Search

Background: Atrophy of thalamic subregions has been observed across the spectrum of frontotemporal dementia (FTD). To gain better insight into underlying tissue changes, we investigated how thalamic subregional fractional anisotropy (FA) and mean diffusivity (MD) derived from diffusion tensor imaging (DTI) are altered in genetic FTD. Method: We used our newly developed thalamus segmentation tool, which jointly combines structural and diffusion input MRI data, to segment thalami and extract thalamic subregional FA and MD values for 163 genetic mutation carriers and 126 non‐carriers with suitable 3T MRI data from the GENetic FTD Initiative (GENFI). Mutation carriers were divided according to their genetic diagnosis and CDR®+NACC FTLD global scores into presymptomatic/prodromal (≤0.5: 41 C9orf72, 59 GRN, 34 MAPT) and symptomatic (≥1: 8 C9orf72, 11 GRN, 10 MAPT) groups. Mean FA and MD values for thalamic subregions were obtained using diffusion tensors interpolated in the log domain and weighted by segmentation posterior probabilities. Thalamic subregional FA and MD values for presymptomatic and symptomatic mutation carriers within each genetic group were compared with non‐carriers using analysis of covariance with bootstrapping, where age, scanner type, and sex were covariates. We corrected for multiple comparisons and calculated percentage changes in adjusted FA and MD mean values for mutation carriers relative to non‐carriers. Result: The only significant change at the presymptomatic stage was found for C9orf72 expansion carriers, who showed FA reduction in the intralaminar subregion (5%) (Figure 1, Table 1). In symptomatic C9orf72 expansion carriers, FA was reduced in the laterodorsal (21%), lateral posterior (13%), anteroventral (13%) and intralaminar (11%) subregions. Symptomatic MAPT mutation carriers also showed FA reduction in the laterodorsal (15%) and anteroventral (11%) subregions. No significant FA reductions were found for GRN mutation carriers and no significant MD changes were observed for any group after correction for multiple comparisons. Conclusion: We detected FA reductions of thalamic subregions only for C9orf72 expansion carriers at the presymptomatic stage, and for C9orf72 and MAPT mutation carriers at the symptomatic stage. Combined with the lack of robust MD changes, our findings may warrant further assessment of thalamic microstructure with more advanced diffusion models. [ABSTRACT FROM AUTHOR]

Background: Semantic and socioemotional knowledge, including person recognition, can be altered in frontotemporal dementia (FTD), and is often associated with the right temporal lobe variant. Using data from the Genetic FTD Initiative, we investigated person recognition deficits in genetic FTD. Method: 901 GENFI participants (279 mutation negative controls, 280 C9orf72 mutation carriers (MCs), 101 MAPTMCs and 241 GRN MCs) were grouped using the Clinical Dementia Rating scale plus National Alzheimer's Coordinating Centre Frontotemporal Lobar Degeneration (CDR plus NACC FTLD) global score where 0 denotes asymptomatic, 0.5 as prodromal, and 1+ as mild to severe symptoms (C9orf72: 135 = 0, 48 = 0.5, 97 = 1+; GRN: 143 = 0, 35 = 0.5, 63 = 1+; MAPT: 50 = 0, 20 = 0.5, 31 = 1+). Person recognition (PR) was assessed using a single question within a structured clinical questionnaire, scoring the ability to recognise people who should be familiar by face or voice to them, with a value between 0 (absent) to 3 (severe), similar to the CDR scale. The percentage of participants with PR deficits was calculated for each group. Logistic regression with bootstrapping compared the PR score between groups with age, gender, and education as covariates. Result: 16.1% of C9orf72 MCs (0 = 0.7%, 0.5 = 2.1%, 1+ = 44.3%), 7.5% of GRN (0 = 0.0%, 0.5 = 8.6%, 1+ = 23.8%) and 17.8% of MAPT carriers (0 = 2%, 0.5 = 10%, 1+ = 48.4%) showed PR deficits. Mean (standard deviation) severity in each group was: C9orf72 0 = 0.0(0.0), 0.5 = 0.0(0.1), 1+ = 0.6(0.9); GRN 0 = 0.0(0.0), 0.5 = 0.0(0.1), 1+ = 0.2(0.6); MAPT 0 = 0.0(0.2), 0.5 = 0.1(0.3), 1+ = 0.6(0.8). Each of the symptomatic genetic groups had a significantly greater PR deficit than the control group (p<0.001), with the prodromal MAPT (p = 0.006) and GRN (p<0.001) groups also showing a greater impairment than controls. There was a trend to significance in the C9orf72asymptomatic and prodromal groups compared with controls (p = 0.058 and p = 0.059 respectively). Symptomatic C9orf72 and MAPT carriers showed greater impairment than the symptomatic GRN carriers (both p = 0.005). Conclusion: Person recognition is a key early marker of disease in some individuals with genetic FTD and further imaging analyses will help to reveal the underlying mechanism of this deficit. [ABSTRACT FROM AUTHOR]

Background: Primary progressive aphasia (PPA) is an atypical neurodegenerative dementia with three main clinical variants: semantic (svPPA), non‐fluent/agrammatic (nfvPPA) and logopenic (lvPPA). While svPPA is typically associated with left anterior temporal lobe atrophy, neuroimaging findings in nfvPPA, lvPPA and PPA not otherwise specified (PPA‐nos) are more variable. We therefore applied unsupervised machine learning to one of the largest databases of PPA magnetic resonance imaging (MRI) scans ever acquired. Method: Baseline MRI scans were collected from 273 people with PPA participating in the longitudinal research programme in the Queen Square Dementia Research Centre (svPPA n = 97; nfvPPA n = 109; lvPPA n = 51; PPA‐nos n = 16). Longitudinal scans were collected from 138 participants. We used Subtype and Stage Inference (SuStaIn), a model that combines disease progression modelling with unsupervised machine learning to identify subgroups of patients following a similar trajectory. We standardised the data with respect to cognitively normal controls (n = 360). Based on clinicians' input we chose 19 brain regions of interest to model from those generated by GIF (Cardoso et al. 2012). The model was fit using baseline data, selecting the appropriate number of subtypes using cross‐validation (Young et al. 2018). Result: Cross‐validation supported four subtypes of spatiotemporal atrophy (Figure 1). Clinical variant representation per subtype assignment (1/2/3/4) was: svPPA (n = 74/15/4/4), nfvPPA (n = 3/35/21/44), lvPPA (n = 1/18/30/2) and PPA‐nos (n = 7/4/3/2) (Figure 2). Subtype 1 showed early decline in the left anterior temporal lobe and sparing of the left frontal lobe. Subtype 2 showed early decline of the left insula and sparing of the temporal pole. Subtype 3 showed initial atrophy of the left temporoparietal junction; Subtype 4 showed early atrophy in the left precentral gyrus. Statistical analysis of posterior distributions found low similarity between subtypes (Hellinger distance > 0.66) (Figure 3), validating the presence of four distinct subtypes. Conclusion: While the data‐driven atrophy subtypes did not align perfectly with clinical variants, we found a correspondence between Subtype 1 and svPPA, the most coherent PPA variant. Subtypes 2,3 and 4 were associated with nfvPPA and lvPPA, corroborating work suggesting these diagnostic terms encompass multiple clinical phenotypes. Future work will investigate the association between clinical diagnosis, longitudinal subtype, and clinical measures. [ABSTRACT FROM AUTHOR]

Background: Connections among brain regions allow pathological perturbations to spread from a single source node to multiple nodes. Patterns of neurodegeneration in multiple diseases, including behavioral variant of frontotemporal dementia (bvFTD), resemble the network architecture (Seeley et al., 2009, Neuron), but how bvFTD‐related atrophy patterns relate to the network organization remains unknown. Here we investigate whether neurodegeneration patterns in sporadic and genetic bvFTD are conditioned by connectome architecture, such that connected regions display similar atrophy patterns. Method: Deformation‐based morphometry (DBM) was used to estimate regional changes in tissue volume density from T1‐weighted magnetic resonance images of 75 genetic bvFTD patients and 247 healthy controls (GENFI, http://genfi.org.uk/). We used linear mixed effects model to obtain a bvFTD‐related atrophy map, controlling for age, sex and aquision site. Structural and functional connectivity (SC and FC), derived from an independent sample of 70 healthy participants (Griffa et al., 2019, Zenodo), were used to estimate mean neighbor atrophy values of each region. Relationship between node and neighbor atrophy was examined by correlating neighbor atrophy with nodal atrophy. Statistical significance of the analyses was assessed using a spatial autocorrelation‐preserving null model. Analyses were replicated in an independent dataset (FTLDNI, AG032306) with 70 sporadic bvFTD patients and 123 healthy controls. Result: Distributed atrophy patterns were observed in bvFTD, mainly targeting areas associated with limbic intrinsic network and insular cytoarchitectonic class (Fig 1a). A node’s atrophy was significantly correlated with atrophy of its connected neighbors (e.g. high resolution: r=0.58, p=0.006 and r=0.54, p=0.0006, for SC‐ and FC‐ defined neighbors respectively) (Fig 2c). Relationship between node and neighbor atrophy was consistent across resolutions and greater in empirical networks compared to null networks. While a number of frontotemporal regions were identified as potential disease epicenters, anterior insula was the most likely one. Results were consistent in the sporadic cohort (Fig 1b&2d). Conclusion: Using connectivity models and rigorous statistical analyses that account for spatial autocorrelation, we demonstrate that bvFTD‐related neurodegeneration is conditioned by connectome architecture, accounting for 30‐40% of variance in atrophy. Atrophy is most profound in regions associated with insular cortex. [ABSTRACT FROM AUTHOR]

Background: Frontotemporal dementia (FTD) is characterized by abnormal white matter (WM) integrity measured with conventional diffusion tensor imaging techniques. No study has yet investigated the microstructural WM changes across different clinical FTD forms using the novel neurite orientation dispersion and density imaging (NODDI). Here, we focused the analyses on symptomatic individuals, investigating the difference between the sporadic and genetic origin within the same diagnostic group. Method: Neurite density index (NDI) and orientation dispersion index (ODI) were extracted from NODDI sequences, processed and then corrected for tissue‐weighted means. Images were acquired on a 3T MRI Siemens Prisma scanner from a cohort of participants seen in clinic at the National Hospital for Neurology and Neurosurgery, University College London, UK. Data were available for 27 individuals with behavioural variant FTD (bvFTD) and 27 with primary progressive aphasias (11 svPPA, 8 nfvPPA, 4 lvPPA, 4 PPA‐NOS). 18 of these individuals (16 bvFTD and 2 PPA‐NOS) carried an FTD‐linked genetic mutation (6 C9orf72, 7 MAPT, 5 GRN). W‐scores for NDI and ODI were computed from a regression model on 62 non‐carrier healthy individuals, adjusting for their age, sex, and total intracranial volumes. Result: The most abnormal values (<2.5th percentile of controls) were found in bvFTD (NDI: anterior corona radiata and left cingulum), lvPPA (NDI and ODI: left superior corona radiata; ODI: splenium of the corpus callosum) and PPA‐NOS (NDI: left uncinate fasciculus) (Figure). Sporadic bvFTD and PPA‐NOS showed more abnormal and widespread WM abnormalities compared to symptomatic mutation carriers. Conclusion: Changes in WM structure in this cohort of patients with symptomatic FTD appear to be related primarily to a reduction in axonal density. Despite similar clinical features, sporadic forms tend to show more severe and widespread WM involvement than genetic cases. [ABSTRACT FROM AUTHOR]

Background: Genetic frontotemporal dementia (FTD) is highly heterogeneous, with carriers of mutations in the same gene manifesting different phenotypes. Using in vivo MR images from the Genetic FTD Initiative (GENFI), we aimed to identify subgroups within the same genetic group whose brains were affected differently. Method: Cortical and subcortical volumes of interest were generated using automated parcellation methods on volumetric 3T T1‐weighted MRI scans for 479 carriers (198 GRN, 202 C9orf72, and 79 MAPT mutation carriers). W‐scores for 85 volumes of interest were computed from a linear regression model carried out on 298 non‐carrier cognitively normal controls adjusting for the effect of age, sex, total intracranial volume and scanner type. Cluster analyses with the Ward agglomerating method were performed on all w‐scores while considering the three genetic groups independently. The identified clusters were then compared for age, estimated years from onset, global and sum of boxes scores of the CDR® plus NACC FTLD (at baseline and after one year), neurofilament light chain (NfL) levels in the plasma and w‐scores in brain regions typically showing early atrophy (Kruskal‐Wallis test). Result: We identified three clusters among the GRN mutation carriers and four in the MAPT and C9orf72 groups, which were all significantly different for the variables reported in the Table (p‐value<0.003). For all three genetic groups, one cluster was formed by patients with a clinical diagnosis of FTD, with more extensive atrophy and increased disease severity. For the remaining clusters, there seemed to be an association with disease severity for MAPT and GRN mutation carriers but not so for C9orf72 expansion carriers where clinical scores were not clearly associated with a specific cluster. Conclusion: By only looking at regional brain volumes, we were able to detect different clusters within carriers of mutations in the same gene, with C9orf72 expansion carriers being the most heterogenous group. Further investigations with specific cognitive, clinical and biomarkers correlates, including further follow‐up visits, are needed. [ABSTRACT FROM AUTHOR]

Background: Mutations in MAPT are associated with frontotemporal dementia (FTD), but little is known about the progression in its early stages. We aimed at identifying the presence of early brain changes in MAPT mutation carriers. Method: We included 3T MRIs from 84 MAPT carriers [27 symptomatic: mean(SD) age 58(8) years; 57 presymptomatic: 40(11) years] from the Genetic FTD Initiative (GENFI) and from 77 age‐matched non‐carrier healthy controls (44(14) years). Based on their expected years to symptom onset (EYO), we divided the presymptomatic carriers into early (n=35, <‐10 years) and late (n=22, >‐10 years) groups. First, we performed voxel‐based morphometry (VBM) comparing 24 symptomatic carriers with 32 controls to identify the regions of interest (ROIs) which were atrophic in the symptomatic stage of MAPT. We then used automated and manual segmentations to extract these ROI volumes in all carriers. To remove the effect of age, gender, total intracranial volume and scanner type, we transformed the volumes into w‐scores, considering the controls as the reference group. A w‐score of <‐1.28 (corresponding to the 10th percentile) was considered abnormal. Result: From the VBM we identified seven structures significantly atrophic in symptomatic carriers: the nucleus accumbens, amygdala, hippocampus, orbitofrontal cortex, temporal pole, anterior insula and hypothalamus. The percentage of early presymptomatic carriers with abnormal w‐scores was 14% for nucleus accumbens and hippocampus, and 20% orbitofrontal cortex. In the late group, 36% showed abnormal amygdala and temporal pole, 27% abnormal hippocampus, 32% abnormal anterior insula and 9% abnormal hypothalamus. In the symptomatic group, the percentages were higher: nucleus accumbens (41%), amygdala, hippocampus and temporal pole (85%), orbitofrontal cortex (56%), anterior insula (93%) and hypothalamus (48%). Conclusion: Abnormal limbic regions are a frequent feature in presymptomatic MAPT carriers, showing early structural changes before symptom onset. Further investigations on the associated cognitive and white matter changes are ongoing. [ABSTRACT FROM AUTHOR]

Background: Genetic frontotemporal dementia is highly heterogeneous, with different progression patterns seen between individuals. Using in vivo MR images from the Genetic FTD Initiative (GENFI), we aimed to identify clinical progression in genetic mutation carriers from their brain volumes at baseline. Method: Cortical and subcortical volumes of interest (VOIs) were generated using automated parcellation methods on volumetric 3T T1‐weighted MRI scans for 480 carriers (198 GRN, 202 C9orf72, 80 MAPT). W‐scores for 79 VOIs were computed from a linear regression model carried out on 298 non‐carrier cognitively normal controls adjusting for the effect of age, sex, total intracranial volume and scanner type. Carriers were divided into three disease stages based on their global CDR® plus NACC FTLD score (CDR‐GS): asymptomatic (0), possibly or mildly symptomatic (0.5) and fully symptomatic (1 or more). Cut‐off points for each VOI were derived from Youden indices estimated with ROC curves to distinguish between CDR‐GS=0 and CDR‐GS≥1 within each gene. CDR‐GS=0.5 carriers (30 GRN, 32 C9orf72, 13 MAPT) were classified as 'normal' or 'abnormal' based on these cut‐off points. We compared the CDR® plus NACC FTLD sum‐of‐boxes scores (CDR‐SOB) at one year follow‐up in these two groups. Result: Compared to those with normal baseline volumes, C9orf72 expansion carriers at CDR‐GS=0.5 showed significantly higher CDR‐SOB scores at follow‐up if they had abnormal volumes in the total frontal (+5 points), orbitofrontal (+3), dorsolateral prefrontal (+6), or anterior cingulate (+4) cortices, the basal‐paralaminar amygdala region (+2), CA1 region of the hippocampus (+4), total hippocampus (+6), cerebellar lobule VIIIb (+4), or lateral ventricles (+8). GRN mutation carriers showed significantly higher CDR‐SOB scores if their volumes were abnormal in the frontal (+10), parietal (+14), insula (+8), orbitofrontal (+12), or medial parietal (+7) cortices, or the total hippocampus (+10). MAPT mutation carriers with abnormal volumes in the lobule VI and dentate nucleus had 1 point higher CDR‐SOB scores at follow‐up. Conclusion: Abnormal baseline volumes in specific VOI within each of the genetic groups were related to worse CDR‐SOB scores over time. Future studies including longer follow‐up intervals and other longitudinal biomarkers are needed to explore this further. [ABSTRACT FROM AUTHOR]

Background: Prior studies of prodromal genetic frontotemporal dementia (FTD) have identified impaired executive function and social cognition in C9orf72 and GRN mutation carriers but few cognitive deficits have been found in MAPT mutation carriers. Method: We investigated performance on the Boston Naming Test (BNT) in the GENetic Frontotemporal dementia Initiative (GENFI) cohort of 499 mutation carriers and 248 mutation negative controls divided across three genetic groups: C9orf72, MAPT and GRN. Mutation carriers were further divided into 3 groups according to their global CDR plus NACC FTLD score: 0 (asymptomatic), 0.5 (prodromal) and 1+ (fully symptomatic). Mutation carrier groups were compared to each other and to mutation‐negative controls using a bootstrapped linear regression model, adjusting for age and education. Result: We found that all of the symptomatic (1+) groups performed significantly worse on the BNT than controls, and worse than the prodromal and asymptomatic mutation carriers within the same genetic group. The MAPT symptomatic group also performed significantly worse than the GRN and C9orf72 symptomatic groups. Furthermore, both the MAPT asymptomatic and prodromal groups performed significantly worse than controls which was not the case in the other genetic groups. Conclusion: These results suggest a specific deficit in naming in MAPT mutation carriers consistent with previous literature, likely to be due to impaired semantic knowledge. Notably, this study shows that the BNT can be used to detect very early deficits in the disease course, allowing dissociation of MAPT mutation carriers from controls and from other genetic groups. Clinical trials focused on the prodromal period within individuals with MAPT mutations should use language tasks such as the BNT for patient stratification and as outcome measures. [ABSTRACT FROM AUTHOR]