ARTICLE OPEN ACCESS

Neuropathologic features associated with basalforebrain atrophy in Alzheimer diseaseStefan J Teipel MD H-Christian Fritz BSc and Michel J Grothe PhD

for the Alzheimerrsquos Disease Neuroimaging Initiative

Neurologyreg 202095e1301-e1311 doi101212WNL0000000000010192

Correspondence

Dr Teipel

stefanteipel

meduni-rostockde

AbstractObjectiveTo study the neuropathologic correlates of cholinergic basal forebrain (BF) atrophy as de-termined using antemortem MRI in the Alzheimer disease (AD) spectrum

MethodsWe determined associations between BF volume from antemortem MRI brain scans andpostmortem assessment of neuropathologic features including neuritic plaques neurofibrillarytangles (NFTs) Lewy body (LB) pathology and TDP-43 in 64 cases of the AlzheimerrsquosDisease Neuroimaging Initiative cohort For comparison we assessed neuropathologic featuresassociated with hippocampal and parahippocampal gyrus atrophy In addition to region ofinterestndashbased analysis we determined the association of neuropathologic features with wholebrain gray matter volume using regionally unbiased voxel-based volumetry

ResultsBF atrophy was associated with Thal amyloid phases (95 confidence interval [CI] minus049 tominus001 p = 0049) and presence of LB pathology (95 CI minus054 to minus006 p = 0015) as well aswith the degree of LB pathology within the nucleus basalisMeynert (95CI minus054 to minus007 p =0025) These effects were no longer significant after false discovery rate (FDR) correctionHippocampal atrophy was significantly associated with the presence of TDP-43 pathology(95 CI minus061 to minus017 p = 0003 surviving FDR correction) in addition to dentate gyrusNFT load (95 CI minus049 to minus001 p = 0044 uncorrected) Voxel-based analysis confirmedspatially restricted effects of Thal phases and presence of LB pathology on BF volume

ConclusionsThese findings indicate that neuropathologic correlates of regional atrophy differ substantiallybetween different brain regions that are typically involved in AD-related neurodegenerationincluding different susceptibilities to common comorbid pathologies

From the German Center for Neurodegenerative Diseases (DZNE) (SJT MJG) Department of Psychosomatic Medicine (SJT H-CF) University Medicine Rostock Germany andInstituto de Biomedicina de Sevilla (IBiS) (MJG) Unidad de Trastornos del Movimiento Hospital Universitario Virgen del RocıoCSICUniversidad de Sevilla Spain

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

Data used in preparation of this article were obtained from the Alzheimerrsquos Disease Neuroimaging Initiative (ADNI) database (adniloniuscedu) As such the investigators within theADNI contributed to the design and implementation of ADNI andor provideddata but did not participate in analysis or writing of this report A complete listing of ADNI investigators canbe found at linkslwwcomWNLB160

The Article Processing Charge was funded by DZNE eV Germany

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

Copyright copy 2020 The Author(s) Published by Wolters Kluwer Health Inc on behalf of the American Academy of Neurology e1301

MRI-based volumetry is recognized as a topographic markerfor staging and monitoring the progression of Alzheimerdisease (AD)1 To understand the pathologic substrate ofMRI-based volumetric changes previous studies have in-vestigated associations of neuropathologic measures of pri-mary AD pathology and common comorbid pathologies withantemortem MRI volumetry A systematic review identified27 studies until mid-2015 on antemortem MRI and neuro-pathology2 The large majority of these and subsequentstudies focused on the hippocampus and other medial tem-poral lobe (MTL) structures such as the parahippocampalgyrus and entorhinal cortex

Besides degeneration of the hippocampus and MTL cholin-ergic system degeneration is a key event in AD pathogenesis3

Over the last decade advances in MRI data analysis have en-abled in vivo volumetry of the cholinergic basal forebrain(BF)45 the central site of origin of cholinergic projections tothe cerebral cortex and limbic system in the human brain6

Studies using in vivo BF volumetry to study BF degeneration inthe course of AD have shown that this brain region is severelyatrophied in AD dementia and may even precede MTL de-generation in the prodromal phase of AD57 Neuropathologicstudies suggest altered expression of neurotrophic receptors oncholinergic neurons as well as neurofibrillary tangle (NFT)accumulation and dystrophic cell shrinkage but not frank lossof cholinergic BF neurons in MCI and early AD cases8

Whereas the main pathologic correlate of cholinergic BF de-generation in AD is considered to be NFT pathology9 thisbrain region also shows high susceptibility to other neurode-generative pathologies particularly Lewy body (LB) pathol-ogy10 but also TDP-43 pathology11 In addition to these localpathologic changes distinct associations between cholinergicsystem degeneration and accumulation of cortical amyloidpathology have been described in postmortem data12 andthese associations could recently also be demonstrated in vivousing MRI-measured BF volume and amyloid-sensitive PETimaging1314 However associations between in vivo BF volumeon MRI and global and local neuropathologic lesions areunexplored

We used available antemortem MRI and postmortem neuro-pathologic examination data from 64 individuals enrolled in theAlzheimerrsquos Disease Neuroimaging Initiative (ADNI) to de-termine neuropathologic features associated with MRI-basedBF volume and compared these to neuropathologic correlates

of more widely studied MRI-based MTL volumes In additionwe extended the currently limited evidence on the regionalspecificity of neuropathology associations with whole brain graymatter (GM) volume using unbiased voxel-based volumetry

MethodsData sourceData used in the preparation of this article were obtained fromthe ADNI database (adniloniuscedu) The ADNI waslaunched in 2003 by the National Institute on Aging the Na-tional Institute of Biomedical Imaging and Bioengineering theFood and Drug Administration private pharmaceutical com-panies and nonprofit organizations with the primary goal oftesting whether neuroimaging neuropsychologic and otherbiologic measurements can be used as reliable in vivo markersof AD pathogenesis A fuller description of ADNI and up-to-date information is available at adni-infoorg

Standard protocol approvals registrationsand patient consentsAll procedures performed in the ADNI studies involving hu-man participants were in accordance with the ethical standardsof the institutional research committees and with the 1964Helsinki declaration and its later amendments Written in-formed consent was obtained from all participants andor au-thorized representatives and the study partners before anyprotocol-specific procedures were carried out in the ADNIstudies

Study participantsWe retrieved the last available MRI scans of 64 ADNI partic-ipants who had come to autopsy between 2007 and 2017Detailed inclusion criteria for the antemortem diagnostic cat-egories can be found at the ADNI web site (adniloniuscedumethods) Briefly cognitively normal (CN) participants hadMini-Mental State Examination (MMSE) scores between 24and 30 (inclusive) had a Clinical Dementia Rating (CDR)score of 0 did not have depression mild cognitive impairment(MCI) or dementia and reported no subjective memoryconcerns (SMCs) One participant fulfilled identical criteria tothe CN cases with the exception that he had a self-reportedpersistent memory complaint quantified by the CognitiveChange Index (with a total score from the first 12 items ge16)As this participant lacked objective cognitive decline he wasintegrated into the CN group Participants with MCI had

GlossaryAD = Alzheimer diseaseADNI = Alzheimerrsquos Disease Neuroimaging Initiative BF = basal forebrainCDR =Clinical DementiaRating CERAD = Consortium to Establish a Registry for Alzheimerrsquos Disease CI = confidence interval CN = cognitivelynormal FDR = false discovery rate GM = gray matter LB = Lewy bodyMCI = mild cognitive impairmentMMSE = Mini-Mental State ExaminationMNI = Montreal Neurological InstituteMPRAGE = magnetization-prepared rapid gradient echoMTL = medial temporal lobeNFT = neurofibrillary tangleNbM = nucleus basalis Meynert ROI = region of interest SMC =subjective memory concern TIV = total intracranial volume WM = white matter

e1302 Neurology | Volume 95 Number 10 | September 8 2020 NeurologyorgN

MMSE scores between 24 and 30 (inclusive) a SMC reportedby the participant informant or clinician objective memoryloss measured by education-adjusted scores on delayed recallCDR = 05 absence of significant levels of impairment in othercognitive domains essentially preserved activities of daily liv-ing and an absence of dementia All MCI cases were classifiedas amnestic MCI according to the ADNI guidelines At in-clusion into the ADNI cohort patients with AD dementia hadinitial MMSE scores between 20 and 26 (inclusive) had a CDR= 05 or 10 with impaired activities of daily living and fulfilledNational Institute of Neurological and Communicative Dis-orders and StrokendashAlzheimerrsquos Disease and Related DisordersAssociation criteria for clinically probable AD15

Neuropathological assessmentsAll neuropathologic evaluations in the ADNI cohort are per-formed through the central laboratory of the ADNI neuropa-thology core (adniloniusceduaboutcore-container) andassess a wide range of AD and non-AD neuropathologic le-sions including both established neuropathologic criteria aswell as detailed regional assessments within 22 cortical andsubcortical brain regions16 The neuropathologic proceduresfollow previously established guidelines17 that are captured inthe format of the NeuropathologyData FormVersion 10 of theNational AlzheimerCoordinating Center (alzwashingtoneduNONMEMBERNPnpform10pdf)

We used established rating scales for AD neuropathologicchange represented by Thal amyloid phases Braak NFT stag-ing and the Consortium to Establish a Registry for AlzheimerrsquosDisease (CERAD) score for density of neuritic plaques andfurther assessed whether comorbid LB pathology TDP-43pathology or hippocampal sclerosis was present or not Inaddition we used regional neuropathologic rating scores for thenucleus basalis Meynert (NbM) dentate gyrus entorhinalcortex and parahippocampal gyrus corresponding to ourMRI-based regions of interest (ROIs) in the BF and the MTL TheADNI autopsy data lack an assessment of CA1 neuron loss sothat we did not include CA1 in our evaluation NbM pathologywas sampled on a coronal slice at the level of the crossing of theanterior commissure and was not selective for cholinergicneuron populations within this area Therefore we did notinclude NbM neuronal loss scores into our analysis Thedentate gyrus and parahippocampal gyrus were both sampledon a coronal section at the level of the lateral geniculate nucleusThe entorhinal cortex was sampled on a separate more anteriorsection Within the target regions we focused on scores forneuronal loss NFT neuritic plaques TDP-43 immunoreactiveneuronal cytoplasmic inclusions and LB pathology assessedon semiquantitative scales (none mild moderate and severe)Further details on the neuropathologic assessments can befound in the supplementary materials section (supplementarydata available from Dryad doiorg105061dryaddfn2z34x6)

Neuropsychological assessmentThe MMSE18 was available to assess degree of cognitive im-pairment and established ADNI composite scores for episodic

memory and executive function were used for the evaluation ofdomain-specific cognitive performance1920

Imaging data acquisitionMRI data were acquired on multiple 15T and 3T MRIscanners using scanner-specific T1-weighted sagittal 3Dmagnetization-prepared rapid gradient echo (MPRAGE) se-quences In order to increase signal uniformity across themulticenter scanner platforms original MPRAGE acquisi-tions in ADNI undergo standardized image preprocessingcorrection steps Detailed information on the different im-aging protocols employed across ADNI sites and standardizedimage preprocessing steps can be found on the ADNI website(adniloniuscedumethods)

Imaging data processingImaging data were processed by using statistical parametricmapping (SPM12 Wellcome Trust Center for Neuro-imaging) and the CAT123 toolbox (dbmneurouni-jenadecat) implemented in MATLAB R2018a (MathWorks NatickMA) First MRI scans were automatically segmented intoGM white matter (WM) and CSF partitions of 15 mmisotropic voxel size using the Adaptive Maximum A Posterior(AMAP) segmentation routine of the CAT12 toolbox Theresulting GM and WM partitions of each participant in nativespace were then high-dimensionally registered to the Mon-treal Neurological Institute (MNI) reference template usingthe DARTEL algorithm21 Individual flow fields resultingfrom the DARTEL registration to the reference template wereused to warp the GM segments and voxel values were mod-ulated for volumetric changes introduced by the high-dimensional normalization such that the total amount ofGM volume present before warping was preserved

Extraction of imaging features from BF andhippocampal ROIsThe cholinergic nuclei are not directly visible on currentstructuralMRI contrasts and no comprehensive set of externallandmarks has been identified that could be used for indirectmanual delineation of the cholinergic BF onMRI scans In thecurrent study we localized the cholinergic space of the BFbased on a cytoarchitectonic map of BF cholinergic nuclei inMNI space derived from combined histology and MRI of apostmortem brain as described previously4

An ROI mask for the hippocampus was obtained by manualdelineation of the hippocampus in the reference templatefollowing the harmonized protocol for hippocampussegmentation22

The parahippocampal gyrus ROI was derived from the cor-responding label in the ldquoHammersrdquo maximum probabilitystructural atlas23 which includes both the parahippocampalgyrus proper and the entorhinal cortex

Individual GM volumes of the ROIs were extracted auto-matically from the warped GM segments by summing up the

NeurologyorgN Neurology | Volume 95 Number 10 | September 8 2020 e1303

modulated GM voxel values within the respective ROI masksin the reference space For further analyses the extractedregional GM volumes were scaled by the total intracranialvolume (TIV) calculated as the sum of total volumes of theGM WM and CSF partitions

StatisticsAll variables used for analysis except the volume measureswere directly retrieved from the ADNI database Statisticalanalyses were conducted with R (The R Foundation forStatistical Computing) as implemented in R Studio Version11453 unless otherwise specified

Volume measurements were compared between clinical di-agnostic groups using analysis of covariance models controlling

for age at MRI sex scanner field strength and the temporalinterval betweenMRI and death (considering a p value of lt 005as significant)

Associations between neuropathologic scores and volumemeasures were analyzed using partial correlations control-ling for age sex MRI scanner field strength and the tem-poral interval between MRI and death In additionassociations between neuropathologic scores and neuro-psychological measures were analyzed using partial correla-tion controlling for age sex and the temporal intervalbetween the last cognitive testing and death We report re-sults at a p value lt 005 uncorrected for multiple compari-sons as well as following false discovery rate (FDR)correction24

Table Sample characteristics at time of last MRI

Clinical diagnosisat last MRI FMa

Age y mean(SD) rangeb

MMSEmean (SD)c

Education ymean (SD)d

Time between MRI and deathy average (SD) rangee

Controls 32 808 (77) 68ndash88 284 (31) 180 (346) 21 (24) 05ndash41

MCI 110 844 (44) 75ndash90 236 (40) 1482 (23) 26 (14) 05ndash71

AD dementia 1036 800 (68) 58ndash92 171 (61) 162 (24) 20 (18) 01ndash78

Abbreviations AD = Alzheimer disease MCI = mild cognitive impairment MMSE = Mini-Mental State Examinationa Not significantly different between groups χ2 (2 df) = 52 p = 008b Not significantly different between groups F259 = 20 p = 015c Significantly different between groups F259 = 132 p lt 0001d Not significantly different between groups F259 = 29 p = 0064e Not significantly different between groups F259 = 041 p = 067

Figure 1 Partial correlations between regional volumes and global neuropathologic scores

Partial correlations between regional volumes and global neuropathologic scores controlling for age sex and interval between MRI and death Numbers inblue indicate nonsignificant effects numbers in red indicate significant effects at a p value lt 005 uncorrected for multiple comparisons and numbers in redwith an indicate significant effects at a false discovery ratendashcorrectedp value lt 005 All volumes are normalized to the individual total intracranial volume BF= basal forebrain volume according to ref 4 Braak stages = Braak stages for neurofibrillary pathology Hip = hippocampal volume according to ref 22 LBbinary = presence of Lewy body pathology (assessed according to dementia with Lewy bodies criteria as brainstem predominant limbic neocorticalamygdala predominant) yesno Neur plaques = Consortium to Establish a Registry for Alzheimerrsquos Disease score for density of neocortical neuritic plaquesParahip = parahippocampal gyrus from the Hammers atlas23 TDP binary = presence of TDP-43 pathology (assessed in the following regions spinal cordamygdala hippocampus entorhinal cortex neocortical) yesno Thal phases = Thal phases for amyloid plaques

e1304 Neurology | Volume 95 Number 10 | September 8 2020 NeurologyorgN

In a complementary voxel-wise analysis in SPM12 we de-termined the effect of neuropathologic scores that showedsignificant associations with ROI volumes on whole brain GMchanges The analyses were controlled for age at MRI sexMRI scanner field strength and the temporal interval betweenMRI and death and results are reported using a voxel-wisethreshold of p lt 0001 uncorrected

Data availabilityNeuropathologic data demographic data and MRI scans areavailable through the ADNI data repository accessible free ofcost after registration (adniloniuscedudata-samplesac-cess-data) Volumetric measures of BF and hippocampus havebeen obtained from the MRI scans using established pro-cessing pipelines as described above The anonymized volu-metric measures containing ADNI cohort identifiers will beshared by request from any qualified investigator

ResultsSample characteristicsTwo participants one with a clinical diagnosis of AD dementiaand one with a clinical diagnosis of MCI were excluded fromfurther analysis because the processing of their MRI scans

yielded inaccurate segmentation most likely due to the severedegree of atrophy The remaining 62 participants had receivedthe following clinical diagnoses at the time of their last MRIscan 5 CN older participants (1 person reported SMCs) 11participants with MCI and 46 participants with AD dementiaDiagnostic groups did not differ significantly with respect toage sex education or the temporal interval between MRI scanand death (on average 21 [SD 18] years range 01ndash78 years)As expected MMSE scores differed significantly betweengroups The demographic details can be found in the tableClinical diagnosis was significantly associated with BF andhippocampal (F259 gt 31 p lt 005) but not parahippocampalgyrus volumes The significant effects were no longer preservedafter controlling for age sex and MRI scanner field strength

Neuropathologic characteristicsWith respect to neuropathologic scores in the selected brainregions the NbM showed a high degree of NFT with a rela-tively low degree of neuritic plaques Similarly the para-hippocampal gyrus had a high degree of NFT but a low degreeof neuronal loss and amoderate degree of neuritic plaques Theentorhinal cortex had a high degree of NFT and a moderate tohigh degree of neuritic plaques and neuronal loss whereas thedentate gyrus had a comparably low degree of NFT neuriticplaques and neuronal loss Local α-synuclein and TDP-43 was

Figure 2 Partial correlations between regional volumes and region-specific neuropathologic scores

Partial correlations between regional volumes andneuropathologic scoreswithin the brain ROIs controlling for age sex scanner field strength and temporalinterval between MRI and death Numbers in blue indicate nonsignificant effects numbers in red indicate significant effects at a p value lt 005 uncorrectedfor multiple comparisons and numbers in red with an indicate significant effects at a false discovery ratendashcorrected p value lt 005 All volumes arenormalized to the individual total intracranial volume BF = basal forebrain volume according to ref 4 Hip = hippocampal volume according to ref 22 DG =dentate gyrus (autopsy region) ENTX = entorhinal cortex (autopsy region) NA = not assessed NMB = nucleus basalis of Meynert (autopsy region) NeurPlaques = neuritic plaques (0 [none] to 3 [ge20 per visual field]) Neuron loss = neuronal loss (0 [none] to 3 [severe]) Parahip = parahippocampal gyrus volumefrom the Hammers atlas23 PHG = parahippocampal gyrus (autopsy region) Synucl = Lewy bodies (α-synucleinndashpositive inclusions) (0 [none] to 3 [gt10 ornumerous LB inclusions per times10 field]) Tau = neurofibrillary tangles (0 [none] to 3 [ge20 per visual field]) TDP NCI = TDP-43 immunoreactive neuronalcytoplasmatic inclusions (0 [none] to 3 [gt10 or numerous TDP-43 inclusions per times10 field])

NeurologyorgN Neurology | Volume 95 Number 10 | September 8 2020 e1305

detected only in a subgroup of cases with TDP-43 being absentin the NbM (with the exception of a single case) and α-synu-clein being absent in the dentate gyrus

Associations between global neuropathologicscores and BF and MTL volumesDetailed nuisance-controlled partial correlations betweenglobal neuropathologic scores and MRI volumes can be foundin figure 1 Thal amyloid phases and presence of LB pathologywere significantly associated with BF volume (r = minus026 [95confidence interval (CI) minus049 to minus001] and r = minus032 [95CI minus054 to minus006] respectively) but effects were not signifi-cant after FDR correction Braak tangle stages and CERADscore for neuritic plaques were not significantly associated withBF or MTL volumes Presence of TDP-43 and LB pathologywere significantly associated with hippocampal volume (r =minus041 [minus061 to minus017] and r = minus036 [minus057 to minus011]) bothsurviving FDR correction at p lt 005 Hippocampal sclerosiswas only present in 5 cases (unilateral in 3 cases laterality notassessed in 2 cases) The association between hippocampalvolume and presence of TDP-43 pathology remained signifi-cant even after additionally controlling for the presence ofhippocampal sclerosis (r = minus035 [minus056 to minus010])

Associations between regionalneuropathologic scores and BF andMTL volumesDetailed associations can be found in figure 2 BF volume wassignificantly associated with the degree of LB pathology in the

NbM (r = minus032 [95 CI minus054 to minus007]) and hippocampalvolume was significantly correlated with dentate gyrus NFT (r =minus027 [95 CI minus049 to minus001]) and TDP-43 scores (r = minus030[95CIminus052 tominus004]) These effectswere no longer significantafter FDR correction The parahippocampal gyrus volume (in-cluding the entorhinal cortex) was significantly associated withneuronal loss ratings in the entorhinal cortex (r = minus053 [95 CIminus069 to minus031]) and parahippocampal gyrus (r = 049 [95 CI067 to minus026]) surviving FDR correction at p lt 005

Neuropathologic scores and whole brainGM analysisIn voxel-based regression analyses across the whole braincontrolling for age sex TIV and interval between MRI anddeath presence of LB pathology was associated with reducedGM volume in anterior cingulate GM and lateral temporaland insular cortex at p lt 0001 Presence of TDP-43 pathology(vs no TDP-43 pathology) was associated with reduced GMvolume in left and right hippocampus (figure 3) We found nosignificant voxel-wise effect of Thal phases on GM volumes atp lt 0001 However in a spatially restricted analysis within theBF mask only and applying a more liberal threshold of p lt001 we found significant effects of both Thal phases andpresence of LB pathology but not of TDP-43 indicatingconsistent effects with the ROI analyses for this region (figure4 A and B)

In a complementary analysis25 we did not find significanteffects of Braak stages on GM volume at p lt 0001 Dentategyrus NFT score showed no significant association with GMvolumes at p lt 0001 Restricting the search region to thebilateral hippocampus and applying a more liberal thresholdof p lt 001 we found significant effects of dentate gyrus NFTscore on bilateral posterior hippocampal volume most pro-nounced in the right hippocampus (figure 4C) Entorhinalcortex neuron loss showed a significant association with re-duced GM in extended MTL areas (figure 5) and very similarregional associations were observed for parahippocampalgyrus neuron loss (data not shown) Adding field strength ofMRI scanner as additional covariate did not markedly alter theresults

Association of neuropathologic scores andneuropsychological measuresWe found a significant association of the degree of LB pa-thology in the NbM with ADNI executive function score (r =minus029 [95 CI minus052 to minus001]) but no effect on ADNImemory or MMSE score NbM NFT scores were not corre-lated with any of the 3 cognitive measures Thal amyloidphases Braak stages CERAD neuritic plaque scores andpresence of LB pathology were significantly associated withglobal cognition memory and executive function Except forthe association between CERAD neuritic plaque score andexecutive function these effects were preserved after FDRcorrection In contrast presence of TDP-43 pathology wasnot significantly associated with any of the cognitive out-comes Detailed results are shown in figure 6

Figure 3 Presence of TDP-43 pathology and gray mattervolume

Sagittal coronal and axial view (right of image is right of brain) of clusters ofat least 20 voxels surpassing a significance threshold of 0001 uncorrectedfor multiple comparisons for the effect of TDP-43 pathology (absent vspresent) on gray matter volume controlling for age sex total intracranialvolume and the temporal interval between MRI scan and death The colorbar in the right lower corner indicates the T values

e1306 Neurology | Volume 95 Number 10 | September 8 2020 NeurologyorgN

DiscussionAt an uncorrected level of significance Thal amyloid phaseswere associated with BF volume At the same time the NbMitself exhibited very little neuritic plaques Thus the degree ofatrophy of the NbMwas associated with amyloid pathology incortical areas but not with plaque load in the NbM itself Thisagrees with postmortem studies that showed distinct associ-ations between cholinergic system degeneration and accu-mulation of cortical amyloid pathology12 Similarly in vivostudies could also demonstrate a regionally focused associa-tion of cortical amyloid accumulation with BF volume usingamyloid-sensitive PET imaging and MRI-measured BFvolume1314 Together these findings suggest that BF atrophyon MRI associates with remote cortical amyloid pathologyrather than amyloid load within the BF itself which couldreflect neurotoxic effects of cortical amyloid pathology me-diated by uptake of β-amyloid proteins into cholinergic nerveterminals or a disruption of nerve growth factor signaling826

In another chain of events findings from experimental animalstudies suggest that cholinergic degeneration may contributeto increased accumulation of cortical amyloid pathology due

to altered amyloid processing and clearance mechanismssubserved by cholinergic signaling27 Future in vivo studiesassessing the respective biomarkers longitudinally may help togain more insight into the directionality of the interactionsbetween cholinergic BF atrophy and cortical amyloid pa-thology in the human brain

BF volume was also related to the general presence of LBpathology as well as to the degree of LB pathology in theNbM One reason for the strong effect of LB pathology maybe its high prevalence in the current cohort where 31 ofcases exhibited neocortical and 19 limbic predominant LBpathology consistent with the reportedly high comorbidity ofAD and LB pathology28 In a combined analysis the effect ofNbM LB pathology on BF volume rendered the effect of Thalphases nonsignificant (supplementary data available fromDryad doiorg105061dryaddfn2z34x6) suggesting astronger association of LB pathology than cortical amyloidpathology with BF volume decline

Of note none of the associations between neuropathologiclesions and BF volume survived FDR correction This

Figure 4 Association of neuropathologic scores with spatially restricted gray matter regions

(A) Association of Thal phases with basal forebrain restricted region at an uncorrected p lt 001 Coronal view (right of image is right of brain) of clusters of atleast 5 voxels (red) surpassing a significance threshold of 001 uncorrected for multiple comparisons for the effect of Thal phases within the basal forebrainsearch region (in green) controlling for age sex total intracranial volume (TIV) and the temporal interval between MRI scan and death Numbers on the topindicate the y-coordinate in Montreal Neurological Institute (MNI) space (anterior posterior direction) (B) Association of Lewy body pathology with basalforebrain restricted region at an uncorrected p lt 001 Coronal view (right of image is right of brain) of clusters of at least 5 voxels (red) surpassing asignificance threshold of 001 uncorrected for multiple comparisons for the effect of Lewy body pathology (absent vs present) within the basal forebrainsearch region (in green) controlling for age sex TIV and the temporal interval betweenMRI scan and death Numbers on the top indicate the y-coordinate inMNI space (anterior posterior direction) (C) Association of dentate gyrus neurofibrillary tangle (NFT) score with bilateral hippocampus restricted region at anuncorrectedp lt 001 Coronal view (right of image is right of brain) of clusters of at least 10 voxels (red) surpassing a significance threshold of 001 uncorrectedfor multiple comparisons for the effect of dentate gyrus NFT score within the hippocampus search region (in green) controlling for age sex TIV and thetemporal interval between MRI scan and death Numbers on the top indicate the y-coordinate in MNI space (anterior posterior direction)

NeurologyorgN Neurology | Volume 95 Number 10 | September 8 2020 e1307

indicates only moderate effect sizes and some risk of false-positive findings Therefore in the absence of previous studieson neuropathologic correlates of BF atrophy these findingsrequire independent confirmation

BF volume was not associated with NFT load in the NbMNFTs have been found to occur already at Braak stage I in theBF cholinergic neurons29 and NFT pathology in the NbMwas found already in persons without dementia in their thirddecade of life followed by increasing morphologic alterationsof axons and cellular tangles preceding neuronal loss30 Inagreement with the early build-up of NFT pathology in theNbM in previous studies 70 of our cases had semi-quantitative ratings indicating moderate to severe NFT pa-thology in the NbM Thus the limited variation in NFT levelsmay have obscured possible associations between NbM NFTload and BF volume BF neuron loss is consistently beingfound in advanced stages of AD but evidence points to dys-trophic shrinkage and dysfunction of cholinergic neuronsrather than to a significant reduction of cholinergic neuronnumbers in the BF in prodromal and early AD stages31 In thecurrent study we could not specifically analyze associationsbetween BF volume and cholinergic cell loss as the availableneuropathologic sampling of the NbM did not differentiatebetween cholinergic and noncholinergic neurons

Presence of LB pathology was associated with worse cognitiveperformance and the degree of LB pathology in the NbMwasspecifically associated with executive dysfunction This agrees

with the role of cholinergic BF integrity for executive functionfound in clinical studies

In contrast to the BF hippocampal and parahippocampalgyrus atrophy was not associated with Thal amyloid phasesThal phases include both diffuse and dense neuritic plaquesand previous studies found no association of diffuse plaqueswith total brain32 or hippocampal33 volumes Thal phaseswere also not found to correlate with antemortem cogni-tion34 However even CERAD score for density of neuriticplaques was not associated with hippocampal or para-hippocampal gyrus volumes in our study different fromprevious studies that reported associations of neuritic plaqueswith whole brain and MTL volumes2 Our current findingshowever agree with findings from in vivo neuroimagingstudies reporting a lack of association between PET-measuredcortical amyloid load and hippocampal atrophy1314

Consistent with one previous study using antemortemMRI35

and several studies using postmortem MRI3637 we found asignificant association between neuronal loss ratings in MTLregions and MTL volumes both using ROI and voxel-basedanalyses In addition NFT score in the dentate gyrus wassignificantly associated with hippocampal volume Significantassociations of NFTs with hippocampal volume agree withprevious studies2538 However both presence of TDP-43pathology in any location and the degree of TDP-43 pathol-ogy in the dentate gyrus were associated with MTL volumesas well in agreement with earlier reports39 Interestingly theassociation with TDP-43 pathology even remained whencontrolling for hippocampal sclerosis and Braak tau stagesand additional voxel-based analysis revealed that the effect ofTDP-43 was regionally specific for the bilateral hippocampusThe effect of TDP-43 was only reduced when accounting forlocal NFT load in the dentate gyrus This finding agrees with acentral role of comorbid TDP-43 pathology for hippocampalatrophy within the AD spectrum4041 In contrast we did notfind an association of TDP-43 pathology with BF atrophy inour study In fact there was only a single patient in our cohortwho exhibited TDP-43 pathology in the NbM and in-terestingly this patient also had hippocampal sclerosisTDP-43 pathology in the BF has recently been reported toaccompany hippocampal sclerosis but not AD or otherneurodegenerative diseases11 In the current cohort TDP-43pathology was present at least to a mild extent in 37 of caseswhich agrees with the prevalence of TDP-43 in previousstudies ranging between 33 and 52 in cases with ADpathology4042ndash44 Also consistent with these previous studiesthe regional distribution of TDP-43 inclusions in the currentsample was almost exclusively focused on MTL regions in-cluding hippocampus entorhinal cortex and amygdala andonly 6 patients had neocortical TDP-43 inclusions in additionto MTL involvement These data underscore the strong as-sociation of regional distribution of TDP-43 inclusions withthe regional pattern of brain atrophy consistent with previousreports on amygdala and hippocampal involvement withTDP-43 pathology45 and the current discussion on limbic

Figure 5 Entorhinal cortex neuron loss and gray mattervolume

Sagittal coronal and axial view (right of image is right of brain) of clusters of atleast 20 voxels surpassing a significance threshold of 0001 uncorrected formultiple comparisons for the effect of entorhinal cortex neuron loss on graymatter volume controlling for age sex total intracranial volume and thetemporal interval betweenMRI scan and death The color bar in the right lowercorner indicates the T values

e1308 Neurology | Volume 95 Number 10 | September 8 2020 NeurologyorgN

predominant TDP-43 pathology in older individuals that hasrecently been termed limbic-predominant age-related TDP-43 encephalopathy (LATE)46

Braak stages had no significant effects in the ROI- or voxel-based analyses A previous voxel-based analysis found a sig-nificant difference in the degree of atrophy between Braak VVI stages and Braak 0ndashII stages but not between Braak IIIIVstages and Braak 0ndashII stages25 Different from this previousstudy we could not find a significant difference in degree ofatrophy between Braak VVI vs Braak 0ndashII stages We cannotexplain this discrepancy The number of participants withinthe different stages was comparable between the previous (n =20 Braak 0ndashII and n = 59 Braak VndashVI) and our study (n = 16Braak 0ndashII and n = 43 Braak VndashVI) Different from the pre-vious study we also controlled for the temporal interval be-tween MRI and death However even when we removed thiscovariate we still did not find robust effects

Our study has strengths and limitations One major limitationis the lack of assessment of CA1 neuron loss in the ADNIautopsy data that hindered a comprehensive evaluation of therole of CA1 degeneration which was found to be a key driverof hippocampal atrophy in previous studies35 Howeveraccording to neuropathologic Braak staging neurofibrillarydegeneration of CA1 and the entorhinal cortex should behighly collinear Another limitation is that the pathologic as-sessment of the NbM was only sampled from a single coronalsection at the level of the crossing of the anterior commissureand did not selectively assess cholinergic neurons preventing

any inference on specific subsections of the NbM A furtherlimitation was the quality of the MRI scans Due to the ad-vanced stage of disease in this autopsy cohort some of thecases exhibited severe brain atrophy which can affect thequality of brain segmentation This led to the exclusion of 2 ofthe 64 cases from the analysis due to insufficient scan qualityfor brain segmentation Finally the overrepresentation ofmenin the current cohort limits its translational value to agedcohorts where women are known to predominate

A strength of our study is the regional assessment ofneuropathologyndashbrain volume associations including stan-dardized global neuropathologic rating scales and detailedlocal assessments of neuropathology within our target re-gions as well as ROI-based and complementary voxel-wiseanalyses of GM volume Particularly specific regional as-sessments of NbM pathology had not been examined yet inneuropathologyndashimaging correlation studies Voxel-basedanalysis has only been used in a few previous studies onassociations between MRI volumes and quantitative neu-ropathologic features25394748 or pathologically confirmeddiagnoses4950 Here we complemented our primary analy-ses focused on BF and MTL ROIs with a comprehen-sive voxel-wise search for associations of AD and non-ADneuropathologic features with brain-wide GM volumereductions

The current findings confirm previous evidence for an asso-ciation of BF volume with deposition of amyloid in corticalassociation areas1314 representing input areas of cholinergic

Figure 6 Partial correlations between cognitive measures and global neuropathologic scores

Partial correlations between cognitive measures and global neuropathologic scores controlling for age sex and the interval between cognitive testing and deathNumbers inblue indicatenonsignificant effectsnumbers in red indicate significant effects at ap valuelt005 uncorrected formultiple comparisons andnumbers inredwith an indicate significant effects at a false discovery ratendashcorrected p value lt 005 ADNI_EF = ADNI composite executive function score ADNI_Mem= ADNIcomposite memory score Braak stages = Braak stages for neurofibrillary pathology LB total = presence of Lewy body pathology (assessed according to dementiawith Lewy bodies criteria as brainstem predominant limbic neocortical amygdala predominant) yesno MMSE =Mini-Mental State Examination Neur plaques =Consortium to Establish a Registry for Alzheimerrsquos Disease score for density of neocortical neuritic plaques TDP total = presence of TDP-43 pathology (assessed inthe following regions spinal cord amygdala hippocampus entorhinal cortex neocortical) yesno Thal phases = Thal phases for amyloid plaques

NeurologyorgN Neurology | Volume 95 Number 10 | September 8 2020 e1309

projections At the same time LB pathology but not NFTpathology was associated with BF atrophy pointing to thehigh relevance of pathologic comorbidity for the expression ofregional atrophy in the AD spectrum The validity of thesefindings is supported by the robust replication of previousreports on the association of TDP-43 pathology with hippo-campal atrophy in AD cases43 The question of traceability ofa biomarker linking the in vivo measured marker with theneuropathologic reference standard is of high relevance forthe AD field to enhance our ability to identify treatmenttargets and measure target engagement with valid biomarkers

AcknowledgmentMJG is supported by the ldquoMiguel Servetrdquo program [CP1900031] of the Spanish Instituto de Salud Carlos III (ISCIII-FEDER) The authors thankManuela Neumann Departmentof Neuropathology of the University Medicine TubingenGermany for discussions

Study fundingData collection and sharing for this project was funded bythe Alzheimerrsquos Disease Neuroimaging Initiative (ADNI)(NIH grant U01 AG024904) and DOD ADNI (Departmentof Defense award number W81XWH-12-2-0012) ADNI isfunded by the National Institute on Aging the NationalInstitute of Biomedical Imaging and Bioengineering andthrough contributions from the following Alzheimerrsquos As-sociation Alzheimerrsquos Drug Discovery Foundation AraclonBiotech BioClinica Inc Biogen Idec Inc Bristol-MyersSquibb Company Eisai Inc Elan Pharmaceuticals Inc EliLilly and Company EuroImmun F Hoffmann-La RocheLtd and its affiliated company Genentech Inc FujirebioGE Healthcare IXICO Ltd Janssen Alzheimer Immuno-therapy Research ampDevelopment LLC Johnson amp JohnsonPharmaceutical Research amp Development LLC MedpaceInc Merck amp Co Inc Meso Scale Diagnostics LLCNeuroRx Research Neurotrack Technologies NovartisPharmaceuticals Corporation Pfizer Inc Piramal ImagingServier Synarc Inc and Takeda Pharmaceutical CompanyThe Canadian Institutes of Health Research is providingfunds to support ADNI clinical sites in Canada Privatesector contributions are facilitated by the Foundation for theNational Institutes of Health (fnihorg) The grantee orga-nization is the Northern California Institute for Researchand Education and the study is coordinated by the Alz-heimerrsquos Disease Cooperative Study at the University ofCalifornia San Diego ADNI data are disseminated by theLaboratory for Neuro Imaging at the University of SouthernCalifornia

DisclosureThe authors report no disclosures relevant to the manuscriptGo to NeurologyorgN for full disclosures

Publication historyReceived by Neurology November 26 2019 Accepted in final formMarch 9 2020

References1 Dubois B Feldman HH Jacova C et al Advancing research diagnostic criteria for

Alzheimerrsquos disease the IWG-2 criteria Lancet Neurol 201413614ndash6292 Dallaire-Theroux C Callahan BL Potvin O Saikali S Duchesne S Radiological-

pathological correlation in Alzheimerrsquos disease systematic review of antemortemmagnetic resonance imaging findings J Alzheimers Dis 201757575ndash601

3 Bohnen NI Grothe MJ Ray NJ Muller M Teipel SJ Recent advances in cholinergicimaging and cognitive decline revisiting the cholinergic hypothesis of dementia CurrGeriatr Rep 201871ndash11

4 Kilimann I Grothe M Heinsen H et al Subregional basal forebrain atrophy inAlzheimerrsquos disease a multicenter study J Alzheimers Dis 201440687ndash700

5 Grothe M Heinsen H Teipel SJ Atrophy of the cholinergic basal forebrain over theadult age range and in early stages of Alzheimerrsquos disease Biol Psychiatry 201271805ndash813

6 Mesulam MM The systems-level organization of cholinergic innervation in the hu-man cerebral cortex and its alterations in Alzheimerrsquos disease Prog Brain Res 1996109285ndash297

7 Schmitz TW Nathan Spreng R Alzheimerrsquos Disease Neuroimaging Initiative Basalforebrain degeneration precedes and predicts the cortical spread of Alzheimerrsquos pa-thology Nat Commun 2016713249

8 Mufson EJ Ginsberg SD Ikonomovic MD DeKosky ST Human cholinergic basalforebrain chemoanatomy and neurologic dysfunction J Chem Neuroanat 200326233ndash242

9 Mesulam M Cholinergic aspects of aging and Alzheimerrsquos disease Biol Psychiatry201271760ndash761

10 Liu AK Chang RC Pearce RK Gentleman SM Nucleus basalis of Meynert revisitedanatomy history and differential involvement in Alzheimerrsquos and Parkinsonrsquos diseaseActa Neuropathol 2015129527ndash540

11 Cykowski MD Takei H Van Eldik LJ et al Hippocampal sclerosis but not normalaging or Alzheimer disease is associated with TDP-43 pathology in the basal forebrainof aged persons J Neuropathol Exp Neurol 201675397ndash407

12 Potter PE Rauschkolb PK Pandya Y et al Pre- and post-synaptic cortical cholinergicdeficits are proportional to amyloid plaque presence and density at preclinical stagesof Alzheimerrsquos disease Acta Neuropathol 201112249ndash60

13 Grothe MJ Ewers M Krause B Heinsen H Teipel SJ Alzheimerrsquos Disease Neuro-imaging Initiative Basal forebrain atrophy and cortical amyloid deposition in non-demented elderly subjects Alzheimers Dement 201410S344ndashS353

14 Kerbler GM Fripp J Rowe CC et al Basal forebrain atrophy correlates with amyloidbeta burden in Alzheimerrsquos disease Neuroimage Clin 20157105ndash113

15 McKhann G Drachman D Folstein M Katzman R Price D Stadlan EM Clinicaldiagnosis of Alzheimerrsquos disease report of the NINCDS-ADRDAWork Group underthe auspices of the Department of Health and Human Services Task Force on Alz-heimerrsquos disease Neurology 198434939ndash944

16 Franklin EE Perrin RJ Vincent B et al Brain collection standardized neuropatho-logic assessment and comorbidity in Alzheimerrsquos Disease Neuroimaging Initiative 2participants Alzheimers Dement 201511815ndash822

17 Montine TJ Phelps CH Beach TG et al National Institute on Aging-AlzheimerrsquosAssociation guidelines for the neuropathologic assessment of Alzheimerrsquos disease apractical approach Acta Neuropathol 20121231ndash11

18 Folstein MF Folstein SE McHugh PR Mini-mental-state a practical method forgrading the cognitive state of patients for the clinician J Psychiatr Res 197512189ndash198

Appendix 1 Authors

Name Location Contribution

Stefan JTeipel MD

University MedicineRostock and DZNEGermany

Designed and conceptualizedstudy analyzed the datadrafted the manuscript forintellectual content

H-ChristianFritz BSc

University MedicineRostock and DZNEGermany

Interpreted the data revisedthe manuscript for intellectualcontent

Michel JGrothePhD

DZNE Germany Designed and conceptualizedstudy helped in analyzing thedata interpreted the datarevised the manuscript forintellectual content

Appendix 2 Coinvestigators

Coinvestigators are listed at linkslwwcomWNLB160

e1310 Neurology | Volume 95 Number 10 | September 8 2020 NeurologyorgN

19 Crane PK Carle A Gibbons LE et al Development and assessment of a compositescore for memory in the Alzheimerrsquos Disease Neuroimaging Initiative (ADNI) BrainImaging Behav 20126502ndash516

20 Gibbons LE Carle AC Mackin RS et al A composite score for executive functioningvalidated in Alzheimerrsquos Disease Neuroimaging Initiative (ADNI) participants withbaseline mild cognitive impairment Brain Imaging Behav 20126517ndash527

21 Ashburner J A fast diffeomorphic image registration algorithm NeuroImage 20073895ndash113

22 Wolf D Bocchetta M Preboske GM Boccardi M Grothe MJ Alzheimerrsquos DiseaseNeuroimaging Initiative Reference standard space hippocampus labels according tothe EADC-ADNI harmonized protocol utility in automated volumetry AlzheimersDement 201713893ndash902

23 Hammers A Allom R KoeppMJ et al Three-dimensional maximum probability atlasof the human brain with particular reference to the temporal lobe Hum Brain Mapp200319224ndash247

24 Benjamini Y Hochberg Y Controlling the false discovery rate a practical and pow-erful approach to multiple testing J Roy Stat Soc B Met 199557289ndash300

25 Whitwell JL Josephs KA Murray ME et al MRI correlates of neurofibrillary tanglepathology at autopsy a voxel-based morphometry study Neurology 200871743ndash749

26 Baker-Nigh A Vahedi S Davis EG et al Neuronal amyloid-beta accumulation withincholinergic basal forebrain in ageing and Alzheimerrsquos disease Brain 20151381722ndash1737

27 Kolisnyk B Al-Onaizi M Soreq L et al Cholinergic surveillance over hippocampalRNA metabolism and Alzheimerrsquos-like pathology Cereb Cortex 2017273553ndash3567

28 Brenowitz WD Keene CD Hawes SE et al Alzheimerrsquos disease neuropathologicchange Lewy body disease and vascular brain injury in clinic- and community-basedsamples Neurobiol Aging 20175383ndash92

29 Mesulam M Shaw P Mash D Weintraub S Cholinergic nucleus basalis tauopathyemerges early in the aging-MCI-AD continuum Ann Neurol 200455815ndash828

30 Geula C Nagykery N Nicholas A Wu CK Cholinergic neuronal and axonal ab-normalities are present early in aging and in Alzheimer disease J Neuropathol ExpNeurol 200867309ndash318

31 Mufson EJ Ikonomovic MD Counts SE et al Molecular and cellular pathophysi-ology of preclinical Alzheimerrsquos disease Behav Brain Res 201631154ndash69

32 Silbert LC Quinn JF Moore MM et al Changes in premorbid brain volume predictAlzheimerrsquos disease pathology Neurology 200361487ndash492

33 Csernansky JG Hamstra J Wang L et al Correlations between antemortem hip-pocampal volume and postmortem neuropathology in AD subjects Alzheimer DisAssoc Disord 200418190ndash195

34 Serrano-Pozo A Qian J Muzikansky A et al Thal amyloid stages do not significantlyimpact the correlation between neuropathological change and cognition in the Alz-heimer disease continuum J Neuropathol Exp Neurol 201675516ndash526

35 Zarow C Vinters HV Ellis WG et al Correlates of hippocampal neuron number inAlzheimerrsquos disease and ischemic vascular dementia Ann Neurol 200557896ndash903

36 Apostolova LG Zarow C Biado K et al Relationship between hippocampal atrophyand neuropathology markers a 7T MRI validation study of the EADC-ADNI Har-monized Hippocampal Segmentation Protocol Alzheimers Dement 201511139ndash150

37 Bobinski M de Leon MJ Wegiel J et al The histological validation of post mortemmagnetic resonance imaging-determined hippocampal volume in Alzheimerrsquos diseaseNeuroscience 200095721ndash725

38 Thaker AA Weinberg BD Dillon WP et al Entorhinal cortex antemortem corticalthickness and postmortem neurofibrillary tangles and amyloid pathology AJNR Am JNeuroradiol 201738961ndash965

39 Josephs KA Whitwell JL Weigand SD et al TDP-43 is a key player in the clinicalfeatures associated with Alzheimerrsquos disease Acta Neuropathol 2014127811ndash824

40 Robinson AC Davidson YS Horan MA Pendleton N Mann DMA Pathologicalcorrelates of cognitive impairment in the University of Manchester LongitudinalStudy of cognition in normal healthy old age J Alzheimers Dis 201864483ndash496

41 Josephs KA Whitwell JL Tosakulwong N et al TAR DNA-binding protein 43 andpathological subtype of Alzheimerrsquos disease impact clinical features AnnNeurol 201578697ndash709

42 James BDWilson RS Boyle PA Trojanowski JQ Bennett DA Schneider JA TDP-43stage mixed pathologies and clinical Alzheimerrsquos-type dementia Brain 20161392983ndash2993

43 Josephs KA Dickson DW Tosakulwong N et al Rates of hippocampal atrophy andpresence of post-mortem TDP-43 in patients with Alzheimerrsquos disease a longitudinalretrospective study Lancet Neurol 201716917ndash924

44 Ihara R Vincent BD BaxterMR et al Relative neuron loss in hippocampal sclerosis ofaging and Alzheimerrsquos disease Ann Neurol 201884741ndash753

45 Makkinejad N Schneider JA Yu J et al Associations of amygdala volume and shapewith transactive response DNA-binding protein 43 (TDP-43) pathology in a com-munity cohort of older adults Neurobiol Aging 201977104ndash111

46 Nelson PT Dickson DW Trojanowski JQ et al Limbic-predominant age-relatedTDP-43 encephalopathy (LATE) consensus working group report Brain 20191421503ndash1527

47 Kantarci K Murray ME Schwarz CG et al White-matter integrity on DTI and thepathologic staging of Alzheimerrsquos disease Neurobiol Aging 201756172ndash179

48 Josephs KA Whitwell JL Knopman DS et al Abnormal TDP-43 immunoreactivity inAD modifies clinicopathologic and radiologic phenotype Neurology 2008701850ndash1857

49 Hornberger M Wong S Tan R et al In vivo and post-mortem memory circuitintegrity in frontotemporal dementia and Alzheimerrsquos disease Brain 20121353015ndash3025

50 Harper L Bouwman F Burton EJ et al Patterns of atrophy in pathologically con-firmed dementias a voxelwise analysis J Neurol Neurosurg Psychiatry 201788908ndash916

NeurologyorgN Neurology | Volume 95 Number 10 | September 8 2020 e1311

DOI 101212WNL0000000000010192202095e1301-e1311 Published Online before print July 6 2020Neurology

Stefan J Teipel H-Christian Fritz Michel J Grothe et al Neuropathologic features associated with basal forebrain atrophy in Alzheimer disease

This information is current as of July 6 2020

ServicesUpdated Information amp

httpnneurologyorgcontent9510e1301fullincluding high resolution figures can be found at

References httpnneurologyorgcontent9510e1301fullref-list-1

This article cites 50 articles 6 of which you can access for free at

Citations httpnneurologyorgcontent9510e1301fullotherarticles

This article has been cited by 1 HighWire-hosted articles

Subspecialty Collections

httpnneurologyorgcgicollectionmriMRI

httpnneurologyorgcgicollectionmemoryMemory

httpnneurologyorgcgicollectionalzheimers_diseaseAlzheimers diseasefollowing 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