad8931zcrkv) suggesting that the longitudinalresults might be slightly underestimating the extent of braininvolvement over time However it is also possible thatpatients with more severe baseline involvement who couldnot undergo follow-up might have reached a plateau of neu-rodegeneration MRI scanners allowing scanning in an up-right position may facilitate inclusion of patients with ALS infollow-up studies

The multiple modalities comprised gray matter data in termsof cortical thickness and subcortical volumes and all whitematter tracts Previous multimodal imaging studies includeda maximum of 64 patients who were assessed at a single timepoint These studies reported both gray and white matteralterations in patients with ALS compared to controls mostprominent in the motor cortex frontal lobe and corticospinaltract32ndash35 Our study demonstrates probably due to the largersample size that besides these regions temporal regions areexposed to cortical thinning subcortical structures (ie basalganglia thalamus and hippocampi) have reduced volumesand ventricles are enlarged By applying connectome-basedanalyses we showed widespread loss of white matter integritynot restricted to the corticospinal tract Future research couldexamine cerebellum connectomes to investigate diseaseeffects in this region

Few previous studies combined a multimodal and longitudi-nal approach32ndash34 Other longitudinal imaging studies in-cluded a single modality (gray or white matter) to examine thepattern of disease progression36ndash40 and revealed mainly pro-gressive gray matter atrophy in the primary motor system andfrontal areas32ndash3436 We also found progressive thinning intemporal regions shrinking of subcortical volumes and en-larging ventricles at follow-up Concerning white matterlongitudinal studies focused mainly on the corticospinal tractshowing minor or no changes32ndash34373840 while assessment ofthe entire connectome in our study demonstrated significantloss of connectivity in the motor system and its linked con-nections in the right hemisphere in the absence of increasingFA Importantly we showed that progressive loss of gray andwhite matter integrity typically takes place in patients withshorter disease duration (lt13 months after disease onset forgray matter and lt10 months for white matter) while previousstudies included patients with a mean disease duration atbaseline at a later stage of at least 18 months Loss of whitematter integrity in previous studies may thus have alreadyoccurred before patients were included indicating the rele-vance of timing of neuroimaging Patients with the shortestdisease duration revealed a distinct connectivity profile and

more extensive changes in cortical thickness over time whilepatients with longer disease duration showed widespreadcortical thinning and loss of connectivity only at baselinepossibly reflecting a kind of brain change end stage Previousresearch has shown by applying simulations that the healthyconnectome provides an anatomical underlying substrate forTDP-43 pathology41 following the stages suggested by theBraak model of corticofugal spread42 Other studies indicatedthat interconnected gray matter regions become graduallyaffected instead of being subject to contiguous spread1043

Future research could therefore examine differences betweenthe Braak stages and observed disease spread in patients Aspatients with early diagnosis may have a more aggressivedisease course44 we investigated whether the observed effectsreflect a more progressive disease course but no clear relationbetween neuroimaging effects and disease progression wasfound in patients with short or long disease duration Hencescanning patients relatively early in the disease course may bemore important than stratifying for progression rate Usingmean diffusivity instead of FA might help prolong the op-portunity to reveal disease effects in white matter

Patterns of brain abnormalities may also be determined by siteof disease onset We observed that both bulbar- and spinal-onset subgroups develop atrophy of similar anatomicalregions which is congruent to neuropathologic studies45

However we found a differential pattern of cerebral in-volvement in patients with bulbar or spinal onset at baselineand at follow-up bulbar-onset patients showed prominentgray matter atrophy and some white matter changes atbaseline whereas spinal-onset patients mainly displayedwidespread loss of connectivity At follow-up gray matterbecame affected in spinal-onset patients whereas bulbar-onset patients showed no additional brain changes One couldspeculate that the prominent white matter changes at base-line followed by additional gray matter involvement in sub-sequent scans might imply a dying-backward process of uppermotor neuron degeneration in spinal-onset patients Forbulbar-onset patients the direction of the disease process isless clear clinically upper motor neuron features in the bulbarregion are more prominent in the early stage of the diseaseimplying a dying-forward process in bulbar-onset patientswhich is at least not contradicted by our neuroimaging resultsIncorporating other phenotypes such as primary lateralsclerosis or progressive muscular atrophy might providemoreinsight in the direction of neurodegeneration

Previous cross-sectional imaging studies have shown extra-motor cerebral involvement of frontal temporal and parietalregions in patients with cognitive or behavioral impairment(often considering abnormal cognition and behavior as oneimpaired group)3546ndash48 In our multimodal study gray andwhite matter changes were widespread mainly in motor andfrontotemporal regions in cognitively and even more ex-tensively in behaviorally impaired patients Subcortical graymatter and white matter changes including frontotemporalregions were found to be progressive over time In patients

NeurologyorgN Neurology | Volume 94 Number 24 | June 16 2020 e2601

with unaffected cognition and behavior cerebral changes wererestricted to motor regions thinning of the motor cortex andreduced ventricle volumes were found to be progressiveThus these findings largely reflect clinical observations andpossibly provide an underlying anatomical substrate for cog-nitive and behavioral impairment in ALS

A limitation of our study may be the missing data and het-erogenous methodology for assessing cognitive or behavioralimpairment as the ECAS was introduced after the studystarted Despite this limitation we were able to determine inone of the largest imaging cohorts of cognitively and behav-iorally impaired patients with ALS according to the currentStrong criteria3 that cognition and behavior are relevantfeatures to consider in follow-up neuroimaging studies inALS Future multimodal neuroimaging studies may need toexplore more detailed patterns of cognitive and behavioralimpairment in a cohort in which cognition and behavior areassessed homogeneously for the entire population35 More-over it might be worthwhile to incorporate level of educationand depression in the analyses as they have a relevant effecton the brain4950

A distinctive MRI pattern of cerebral involvement was foundin C9orf72-positive patients showing more extensive corticaland subcortical changes and reduced white matter integrity atbaseline which is in line with previous studies1112 and ex-tensive additional white matter effects over time that weremore prominent compared to the effects in C9orf72-negativepatients Longitudinal gray matter changes in C9orf72-posi-tive patients were limited to enlargement of the ventriclesOur results indicate that connectome analyses comprising allwhite matter tracts may serve as a potential biomarker forupcoming C9orf72 gene therapy clinical trials Moreover thedifferential patterns of brain involvement emphasize the rel-evance of screening for theC9orf72mutation in neuroimagingstudies

Our study underlines the potential of imaging studies toprovide insight into structural brain involvement in relation tothe heterogeneity of ALS By conducting this study in a largecohort of C9orf72-positive and -negative patients with ALSwith carefully defined phenotypes and systematic follow-upincluding controls we were able to show differential effects ofgray and white matter changes at baseline and during follow-up The observed results might indicate that cerebral changesin ALS occur at a relatively early stage of the disease leastwiseat the level at which current multimodal techniques are able todetect abnormalities This may imply that when designingclinical trials inclusion of patients at an early stage of thedisease is necessary to prevent cerebral neurodegenerationThese cerebral changes could be monitored by MRI providedthat the appropriate modalities are applied to the right sub-groups based on phenotypic characteristics or genotype In-corporating innovative modalities such as spinal cordimaging EEG or other neurophysiologic techniques toquantify upper or lower motor neuron involvement in the

analyses of future studies may improve our ability to in-vestigate propagation of neurodegeneration in vivo at differ-ent stages of ALS

AcknowledgmentThe authors thank Ruben Schmidt for his early work on thisproject and Marcel de Reus for using software to create thewhite matter connectomes which were visualized withBrainNet Viewer (nitrcorgprojectsbnv)

Study fundingThis work was supported by the ALS Foundation Nether-lands MPvdH was further supported by the NetherlandsOrganization for Scientific Research VIDI grant (VIDI-452-16-015) MAvE was supported by the Netherlands Orga-nization for Scientific Research VENI grant (91614039)LHvdB further received funding from the NetherlandsOrganization for Scientific Research VICI grant and theNetherlands Organization for Health Research and De-velopment funded through the EU Joint ProgrammendashNeurodegenerative Disease Research JPND LHvdB re-ceived travel grants and consultancy fees from Shire andserves on scientific advisory boards for Treeway Cytokineticsand Biogen Idec JHV has received funding from the Eu-ropean Research Council under the European Unionrsquos Ho-rizon 2020 research and innovation programme (grantagreement 772376ndashEScORIAL)

DisclosureH van der Burgh H-J Westeneng R Walhout K vanVeenhuijzen H Tan J Meier L Bakker J Hendrikse Mvan Es J Veldink and M van den Heuvel report no dis-closures L van den Berg reports grants from ALS Foun-dation Netherlands The Netherlands Organization forHealth Research and Development (VICI scheme fundedthrough the EU Joint ProgrammendashNeurodegenerative Dis-ease Research JPND [SOPHIA STRENGTH ALS-CarEprojects]) and personal fees from Shire Cytokinetics andTreeway outside the submitted work Go to NeurologyorgNfor full disclosures

Publication historyReceived by Neurology September 8 2019 Accepted in final formDecember 5 2019

Appendix Authors

Name Location Contribution

Hannelore Kvan der BurghMSc

UMC Utrecht theNetherlands

Design and conception of studyacquisition and analysis of datadrafting the manuscript forintellectual content

Henk-JanWestenengMD

UMC Utrecht theNetherlands

Design and conception of studyacquisition and analysis of datadrafting the manuscript forintellectual content

e2602 Neurology | Volume 94 Number 24 | June 16 2020 NeurologyorgN

References1 van Es MA Hardiman O Chio A et al Amyotrophic lateral sclerosis Lancet 2017

3902084ndash20982 Ringholz GM Appel SH Bradshaw M Cooke NA Mosnik DM Schulz PE Preva-

lence and patterns of cognitive impairment in sporadic ALS Neurology 200565586ndash590

3 Strong MJ Abrahams S Goldstein LH et al Amyotrophic lateral sclerosisndashfrontotemporal spectrum disorder (ALS-FTSD) revised diagnostic criteria Amyo-troph Lateral Scler Front Degener 201718153ndash174

4 DeJesus-Hernandez M Mackenzie IR Boeve BF et al Expanded GGGGCC hex-anucleotide repeat in noncoding region of C9ORF72 causes chromosome 9p-linkedFTD and ALS Neuron 201172245ndash256

5 Renton AEMajounie EWaite A et al A hexanucleotide repeat expansion in C9ORF72is the cause of chromosome 9p21-linked ALS-FTD Neuron 201172257ndash268

6 Chio A Pagani M Agosta F Calvo A Cistaro A Filippi M Neuroimaging inamyotrophic lateral sclerosis insights into structural and functional changes LancetNeurol 2014131228ndash1240

7 Grosskreutz J Kaufmann J Fradrich J Dengler R Heinze HJ Peschel T Widespreadsensorimotor and frontal cortical atrophy in amyotrophic lateral sclerosis BMCNeurol 2006617

8 Buchanan CR Pettit LD Storkey AJ Abrahams S Bastin ME Reduced structuralconnectivity within a prefrontal-motor-subcortical network in amyotrophic lateralsclerosis J Magn Reson Imaging 2015411342ndash1352

9 Bede P Elamin M Byrne S et al Basal ganglia involvement in amyotrophic lateralsclerosis Neurology 2013812107ndash2115

10 Westeneng H-J Verstraete E Walhout R et al Subcortical structures in amyotrophiclateral sclerosis Neurobiol Aging 2015361075ndash1082

11 Bede P Bokde AL Byrne S et al Multiparametric MRI study of ALS stratified for theC9orf72 genotype Neurology 201381361ndash369

12 Westeneng HJ Walhout R Straathof M et al Widespread structural brain in-volvement in ALS is not limited to the C9orf72 repeat expansion J Neurol NeurosurgPsychiatry 2016871354ndash1360

13 Huisman MHB de Jong SW van Doormaal PTC et al Population based epidemi-ology of amyotrophic lateral sclerosis using capture-recapture methodology J NeurolNeurosurg Psychiatry 2011821165ndash1170

14 Cedarbaum JM Stambler N Malta E et al The ALSFRS-R a revised ALS functionalrating scale that incorporates assessments of respiratory function J Neurol Sci 199916913ndash21

15 Beeldman E Jaeger B Raaphorst J et al The verbal fluency index Dutch normativedata for cognitive testing in ALS Amyotroph Lateral Scler Front Degener 201415388ndash391

16 Bakker LA Schroder CD Spreij LA et al Derivation of norms for the Dutch versionof the Edinburgh cognitive and behavioural ALS screen Amyotroph Lateral SclerFront Degener 20192019ndash27

17 Abrahams S Newton J Niven E Foley J Bak TH Screening for cognition andbehaviour changes in ALS Amyotroph Lateral Scler Front Degener 2014159ndash14

18 Terada T Miyata J Obi T et al Frontal assessment battery and frontal atrophy inamyotrophic lateral sclerosis Brain Behav 20177e00707

19 Raaphorst J Beeldman E Schmand B et al The ALS-FTD-Q a new screening toolfor behavioral disturbances in ALS Neurology 2012791377ndash1383

20 Kimura F Fujimura C Ishida S et al Progression rate of ALSFRS-R at time ofdiagnosis predicts survival time in ALS Neurology 200666265ndash267

21 Van Rheenen W Van Blitterswijk M Huisman MHB et al Hexanucleotide repeatexpansions in C9ORF72 in the spectrum of motor neuron diseases Neurology 201279878ndash882

22 Verstraete E Veldink JH Mandl RCW van den Berg LH van den Heuvel MPImpaired structural motor connectome in amyotrophic lateral sclerosis PLoS One20116e24239

23 van der Burgh HK Schmidt RWesteneng HJ de ReusMA van den Berg LH van denHeuvel MP Deep learning predictions of survival based on MRI in amyotrophiclateral sclerosis Neuroimage Clin 201713361ndash369

24 Fischl B van der Kouwe A Destrieux C et al Automatically parcellating the humancerebral cortex Cereb Cortex 20041411ndash22

25 Desikan RS Segonne F Fischl B et al An automated labeling system for subdividingthe human cerebral cortex on MRI scans into gyral based regions of interest Neu-roimage 200631968ndash980

26 Reuter M Rosas HD Fischl B Highly accurate inverse consistent registration a ro-bust approach Neuroimage 2010531181ndash1196

27 de Reus MA van den Heuvel MP Estimating false positives and negatives in brainnetworks Neuroimage 201370402ndash409

28 Bernal-Rusiel JL Greve DN Reuter M Fischl B Sabuncu MR Alzheimerrsquos DiseaseNeuroimaging Initiative Statistical analysis of longitudinal neuroimage data withlinear mixed effects models Neuroimage 201366249ndash260

29 Cnaan A Laird NM Slasor P Tutorial in biostatistics using the general linear mixedmodel to analyse unbalanced repeated measures and longitudinal data Stat Med1997162349ndash2380

30 Zalesky A Fornito A Bullmore ET Network-based statistic identifying differences inbrain networks Neuroimage 2010531197ndash1207

31 von Elm E Altman DG Egger M et al The Strengthening the Reporting of Ob-servational Studies in Epidemiology (STROBE) statement guidelines for reportingobservational studies J Clin Epidemiol 200861344ndash349

32 Menke RAL Korner S Filippini N et al Widespread grey matter pathology domi-nates the longitudinal cerebral MRI and clinical landscape of amyotrophic lateralsclerosis Brain 20141372546ndash2555

33 Cardenas-Blanco A Machts J Acosta-Cabronero J et al Structural and diffusionimaging versus clinical assessment to monitor amyotrophic lateral sclerosis Neuro-image Clin 201611408ndash414

34 Bede P Hardiman O Longitudinal structural changes in ALS a three time-pointimaging study of white and gray matter degeneration Amyotroph Lateral Scler FrontDegener 201819232ndash241

35 Bede P Omer T Finegan E et al Connectivity-based characterisation of subcorticalgrey matter pathology in frontotemporal dementia and ALS a multimodal neuro-imaging study Brain Imaging Behav 2018121696ndash1707

36 Agosta F Gorno-Tempini ML Pagani E et al Longitudinal assessment of grey mattercontraction in amyotrophic lateral sclerosis a tensor based morphometry studyAmyotroph Lateral Scler 200910168ndash174

37 Agosta F Rocca MA Valsasina P et al A longitudinal diffusion tensor MRI study ofthe cervical cord and brain in amyotrophic lateral sclerosis patients J Neurol Neu-rosurg Psychiatry 20098053ndash55

38 Keil C Prell T Peschel T Hartung V Dengler R Grosskreutz J Longitudinaldiffusion tensor imaging in amyotrophic lateral sclerosis BMC Neurosci 201213141

39 Schuster C Kasper E Machts J et al Longitudinal course of cortical thickness declinein amyotrophic lateral sclerosis J Neurol 20142611871ndash1880

40 Baldaranov D Khomenko A Kobor I et al Longitudinal diffusion tensor imaging-based assessment of tract alterations an application to amyotrophic lateral sclerosisFront Hum Neurosci 2017111ndash8

41 Schmidt R de Reus MA Scholtens LH van den Berg LH van den Heuvel MPSimulating disease propagation across white matter connectome reveals anatomicalsubstrate for neuropathology staging in amyotrophic lateral sclerosis Neuroimage2016124(part A)762ndash769

42 Braak H Brettschneider J Ludolph AC Lee VM Trojanowski JQ Del Tredici KAmyotrophic lateral sclerosis a model of corticofugal axonal spread Nat Rev Neurol20139708ndash714

43 Christidi F Karavasilis E Rentzos M et al Hippocampal pathology in amyotrophiclateral sclerosis selective vulnerability of subfields and their associated projectionsNeurobiol Aging 201984178ndash188

44 Westeneng H-J Debray TPA Visser AE et al Prognosis for patients with amyo-trophic lateral sclerosis development and validation of a personalised predictionmodel Lancet Neurol 201817423ndash433

Appendix (continued)

Name Location Contribution

ReneeWalhout MDPhD

UMC Utrecht theNetherlands

Design and conception of studyacquisition and analysis of datadrafting the manuscript forintellectual content

Kevin vanVeenhuijzenMD

UMC Utrecht theNetherlands

Major role in acquisition of dataand preparation of figures

Harold HGTan MD

UMC Utrecht theNetherlands

Major role in acquisition of data

Jil M MeierPhD

UMC Utrecht theNetherlands

Major role in acquisition of datarevised the manuscript forintellectual content

Leonhard ABakker MSc

UMC Utrecht theNetherlands

Major role in acquisition ofcognitive data

JeroenHendrikseMD PhD

UMC Utrecht theNetherlands

Major role in acquisition of data

Michael A vanEs MD PhD

UMC Utrecht theNetherlands

Design and conception of study

Jan H VeldinkMD PhD

UMC Utrecht theNetherlands

Design and conception of study

Martijn P vanden HeuvelPhD

VU UniversityAmsterdam theNetherlands

Design and conception of study

Leonard H vanden Berg MDPhD

UMC Utrecht theNetherlands

Design and conception of studyrevised the manuscript forintellectual content

NeurologyorgN Neurology | Volume 94 Number 24 | June 16 2020 e2603

45 Shellikeri S Karthikeyan V Martino R et al The neuropathological signature ofbulbar-onset ALS a systematic review Neurosci Biobehav Rev 201775378ndash392

46 Rajagopalan V Pioro EP Distinct patterns of cortical atrophy in ALS patients with orwithout dementia an MRI VBM study Amyotroph Lateral Scler Front Degener201415216ndash225

47 Agosta F Ferraro PM Riva N et al Structural brain correlates of cognitive andbehavioral impairment in MND Hum Brain Mapp 2016371614ndash1626

48 Lule D Bohm S Muller HP et al Cognitive phenotypes of sequential staging inamyotrophic lateral sclerosis Cortex 2018101163ndash171

49 Cox SR Dickie DA Ritchie SJ et al Associations between education and brainstructure at age 73 years adjusted for age 11 IQ Neurology 2016871820ndash1826

50 Frodl TS Koutsouleris N Bottlender R et al Depression-related variation in brainmorphology over 3 years effects of stress Arch Gen Psychiatry 2008651156ndash1165

e2604 Neurology | Volume 94 Number 24 | June 16 2020 NeurologyorgN

DOI 101212WNL0000000000009498202094e2592-e2604 Published Online before print May 15 2020Neurology

Hannelore K van der Burgh Henk-Jan Westeneng Reneacutee Walhout et al sclerosis

Multimodal longitudinal study of structural brain involvement in amyotrophic lateral

This information is current as of May 15 2020

ServicesUpdated Information amp

httpnneurologyorgcontent9424e2592fullincluding high resolution figures can be found at

References httpnneurologyorgcontent9424e2592fullref-list-1

This article cites 50 articles 10 of which you can access for free at

Citations httpnneurologyorgcontent9424e2592fullotherarticles

This article has been cited by 1 HighWire-hosted articles

Subspecialty Collections

httpnneurologyorgcgicollectionvolumetric_mriVolumetric MRI

httpnneurologyorgcgicollectionmriMRI

httpnneurologyorgcgicollectiondwiDWI

httpnneurologyorgcgicollectionamyotrophic_lateral_sclerosis_Amyotrophic lateral sclerosisfollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpwwwneurologyorgaboutabout_the_journalpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnneurologyorgsubscribersadvertiseInformation about ordering reprints can be found online

ISSN 0028-3878 Online ISSN 1526-632XWolters Kluwer Health Inc on behalf of the American Academy of Neurology All rights reserved Print1951 it is now a weekly with 48 issues per year Copyright Copyright copy 2020 The Author(s) Published by

reg is the official journal of the American Academy of Neurology Published continuously sinceNeurology