Review: Modelling the pathology and behaviour of frontotemporal dementia

Frontotemporal dementia (FTD) encompasses a collection of clinically and pathologically diverse neurological disorders. Clinical features of behavioural and language dysfunction are associated with neurodegeneration, predominantly of frontal and temporal cortices. Over the past decade, there have been significant advances in the understanding of the genetic aetiology and neuropathology of FTD which have led to the creation of various disease models to investigate the molecular pathways that contribute to disease pathogenesis. The generation of in vivo models of FTD involves either targeting genes with known disease‐causative mutations such as GRN and C9orf72 or genes encoding proteins that form the inclusions that characterize the disease pathologically, such as TDP‐43 and FUS. This review provides a comprehensive summary of the different in vivo model systems used to understand pathomechanisms in FTD, with a focus on disease models which reproduce aspects of the wide‐ranging behavioural phenotypes seen in people with FTD. We discuss the emerging disease pathways that have emerged from these in vivo models and how this has shaped our understanding of disease mechanisms underpinning FTD. We also discuss the challenges of modelling the complex clinical symptoms shown by people with FTD, the confounding overlap with features of motor neuron disease, and the drive to make models more disease‐relevant. In summary, in vivo models can replicate many pathological and behavioural aspects of clinical FTD, but robust and thorough investigations utilizing shared features and variability between disease models will improve the disease‐relevance of findings and thus better inform therapeutic development.     

Review: Clinical,genetic and neuroimaging features of frontotemporal dementia

Frontotemporal dementia (FTD) is a heterogeneous group of disorders causing neurodegeneration within a network of areas centred on the frontal and temporal lobes. Clinically, patients present with behavioural symptoms (behavioural variant FTD) or language disturbance (primary progressive aphasia), although there is an overlap with motor neurone disease and atypical parkinsonian disorders. Whilst neuroimaging commonly reveals abnormalities in the frontal and temporal lobes, a closer review identifies a more complex picture with variable asymmetry of neuronal loss, widespread subcortical involvement and in many cases more posterior cortical atrophy. An autosomal‐dominant genetic disorder is found in around a third of people with mutations in progranulin, C9orf72 and the microtubule‐associated protein tau being the commonest causes. In the other two‐thirds, the disorder is sporadic, although recent genome‐wide association studies have started to identify genetic risk factors within this group. Much of this knowledge has been understood only in the past 10 years and so this review will discuss the current knowledge about the clinical, genetic and neuroimaging features of FTD.     

Frontotemporal dementia: Pick type

The status of Pick's disease within the concept of frontotemporal dementia (FTD) is unclear. Some researchers have defined Pick's disease as FTD with Pick bodies. Alternatively, the confusion may be clarified by using the term Pick body dementia (PBD) rather than by using the term Pick's disease in a narrow sense. Pick body dementia is characterized by a prominent frontotemporal lobar atrophy, gliosis, severe neuronal loss, ballooned neurons, and the presence of neuronal inclusions called Pick bodies. In recent years, studies of Pick body dementia have advanced from the standpoint of the tau pathology. Tau-positive glial inclusions as well as neuronal inclusions have been observed in PBD and its related disorders. Various forms of FTD have been proposed based on the presence of neuronal or glial inclusions. We propose a new variant of FTD termed ‘glial tangle-predominant type’. More research is required to understand the tau abnormalities in the various forms of FTD.     

Clinic-Based Cases with Frontotemporal Dementia Show Increased Cerebrospinal Fluid Tau and High Apolipoprotein E 4 Frequency, but No Tau Gene Mutations

Frontotemporal dementia (FTD) belongs to a group of neurodegenerative disorders known as tauopathies, characterized by intracellular aggregation of hyperphosphorylated tau protein in the brain. Some tauopathies, like Alzheimer's disease (AD), consistently show increased levels of tau protein in cerebrospinal fluid (CSF). However, similar studies in FTD populations have shown variable results, although mutations in the tau gene are identified as causes of disease in certain FTD families. In the present study, a Swedish clinic-based FTD population was investigated with respect to CSF tau levels, apolipoprotein E (APOE) genotype distribution and occurrence of mutations in the tau gene. CSF tau levels were significantly increased among FTD patients (534 ± 235 pg tau/ml, P < 0.001) (n = 47) compared to controls (316 ± 137 pg tau/ml) (n = 51). Furthermore, a strong increase in the APOE 4 allele frequency was found in the FTD population, as 52% were 4 carriers, compared to 21% of the controls. However, no mutations in the tau gene were identified. These findings support the present notion of a common pathogenic pathway in the disease processes for several tauopathies, with both APOE 4 and CSF tau being a pathological link between the different disorders. Furthermore, we conclude that mutations in the tau gene are a rare cause of FTD.     

Frontotemporal degenerative dementias

Frontotemporal dementia (FTD) is a clinical term now used for a group of dementing neurodegenerative disorders marked clinically by general predominance of symptoms referable to the frontal lobes. Symptoms may include changes in language ability with expressive or receptive aphasia, and changes in personality, including disinhibited, obsessive, hyperoral, hypersexual, repetitive, and perseverative behaviors. Memory is often affected, but is usually not a primary symptom. Characteristically, analytic, navigational, and other visuospatial abilities may be quite preserved. The clinical symptomatology of FTD contrasts with the general predominance of temporal and parietal symptoms in the most common neurodegenerative disorder, Alzheimer's disease (AD). FTD symptoms also differ from those seen with the second most common neurodegenerative disorder, Lewy Body Dementia (LBD), which while having substantial clinical and pathological overlap with AD, often shows additional clinical visuospatial (occipital) and parkinsonian symptomatology. FTD likely constitutes the third most-common cause of degenerative dementia. Clinical overlap can sometimes make it difficult to distinguish individual cases from AD. Genetic studies have shown that FTD, like AD, occurs in both familial and nonfamilial types, and a proportion of FTD cases arise from autosomal dominant genetic disorders due to mutations in the tau gene. Some recent formulations include in the family of frontal degenerative disorders other dementias with frontal symptomatology such as progressive supranuclear palsy (PSP) and corticobasal degeneration (CBD). Lending support to such a grouping, neuropathological studies show a group of histologies, sometimes including PSP and CBD, to be the substrates for clinical FTD.     

Is early‐onset clinically different from late‐onset frontotemporal dementia?

Background and purpose: Frontotemporal dementia (FTD) is the second most common neurodegenerative dementia in the young age after Alzheimer disease. Recent improvement in diagnostic assessment suggests that it is more common than previously, although with a great heterogeneity in clinical presentation. The different clinical patterns related to age of disease onset in behavioural variant FTD (bvFTD) have been fairly studied. Aim of the study was to evaluate whether age at disease onset modulate the heterogeneity of either cognitive impairment or behavioural disturbances in patients affected by bvFTD. Methods: One hundred and thirty‐four patients with bvFTD entered the study. Age at onset and demographic characteristics were carefully recorded. Each patient underwent a wide neuropsychological and behavioural standardized assessment, as well as a brain SPECT perfusion imaging study. Results: Behavioural variant FTD were subdivided into four groups according to the age at onset. The four quartile groups did not differ for demographic characteristics and family history for dementia. Global cognitive impairment as well as analysis of the different cognitive domains and behavioural patterns were comparable. Conclusions: These findings provide evidence that the clinical heterogeneity of bvFTD is not explained by age at disease onset. Further studies are needed.     

Clinical entity of frontotemporal dementia with motor neuron disease

Non‐Alzheimer‐type dementias occur in association with a variety of pathological conditions that include a group of diseases characterized by atrophy of the frontal and temporal lobes. Frontotemporal dementia (FTD) is a clinical entity that comprises at least two distinct diseases: Pick's disease with Pick bodies and frontotemporal lobar degeneration with ubiquitin‐positive inclusions (FTLD‐U). The vast majority of FTLD‐U is now referred to as FTLD‐TDP, following the recent discovery of TAR DNA‐binding protein of 43 kDa (TDP‐43) as the major constituent of the ubiquitin‐positive inclusions. FTLD‐TDP, but not Pick's disease with Pick bodies, is often associated with motor neuron disease (MND). MND is a group of diseases in which the central nervous system lesions were long believed to be confined to the motor neuron system. In other words, MND was not considered to be associated with other neurological symptoms such as dementia. Nevertheless, more than 200 FTD cases associated with clinical MND have been reported in Japan since 1964. Neuropathologically, MND in such FTD cases was essentially similar to MND in cases without dementia. The combination of FTD and MND was so characteristic that we considered these cases comprise a unique clinicopathological subgroup of FTD. FTD with MND and the classical MND without dementia share the occurrence of ubiquitinated TDP‐43‐positive inclusions, a finding that could be a key to unlock the pathological backgrounds of both diseases.     

‘Limits and current knowledge of Pick’s disease: its differential diagnosis’ A translation of the 1957 Delay,Brion, Escourolle article

This article is a translation of a French article by Delay, Brion, and Escourolle. In a seminal article published in French in 1957 these authors summarized the work of previous researchers and reviewed a wide sample of frontotemporal dementia (FTD) cases formerly referred to as Pick’s disease. The authors were among the first to define the critical clinical and anatomical differences between Alzheimer’s disease (AD) and FTD and they even delineated distinctive FTD subtypes making possible the advances that now constitute the base of our studies. Reviewing their work allows us to appreciate the progress research has made.     

Typical cerebral metabolic patterns in neurodegenerative brain diseases

The differential diagnosis of neurodegenerative brain diseases on clinical grounds is difficult, especially at an early disease stage. Several studies have found specific regional differences of brain metabolism applying [¹⁸F]‐fluoro‐deoxyglucose positron emission tomography (FDG‐PET), suggesting that this method can assist in early differential diagnosis of neurodegenerative brain diseases. We have studied patients who had an FDG‐PET scan on clinical grounds at an early disease stage and included those with a retrospectively confirmed diagnosis according to strictly defined clinical research criteria. Ninety‐six patients could be included of which 20 patients with Parkinson's disease (PD), 21 multiple system atrophy (MSA), 17 progressive supranuclear palsy (PSP), 10 corticobasal degeneration (CBD), 6 dementia with Lewy bodies (DLB), 15 Alzheimer's disease (AD), and 7 frontotemporal dementia (FTD). FDG PET images of each patient group were analyzed and compared to18 healthy controls using Statistical Parametric Mapping (SPM5). Disease‐specific patterns of relatively decreased metabolic activity were found in PD (contralateral parietooccipital and frontal regions), MSA (bilateral putamen and cerebellar hemispheres), PSP (prefrontal cortex and caudate nucleus, thalamus, and mesencephalon), CBD (contralateral cortical regions), DLB (occipital and parietotemporal regions), AD (parietotemporal regions), and FTD (frontotemporal regions). The integrated method addressing a spectrum of various neurodegenerative brain diseases provided means to discriminate patient groups also at early disease stages. Clinical follow‐up enabled appropriate patient inclusion. This implies that an early diagnosis in individual patients can be made by comparing each subject's metabolic findings with a complete database of specific disease related patterns. © 2010 Movement Disorder Society     

Mouse models of frontotemporal dementia

The pace of discovery in frontotemporal dementia (FTD) has accelerated dramatically with the discovery of new genetic causes and pathological substrates of the disease. MAPT/tau, GRN/progranulin, and C9ORF72 have emerged as common FTD genes, and TARDBP/TDP‐43, VCP, FUS, and CHMP2B have been identified as less common genetic causes. TDP‐43 and FUS have joined tau as common neuropathological substrates of the disease. Mouse models provide an important tool for understanding the role of these molecules in FTD pathogenesis. Here, we review recent progress with mouse models based on tau, TDP‐43, progranulin, VCP, and CHMP2B. We also consider future prospects for FTD models, including developing new models to address unanswered questions. There are also opportunities for capitalizing on conservation of the salience network, which is selectively vulnerable in FTD, and the availability of FTD‐related behavioral paradigms to analyze mouse models of the disease. Ann Neurol ANN NEUROL 2012;72:837–849     

Frontotemporal dementia: Its definition,differential diagnosis,and management

Background: Frontotemporal dementia (FTD) is defined as primary neurodegeneration of the anterior temporal and/or frontal lobes resulting in a group of associated conditions marked by changes in cognition, language, personality, and social functioning. FTD was previously thought to be a rare disease. However, researchers report that FTD is the third most common form of dementia. Because adults with FTD have deficits in language, cognition, and behaviour, familiarity with FTD subtypes, associated deficits, and currently available management strategies is warranted.

Aims: The aims of this tutorial are (a) to define frontotemporal dementia including behavioural and language characteristics of the three clinically distinct FTD subtypes (frontotemporal variant, nonfluent progressive aphasia, semantic dementia); (b) to identify similarities and differences between FTD and Alzheimer's dementia; and (c) to discuss management strategies for patients with FTD.

Main Contribution: Different subtypes and presentations of FTD as well as the neurological, behavioural, and language symptoms that have been consistently identified are reviewed. Behavioural and language symptoms of the two FTD subtypes with primary language disturbances (nonfluent progressive aphasia and semantic dementia) are also reviewed. Patients with FTD are frequently misdiagnosed as presenting with Alzheimer's dementia due to limitations in the literature describing the differing profiles of the two populations. When considering neurological changes, behavioural changes, language and communication behaviours, and disease progression, these patient populations are distinct and easily differentiated. Finally, management strategies are discussed. Although there is no cure for FTD, medical intervention can address some of the associated symptoms, and behavioural techniques may manage the client's environment and prolong communication abilities.

Conclusions: General discussion seeks to differentially diagnose FTD dementia from Alzheimer's dementia as well as to clarify the language and communication symptoms of FTD subtypes. Future research directions are suggested for developing evidence‐based direct and indirect management strategies.     

You stole my food! Eating alterations in frontotemporal dementia

ABSTRACT

Patients with different types of dementia may exhibit pathological eating habits, including food fads, hyperphagia, or even ingestion of inanimate objects. Several findings reveal that such eating alterations are more common in patients with frontotemporal dementia (FTD) than other types of dementia. Moreover, eating alterations may differ between the two variants of the disease, namely the behavioral variant and semantic dementia (SD). In this review, we summarized evidences regarding four areas: eating and body weight alterations in FTD, the most common assessment methods, anatomical correlates of eating disorders, and finally, proposed underlying mechanisms. An increasing understanding of the factors that contribute to eating abnormalities may allow first, a better comprehension of the clinical features of the disease and second, shed light on the mechanism underlying eating behaviors in the normal population.     

Is motor neuron disease‐inclusion dementia a forme fruste of amyotrophic lateral sclerosis with dementia? An autopsy case further supporting the disease concept

We report the autopsy findings of a 62‐year‐old man who exhibited progressive FTD 10 years before the appearance of muscle weakness and wasting, and who died approximately 11 years after onset of the symptoms. Degeneration and atrophy of the frontal and temporal lobes, which contained ubiquitin‐positive neuronal inclusions and dystrophic neurites, were evident. Circumscribed degeneration affecting the hippocampal CA1‐subiculum border zone was also a feature. Moreover, degeneration was present in both the upper and lower motor neuron systems, the latter being more severely affected. A few lower motor neurons were found to contain the cytoplasmic inclusions characteristic of ALS (i.e. Bunina bodies and ubiquitin‐positive skeins). Also of interest was the presence of pallidonigroluysian atrophy, which appeared to be responsible for the chorea‐like involuntary movements that developed in this patient approximately 2 months before death. The clinical and pathological features of our patient further support the idea that motor neuron disease‐inclusion dementia (MND‐ID), which has been classified as a pathological subgroup of FTD, is a forme fruste of ALS with dementia. In other words, if patients with MND‐ID live long enough, they may develop ALS.     

Self-injurious behavior in frontotemporal dementia

Abstract

Frontotemporal dementia (FTD) may be one of the most common neurodegenerative dementias with onset before age 65. Investigators have only recently characterized FTD, which encompasses Pick's disease as well as non-specific degeneration of the frontal lobes. Neuropsychiatric symptoms are the main features of FTD and include disinhibition, apathy, obsessive-compulsive behavior, and the Kluver-Bucy syndrome. We report a patient with FTD who developed prominent self-injurious behavior (SIB) consisting of persistent self-biting and hair-pulling. Stereotypical, repetitive SIB usually occurs in patients with autism or mental retardation; SIB is rarely discussed in patients with dementia. This report extends the neuropsychiatric spectrum of FTD to encompass SIB and discusses the potential mechanisms for SIB in these patients. The most likely contributing source for their self-injury is the development of obsessive-compulsive behaviors in the presence of coincident self-directed behaviors. Hyperorality from anterior temporal Involvement and the release of primitive grooming behavior from frontal degeneration are self-directed behaviors that contribute to the specific manifestations of SIB among FTD patients. The pharmacological management of SIB emphasizes drugs that work through opioid, serotonergic, or dopaminergic systems.     

Brain 18F-FDG and 11C-PiB PET findings in two siblings with FTD/ALS associated with the C9ORF72 repeat expansion

The C9ORF72 hexanucleotide expansion is a major pathological expansion pattern found in patients with frontotemporal dementia (FTD) and amyotrophic lateral sclerosis (C9FTD/ALS). We describe a patient in whom early clinical evaluation, MRI and fluorodeoxyglucose (FDG) positron emission tomography (PET) findings failed to definitively differentiate between FTD and Alzheimer’s disease (AD), whereas ¹¹C-Pittsburgh compound B (PiB) PET was negative for amyloid pathology. He later developed ALS symptoms, and post mortem neuropathological findings were diagnostic of FTD-ALS, while no findings suggested AD. His sister was diagnosed with FTD, and the C9ORF72 expansion was detected in both siblings. We conclude that ¹¹C-PiB PET imaging may help the early differential diagnosis between AD and FTD, including C9FTD/ALS.     

Neuropsychiatric symptoms of progressive supranuclear palsy in a dementia clinic

Background: Progressive supranuclear palsy (PSP) is a neurodegenerative disease characterized by supranuclear gaze palsy, postural instability, akinesia and other parkinsonism. Recently, the relationship between PSP and frontotemporal dementia (FTD) has been recognized, which includes clinical, pathological, biochemical and genetic features. However, there have been few studies that directly compared neuropsychiatric symptoms between PSP and FTD. The aim of the present study was to investigate comprehensive psychiatric and behavioural symptoms in PSP and compared them with those in FTD. Methods: Patients with PSP (n = 10) and FTD (n = 13) were selected on the basis of inclusion/exclusion criteria from a consecutive series in the dementia clinic of Kumamoto University Hospital. We assessed their comprehensive neuropsychiatric features by using the Neuropsychiatric Inventory (NPI), the Stereotypy Rating Inventory (SRI) and a specific antisocial behaviour checklist. Results: There were no significant differences in the total NPI and NPI subscale scores between the two groups. Both groups showed quite a similar pattern in the features of neuropsychiatric symptoms: apathy showed the highest score, followed by aberrant motor behaviour and disinhibition. The PSP group was significantly lower in the total SRI and eating and cooking behaviour scores than those in the FTD group. The prevalence of antisocial behaviours in PSP (50%) was equal to those in the FTD group (46%). Conclusions: In a dementia clinic, the neuropsychiatric profile in patients with PSP closely resembled those in the FTD group. The present results suggest that PSP should be considered as not only a movement disorder, but also a disorder with a wide range of neuropsychiatric symptoms.     

Frontotemporal dementia linked to chromosome 3 (FTD-3) – current concepts and the detection of a previously unknown branch of the Danish FTD-3 family

Background:  Among patients with onset of dementia below the age of 65 years, frontotemporal dementia (FTD) is the second most prevalent cause, secondary only to Alzheimer's disease. Recent advances in understanding the heterogeneous genetic background for different clinical and neuropathological entities of FTD have involved identification of several new causative genes.
Methods and results:  We report the finding of a truncating mutation in the CHMP2B gene (c.532-1G>C) in a patient with early onset dementia. The patient was previously not known to be related to the single Danish pedigree known to have this specific mutation. Subsequently he has turned out to represent a new branch of the family with several affected individuals.
Discussion:  Our findings highlight the need for awareness of the CHMP2B mutation and associated clinical phenotype for neurological assessment in Denmark. Further, we discuss recent advances and current concepts in the understanding of CHMP2B -related dementia.     

Frontotemporal lobar degeneration: clinical and pathological relationships

Frontotemporal lobar degeneration (FTLD) encompasses a heterogeneous group of clinical syndromes that include frontotemporal dementia (FTD), frontotemporal dementia with motor neurone disease (FTD/MND), progressive non-fluent aphasia (PNFA), semantic dementia (SD) and progressive apraxia (PAX). Clinical phenotype is often assumed to be a poor predictor of underlying histopathology. Advances in immunohistochemistry provide the opportunity to re-examine this assumption. We classified pathological material from 79 FTLD brains, blind to clinical diagnosis, according to topography of brain atrophy and immunohistochemical characteristics. There were highly significant relationships to clinical syndrome. Atrophy was predominantly frontal and anterior temporal in FTD, frontal in FTD/MND, markedly asymmetric perisylvian in PNFA, asymmetric bitemporal in SD and premotor, parietal in PAX. Tau pathology was found in half of FTD and all PAX cases but in no FTD/MND or SD cases and only rarely in PNFA. FTD/MND, SD and PNFA cases were ubiquitin and TDP-43 positive. SD cases were associated with dystrophic neurites without neuronal cytoplasmic or intranuclear inclusions (FTLD-U, type 1), FTD/MND with numerous neuronal cytoplasmic inclusions (FTLD-U, type 2 ) and PNFA with neuronal cytoplasmic inclusions, dystrophic neurites and neuronal intranuclear inclusions (FTLD-U, type 3). MAPT mutations were linked to FTD and PGRN mutations to FTD and PNFA. The findings demonstrate predictable relationships between clinical phenotype and both topographical distribution of brain atrophy and immunohistochemical characteristics. The findings emphasise the importance of refined delineation of both clinical and pathological phenotype in furthering understanding of FTLD and its molecular substrate.     

Diagnostic patterns of regional atrophy on MRI and regional cerebral blood flow change on SPECT in young onset patients with Alzheimer's disease, frontotemporal dementia and vascular dementia

OBJECTIVES: Alzheimer's disease (AD), frontotemporal dementia (FTD) and vascular dementia (VaD) are the three most common causes of young onset dementias. Most neuroimaging studies of these disorders have involved comparisons with normal controls. The aims of this study were to examine the clinical diagnostic value of magnetic resonance imaging (MRI) and single photon emission computed tomography (SPECT) (in combination and in isolation) in the differentiation of one form of dementia from another from amongst a group of AD, FTD and VaD. METHODS: T1 weighted MRI images and 99mTc-HMPAO SPECT images were obtained from consecutive patients with FTD (n=21), AD (n=23) and VaD (n=20) and rated visually by experienced neuroradiologists and nuclear medicine physicians. RESULTS: Asymmetrical atrophy was seen only in FTD. Frontotemporal dementia patients were the most atrophic whereas severe atrophy was rarely observed in VaD. Severe frontal atrophy (unilaterally or bilaterally) and/or asymmetrical atrophy on MRI is highly diagnostic (sensitivity 0.71, specificity 0.93, LR 10.24) of FTD from within a group of FTD and non-FTD (AD, VaD) patients. Mild or severe parietal atrophy with severe reduction in parietal regional cerebral blood flow on SPECT is diagnostic (sensitivity 0.71, specificity 0.76, LR 3.02) of AD from within a group of AD and non-AD (VaD, FTD) patients. CONCLUSION: Anatomical (MRI) and functional (SPECT) imaging provide different information and a combination of these modalities improves diagnostic specificity.     

Longitudinal white matter changes in frontotemporal dementia subtypes

Frontotemporal dementia is a degenerative brain condition characterized by focal atrophy affecting the frontal and temporal lobes predominantly. Changes in white matter with disease progression and their relationship to grey matter atrophy remain unknown in FTD. This study aimed to establish longitudinal white matter changes and compare these changes to regional grey matter atrophy in the main FTD subtypes. Diffusion and T₁‐weighted images were collected from behavioral‐variant FTD (bvFTD: 12), progressive non‐fluent aphasia (PNFA: 10), semantic dementia (SD: 11), and 15 controls at baseline and 12 months apart. Changes in white matter integrity were established by fractional anisotropy, mean, axial and radial diffusivity measurements using tract‐based spatial statistics. Patterns of cortical grey matter atrophy were measured using voxel‐based morphometry. At baseline, bvFTD showed severe cross‐sectional changes in orbitofrontal and anterior temporal tracts, which progressed to involve posterior temporal and occipital white matter over the 12‐month. In PNFA, cross‐sectional changes occurred bilaterally in frontotemporal white matter (left > right), with longitudinal changes more prominent on the right. Initial white matter changes in SD were circumscribed to the left temporal lobe, with longitudinal changes extending to bilateral frontotemporal tracts. In contrast, progression of grey matter change over time was less pronounced in all FTD subtypes. Mean diffusivity was most sensitive in detecting baseline changes while fractional anisotropy and radial diffusivity revealed greatest changes over time, possibly reflecting different underlying pathological processes with disease progression. Our results indicate that investigations of white matter changes reveal important differences across FTD syndromes with disease progression. Hum Brain Mapp 35:3547–3557, 2014. © 2013 Wiley Periodicals, Inc .