White Matter Hyperintensities Load Research Articles

Assessing brain involvement in Fabry disease with deep learning and the brain-age paradigm.

While neurological manifestations are core features of Fabry disease (FD), quantitative neuroimaging biomarkers allowing to measure brain involvement are lacking. We used deep learning and the brain-age paradigm to assess whether FD patients' brains appear older than normal and to validate brain-predicted age difference (brain-PAD) as a possible disease severity biomarker. MRI scans of FD patients and healthy controls (HCs) from a single Institution were, retrospectively, studied. The Fabry stabilization index (FASTEX) was recorded as a measure of disease severity. Using minimally preprocessed 3D T1-weighted brain scans of healthy subjects from eight publicly available sources (N = 2160; mean age = 33 years [range 4-86]), we trained a model predicting chronological age based on a DenseNet architecture and used it to generate brain-age predictions in the internal cohort. Within a linear modeling framework, brain-PAD was tested for age/sex-adjusted associations with diagnostic group (FD vs. HC), FASTEX score, and both global and voxel-level neuroimaging measures. We studied 52 FD patients (40.6 ± 12.6 years; 28F) and 58 HC (38.4 ± 13.4 years; 28F). The brain-age model achieved accurate out-of-sample performance (mean absolute error = 4.01 years, R2 = .90). FD patients had significantly higher brain-PAD than HC (estimated marginal means: 3.1 vs. -0.1, p = .01). Brain-PAD was associated with FASTEX score (B = 0.10, p = .02), brain parenchymal fraction (B = -153.50, p = .001), white matter hyperintensities load (B = 0.85, p = .01), and tissue volume reduction throughout the brain. We demonstrated that FD patients' brains appear older than normal. Brain-PAD correlates with FD-related multi-organ damage and is influenced by both global brain volume and white matter hyperintensities, offering a comprehensive biomarker of (neurological) disease severity.

Hemorrhagic transformation of cerebral infarction: risk factors, diagnosis, and new approaches to treatment

Background. Hemorrhagic transformation (HT) is a frequent complication of acute cerebral infarction, which is unfavo­rable for the treatment and prognosis of patients. It can be divided into two main subtypes, hemorrhagic infarction, and parenchymal hematoma, depending on the type of hemorrhage. Although the pathophysiological mechanism of HT is still unclear, hypotheses have been proposed about the loss of microvascular integrity and impaired neurovascular homeostasis. The purpose of the study was to analyze the current scientific literature on risk factors, diagnosis, and new approaches to the treatment of hemorrhagic transformation of cerebral infarction. Materials and methods. A literature search using keywords was conducted in Web of Science, Scopus, PubMed, and Elsevier databases. Results. Ischemic tissues have a natural tendency to bleed. In addition, the first trials of recanalization using intravenous thrombolysis showed an increase in the incidence of mild to severe intracranial hemorrhages. Symptomatic intracerebral hemorrhage is closely associated with poor outcomes and is an important factor in recanalization decisions. The development of HT after stroke involves numerous interrelated pathological processes from peripheral blood cells to neurovascular units such as hyperactive ischemic cascades with increased levels of matrix metalloproteinases, excessive reactive oxygen species, coagulopathy, blood-brain barrier breakdown, and reperfusion injury. A number of risk factors or prognostic factors for HT after cerebral infarction have been identified, namely the use of antiplatelet agents or anticoagulants, infarct size, atrial fibrillation, hypertension, age, gender, heart failure, coronary artery disease, diabetes mellitus, lipid profile, reperfusion therapy, and white matter hyperintensity load. The incidence of HT is reported mainly in clinical studies using brain imaging techniques such as computed tomography or magnetic resonance imaging, diffusion weighted imaging, and perfusion weighted imaging. Recombinant tissue plasminogen activator therapy and mechanical thrombectomy are currently the main treatments for ischemic stroke, but reperfusion injury due to revascula­rization increases the risk of cerebral hemorrhage. Understanding the risk factors and in-depth evaluation of predictors can significantly help physicians develop strategies to reduce the occurrence of HT, as well as provide insight into the pathophysiological mechanism of the disease. Conclusions. Patients at risk of hemorrhagic transformation require invasive and non-invasive neuromonitoring to help make decisions in decompressive neurosurgery for large cerebellar infarction, specific cardiorespiratory treatment, nutrition, blood pressure control, biochemical parameters, and the choice of an individual management strategy.

Genetic and Clinical Correlates of AI-Based Brain Aging Patterns in Cognitively Unimpaired Individuals

Brain aging elicits complex neuroanatomical changes influenced by multiple age-related pathologies. Understanding the heterogeneity of structural brain changes in aging may provide insights into preclinical stages of neurodegenerative diseases. To derive subgroups with common patterns of variation in participants without diagnosed cognitive impairment (WODCI) in a data-driven manner and relate them to genetics, biomedical measures, and cognitive decline trajectories. Data acquisition for this cohort study was performed from 1999 to 2020. Data consolidation and harmonization were conducted from July 2017 to July 2021. Age-specific subgroups of structural brain measures were modeled in 4 decade-long intervals spanning ages 45 to 85 years using a deep learning, semisupervised clustering method leveraging generative adversarial networks. Data were analyzed from July 2021 to February 2023 and were drawn from the Imaging-Based Coordinate System for Aging and Neurodegenerative Diseases (iSTAGING) international consortium. Individuals WODCI at baseline spanning ages 45 to 85 years were included, with greater than 50 000 data time points. Individuals WODCI at baseline scan. Three subgroups, consistent across decades, were identified within the WODCI population. Associations with genetics, cardiovascular risk factors (CVRFs), amyloid β (Aβ), and future cognitive decline were assessed. In a sample of 27 402 individuals (mean [SD] age, 63.0 [8.3] years; 15 146 female [55%]) WODCI, 3 subgroups were identified in contrast with the reference group: a typical aging subgroup, A1, with a specific pattern of modest atrophy and white matter hyperintensity (WMH) load, and 2 accelerated aging subgroups, A2 and A3, with characteristics that were more distinct at age 65 years and older. A2 was associated with hypertension, WMH, and vascular disease-related genetic variants and was enriched for Aβ positivity (ages ≥65 years) and apolipoprotein E (APOE) ε4 carriers. A3 showed severe, widespread atrophy, moderate presence of CVRFs, and greater cognitive decline. Genetic variants associated with A1 were protective for WMH (rs7209235: mean [SD] B = -0.07 [0.01]; P value = 2.31 × 10-9) and Alzheimer disease (rs72932727: mean [SD] B = 0.1 [0.02]; P value = 6.49 × 10-9), whereas the converse was observed for A2 (rs7209235: mean [SD] B = 0.1 [0.01]; P value = 1.73 × 10-15 and rs72932727: mean [SD] B = -0.09 [0.02]; P value = 4.05 × 10-7, respectively); variants in A3 were associated with regional atrophy (rs167684: mean [SD] B = 0.08 [0.01]; P value = 7.22 × 10-12) and white matter integrity measures (rs1636250: mean [SD] B = 0.06 [0.01]; P value = 4.90 × 10-7). The 3 subgroups showed distinct associations with CVRFs, genetics, and subsequent cognitive decline. These subgroups likely reflect multiple underlying neuropathologic processes and affect susceptibility to Alzheimer disease, paving pathways toward patient stratification at early asymptomatic stages and promoting precision medicine in clinical trials and health care.

End-to-end volumetric segmentation of white matter hyperintensities using deep learning

Background and objectivesReliable detection of white matter hyperintensities (WMH) is crucial for studying the impact of diffuse white-matter pathology on brain health and monitoring changes in WMH load over time. However, manual annotation of 3D high-dimensional neuroimages is laborious and can be prone to biases and errors in the annotation procedure. In this study, we evaluate the performance of deep learning (DL) segmentation tools and propose a novel volumetric segmentation model incorporating self-attention via a transformer-based architecture. Ultimately, we aim to evaluate diverse factors that influence WMH segmentation, aiming for a comprehensive analysis of the state-of-the-art algorithms in a broader context. MethodsWe trained state-of-the-art DL algorithms, and incorporated advanced attention mechanisms, using structural fluid-attenuated inversion recovery (FLAIR) image acquisitions. The anatomical MRI data utilized for model training was obtained from healthy individuals aged 62–70 years in the Live active Successful Aging (LISA) project. Given the potential sparsity of lesion volume among healthy aging individuals, we explored the impact of incorporating a weighted loss function and ensemble models. To assess the generalizability of the studied DL models, we applied the trained algorithm to an independent subset of data sourced from the MICCAI WMH challenge (MWSC). Notably, this subset had vastly different acquisition parameters compared to the LISA dataset used for training. ResultsConsistently, DL approaches exhibited commendable segmentation performance, achieving the level of inter-rater agreement comparable to expert performance, ensuring superior quality segmentation outcomes. On the out of sample dataset, the ensemble models exhibited the most outstanding performance. ConclusionsDL methods generally surpassed conventional approaches in our study. While all DL methods performed comparably, incorporating attention mechanisms could prove advantageous in future applications with a wider availability of training data. As expected, our experiments indicate that the use of ensemble-based models enables the superior generalization in out-of-distribution settings. We believe that introducing DL methods in the WHM annotation workflow in heathy aging cohorts is promising, not only for reducing the annotation time required, but also for eventually improving accuracy and robustness via incorporating the automatic segmentations in the evaluation procedure.

Early detection of white matter hyperintensities using SHIVA-WMH detector.

White matter hyperintensities (WMHs) are well-established markers of cerebral small vessel disease, and are associated with an increased risk of stroke, dementia, and mortality. Although their prevalence increases with age, small and punctate WMHs have been reported with surprisingly high frequency even in young, neurologically asymptomatic adults. However, most automated methods to segment WMH published to date are not optimized for detecting small and sparse WMH. Here we present the SHIVA-WMH tool, a deep-learning (DL)-based automatic WMH segmentation tool that has been trained with manual segmentations of WMH in a wide range of WMH severity. We show that it is able to detect WMH with high efficiency in subjects with only small punctate WMH as well as in subjects with large WMHs (i.e., with confluency) in evaluation datasets from three distinct databases: magnetic resonance imaging-Share consisting of young university students, MICCAI 2017 WMH challenge dataset consisting of older patients from memory clinics, and UK Biobank with community-dwelling middle-aged and older adults. Across these three cohorts with a wide-ranging WMH load, our tool achieved voxel-level and individual lesion cluster-level Dice scores of 0.66 and 0.71, respectively, which were higher than for three reference tools tested: the lesion prediction algorithm implemented in the lesion segmentation toolbox (LPA: Schmidt), PGS tool, a DL-based algorithm and the current winner of the MICCAI 2017 WMH challenge (Park et al.), and HyperMapper tool (Mojiri Forooshani et al.), another DL-based method with high reported performance in subjects with mild WMH burden. Our tool is publicly and openly available to the research community to facilitate investigations of WMH across a wide range of severity in other cohorts, and to contribute to our understanding of the emergence and progression of WMH.

Association of white matter hyperintensity burden and cognitive performance in young‐aged, middle‐aged and old‐aged participants: Results of the population‐based 1000BRAINS study

AbstractBackgroundSome individuals seem less susceptible to the effect of high white matter hyperintensity (WMH) load on cognition reflecting differences in individual cognitive reserve (CR). Little is known about young‐ or middle‐aged participants. The aim of the present study was to examine (1) the effect of WMH on global cognition in three different age groups and (2) if education (as proxy for CR) moderates this association.MethodWe included 707 healthy participants (18‐85 years) without evidence of cardiovascular/neurological disease (young‐aged: 18‐44 years (Ø33.5±6.7): n = 108; middle‐aged: 45‐65 years (Ø57.9±5.5): n = 341; old‐aged: >65 years (Ø72.0±4.1): n = 258) from the population‐based 1000BRAINS study. An extensive cognitive assessment was conducted. The sum of all cognitive domain z‐scores defined the global score. Education was classified according to the International Standard Classification of Education as total years of formal education, combining school and vocational training. Magnetic resonance imaging (MRI) was carried out on a 3‐Tesla‐MR‐scanner (Tim‐TRIO, Siemens Medical Systems, Erlangen, Germany). WMH volume was determined using the Brain‐Intensity‐Abnormality‐Classification‐Algorithm. The associations of global cognition as outcome with WMH volume (in cm3) as predictor were analyzed using linear models (PROCESS v4.1 macro for SPPS) stratified by age group resulting in regression coefficient b with 95% confidence intervals (CI; adjusted for age, sex, depression, diabetes mellitus). To examine moderation effects of education, all models contained an interaction term (WMH x education).ResultHigher WMH volume was associated with lower global cognition in middle‐aged participants (b:‐0.27 (‐0.52 to ‐0.03, all reported results are fully adjusted). This effect was moderated by education (interaction term: b:‐0.07 (‐0.014 to ‐0.01)). In the young‐aged group, the association between WMH load and cognition was ‐1.27 (‐3.30 to 0.77). No association was found in the old‐aged group (b: ‐0.03 (‐0.20 to 0.14)).ConclusionHigher WMH load was associated with lower cognitive performance only in middle‐aged participants and was moderated to a small degree by education. Overall, the influence of WMH on global cognition in our cohort of healthy participants seems limited. Future analyses will focus on specific cognitive domains that might be more vulnerable to higher WMH load and will examine participants with certain cardiovascular risk profile.

MicroRNA expression based on toxic Oligomeric Amyloid Beta and Burden of Cerebrovascular Disease

AbstractBackgroundMicroRNAs(miRNAs) form the most intensively characterised branch of non‐coding RNAs, with universal expression in a repertoire of biofluids and active regulatory roles in multiple biological pathways(Blount, Coursey, and Kocerha 2022). Therefore, any perturbance in miRNA expression produces pronounced biological effects, due to its pleiotropic nature(Blount, Coursey, and Kocerha 2022). A variety of miRNAs have been investigated in multiple dementia syndromes (Blount, Coursey, and Kocerha 2022). However, what remains to be evaluated is differential miRNA expression in dementia classified using the more toxic oligomeric amyloid beta(OAβ) as well as based on burden of cerebrovascular disease(CVD).MethodA cross‐sectional pilot study was conducted in a cohort of 24 research participants(Table 1), stratified on the basis of their plasma OAβ levels [Multimer Detection System based plasma OAβ analysis(PeopleBio Inc, South Korea)](Dominguez et al. 2022). T3 ‐ based MRI data was evaluated for CVD status ‐ White Matter Hyperintensity load, was scored using the Fazekas scale(Table 1). Blood plasma‐based expression profiling of 376 miRNAs was conducted using the ID3EAL™ PanoramiR miRNA Knowledge Panel(MiRXES Pte Ltd, Singapore) and normalised miRNA expression was defined via the delta‐delta(d) Ct method. MiRNA target interaction analyses was conducted using miRTarBase 9.0, revealing potential reactome pathways of interest(Huang et al. 2022).ResultMean age and MoCA scores are 60.05 and 25.42 respectively (Table 1). 10 miRNAs(upregulated : hsa‐miR‐42, hsa‐miR‐378c, hsa‐miR‐938; downregulated: hsa‐miR‐545‐3p, hsa‐miR‐449c‐5p, hsa‐miR‐657, hsa‐miR‐23c, hsa‐miR‐888‐5p, hsa‐miR‐599, hsa‐let‐7d‐5p) were significantly differentially expressed in OAβ positivity(Table 2). 6 miRNAs(upregulated: hsa‐miR‐513c‐5p, hsa‐miR‐592; downregulated: hsa‐miR‐98‐5p, hsa‐miR‐744‐5p, hsa‐let‐7d‐5p, hsa‐miR‐517‐5p were significantly differentially expressed in CVD(Table 2). Reactome pathways demonstrating putative biological relevance are the FOXO/p53 mediated transcriptional regulation (metabolism/senescence/neuronal genes),Caspase mediated inflammatory activation, Insulin/IGF‐1/PI3K/Akt/GSK3β mediated metabolic signalling, IL‐4/IL‐13 mediated immune(inflammatory) signalling, Nervous System/Axonal Development and WNT Signalling(Table 2).ConclusionFindings from this cross‐sectional pilot study, have revealed 16 miRNAs to be promising biomarker candidates within the OAβ positivity and CVD burden setting. Longitudinal study with larger sample size is on‐going to confirm and validate these findings.The authors wish to express their sincere gratitude to MiRXES Pte Ltd for their technical expertise and efforts towards the miRNA expression profiling and associated analyses.

Accelerated brain ageing in participants with higher cerebrovascular lesion load across aging, cognitive impairment, and mixed presentation

AbstractBackgroundWe aimed to investigate the association between cortical brain age and white matter hyperintensities (WMH) in diverse forms of clinically‐defined vascular conditions. Our hypothesis was that a higher WMH burden would be associated with higher cortical brain age in all clinical subtypes.MethodWe used standardized MRI data obtained from participants in the COMPASS‐ND cohort of the CCNA. We used a standard linear support vector regression algorithm to estimate brain age, matching chronological age to cortical anatomical measurements obtained with the FreeSurfer toolbox on T1‐weighted MRI, sex, and intracranial volume. We calculated brain‐PAD (i.e., predicted brain age minus real age) and applied bias adjustment to remove age‐dependency in the estimates. A validated automated technique utilizing T2‐weighted and fluid attenuated inversion recovery MRIs was used to compute WMH loads.ResultParticipants included 107 CIE, 240 mild cognitive impairment (MCI), 115 vascular‐MCI (V‐MCI), 81 probable Alzheimer’s disease (AD), and 50 V‐AD. There was a significant difference in brain‐PAD [ F (4,599) = 56, P < 0.001, ANCOVA test] among groups, whilst adjusting for sex and chronological age. All four categories of patients exhibited a significantly higher mean brain‐PAD than CIE (P < 0.001), with the AD cohort having the highest brain‐PAD. There was a significant difference in WMH loads [ F (4,586) = 48, P < 0.001, ANCOVA test] between groups, whilst adjusting for sex and age. All cohorts showed a positive correlation between brain‐PAD and WMH, indicating accelerated ageing in people with higher WMHs. We observed moderate and significant correlations between brain‐PAD and WMH loads for V‐MCI (r = 0.34, P < 0.001) and V‐AD (r = 0.36, P = 0.001) but weaker, while significant correlations in CIE (r = 0.20, P = 0.035) and AD cohorts (r = 0.24, P = 0.025).ConclusionWe observed a positive and significant link between brain‐PAD and WMH in all categories. This finding indicates on the importance of treatment and prevention strategies for vascular risk factors, which might be able to slow down the progression of cerebrovascular lesions and delay the effect on cortical thickness.

The influence of vascular pathology on the generalizability of machine learning models for cognitive performance estimation

AbstractBackgroundImage‐based machine learning models are promising to aid diagnosis in dementia, e.g. by diagnostic classification or estimating cognitive performance. However, the contribution of vascular pathology is often ignored, although many patients have mixed pathologies. This may hamper model generalizability from research setting to clinical practice, as the models are generally developed on populations with lower vascular burden than is encountered clinically. This study aims to evaluate whether model generalizability is influenced by differences in vascular pathology load between training and test groups, represented by white matter hyperintensities (WMHs).MethodWe developed a convolutional neural network to estimate cognitive performance through ADAS13. We included 4791 timepoints from 1632 ADNI subjects. To mimic differences in vascular burden between research and clinical populations, we split the data into a low, middle and high WMH load group using WMH ratio tertiles (WMH volume corrected for intracranial volume). The model was applied to maps derived from T1‐weighted and FLAIR brain MRI using voxel‐based morphometry: model 1 (gray matter density map (GMDM), age, sex), model 2 (GMDM, WMH density map (WMHDM), age, sex) and model 3 (GMDM, log(WMH ratio), age, sex) (Fig. 1). First, we trained all models on the low WMH load group, meaning to test on the high WMH load group to assess generalizability. Second, we trained models 1 and 2 on all data. Mean Absolute Error (MAE) served as the performance metric.ResultThe models trained on the low WMH load group (MAE = 5.08 – 5.41) performed similar to constant value prediction (MAE = 5.63) (Fig. 2, circles). Using all data, models did train successfully (model 1: MAE = 5.64, model 2: MAE = 6.31, constant value: MAE = 7.39) (Fig. 2, crosses). Because of the suboptimal model training in the low WMH load group, especially for individuals with higher ADAS13 scores which were underrepresented in the data (Fig. 3), we could not evaluate generalizability to the high WMH load group.ConclusionThis study was unable to provide new insight into the influence of vascular pathology on the generalizability of machine learning models in dementia. More clinically representative data is needed, covering the full cognitive performance spectrum and higher vascular pathology loads.

Comparison of white matter hyperintensities between Latin American and American individuals with Alzheimer’s disease and cognitively normal older adults

AbstractBackgroundWhite matter hyperintensities (WMH) are Flair signal abnormalities frequently seen in periventricular and subcortical brain regions in the elderly. They are presumed to be, at least in part, a sign of cardiovascular disease (CVD). WMHs are associated with stroke, cognitive decline, and dementia. Given that most of the literature on dementia comes from high‐income countries, little is known about dementia in Latin American countries (LAC), which have an increased risk of dementia and CVD due to unique demographic, socioeconomic, cultural and ethnic factors. Our study aims to characterize the effects of white matter hyperintensities (WMH) in patients with AD within LAC and to compare these findings with a sample from the USA.MethodA total of 491 latin american subjects (235 cognitively normal (CN), 256 AD) and 271 american subjects (165 CN, 106 AD), underwent a brain MRI. WMH were extracted from T2‐FLAIR images using the Lesion Segmentation Toolbox and were divided into periventricular (pWMH) or deep (dWMH) depending on their distance from the ventricles. Linear mixed‐effects were used to model the volumes of WMH as a function of age, gender, years of education, intracranial volume, average cortical thickness, and region (LAC or USA).ResultAD groups from both regions showed larger total, dWMH and pWMH volumes compared to HC. Higher pWMH from AD and HC was associated with age, female sex and inversely associated with years of education. On the other hand, the increase in dWMH was found to be associated with female sex for the HC group and with age for AD. Cortical thinning was associated with an increase in total, subcortical and periventricular WMH for both groups. No statistically significant association was found between the region (USA or LAC) and WMH.ConclusionAD showed higher WMH compared to HC. Our findings suggest that pWMH and dWMH have different clinical correlates. No associations between region and WMH load or distribution were found. Future studies need to elucidate if there is an increase in other signs of CVD (such as cerebral microbleeds, brain lacunae, enlarged perivascular spaces) in Latin America and if they are linked to cognition and social determinants of health.

APOE‐dependent and ‐independent polygenic pathways determining early Alzheimer’s Disease pathological changes in CSF and MRI

AbstractBackgroundLittle is known about the genetic factors and downstream molecular pathways determining individual variability in fluid and imaging biomarkers associated with Alzheimer’s disease(AD). We studied polygenic risk scores (PRS) and pathway‐specific PRS in relationship with AD fluid and imaging biomarkers, in non‐demented individuals from the European Prevention of Alzheimer’s Dementia (EPAD) cohort.MethodEPAD inclusion criteria were age>50 and Clinical Dementia Rating = 0.5 (n = 1886, Table 1). AD‐PRS was based on 85 previously identified loci, including and excluding APOE (PRSAPOE, PRSnoAPOE)[1]. Using gene‐variant and variant‐pathway mapping[2], six pathway‐specific PRSnoAPOE were identified:1) immune‐activation, 2) signal‐transduction, 3) inflammatory‐response, 4) migration, 5) amyloid‐production, and 6) clearance. Linear models were used to assess the relationship of all PRS with fluid AD biomarkers, including Aß1‐42, p‐Tau181, and t‐tau; and several imaging biomarkers, including hippocampal volume, global and lobar white matter hyperintensities (WMH) volumes, fractional anisotropy (FA) in 14 regions of interest from diffusion tensor imaging, and 10 resting‐state network connectivity from functional MRI. Models were adjusted for age, sex, site, and multiple comparisons.ResultModels’ coefficients are reported in Table 2. PRSAPOE was significantly associated with decreased Aß1‐42, and increased p‐Tau181 and t‐Tau. PRSnoAPOE showed a weaker, but still significant, negative association with Aß1‐42 levels, and a significant positive association with p‐Tau181 and t‐Tau. Aß1‐42 was only significantly associated with the migration, amyloid, and clearance pathways, while all pathways were significantly positively associated with p‐Tau181 and t‐Tau. Hippocampal volume was only significantly associated with PRSAPOE. Regarding WMH load, PRSAPOE showed a positive association with global, frontal periventricular, and parietal deep lobes. Furthermore, the clearance pathway was related to WMH load in all regions, most strongly in periventricular areas. Only the migration pathway was related to increases in FA in the splenium, body, and genu of the corpus callosum. Only the PRSAPOE was related to reduced functional connectivity (FC) in the default mode, control, and visual networks.ConclusionWe show that genetic risk beyond APOE facilitates the manifestation of AD related pathologies. Moreover, clearance and migration pathways were associated with neuroimaging measures of white matter integrity, while FC and hippocampal volume were associated with overall AD genetic risk.

APOE‐dependent and ‐independent polygenic pathways determining early Alzheimer’s Disease pathological changes in CSF and MRI

AbstractBackgroundLittle is known about the genetic factors and downstream molecular pathways determining individual variability in fluid and imaging biomarkers associated with Alzheimer’s disease(AD). We studied polygenic risk scores (PRS) and pathway‐specific PRS in relationship with AD fluid and imaging biomarkers, in non‐demented individuals from the European Prevention of Alzheimer’s Dementia (EPAD) cohort.MethodEPAD inclusion criteria were age>50 and Clinical Dementia Rating≤0.5 (n = 1886, Table 1). AD‐PRS was based on 85 previously identified loci, including and excluding APOE (PRSAPOE, PRSnoAPOE)[1]. Using gene‐variant and variant‐pathway mapping[2], six pathway‐specific PRSnoAPOE were identified:1) immune‐activation, 2) signal‐transduction, 3) inflammatory‐response, 4) migration, 5) amyloid‐production, and 6) clearance. Linear models were used to assess the relationship of all PRS with fluid AD biomarkers, including Aβ1‐42, p‐Tau181, and t‐tau; and several imaging biomarkers, including hippocampal volume, global and lobar white matter hyperintensities (WMH) volumes, fractional anisotropy (FA) in 14 regions of interest from diffusion tensor imaging, and 10 resting‐state network connectivity from functional MRI. Models were adjusted for age, sex, site, and multiple comparisons.ResultModels’ coefficients are reported in Table 2. PRSAPOE was significantly associated with decreased Aβ1‐42, and increased p‐Tau181 and t‐Tau. PRSnoAPOE showed a weaker, but still significant, negative association with Aβ1‐42 levels, and a significant positive association with p‐Tau181 and t‐Tau. Aβ1‐42 was only significantly associated with the migration, amyloid, and clearance pathways, while all pathways were significantly positively associated with p‐Tau181 and t‐Tau.Hippocampal volume was only significantly associated with PRSAPOE. Regarding WMH load, PRSAPOE showed a positive association with global, frontal periventricular, and parietal deep lobes. Furthermore, the clearance pathway was related to WMH load in all regions, most strongly in periventricular areas.Only the migration pathway was related to increases in FA in the splenium, body, and genu of the corpus callosum. Only the PRSAPOE was related to reduced functional connectivity (FC) in the default mode, control, and visual networks.ConclusionWe show that genetic risk beyond APOE facilitates the manifestation of AD related pathologies. Moreover, clearance and migration pathways were associated with neuroimaging measures of white matter integrity, while FC and hippocampal volume were associated with overall AD genetic risk.

Characterizing cerebral microhemorrhage associations in dementia subtypes in the UK Biobank

AbstractBackgroundCerebral microbleeds (CMBs) are associated with neurodegenerative diseases (Charidimou, 2011) and have been identified as an adverse amyloid‐related imaging abnormality event (ARIA‐H) related to amyloid clearing medications (Sperling, 2011). Identifying risk factors for CMBs can help refine inclusion criteria for clinical trials and mitigate the risk of ARIA‐H. CMBs are readily detected on T2*‐GRE MRI sequences, including susceptibility weighted imaging (SWI) – a modality used to detect magnetic field distortions caused by paramagnetic products like iron (Haacke, 2004, Liu, 2017). We explore CMB associations with demographic, disease, and imaging markers in a subset of participants with varying dementia subtypes and complaints in the UK Biobank (Miller, 2016).MethodA total of 75 participants who had clinically diagnosed or self‐reported dementia and SWI imaging were selected; 4 participants were excluded due to extreme motion. We used MARS criteria (Gregoire, 2009) to identify CMBs on SWI and validated our findings using other sequences to exclude CMB ‘mimics’ (Figure 1A). The number, classification, and position of CMBs were recorded. Logistic regressions modeling the outcome of at least one ‘definite’ or ‘total’ (‘definite’ + ‘possible’) CMB were performed and effect sizes were calculated across hippocampal volume, white matter hyperintensity load (WMH), and global WM diffusion MRI (dMRI) microstructural measures.ResultParticipants were predominantly male (65%) with a mean age of 69.8 (± 7.1) years. 9 participants had at least one ‘definite’ CMB and 8 more had at least one ‘possible’ CMB for a ‘total’ of 17 individuals with CMBs. Individuals with CMB(s) were significantly older than those without. CMBs were mostly ‘strictly lobar’ for both definite (78%) and total (71%) classifications (Table 1). The presence of CMBs were found across a spectrum of dementia diagnoses (Figure 1B). Microstructural dMRI measures including global WM FA, MD, and ISOVF were significantly associated with presence of total CMBs (q<0.05) (Figure 2).ConclusionMeasures reflecting WM organization were sensitive to the pathophysiology of CMBs in those with dementia. Future work should investigate whether WM microstructure can predict the development of CMBs in those with dementia, particularly in individuals undergoing amyloid clearing treatments.

White matter hyperintensities associate with pathological stages of neurodegenerative disease.

AbstractBackgroundWhite matter hyper‐intensities (WMH) are observed on FLAIR MRI, increase with age and are more prominent in neurodegenerative diseases such as in Alzheimer’s disease (AD) and Parkinson’s disease (PD). The aim of this study was to assess the association between WMH and neuropathological staging of amyloid‐beta, neurofibrillary tangles (NFT) and α‐synuclein (α‐syn).MethodIn‐situ 3T‐MRI 3DT1‐weighted and 3D‐FLAIR data were collected for 77 clinically defined and pathologically‐confirmed brain donors; 31 AD, 21 PD and PD dementia (PDD), and 25 controls. WMH were segmented on FLAIR using an in‐house developed multiview convolutional neural network, followed by manual checking, and volume was obtained with FSL. Fifteen brain regions were dissected at autopsy, immunostained and evaluated for neuropathological diagnosis, including Thal amyloid phase, CAA type, and Braak stages for NFT and α‐syn. Group differences were assessed with linear models (corrected for age, gender, post‐mortem delay), and associations with Spearman and Pearson partial correlations.ResultCompared to controls, WMH load was higher in AD (p<0.001), specifically amnestic AD (p = 0.003) rather than non‐amnestic AD (p = 0.053), and higher in PD (p = 0.016), specifically non‐demented PD (p = 0.015) rather than PDD (p = 1.0). No difference in gender was observed (p = 0.38). Without the influence of pathology (in controls), WMH was associated with age (r = 0.49,p = 0.016).In the whole cohort, WMH was associated with Fazekas score (r = 0.67; p<0.001), normalized brain volume (NBV; r = ‐0.31,p = 0.007), gray matter volume (NGMV; r = ‐0.36,p = 0.002), Thal phase (r = 0.24,p0.034), Braak NFT stage (r = 0.33,p = 0.004), and Braak α‐syn stage (r = 0.27, p0.018).Across controls and AD, WMH associated with NBV (r = ‐0.39,p = 0.004), NGMV (r = ‐0.42,p = 0.002), left and right hippocampal volume (r = ‐0.36,p = 0.008 and r = ‐0.32,p = 0.022), Braak NFT (r = 0.41,p<0.001), Thal phase (r = 0.41,p = 0.002), and CAA (r = 0.34,p = 0.01). Amnestic and non‐amnestic AD showed similar results, except for CAA, which was associated with WMH in non‐amnestic AD only (r = 0.33,p = 0.04).Across controls and PD(D), WMH associated with Braak NFT stage (r = 0.39,p = 0.008) and Braak α‐syn stage (r = 0.60,p<0.001), but not with MRI outcome measures. In this respect PD and PDD showed different results; in PD WMH associated with NGMV (r = ‐0.37,p = 0.031), in PDD with left and right hippocampal volume (r = ‐0.44,p = 0.014 and r = ‐0.44,p = 0.015).ConclusionWMH associate with pathological stages of AD and PD(D), and contributes to MRI atrophy.

Artificial intelligence reveals brain aging patterns in 27,402 individuals without diagnosed cognitive impairment that are linked to genetics, biomedical measures, and cognitive decline

AbstractBackgroundUnderstanding heterogeneity of structural brain changes in aging may provide insights into susceptibility to neurodegenerative diseases. We characterize the genetics underlying brain structural heterogeneity within cognitively unimpaired (CU) individuals using data‐driven machine learning applied to a diverse dataset of 27,402 individuals from 11 neuroimaging studies from the iSTAGING consortium.MethodStructural brain morphologic patterns of CU individuals were independently examined in four decade‐long intervals spanning ages 45 to 85. Within each interval, Smile‐GAN (Yang et al., 2021) was trained on baseline anatomic and white matter hyperintensity (WMH) volumes. Smile‐GAN probability scores were used as phenotypes in genome‐wide association studies (GWAS). Specifically, we performed multiple linear regressions controlling for confounders (e.g., age) via Plink (Purcell et al., 2007). We observed longitudinal clustering stability across decades, so individuals from adjacent age groups were combined into broader age groups ([45,65), [65,85)) due to the large sample requirement of GWAS. Genomic loci, represented by the top leading single nucleotide polymorphisms (SNPs), were defined considering linkage disequilibrium. We investigated associations of SNPs with clinical traits and mapped them to genes using the GWAS Catalog (Buniello et al., 2019).ResultThree structural brain aging patterns, relative to resilient agers (A0), consistent across decades, emerged: A1, or ‘typical’ aging with low atrophy and WMHs, and two ‘advanced’ aging patterns, one showing elevated WMHs and modest atrophy (A2) and the other displaying severe, widespread atrophy and moderate WMH load (A3) (Figure 1). GWAS discovered eight and six genomic loci in [45,65) and [65,85) age groups, respectively (Table 1, Figure 2). The lead SNPs for A1 and A2 were previously associated with several cardiometabolic risk factors, WMHs, and regional brain volumes. Interestingly, rs4843552, previously associated with white matter microstructure and regional brain volumes, showed opposite effects for A1 (protection) and A2 (risk), consistent with the neuroimaging patterns. WDR41 in A3 group was previously associated with the age of Alzheimer’s disease onset (Herold et al., 2016).ConclusionReproducible neuroimaging patterns defined by regional atrophy and WMH burden were identified across CU individuals and demonstrated unique genetics. Further research is needed to elucidate the neuropathological pathways that mediate these relationships.

Differences in pathology, brain atrophy, and WMHs across APOE subtypes

AbstractBackgroundThe apolipoprotein (APOE) e4 allele is a known risk factor for Alzheimer’s disease (AD), while the e2 allele is thought to be protective. As few studies have examined the relationship between brain pathologies, atrophy, and white matter hyperintensities (WMHs) and APOE status in those with the e2e4 genotype, we do so here.MethodsWe analyzed Alzheimer’s Disease Neuro Imaging participants that had APOE genotyping and at least one of the following metrics: regional WMH load (obtained using Dadar 2018); ventricle size, hippocampal (HC) and entorhinal cortex (EC) volume, amyloid level (i.e., AV‐45), and phosphorylated tau (pTau), all downloaded from LONI. Participants were divided into one of four APOE allele profiles (E4 = e4e4 or e3e4; E2 = e2e2 or e2e3; E3 = e3e3; or E24 = e2e4, Fig.1). Linear mixed effects models examined the relationship between APOE profiles and each pathology while controlling for age, sex, education, and diagnostic status.ResultsAt baseline (Fig.2), E4 exhibited increased pTau (p<.001) compared to E2 and E3 and smaller HC and EC volumes than all other APOE profiles (p<.01). The E24 group exhibited smaller ventricles than E3 and E4 (p<.05). The E2 group exhibited less amyloid than all other APOE profiles (p<.001), while E4 and E24 had more amyloid compared to E3 (p<.001).Longitudinally (Fig.3), ventricular enlargement and HC atrophy significantly differed between all profiles (p<.001). Furthermore, E2 was observed to have slower WMH accumulation compared to all other profiles (p<.001), while E4 exhibited faster accumulation compared to E3 (p<.001). The E4 group exhibited increased EC atrophy compared to E3 and E2 (p<.001), and E24 exhibited more atrophy than E2 (p = .015). Amyloid increased faster in the E4 group compared to E3 and E2 (p<.001), while E2 amyloid progressed slower than E3 (p = .005).ConclusionsAPOE E4 positivity is associated with increased baseline and greater accumulation of pathologies and rates of neurodegeneration. APOE E2 positivity is associated with reduced pathology burden (i.e., WMH and amyloid) and less neurodegeneration as measured by ventricle size and hippocampal and entorhinal cortex volume over time. APOE E2E4 is similar to E4 (i.e., increased neurodegeneration and pathology burden) but with a slightly slower rate of change.

Longitudinal white matter hyperintensities in Down’s syndrome

AbstractBackgroundDown’s syndrome (DS) is a genetically determined form of Alzheimer’s disease (AD). Typically not complicated by systemic vascular risk factors, the DS population represents a unique opportunity to unravel the complex relationship between AD and cerebrovascular disease. Here, we aim at studying the evolution over time of white matter hyperintensities (WMHs), a key hallmark of cerebrovascular disease frequently found in AD, and to assess their topographical progression and associations with AD biomarkers in adults with DS.MethodThis longitudinal study included 43 DS adults (age = 39.8±10.4y; females = 41.9%; n = 37 asymptomatic AD, n = 6 symptomatic AD) and 81 euploid controls (age = 53.7±6.2y; females = 67.3%) from the Down‐Alzheimer Barcelona Neuroimaging Initiative, with at least 2 MRI visits with T1 and FLAIR acquisitions (time delay = 3.4±1.6y). WMHs were segmented using the Lesion Prediction Algorithm implemented in the Lesion Segmentation Toolbox (SPM12). WMH volumes were extracted in 36 regions of interest and used to compute total and regional annual volume changes. Non‐parametric statistical tests were performed to assess the effect of disease severity on WMH volume changes over time and define associations between longitudinal WMHs and baseline demographic and genetic data, and fluid biomarkers.ResultTotal annual WMH volume change increased in symptomatic DS compared to both asymptomatic DS and controls (Figure 1A). Regionally, WMH load increased in regions showing WMH at baseline in all groups (Figure 2). These changes were greater and more distributed, involving layers more distant from the ventricles (especially in occipito‐parietal regions) in symptomatic DS. Age, but not sex, intellectual disability, or APOEε4 status, was associated with longitudinal WMH in DS (Figure 1B‐F): the annual WMH volume changes in DS differed from controls at age ∼45y. Finally, WMH volume changes were negatively related to CSF‐Aβ42/40 and positively related to plasma and CSF ptau‐181 and neurofilament light (Figure 3).ConclusionPathological increases in WMH volume start ∼10 years before the mean age at onset for AD (53.8y) in DS and relate to underlying AD pathology and neurodegeneration. Regional analyses suggest that WMH volume changes increase with disease progression and progressively involve deeper white matter areas. Together, these results support that WMHs are tightly related to AD pathogenesis.

Association of white matter hyperintensities with long-term EGFR-TKI treatment and prediction of progression risk.

The purpose of this study was to test the hypothesis that brain white matter hyperintensities (WMH) are more common in patients receiving epidermal growth factor receptor tyrosine kinase inhibitor (EGFR-TKI) and identify clinical risk factors associated with WMH. This multiple-center, prospective cohort study was conducted from March 2017 to July 2020. Two groups of patients with non-small cell lung cancer (NSCLC) who received or did not receive EGFR-TKI were included and followed up for more than 24 months. The progression of WMH was defined as an increase of ≥1 point on the Fazekas visual rating scale between the baseline and at the 2-year follow-up. A modified Poisson regression model was performed to evaluate risk factors on increased WMH load. Among 286 patients with NSCLC, 194 (68%) patients with NSCLC who received EGFR-TKI and 92 (32%) patients with NSCLC without EGFR-TKI treatment were analyzed. Modified Poisson regression analysis showed that EGFR-TKI treatment was independently associated with the WMH progression (EGFR-TKI: aRR 2.72, 95% confidence interval [CI] 1.46-5.06, p=.002). Interleukin (IL)-2, IL-4, and IL-10 were associated with increased WMH in the adjusted model (IL-2: aRR 1.55 [95% CI 1.06-2.25], p = .023; IL-4: aRR 1.66 [95% CI 1.13-2.43], p = .010; IL-10: aRR 1.48 [95% CI 1.06-2.06], p = .020). Patients with NSCLC who received EGFR-TKI may be at higher risk of developing WMH or worsening of WMH burden. The impact of increased WMH lesions in these patients is to be further assessed. IL-2, IL-4, and IL-10 may be used as potential biomarkers to monitor the risk of increased WMH burden.

Machine‐learning derived white matter signatures and the risk of dementia, stroke and mortality: the Rotterdam Study

AbstractBackgroundWhite matter hyperintensities (WMH), loss of white matter (WM) volume and WM microstructural integrity are important risk indicators of stroke and dementia, albeit of heterogeneous nature. Use of machine learning techniques may unravel different patterns of white matter injury, with distinct underlying pathologies and disease risk.MethodWe measured WMH volume, WM volume, global fractional anisotropy (FA) and global mean diffusivity (MD) in 5205 participants (mean age 64.9 years, 56.0% women) of the population‐based Rotterdam study with brain MRI between 2005 and 2016. We performed hierarchical clustering on age‐standardized imaging parameters to identify separate clusters of white matter injury, and compare determinants across clusters. We then determined the association between clusters and risk of dementia, stroke and mortality, using Cox proportional hazard models adjusted for sex, education, APOE‐ε4 and cardiovascular risk factors.ResultWe identified four distinct white matter signatures: (i) cluster with above‐average microstructural integrity and little WM atrophy and WMH, (ii) cluster with above‐average microstructural integrity and little WMH, yet substantial WM atrophy, (iii) cluster with poor microstructural integrity and substantial WMH, yet little WM atrophy, and (iv) cluster with poor microstructural integrity, substantial WM atrophy, and average WMH load. There were no clear differences between clusters in APOE genotype and cardiometabolic risk factors, except for higher prevalence of hypertension in cluster (iv). Cluster (iii) contained more women, and participants in clusters (iii) and (iv) had more microbleeds and lacunes. During a median 10.7 years of follow‐up, 272, 210, and 844 cases of dementia, stroke, death occurred, respectively. In fully adjusted models, dementia risk was increased for clusters (ii) (HR 1.57 [95%CI:1.04‐2.37]), (iii) (HR 2.43 [95%CI:1.68‐3.52]) and (iv) (HR 1.78 [95%CI:1.22‐2.62], compared to cluster (i). Moderate associations of clusters (iii) and (iv) with incidence stroke attenuated and were no longer statistically significant after adjustment for cardiovascular risk factors, while only cluster (iii) was associated with a small increase in mortality (HR 1.30 [95%CI:1.07‐1.58]).ConclusionMachine‐learning derived white matter signatures are differentially associated with dementia, stroke and mortality. Further research is needed to pinpoint differences in underlying pathology, and assess potential benefit of clusters for risk stratification.

2 Higher White Matter Hyperintensity Load Adversely Affects Pre-Post Proximal Cognitive Training Performance in Healthy Older Adults

Objective:Cognitive training has shown promise for improving cognition in older adults. Aging involves a variety of neuroanatomical changes that may affect response to cognitive training. White matter hyperintensities (WMH) are one common age-related brain change, as evidenced by T2-weighted and Fluid Attenuated Inversion Recovery (FLAIR) MRI. WMH are associated with older age, suggestive of cerebral small vessel disease, and reflect decreased white matter integrity. Higher WMH load associates with reduced threshold for clinical expression of cognitive impairment and dementia. The effects of WMH on response to cognitive training interventions are relatively unknown. The current study assessed (a) proximal cognitive training performance following a 3-month randomized control trial and (b) the contribution of baseline whole-brain WMH load, defined as total lesion volume (TLV), on pre-post proximal training change.Participants and Methods:Sixty-two healthy older adults ages 65-84 completed either adaptive cognitive training (CT; n=31) or educational training control (ET; n=31) interventions. Participants assigned to CT completed 20 hours of attention/processing speed training and 20 hours of working memory training delivered through commercially-available Posit Science BrainHQ. ET participants completed 40 hours of educational videos. All participants also underwent sham or active transcranial direct current stimulation (tDCS) as an adjunctive intervention, although not a variable of interest in the current study. Multimodal MRI scans were acquired during the baseline visit. T1- and T2-weighted FLAIR images were processed using the Lesion Segmentation Tool (LST) for SPM12. The Lesion Prediction Algorithm of LST automatically segmented brain tissue and calculated lesion maps. A lesion threshold of 0.30 was applied to calculate TLV. A log transformation was applied to TLV to normalize the distribution of WMH. Repeated-measures analysis of covariance (RM-ANCOVA) assessed pre/post change in proximal composite (Total Training Composite) and sub-composite (Processing Speed Training Composite, Working Memory Training Composite) measures in the CT group compared to their ET counterparts, controlling for age, sex, years of education and tDCS group. Linear regression assessed the effect of TLV on post-intervention proximal composite and sub-composite, controlling for baseline performance, intervention assignment, age, sex, years of education, multisite scanner differences, estimated total intracranial volume, and binarized cardiovascular disease risk.Results:RM-ANCOVA revealed two-way group*time interactions such that those assigned cognitive training demonstrated greater improvement on proximal composite (Total Training Composite) and sub-composite (Processing Speed Training Composite, Working Memory Training Composite) measures compared to their ET counterparts. Multiple linear regression showed higher baseline TLV associated with lower pre-post change on Processing Speed Training sub-composite (ß = -0.19, p = 0.04) but not other composite measures.Conclusions:These findings demonstrate the utility of cognitive training for improving postintervention proximal performance in older adults. Additionally, pre-post proximal processing speed training change appear to be particularly sensitive to white matter hyperintensity load versus working memory training change. These data suggest that TLV may serve as an important factor for consideration when planning processing speed-based cognitive training interventions for remediation of cognitive decline in older adults.