Ageing is commonly associated with changes to segregation and integration of functional brain networks, but, in isolation, current network-based approaches struggle to elucidate changes across the many axes of functional organisation. However, the advent of gradient mapping techniques in neuroimaging provides a new means of studying functional organisation in a multi-dimensional connectivity space. Here, we studied ageing and behaviourally-relevant differences in a three-dimensional connectivity space using the Cambridge Centre for Ageing Neuroscience cohort (n = 643). Building on gradient mapping techniques, we developed a set of measures to quantify the dispersion within and between functional communities. We detected a strong shift of the visual network across the adult lifespan from an extreme to a more central position in the 3D gradient space. In contrast, the dispersion distance between transmodal communities (dorsal attention, ventral attention, frontoparietal and default mode) did not change. However, these communities themselves were increasingly dispersed with increasing age, reflecting more dissimilar functional connectivity profiles within each community. Increasing dispersion of frontoparietal, attention and default mode networks, in particular, were associated negatively with cognition, measured by fluid intelligence. By using a technique that explicitly captures the ordering of functional systems in a multi-dimensional hierarchical framework, we identified behaviorally-relevant age-related differences of within and between network organisation. We propose that the study of functional gradients across the adult lifespan could provide insights that may facilitate the development of new strategies to maintain cognitive ability across the lifespan in health and disease.

Neuroinflammation is a key part of the etio-pathogenesis of Alzheimer9s disease (AD). We tested the relationship between neuroinflammation and the disruption of functional connectivity in large-scale networks, and their joint influence on cognitive impairment. We combined [¹¹C]PK11195 positron emission tomography (PET) and resting-state functional magnetic resonance imaging (rs-fMRI) in 28 patients (12 females/16 males) with clinical diagnosis of probable AD or mild cognitive impairment with positive PET biomarker for amyloid, and 14 age-, sex-, and education-matched healthy controls (8 females/6 males). Source-based “inflammetry” was used to extract principal components of [¹¹C]PK11195 PET signal variance across all participants. rs-fMRI data were preprocessed via independent component analyses to classify neuronal and non-neuronal signals. Multiple linear regression models identified sources of signal covariance between neuroinflammation and brain connectivity profiles, in relation to the diagnostic group (patients, controls) and cognitive status. Patients showed significantly higher [¹¹C]PK11195 binding relative to controls, in a distributed spatial pattern including the hippocampus, frontal, and inferior temporal cortex. Patients with enhanced loading on this [¹¹C]PK11195 binding distribution displayed diffuse abnormal functional connectivity. The expression of a stronger association between such abnormal connectivity and higher levels of neuroinflammation correlated with worse cognitive deficits. Our study suggests that neuroinflammation relates to the pathophysiological changes in network function that underlie cognitive deficits in Alzheimer9s disease. Neuroinflammation, and its association with functionally-relevant reorganization of brain networks, is proposed as a target for emerging immunotherapeutic strategies aimed at preventing or slowing the emergence of dementia. SIGNIFICANCE STATEMENT Neuroinflammation is an important aspect of Alzheimer9s disease (AD), but it was not known whether the influence of neuroinflammation on brain network function in humans was important for cognitive deficit. Our study provides clear evidence that in vivo neuroinflammation in AD impairs large-scale network connectivity; and that the link between neuro inflammation and functional network connectivity is relevant to cognitive impairment. We suggest that future studies should address how neuroinflammation relates to network function as AD progresses, and whether the neuroinflammation in AD is reversible, as the basis of immunotherapeutic strategies to slow the progression of AD.

In functional magnetic resonance imaging (fMRI) research one is typically interested in neural activity. However, the blood‐oxygenation level‐dependent (BOLD) signal is a composite of both neural and vascular activity. As factors such as age or medication may alter vascular function, it is essential to account for changes in neurovascular coupling when investigating neurocognitive functioning with fMRI. The resting‐state fluctuation amplitude (RSFA) in the fMRI signal (rsfMRI) has been proposed as an index of vascular reactivity. The RSFA compares favourably with other techniques such as breath‐hold and hypercapnia, but the latter are more difficult to perform in some populations, such as older adults. The RSFA is therefore a candidate for use in adjusting for age‐related changes in vascular reactivity in fMRI studies. The use of RSFA is predicated on its sensitivity to vascular rather than neural factors; however, the extent to which each of these factors contributes to RSFA remains to be characterized. The present work addressed these issues by comparing RSFA (i.e., rsfMRI variability) to proxy measures of (i) cardiovascular function in terms of heart rate (HR) and heart rate variability (HRV) and (ii) neural activity in terms of resting state magnetoencephalography (rsMEG). We derived summary scores of RSFA, a sensorimotor task BOLD activation, cardiovascular function and rsMEG variability for 335 healthy older adults in the population‐based Cambridge Centre for Ageing and Neuroscience cohort (Cam‐CAN; www.cam-can.com). Mediation analysis revealed that the effects of ageing on RSFA were significantly mediated by vascular factors, but importantly not by the variability in neuronal activity. Furthermore, the converse effects of ageing on the rsMEG variability were not mediated by vascular factors. We then examined the effect of RSFA scaling of task‐based BOLD in the sensorimotor task. The scaling analysis revealed that much of the effects of age on task‐based activation studies with fMRI do not survive correction for changes in vascular reactivity, and are likely to have been overestimated in previous fMRI studies of ageing. The results from the mediation analysis demonstrate that RSFA is modulated by measures of vascular function and is not driven solely by changes in the variance of neural activity. Based on these findings we propose that the RSFA scaling method is articularly useful in large scale and longitudinal neuroimaging studies of ageing, or with frail participants, where alternative measures of vascular reactivity are impractical. Hum Brain Mapp 36:2248–2269, 2015. © 2015 The Authors Human Brain Mapping Published by Wiley Periodicals, Inc.

The maintenance of wellbeing across the lifespan depends on the preservation of cognitive function. We propose that successful cognitive aging is determined by interactions both within and between large-scale functional brain networks. Such connectivity can be estimated from task-free functional magnetic resonance imaging (fMRI), also known as resting-state fMRI (rs-fMRI). However, common correlational methods are confounded by age-related changes in the neurovascular signaling. To estimate network interactions at the neuronal rather than vascular level, we used generative models that specified both the neural interactions and a flexible neurovascular forward model. The networks' parameters were optimized to explain the spectral dynamics of rs-fMRI data in 602 healthy human adults from population-based cohorts who were approximately uniformly distributed between 18 and 88 years (www.cam-can.com). We assessed directed connectivity within and between three key large-scale networks: the salience network, dorsal attention network, and default mode network. We found that age influences connectivity both within and between these networks, over and above the effects on neurovascular coupling. Canonical correlation analysis revealed that the relationship between network connectivity and cognitive function was age-dependent: cognitive performance relied on neural dynamics more strongly in older adults. These effects were driven partly by reduced stability of neural activity within all networks, as expressed by an accelerated decay of neural information. Our findings suggest that the balance of excitatory connectivity between networks, and the stability of intrinsic neural representations within networks, changes with age. The cognitive function of older adults becomes increasingly dependent on these factors.SIGNIFICANCE STATEMENTMaintaining cognitive function is critical to successful aging. To study the neural basis of cognitive function across the lifespan, we studied a large population-based cohort (n= 602, 18–88 years), separating neural connectivity from vascular components of fMRI signals. Cognitive ability was influenced by the strength of connection within and between functional brain networks, and this positive relationship increased with age. In older adults, there was more rapid decay of intrinsic neuronal activity in multiple regions of the brain networks, which related to cognitive performance. Our data demonstrate increased reliance on network flexibility to maintain cognitive function, in the presence of more rapid decay of neural activity. These insights will facilitate the development of new strategies to maintain cognitive ability.

Tau pathology, neuroinflammation, and neurodegeneration are key aspects of Alzheimer’s disease. Understanding whether these features predict cognitive decline, alone or in combination, is crucial to develop new prognostic measures and enhanced stratification for clinical trials. Here, we studied how baseline assessments of in vivo tau pathology (measured by ¹⁸F-AV-1451 PET), neuroinflammation (measured by ¹¹C-PK11195 PET) and brain atrophy (derived from structural MRI) predicted longitudinal cognitive changes in patients with Alzheimer’s disease pathology. Twenty-six patients (n = 12 with clinically probable Alzheimer’s dementia and n = 14 with amyloid-positive mild cognitive impairment) and 29 healthy control subjects underwent baseline assessment with ¹⁸F-AV-1451 PET, ¹¹C-PK11195 PET, and structural MRI. Cognition was examined annually over the subsequent 3 years using the revised Addenbrooke’s Cognitive Examination. Regional grey matter volumes, and regional binding of ¹⁸F-AV-1451 and ¹¹C-PK11195 were derived from 15 temporo-parietal regions characteristically affected by Alzheimer’s disease pathology. A principal component analysis was used on each imaging modality separately, to identify the main spatial distributions of pathology. A latent growth curve model was applied across the whole sample on longitudinal cognitive scores to estimate the rate of annual decline in each participant. We regressed the individuals’ estimated rate of cognitive decline on the neuroimaging components and examined univariable predictive models with single-modality predictors, and a multi-modality predictive model, to identify the independent and combined prognostic value of the different neuroimaging markers. Principal component analysis identified a single component for the grey matter atrophy, while two components were found for each PET ligand: one weighted to the anterior temporal lobe, and another weighted to posterior temporo-parietal regions. Across the whole-sample, the single-modality models indicated significant correlations between the rate of cognitive decline and the first component of each imaging modality. In patients, both stepwise backward elimination and Bayesian model selection revealed an optimal predictive model that included both components of ¹⁸F-AV-1451 and the first (i.e. anterior temporal) component for ¹¹C-PK11195. However, the MRI-derived atrophy component and demographic variables were excluded from the optimal predictive model of cognitive decline. We conclude that temporo-parietal tau pathology and anterior temporal neuroinflammation predict cognitive decline in patients with symptomatic Alzheimer’s disease pathology. This indicates the added value of PET biomarkers in predicting cognitive decline in Alzheimer’s disease, over and above MRI measures of brain atrophy and demographic data. Our findings also support the strategy for targeting tau and neuroinflammation in disease-modifying therapy against Alzheimer’s disease.

The syndromes caused by frontotemporal lobar degeneration have highly heterogeneous and overlapping clinical features. There has been great progress in the refinement of clinical diagnostic criteria in the past decade, but we propose that a better understanding of aetiology, pathophysiology and symptomatic treatments can arise from a transdiagnostic approach to clinical phenotype and brain morphometry. In a cross-sectional epidemiological study, we examined 310 patients with a syndrome likely to be caused by frontotemporal lobar degeneration, including behavioural variant frontotemporal dementia, non-fluent, and semantic variants of primary progressive aphasia (PPA), progressive supranuclear palsy and corticobasal syndrome. We included patients with logopenic PPA and those who met criteria for PPA but not a specific subtype. To date, 49 patients have a neuropathological diagnosis. A principal component analysis identified symptom dimensions that broadly recapitulated the core features of the main clinical syndromes. However, the subject-specific scores on these dimensions showed considerable overlap across the diagnostic groups. Sixty-two per cent of participants had phenotypic features that met the diagnostic criteria for more than one syndrome. Behavioural disturbance was prevalent in all groups. Forty-four per cent of patients with corticobasal syndrome had progressive supranuclear palsy-like features and 30% of patients with progressive supranuclear palsy had corticobasal syndrome-like features. Many patients with progressive supranuclear palsy and corticobasal syndrome had language impairments consistent with non-fluent variant PPA while patients with behavioural variant frontotemporal dementia often had semantic impairments. Using multivariate source-based morphometry on a subset of patients (n = 133), we identified patterns of covarying brain atrophy that were represented across the diagnostic groups. Canonical correlation analysis of clinical and imaging components found three key brain-behaviour relationships, with a continuous spectrum across the cohort rather than discrete diagnostic entities. In the 46 patients with follow-up (mean 3.6 years) syndromic overlap increased with time. Together, these results show that syndromes associated with frontotemporal lobar degeneration do not form discrete mutually exclusive categories from their clinical features or structural brain changes, but instead exist in a multidimensional spectrum. Patients often manifest diagnostic features of multiple disorders while deficits in behaviour, movement and language domains are not confined to specific diagnostic groups. It is important to recognize individual differences in clinical phenotype, both for clinical management and to understand pathogenic mechanisms. We suggest that a transdiagnostic approach to the spectrum of frontotemporal lobar degeneration syndromes provides a useful framework with which to understand disease aetiology, progression, and heterogeneity and to target future treatments to a higher proportion of patients.