APOE ε4 positivity predicts centrality of episodic memory nodes in patients with mild cognitive impairment: A cohort-based, graph theory-informed study of cognitive networks.

As network neuroscience can capture the systemic impact of APOE variability at a neuroimaging level, this study investigated the network-based cognitive endophenotypes of ε4-carriers and non-carriers across the continuum between normal ageing and Alzheimer's dementia (AD). We hypothesised that the impact of APOE-ε4 on cognitive functioning can be reliably captured by the measurement of graph-theory centrality. Cognitive networks were calculated in 8118 controls, 3482 MCI patients and 4573 AD patients, recruited in the National Alzheimer's Coordinating Center (NACC) database. Nodal centrality was selected as the neurofunctional readout of interest. ε4-carrier-vs.-non-carrier differences were tested in two independent NACC sub-cohorts assessed with either Version 1 or Version 2 of the Uniform Data Set neuropsychological battery. A significant APOE-dependent effect emerged from the analysis of the Logical-Memory nodes in MCI patients in both sub-cohorts. While non-carriers showed equal centrality in immediate and delayed recall, the latter was significantly less central among carriers (v1: bootstrapped confidence interval 0.107-0.667, p < 0.001; v2: bootstrapped confidence interval 0.018-0.432, p < 0.001). This indicates that, in carriers, delayed recall was, overall, significantly more weakly correlated with the other cognitive scores. These findings were replicated in the sub-groups of sole amnestic-MCI patients (n = 2971), were independent of differences in network communities, clinical severity or other demographic factors. No effects were found in the other two diagnostic groups. APOE-ε4 influences nodal properties of cognitive networks when patients are clinically classified as MCI. This highlights the importance of characterising the impact of risk factors on the wider cognitive network via network-neuroscience methodologies.

Accelerated atrophy in dopaminergic targets and medial temporo-parietal regions precedes the onset of delusions in patients with Alzheimer's disease.

People with Alzheimer's disease (AD) and delusions have worse quality of life and prognosis. However, early markers of delusions have not been identified yet. The present study investigated whether there are any detectable differences in grey matter (GM) volume and cognitive changes in the year before symptom onset between patients with AD who did and did not develop delusions. Two matched samples of AD patients, 63 who did (PT-D) and 63 who did not develop delusions (PT-ND) over 1 year, were identified from the Alzheimer's Disease Neuroimaging Initiative database. The Neuropsychiatric Inventory (NPI) was used to assess the presence of delusions. Sixty-three additional matched healthy controls (HC) were selected. Repeated-measures ANCOVA models were used to investigate group-by-time effects on the volume of selected GM regions of interest and on cognitive performance. No neurocognitive differences were observed between patient groups prior to symptom onset. Greater episodic memory decline and GM loss in bilateral caudate nuclei, medio-temporal and midline cingulo-parietal regions were found in the PT-D compared with the PT-ND group. A pattern of faster GM loss in brain areas typically affected by AD and in cortical and subcortical targets of dopaminergic pathways, paralleled by worsening of episodic memory and behavioural symptoms, may explain the emergence of delusions in patients with AD.

Large-Scale Functional Networks, Cognition and Brain Structures Supporting Social Cognition and Theory of Mind Performance in Prodromal to Mild Alzheimer's Disease.

Impairment of social cognition (SC) skills such as recognition and attribution of intentions and affective states of others (Theory of Mind, ToM) has been evidenced in Alzheimer's Disease (AD). This study investigated the neuropsychological, neuroanatomical and brain-functional underpinnings of SC processing to obtain an understanding of the social neurophenotype in early probable AD. Forty-six patients with mild cognitive impairment and mild probable AD underwent SC assessment including emotion recognition (Ekman-60-faces task) and cognitive and affective ToM (Reading-the-Mind-in-the-Eyes test and Story-based Empathy task). Linear models tested the association between SC scores and neuropsychological measures, grey matter maps and large-scale functional networks activity. The executive domain had the most predominant association with SC scores in the cognitive profile. Grey matter volume of the anterior cingulate, orbitofrontal, temporoparietal junction (TPJ), superior temporal, and cerebellar cortices were associated with ToM. Social cognition scores were associated with lower connectivity of the default-mode network with the prefrontal cortex. The right fronto-parietal network displayed higher inter-network connectivity in the right TPJ and insula while the salience network showed lower inter-network connectivity with the left TPJ and insula. Connectivity coupling alterations of executive-attentional networks may support default mode social-cognitive-associated decline through the recruitment of frontal executive mechanisms.

Altered Interplay Among Large-Scale Brain Functional Networks Modulates Multi-Domain Anosognosia in Early Alzheimer's Disease.

Decline in self-awareness is a prevalent symptom in Alzheimer's disease (AD). Current data suggest that an early breakdown in the brain's default mode network (DMN) is closely associated with the main symptomatic features in AD patients. In parallel, the integrity of the DMN has been shown to be heavily implicated in retained self-awareness abilities in healthy individuals and AD patients. However, the global contribution to awareness skills of other large-scale networks is still poorly understood. Resting-state functional magnetic resonance imaging (rs-fMRI) scans were acquired and pre-processed from 53 early-stage AD individuals. A group-level independent component analysis was run to isolate and reconstruct four intrinsic connectivity large-scale brain functional networks, namely left and right central executive fronto-parietal networks (FPN), salience network, and anterior and posterior DMN. Hypothesis-driven seed-based connectivity analyses were run to clarify the region-specific underpinnings of multi-domain anosognosia. Multiple regression models were run on large-scale network- and seed-based connectivity maps, including scores of memory, non-memory and total anosognosia obtained via the Measurement of Anosognosia Questionnaire. Memory anosognosia scores were associated with selective lower fronto-temporal connectivity and higher parieto-temporal connectivity. Non-memory anosognosia scores were associated with higher connectivity between the anterior DMN and the cerebellum, between the left medial prefrontal seeds and the contralateral prefrontal cortex, and between the left hippocampal seed and the left insula; lower connectivity was observed between the right prefrontal cortex and the right lingual seed. Lastly, total anosognosia scores were associated with large-scale network alterations, namely reduced left-FPN expression in the left posterior cingulate, reduced right-FPN expression in the left inferior lingual gyrus and adjacent inferior occipital cortex, and increased right-FPN expression in the right anterior cingulate. Seed-based analyses yielded significant connectivity differences only in the connectivity pattern associated with the left hippocampal seed by displaying lower intercommunication with the right prefrontal cortex, but higher connectivity with the left caudate nucleus. These findings support the hypothesis that alterations in functional connectivity of frontal lobe regions involved in executive-related mechanisms represent the neural correlates of domain-specific anosognosia in early AD. Up-regulated connectivity with subcortical structures appears to contribute to changes in the network dynamics interplay and fosters the appearance of anosognosia.

Neuroanatomical and cognitive correlates of domain-specific anosognosia in early Alzheimer's disease.

Anosognosia in Alzheimer's disease (AD) is defined as a lack of awareness for cognitive deficits or severity of disease. Previous studies have highlighted the link between anosognosia and damage to prefrontal functioning, i.e., executive functions. This study investigated the neuropsychological and neurostructural substrates of domain specific anosognosia in early AD. Fifty-three patients with a clinical diagnosis of early-AD and a reliable informant were administered the Measurement of Anosognosia Instrument, a validated tool to quantify anosognosia. Linear models were devised to test the association between the patient-informant discrepancy scores in the memory and non-memory domains and clinical profiles inclusive of cognitive scores and maps of grey matter. Total anosognosia scores were associated with episodic memory, semantic memory, visuoconstructive skills and volume of the anterior cingulate cortex (ACC), lingual gyrus, fusiform gyrus and thalamus. Memory anosognosia was associated with episodic memory and visuoconstructive skills. Non-memory anosognosia was associated with episodic and semantic memory and with volume of the ACC, precentral gyrus, superior frontal gyrus, postcentral gyrus, fusiform gyrus and lingual gyrus. Known as a region responsible for self-regulation and monitoring, reduction of grey matter in the frontal lobe was highlighted as crucial for the presence of anosognosia. Based on our findings, we argue that specific regions based in the frontal lobe could contribute to the functioning of the mnemonic comparator systems postulated by theoretical models of anosognosia. The cross-domain variability of cognitive correlates indicates that various computational mechanisms are at play in the presence of anosognosia.