Clinical, Cortical, Subcortical, and White Matter Features of Right Temporal Variant FTD

The distinct clinical and radiological characteristics of right temporal variant FTD have only been recently recognized. Methods: Eight patients with right temporal variant FTD were prospectively recruited and underwent a standardised neuropsychological assessment, clinical MRI, and quantitative neuroimaging. Results: Our voxelwise grey analyses captured bilateral anterior and mesial temporal grey matter atrophy with a clear right-sided predominance. Bilateral hippocampal involvement was also observed, as well as disease burden in the right insular and opercula regions. White matter integrity alterations were also bilateral in anterior temporal and sub-insular regions with a clear right-hemispheric predominance. Extra-temporal white matter alterations have also been observed in orbitofrontal and parietal regions. Significant bilateral but right-predominant thalamus, putamen, hippocampus, and amygdala atrophy was identified based on subcortical segmentation. The clinical profile of our patients was dominated by progressive indifference, decline in motivation, loss of interest in previously cherished activities, incremental social withdrawal, difficulty recognising people, progressive language deficits, increasingly rigid routines, and repetitive behaviours. Conclusions: Right temporal variant FTD has an insidious onset and may be mistaken for depression at symptom onset. It manifests in a combination of apathy, language, and behavioural features. Quantitative MR imaging captures a characteristic bilateral but right-predominant temporal imaging signature with extra-temporal frontal and parietal involvement.


Introduction
Frontotemporal dementia (FTD) encompasses a wide spectrum of neurodegenerative disorders that may be further stratified according to clinical phenotype, genotype, or the underlying molecular pathology [1][2][3][4][5][6].The striking clinical, radiological, genetic, and molecular heterogeneity of FTD is well recognised and clinical subtypes are defined based on unique clinical and radiological features [3,7].On clinical grounds, language-variant and behavioural-variant phenotypes are typically distinguished first before subcategorising patients into specific categories based on detailed neuropsychological data.The overlap with Amyotrophic Lateral Sclerosis is also well recognised with a number of shared radiological and pathological features [8][9][10][11][12][13].Individuals harbouring hexanucleotide GGGCC repeat expansions in C9orf72, in particular, are at risk of developing either ALS, FTD, or ALS-FTD [14,15].ALS-FTD as a distinct entity has been recognised before the discovery of C9orf72 repeat expansions and has been extensively studied through robust neuropsychology, post mortem, and neuroimaging studies [16][17][18].
Semantic variant primary progressive aphasia (svPPA) is an FTD phenotype that clinically manifests as anomia and impaired single-word comprehension [1,19], is radiologically Brain Sci.2024, 14, 806 2 of 20 defined by dominant-hemispheric anterior temporal lobe atrophy [1], and is pathologically characterised by frontotemporal lobar degeneration transactive response DNA-binding protein 43 (FTLD-TDP-43) pathology type C in the majority of cases [20].In recent times, it has become apparent that non-dominant anterior temporal lobe atrophy presents with a distinct clinical phenotype that initially does not meet the classification criteria for svPPA [1,21].A vast range of alternative terminology has been used to describe this entity: 'right temporal variant FTD', 'right temporal semantic dementia', 'right temporal svPPA', and 'right temporal behavioural variant FTD (bvFTD)'.Clinical algorithms have been proposed to differentiate this presentation from other FTD phenotypes and other neurodegenerative disorders [22].A recent study outlined the longitudinal clinical characteristics of this cohort, proposing dedicated classification criteria with streamlined nomenclature highlighting the main symptomatology: 'semantic behavioural variant FTD' (sbvFTD) [23].
The proposed classification criteria for sbvFTD require at least two core criteria: loss of empathy; difficulty identifying and naming people; rigid thought processes or complex compulsions; and at least 2 supportive criteria: object-naming difficulties, spared visuospatial functions, and spared motor speech and phonology [23].It may be a particularly difficult diagnosis early in the course of the disease and often mistaken for psychiatric illnesses [23].The behavioural and language manifestations later progress and overlap with other FTD phenotypes, particularly svPPA and bvFTD [24][25][26][27].It is radiologically defined by non-dominant anterior temporal lobe atrophy with progressive bilateral orbitofrontal cortex, anterior cingulate, and contralateral anterior temporal lobe atrophy [24].FTLD TDP-43 type C is the most commonly reported pathology [21,23,25].While sbvFTD is increasingly recognised as a distinct phenotype, relatively few case series have been published, and it is a particularly challenging diagnosis to establish as atrophy patterns can be challenging to appreciate on standard clinical imaging.Accordingly, our objective is the detailed clinical and radiological profiling of a cohort of patients with sbvFTD using standardised clinical instruments and a standardised quantitative neuroimaging protocol.

Ethics Approval
All aspects of this project were approved by the Ethics Committee of Beaumont Hospital Dublin (REC reference: 08/90), and each participant gave informed consent prior to study enrolment.

Participants
A total of 8 participants with right temporal variant FTD and 100 healthy controls were included in this study.All patients first had standard clinical T1-weighted, FLAIR and DWI MRI imaging, and 4 patients also underwent [ 18 F] FDG PET-CT imaging.A total of 7 patients and 100 healthy controls (Table 1) additionally underwent high-resolution 3D T1-weighted imaging to map patterns of grey matter atrophy and diffusion tensor imaging for quantitative white matter analyses using the same scanner and a standardised neuroimaging protocol described below.Seven patients had a standardised neuropsychological evaluation with the Edinburgh Cognitive and Behavioural ALS Screen (ECAS) [28].Three patients had comorbid Amyotrophic Lateral Sclerosis, fulfilling the El Escorial criteria [29].Exclusion criteria for both patients and controls included prior cerebrovascular events (strokes), prior neurosurgery, traumatic brain injury, malignancies, demyelination, and paraneoplastic syndromes.None of the healthy controls had a family history of dementia, psychiatric diagnoses, or motor neuron disease among their first-and second-degree relatives.

Morphometric analyses
Patterns of grey matter atrophy in the semantic behavioural variant frontotemporal dementia cohort were evaluated by voxel-based morphometry (VBM) in contrast to the cohort of the age-and sex-matched healthy controls.FMRIB's FSL suite was utilised to conduct the VBM analyses.[30][31][32] Standard pre-processing pipelines were implemented with skull-removal (BET) [33], motion-correction, and tissue-type segmentation.Subsequently, grey matter partial volume images were aligned to the MNI152 standard space using affine registration.A study-specific GM template was created thereafter to which the grey matter images from each subject were non-linearly co-registered.Permutationbased non-parametric inference [34] was used for group comparisons, controlling for total intracranial volumes (TIV), sex, age, and education.TIV was calculated by linearly aligning each participant's brain image to the MNI152 standard, and the inverse of the determinant of the affine registration matrix was calculated and multiplied by the size of the template.The threshold-free cluster enhancement (TFCE) method [35] was implemented to correct for multiple comparisons.Resulting statistical maps were visualised in FSLEYES and thresholded at p < 0.01 to characterise focal grey matter vulnerability patterns.

White matter analyses
Pre-processing of the raw diffusion data included eddy current corrections and skull removal before a tensor model was fitted to the data to generate maps of axial diffusivity (AD), fractional anisotropy (FA), radial diffusivity (RD), and mean diffusivity (MD).The tract-based statistics (TBSS) stream [36]of FMRIB's FSL image analysis suite was implemented for non-linear registration and skeletonisation of each subject's images.FA, AD, MD, and RD images were merged into a single 4D image file, and a mean FA mask was created.Permutation-based non-parametric inference was used for the voxelwise comparison of diffusivity parameters between patients with semantic behavioural variant frontotemporal dementia and healthy controls using design matrix-defined contrasts incorporating age, sex, and education as covariates.The threshold-free cluster enhancement (TFCE) method was implemented, and resulting statistical maps were thresholded at p < 0.0125 TFCE family-wise error (FWE).

Subcortical segmentation
The standard anatomical segmentation pipeline of the FreeSurfer image analysis suite [37] was first implemented with 'recon-all', which includes non-parametric nonuniform intensity normalisation, affine registration to the MNI305 atlas, intensity normalisation, skull stripping, automatic subcortical segmentation, linear volumetric registration, neck removal, tessellation of the grey matter-white matter boundary, surface smoothing, inflation to minimise metric distortion, and automated topology correction [38].The automated subcortical segmentation pipeline of FreeSurfer relies on a probabilistic atlas [39].Segmentation accuracy has been individually reviewed for all subjects.Estimated total intracranial volumes (eTIV) were calculated in FreeSurfer using Buckner's approach [40] and subsequently used as a covariate in group comparisons in addition to age, sex, and education.

Data availability
Clinical, genetic, or neuroimaging data from individual patients cannot be made available due to departmental policies, but additional information on data-processing pipelines can be requested from the corresponding author.

Statistics
Group differences in age and education were examined with one-way analysis of variance, and differences in sex and handedness distributions between patients with semantic-behavioural variants and healthy controls were contrasted with chi-square (χ 2 ) tests (Table 1).As described above, non-parametric permutation-based testing was utilised for voxelwise grey and white matter analyses.The design matrices included age, sex, and education as covariates and total intracranial volumes (eTIV) were included as additional covariates as for voxel-based morphometry.The resulting statistical maps were corrected for family-wise error.No statistical analyses were run on PET data.
To test the effect of group on subcortical volumes, a multivariate analysis of covariance (MANCOVA) was conducted with the volumes of individual structures as dependent variables, the study group (HC, sbvFTD) as independent factor and age, sex, education, and eTIV as covariates.In case of a significant multivariate omnibus test, post hoc univariate comparisons were considered significant at p < 0.05, following Bonferroni corrections for multiple comparisons to reduce Type I error.These analyses were conducted using IBM SPSS v. 29.

Grey Matter Atrophy Patterns
The demographic profiles of those included in quantitative neuroimaging analyses are summarised in Table 1.Voxel-based morphometry revealed bilateral anterior and mesial temporal atrophy with right-sided predominance.Patterns of atrophy also affected the bilateral hippocampi and the right insular and opercula region (Figure 1).

Grey Matter Atrophy Patterns
The demographic profiles of those included in quantitative neuroimaging analyse are summarised in Table 1.Voxel-based morphometry revealed bilateral anterior and mesial temporal atrophy with right-sided predominance.Patterns of atrophy als affected the bilateral hippocampi and the right insular and opercula region (Figure 1).

White Matter Patterns
Tract-based spatial statistics captured increased axial diffusivity (AD) in the righ anterior temporal lobe in patients with semantic behavioural variant frontotempora dementia (Figure 2).Similar to the voxel-based morphometry results, fractional anisot ropy (FA) reductions were noted bilaterally in both anterior temporal and sub-insula regions with a marked predominance to the right hemisphere (Figure 3).In addition t the temporal and insular white matter integrity changes, right-predominant orbitofronta and parietal FA alterations were also noted at p < 0.0125 FEW-corrected.Increased radia diffusivity was noted in the right temporal lobe with a clear anterior predominanc (Figure 4), but also some orbitofrontal and sub-insular involvement.Mean diffusivit increases were confined to the right anterior temporal lobe and right insular region (Figure 5).

White Matter Patterns
Tract-based spatial statistics captured increased axial diffusivity (AD) in the right anterior temporal lobe in patients with semantic behavioural variant frontotemporal dementia (Figure 2).Similar to the voxel-based morphometry results, fractional anisotropy (FA) reductions were noted bilaterally in both anterior temporal and sub-insular regions with a marked predominance to the right hemisphere (Figure 3).In addition to the temporal and insular white matter integrity changes, right-predominant orbitofrontal and parietal FA alterations were also noted at p < 0.0125 FEW-corrected.Increased radial diffusivity was noted in the right temporal lobe with a clear anterior predominance (Figure 4), but also some orbitofrontal and sub-insular involvement.Mean diffusivity increases were confined to the right anterior temporal lobe and right insular regions (Figure 5).

Grey Matter Atrophy Patterns
The demographic profiles of those included in quantitative neuroimaging analyses are summarised in Table 1.Voxel-based morphometry revealed bilateral anterior and mesial temporal atrophy with right-sided predominance.Patterns of atrophy also affected the bilateral hippocampi and the right insular and opercula region (Figure 1).

White Matter Patterns
Tract-based spatial statistics captured increased axial diffusivity (AD) in the right anterior temporal lobe in patients with semantic behavioural variant frontotemporal dementia (Figure 2).Similar to the voxel-based morphometry results, fractional anisotropy (FA) reductions were noted bilaterally in both anterior temporal and sub-insular regions with a marked predominance to the right hemisphere (Figure 3).In addition to the temporal and insular white matter integrity changes, right-predominant orbitofrontal and parietal FA alterations were also noted at p < 0.0125 FEW-corrected.Increased radial diffusivity was noted in the right temporal lobe with a clear anterior predominance (Figure 4), but also some orbitofrontal and sub-insular involvement.Mean diffusivity increases were confined to the right anterior temporal lobe and right insular regions (Figure 5).

Concordance with PET
Four patients had additional PET imaging, which revealed a relative concordance between focal PET hypometabolism and volume loss based on visual inspection (Figure 6).

Concordance with PET
Four patients had additional PET imaging, which revealed a relative concordance between focal PET hypometabolism and volume loss based on visual inspection (Figure 6).

Concordance with PET
Four patients had additional PET imaging, which revealed a relative concordance between focal PET hypometabolism and volume loss based on visual inspection (Figure 6).

Subcortical Grey Matter Volume Alterations
There was a significant group effect on subcortical grey matter volumes (Pillai's Trace = 0.684, F (14,88) = 13.608,p < 0.001, η 2 p = 0.684).Significant differences were detected in the volumes of most subcortical structures between patients with sbvFTD and controls after Bonferroni corrections for multiple comparisons: left thalamus (p < 0.001), left caudate (p = 0.014), left putamen (p < 0.001), left pallidum (p < 0.001), left hippocampus (p < 0.001), left amygdala (p < 0.001), right thalamus (p < 0.001), right caudate (p < 0.001), right putamen (p < 0.001), right hippocampus (p < 0.001), right amygdala (p < 0.001), and right accumbens area (p < 0.001).Based on η 2 p values, larger effect sizes were identified for most right hemispheric structures compared to the left ones.The volumetric profile of the two study groups is summarised in Table 2, reporting estimated marginal means and standard error in patients with sbvFTD and healthy controls, as well as univariate effect sizes and corrected p-values (Table 2).

Clinical Profiles
Six patients had a detailed clinical assessment with neuropsychological screening.The main findings are summarised in Table 3. -

Discussion
This study highlights the core and radiological features of sbvFTD.The systematic assessment of a cohort of patients with sbvFTD in a single-centre setting and the computational analyses of MRI data acquired with a standardised radiological protocol allow the description of unifying anatomical features.

Clinical observations
Despite differences in symptom duration, our cohort exhibited relatively unifying clinical features.All cases had initial insight into their deficits.All cases presented with rigid thought processes, executive dysfunction, and varying degrees of prosopagnosia; the majority had verbal semantic loss, obsessive repetitive behaviours, and episodic memory impairment; and some also had a loss of empathy, apathy, disinhibition, alexithymia, and dietary changes.The main cognitive domains affected were executive, language, fluency, and memory.Most cases had anomia with varying levels of impaired comprehension.Surface dyslexia was also observed.All of the patients exhibited progressive indifference and a decline in motivation.The initial indifference and lack of motivation in 3 out of 8 cases were initially mistaken for depression.A loss of interest in previously cherished activities, such as reading, playing golf, and gardening, was a common theme.Incremental social withdrawal was invariably reported, which is likely to be multifactorial in the majority of cases due to loss of interest in friends and relatives, difficulty recognising people, and language deficits.Many of our patients developed an increasingly rigid routine, such as repetitively watching the same movie, going for long drives to the same location, listening to the same music, and taking the train to a specific destination to get an ice cream every day.In addition to the development of a regimented routine, increasingly rigid and obsessive behaviours were exhibited by some, such as only charging their mobile phone to exactly 100%.Difficulty recognising people, including familiar faces, is a common complaint in this cohort, and one patient explained that she mostly identifies people by their voices.Relentlessly evolving language deficits have particularly severe quality of life ramifications, impacting employment, job fulfilment, and enjoyment of social interactions.Increasing difficulty with both low-and high-frequency objects, variable levels of semantic deficits, frequent circumlocution, and perseveration increasingly affect both professional communication at work and informal communication in the community.Even with relatively preserved verbal fluency, digression into tangential anecdotes was noted in three patients.Family members of affected patients reported notable changes in the character of their loved ones.Three patients developed a preference for sweet foods, a symptom commonly observed in both FTD phenotypes and ALS [41].One patient described a marked change in musical taste at the onset of symptoms, which the family found unusual.Inappropriate comments, reference to a passive death wish, and perseverative thoughts ruminating on previous work issues have also been observed in our cohort.A relative lack of empathy was noted in two patients when their partner became tearful during the consultation.Increasing difficulty using everyday technology, such as mobile phones and laptops, is also commonly reported.One patient became a victim of online fraud due to impaired judgment.

Radiological considerations
Our voxelwise grey analyses captured bilateral anterior and mesial temporal grey matter atrophy with a clear right-sided predominance.Bilateral hippocampal involvement was also observed, as well as disease burden in the right insular and opercula region (Figure 1).Similar to the bilateral grey matter patterns, bilateral white matter integrity alterations were noted bilaterally in anterior temporal and sub-insular regions with a clear right-hemispheric predominance.Interestingly, extra-temporal white matter alterations have also been observed in orbitofrontal (FA and RD) and parietal (FA) regions.The sensitivity profiles of the various diffusivity metrics to capture sbvFTD-associated changes differed significantly.While both FA and RD detected bilateral temporal and extra-temporal changes in orbitofrontal regions, parietal change was only detected by FA, and MD only detected right anterior temporal lobe and right insular white matter changes.FA is the most widely used diffusivity metric, which is generally considered sensitive to focal white matter integrity alterations, albeit histopathologically relatively non-specific.Traditionally, axial diffusivity is often conceptualised as an 'axonal' measure [42], whereas RD has been classically regarded as a myelin-related marker [43], although this interpretation of diffusivity alteration is likely to be simplistic.Notwithstanding the differences between the statistical maps generated based on the different diffusivity metrics, the integrative interpretation of the voxel-based morphometry and tract-based statistics results is that sbvFTD affects the bilateral temporal lobes, and extra-temporal frontal and parietal changes can also be readily captured.In addition to the anatomical concordance of our grey and white matter analyses, metabolic profiles on PET imaging have also been strikingly concurring (Figure 6), highlighting bilateral but right-predominant hypometabolism with varying degrees of extra-temporal involvement.The review of individual PET images reveals a degree of heterogeneity in the anatomical extent and symmetry of neurodegeneration.While some patients in our cohort exhibit relatively focal right temporal-predominant degeneration, others exhibit more widespread changes.Furthermore, while all patients had a standard clinical MRI before enrolling in this imaging study, the visual review of their original clinical T1-weighted clinical images did not permit the detection of extra-temporal atrophy.The review of their initial clinical DWI/ADC and FLAIR images was unsuitable for the detection of the underlying white matter degeneration.This highlights the importance of acquiring high-resolution 3D T1w images in the clinical setting, as these permit quantitative post hoc analyses, as demonstrated in this study.We would also advocate the inclusion of a short DTI protocol so that underlying white matter integrity changes can be mapped computationally.

Subcortical involvement in ALS-FTD and FTD phenotypes
The physiology role of specific subcortical nuclei in relaying distinct cortico-cortical and cortico-basal networks is well described [88,89], and these networks have been implicated in disease propagation in ALS and ALS-FTD [90,91].Thalamic degeneration is particularly well characterised across the spectrum of ALS-FTD phenotypes [86,[92][93][94][95]. Subcortical involvement and frontotemporal dysfunction are also recognised in less common motor neuron disease phenotypes, such as primary lateral sclerosis [96,97].Presymptomatic thalamic changes have been described in GGGGCC hexanucleotide repeat expansion carriers by several groups long before projected symptom onset [98][99][100].However, hexanucleotide repeat expansion status is not the sole determinant for subcortical degeneration in ALS; considerable thalamic and subcortical grey matter degeneration can also be observed in C9orf72 negative ALS cohorts [101,102].Hippocampal degeneration also has considerable literature in both FTD [103] and ALS [83,104], and more recent studies have examined specific hippocampal subfields separately as they relay different networks and contribute to the function of specific limb circuits including Papez circuit [105][106][107].Accumbens nucleus and amygdalar degeneration [84] are also well recognised along the ALS-FTD spectrum and have been linked to specific neurocognitive manifestations [108], but differences in left and right hemispheric pathology are seldom specifically evaluated.While the cortical profile of sbvFTD is well characterised [47,56], subcortical involvement is less well studied.

Clinical implications
As demonstrated by the clinical profile of our cohort, right temporal variant ALS-FTD/FTD may be initially mistaken for depression or psychiatric conditions, and there may be a very significant interval between symptom onset and definite diagnosis.Misdiagnoses and diagnostic delays not only cause frustration in affected families but also delay supportive interventions, genetic screening, timely resource allocation, capacity testing, etc.The wider recognition of sbvFTD will eventually lead to an earlier diagnosis of these patients, a better understanding of disease mechanisms [23], and, ultimately, the development of robust research frameworks to accurately stratify FTD phenotypes [21].Our report also highlights the role of PET imaging in suspected cases and the limitations of visual inspection of standard clinical MRI.While in advanced disease, visual inspection of structural images may reveal ex vacuo ventricular dilation and widened sulci, these images are not suitable to appraise cortical thickness and density reductions, and visual inspection also precludes the assessment of deep subcortical grey matter volume reductions.Similarly, while FLAIR or T2w imaging would capture microvascular disease burden, degenerative changes in specific white matter tracts can only be evaluated by 3D DTI-derived or more advanced (NODDI, HARDI) white matter imaging techniques [109,110].While until recently, quantitative neuroimaging has primarily offered group-level imaging signatures in various ALS-FTD phenotypes, with the advent of novel machine-learning [111][112][113][114][115][116] and robust z-score-based approaches [117][118][119][120], single MRI data sets can now be meaningfully interpreted [121].

Study limitations
This study is not without limitations.Despite the relative clinical and radiological homogeneity of our cohort, there were considerable differences in the symptom duration profile of our patients.A key limitation of this study stems from its cross-sectional design.While we capture and describe a unifying radiological signature, the cross-sectional analyses preclude the characterisation of the clinical and radiological evolution of disease trajectories.Only a prospective, longitudinal, multi-time point study would elucidate anatomical propagation patterns and confirm whether progressive contralateral temporal lobe atrophy ensues eventually.Finally, we do not have accompanying neuropathology data on this cohort to offer histological descriptions and TDP-43 subtyping.Owing to the small sample size of our cohort, the imaging profile of the sbvFTD group has not been systematically contrasted to other ALS-FTD phenotypes.

Future directions
Large longitudinal studies are needed with a comprehensive neuropsychological battery to assess the clinical and radiological trajectory of this entity with subsequent post mortem examination and TDP-43 subtyping.In light of the heterogeneity of temporal lobe pathology in other neurodegenerative disorders, the clinical correlates of right temporal lobe pathology should also be studied in other motor neuron diseases (PLS, HSP, and SBMA), Alzheimer's disease, and movement disorders [122][123][124][125][126][127][128][129][130].In the assessment of patients with suspected neurodegenerative conditions, clinical MRI protocols should routinely incorporate a 3D T1-weighted sequence and a DTI sequence to enable post hoc quantitative grey and white matter analyses, which are useful for both diagnostic and monitoring purposes.

Conclusions
Our data suggest a unifying imaging signature in sbvFTD encompassing right-predominant but bilateral temporal lobe degeneration.In addition to the striking temporal disease burden, we have also demonstrated considerable subcortical grey matter pathology, as well as insular, frontal, and parietal involvement.Due to its distinguishing clinical features, the associated diagnostic challenges and singular metabolic, diffusion, and structural signature, sbvFTD should be considered a distinct clinical phenotype along the FTLD continuum.The increasing recognition of this phenotype and increased research efforts dedicated to sbvFTD may ultimately enhance the development of consensus management protocols.

Figure 6 .
Figure 6.Concordance of structural changes on clinical MRIs and metabolic patterns on [ 18 F] FDG PET-CT.

Figure 6 .
Figure 6.Concordance of structural changes on clinical MRIs and metabolic patterns on [ 18 F] FDG PET-CT.Figure 6. Concordance of structural changes on clinical MRIs and metabolic patterns on [ 18 F] FDG PET-CT.

Figure 6 .
Figure 6.Concordance of structural changes on clinical MRIs and metabolic patterns on [ 18 F] FDG PET-CT.Figure 6. Concordance of structural changes on clinical MRIs and metabolic patterns on [ 18 F] FDG PET-CT.

Table 1 .
The demographic profile of patients with semantic behavioural variant FTD (sbvFTD) and healthy controls (HC) who underwent quantitative grey and white matter neuroimaging.
Notes.Group-differences in age and education were examined with one-way analysis of variance and differences in sex and handedness distributions between patients with semantic-behavioural variant and healthy controls were contrasted with chi-square test (χ2) tests.Abbreviations: HC-healthy controls, Lt-left, M-mean, Rt-right, sbvFTD-semantic-behavioural variant FTD, SD-standard deviation.

Table 3 .
The clinical profile of eight representative cases with semantic behavioural-variant FTD (sbvFTD).