Genetic architecture of brain age and its causal relations with brain and mental disorders.

The difference between chronological age and the apparent age of the brain estimated from brain imaging data-the brain age gap (BAG)-is widely considered a general indicator of brain health. Converging evidence supports that BAG is sensitive to an array of genetic and nongenetic traits and diseases, yet few studies have examined the genetic architecture and its corresponding causal relationships with common brain disorders. Here, we estimate BAG using state-of-the-art neural networks trained on brain scans from 53,542 individuals (age range 3-95 years). A genome-wide association analysis across 28,104 individuals (40-84 years) from the UK Biobank revealed eight independent genomic regions significantly associated with BAG (p < 5 × 10-8) implicating neurological, metabolic, and immunological pathways - among which seven are novel. No significant genetic correlations or causal relationships with BAG were found for Parkinson's disease, major depressive disorder, or schizophrenia, but two-sample Mendelian randomization indicated a causal influence of AD (p = 7.9 × 10-4) and bipolar disorder (p = 1.35 × 10-2) on BAG. These results emphasize the polygenic architecture of brain age and provide insights into the causal relationship between selected neurological and neuropsychiatric disorders and BAG.

Individual differences in brain aging: heterogeneity in cortico-hippocampal but not caudate atrophy rates.

It is well documented that some brain regions, such as association cortices, caudate, and hippocampus, are particularly prone to age-related atrophy, but it has been hypothesized that there are individual differences in atrophy profiles. Here, we document heterogeneity in regional-atrophy patterns using latent-profile analysis of 1,482 longitudinal magnetic resonance imaging observations. The results supported a 2-group solution reflecting differences in atrophy rates in cortical regions and hippocampus along with comparable caudate atrophy. The higher-atrophy group had the most marked atrophy in hippocampus and also lower episodic memory, and their normal caudate atrophy rate was accompanied by larger baseline volumes. Our findings support and refine models of heterogeneity in brain aging and suggest distinct mechanisms of atrophy in striatal versus hippocampal-cortical systems.

Handedness and its genetic influences are associated with structural asymmetries of the cerebral cortex in 31,864 individuals.

Roughly 10% of the human population is left-handed, and this rate is increased in some brain-related disorders. The neuroanatomical correlates of hand preference have remained equivocal. We resampled structural brain image data from 28,802 right-handers and 3,062 left-handers (UK Biobank population dataset) to a symmetrical surface template, and mapped asymmetries for each of 8,681 vertices across the cerebral cortex in each individual. Left-handers compared to right-handers showed average differences of surface area asymmetry within the fusiform cortex, the anterior insula, the anterior middle cingulate cortex, and the precentral cortex. Meta-analyzed functional imaging data implicated these regions in executive functions and language. Polygenic disposition to left-handedness was associated with two of these regional asymmetries, and 18 loci previously linked with left-handedness by genome-wide screening showed associations with one or more of these asymmetries. Implicated genes included six encoding microtubule-related proteins: TUBB, TUBA1B, TUBB3, TUBB4A, MAP2, and NME7-mutations in the latter can cause left to right reversal of the visceral organs. There were also two cortical regions where average thickness asymmetry was altered in left-handedness: on the postcentral gyrus and the inferior occipital cortex, functionally annotated with hand sensorimotor and visual roles. These cortical thickness asymmetries were not heritable. Heritable surface area asymmetries of language-related regions may link the etiologies of hand preference and language, whereas nonheritable asymmetries of sensorimotor cortex may manifest as consequences of hand preference.

Deep neural networks learn general and clinically relevant representations of the ageing brain.

The discrepancy between chronological age and the apparent age of the brain based on neuroimaging data - the brain age delta - has emerged as a reliable marker of brain health. With an increasing wealth of data, approaches to tackle heterogeneity in data acquisition are vital. To this end, we compiled raw structural magnetic resonance images into one of the largest and most diverse datasets assembled (n=53542), and trained convolutional neural networks (CNNs) to predict age. We achieved state-of-the-art performance on unseen data from unknown scanners (n=2553), and showed that higher brain age delta is associated with diabetes, alcohol intake and smoking. Using transfer learning, the intermediate representations learned by our model complemented and partly outperformed brain age delta in predicting common brain disorders. Our work shows we can achieve generalizable and biologically plausible brain age predictions using CNNs trained on heterogeneous datasets, and transfer them to clinical use cases.

Associations of circulating C-reactive proteins, APOE ε4, and brain markers for Alzheimer's disease in healthy samples across the lifespan.

The apolipoprotein E gene ε4 allele (APOE ε4) and higher circulating level of C-reactive protein (CRP) have been extensively investigated as risk factors for Alzheimer's disease (AD). Paradoxically, APOE ε4 has been associated with lower levels of blood CRP in middle-aged and older populations. However, few studies have investigated this intriguing relation and its impact on neurological markers for AD in younger ages, nor across the whole lifespan. Here, we examine associations of blood CRP levels, APOE ε4, and biomarkers for AD in a cognitively healthy lifespan cohort (N up to 749; 20-81 years of age) and replicate the findings in UK Biobank (N = 304 322; 37-72 years of age), the developmental ABCD study (N = 10 283; 9-11 years of age), and a middle-aged sample (N = 339; 40-65 years of age). Hippocampal volume, brain amyloid-ß (Aß) plaque levels, cerebrospinal fluid (CSF) levels of Aß and tau species, and neurofilament protein light protein (NFL) were used as AD biomarkers in subsamples. In addition, we examined the genetic contribution to the variation of CRP levels over different CRP ranges using polygenic scores for CRP (PGS-CRP). Our results show APOE ε4 consistently associates with low blood CRP levels across all age groups (p < 0.05). Strikingly, both ε4 and PGS-CRP associated mainly with blood CRP levels within the low range (<5mg/L). We then show both APOE ε4 and high CRP levels associate with smaller hippocampus volumes across the lifespan (p < 0.025). APOE ε4 was associated with high Aß plaque levels in the brain (FDR-corrected p = 8.69x10-4), low levels of CSF Aß42 (FDR-corrected p = 6.9x10-2), and lower ratios of Aß42 to Aß40 (FDR-corrected p = 5.08x10-5). Blood CRP levels were weakly correlated with higher ratio of CSF Aß42 to Aß40 (p = 0.03, FDR-corrected p = 0.4). APOE ε4 did not correlate with blood concentrations of another 9 inflammatory cytokines, and none of these cytokines correlated with AD biomarkers. CONCLUSION: The inverse correlation between APOEε 4 and blood CRP levels existed before any pathological AD biomarker was observed, and only in the low CRP level range. Thus, we suggest to investigate whether APOEε 4 can confer risk by being associated with a lower inflammatory response to daily exposures, possibly leading to greater accumulation of low-grade inflammatory stress throughout life. A lifespan perspective is needed to understand this relationship concerning risk of developing AD.

The Functional Foundations of Episodic Memory Remain Stable Throughout the Lifespan.

It has been suggested that specific forms of cognition in older age rely largely on late-life specific mechanisms. Here instead, we tested using task-fMRI (n = 540, age 6-82 years) whether the functional foundations of successful episodic memory encoding adhere to a principle of lifespan continuity, shaped by developmental, structural, and evolutionary influences. We clustered regions of the cerebral cortex according to the shape of the lifespan trajectory of memory activity in each region so that regions showing the same pattern were clustered together. The results revealed that lifespan trajectories of memory encoding function showed a continuity through life but no evidence of age-specific mechanisms such as compensatory patterns. Encoding activity was related to general cognitive abilities and variations of grey matter as captured by a multi-modal independent component analysis, variables reflecting core aspects of cognitive and structural change throughout the lifespan. Furthermore, memory encoding activity aligned to fundamental aspects of brain organization, such as large-scale connectivity and evolutionary cortical expansion gradients. Altogether, we provide novel support for a perspective on memory aging in which maintenance and decay of episodic memory in older age needs to be understood from a comprehensive life-long perspective rather than as a late-life phenomenon only.

Poor Self-Reported Sleep is Related to Regional Cortical Thinning in Aging but not Memory Decline-Results From the Lifebrain Consortium.

We examined whether sleep quality and quantity are associated with cortical and memory changes in cognitively healthy participants across the adult lifespan. Associations between self-reported sleep parameters (Pittsburgh Sleep Quality Index, PSQI) and longitudinal cortical change were tested using five samples from the Lifebrain consortium (n = 2205, 4363 MRIs, 18-92 years). In additional analyses, we tested coherence with cell-specific gene expression maps from the Allen Human Brain Atlas, and relations to changes in memory performance. "PSQI # 1 Subjective sleep quality" and "PSQI #5 Sleep disturbances" were related to thinning of the right lateral temporal cortex, with lower quality and more disturbances being associated with faster thinning. The association with "PSQI #5 Sleep disturbances" emerged after 60 years, especially in regions with high expression of genes related to oligodendrocytes and S1 pyramidal neurons. None of the sleep scales were related to a longitudinal change in episodic memory function, suggesting that sleep-related cortical changes were independent of cognitive decline. The relationship to cortical brain change suggests that self-reported sleep parameters are relevant in lifespan studies, but small effect sizes indicate that self-reported sleep is not a good biomarker of general cortical degeneration in healthy older adults.

Age-Related Differences in Functional Asymmetry During Memory Retrieval Revisited: No Evidence for Contralateral Overactivation or Compensation.

Brain asymmetry is inherent to cognitive processing and seems to reflect processing efficiency. Lower frontal asymmetry is often observed in older adults during memory retrieval, yet it is unclear whether lower asymmetry implies an age-related increase in contralateral recruitment, whether less asymmetry reflects compensation, is limited to frontal regions, or predicts neurocognitive stability or decline. We assessed age-related differences in asymmetry across the entire cerebral cortex, using functional magnetic resonance imaging data from 89 young and 76 older adults during successful retrieval, and surface-based methods allowing direct homotopic comparison of activity between cortical hemispheres . An extensive left-asymmetric network facilitated retrieval in both young and older adults, whereas diverse frontal and parietal regions exhibited lower asymmetry in older adults. However, lower asymmetry was not associated with age-related increases in contralateral recruitment but primarily reflected either less deactivation in contralateral regions reliably signaling retrieval failure in the young or lower recruitment of the dominant hemisphere-suggesting that functional deficits may drive lower asymmetry in older brains, not compensatory activity. Lower asymmetry predicted neither current memory performance nor the extent of memory change across the preceding ~ 8 years in older adults. Together, these findings are inconsistent with a compensation account for lower asymmetry during retrieval and aging.

Genetic evidence for the causal effects of C-reactive protein on self-reported habitual sleep duration.

Inflammatory responses to acute stimuli are proposed to regulate sleep, but the relationship between chronic inflammation and habitual sleep duration is elusive. Here, we study this relation using genetically predicted level of chronic inflammation, indexed by CRP and IL6 signaling, and self-reported sleep duration. By Mendelian randomization analysis, we show that elevated CRP level within <10 mg/L has a homeostatic effect that facilitates maintaining 7-8 h sleep duration per day - making short-sleepers sleep longer (p = 2.42 × 10-2) and long-sleepers sleep shorter (1.87 × 10-7); but it is not associated with the overall sleep duration (p = 0.17). This homeostatic effect replicated in an independent CRP dataset. We observed causal effects of the soluble interleukin 6 receptor and gp130 on overall sleep duration (p = 1.62 × 10-8, p = 2.61 × 10-58, respectively), but these effects disappeared when CRP effects were accounted for in the model. Using polygenic score analysis, we found that the homeostatic effect of CRP on sleep duration stems primarily from the genetic variants within the CRP gene region: when genetic variants outside of this region were used to predict CRP levels, the opposite direction of effect was observed. In conclusion, we show that elevated CRP level may causally facilitate maintaining an optimal sleep duration that is beneficial to health, thus updating our current knowledge of immune regulation on sleep.

Constructing personalized characterizations of structural brain aberrations in patients with dementia using explainable artificial intelligence.

Deep learning approaches for clinical predictions based on magnetic resonance imaging data have shown great promise as a translational technology for diagnosis and prognosis in neurological disorders, but its clinical impact has been limited. This is partially attributed to the opaqueness of deep learning models, causing insufficient understanding of what underlies their decisions. To overcome this, we trained convolutional neural networks on structural brain scans to differentiate dementia patients from healthy controls, and applied layerwise relevance propagation to procure individual-level explanations of the model predictions. Through extensive validations we demonstrate that deviations recognized by the model corroborate existing knowledge of structural brain aberrations in dementia. By employing the explainable dementia classifier in a longitudinal dataset of patients with mild cognitive impairment, we show that the spatially rich explanations complement the model prediction when forecasting transition to dementia and help characterize the biological manifestation of disease in the individual brain. Overall, our work exemplifies the clinical potential of explainable artificial intelligence in precision medicine.

Circulating S100B levels at birth and risk of six major neuropsychiatric or neurological disorders: a two-sample Mendelian Randomization Study.

Circulating levels of the astrocytic marker S100B have been associated with risk of neuropsychiatric or neurological disorders. However, reported effects have been inconsistent, and no causal relations have yet been established. We applied two-sample Mendelian Randomization (MR) on the association statistics from genome-wide association studies (GWAS) for circulating S100B levels measured 5-7 days after birth (the iPSYCH sample) and in an older adult sample (mean age, 72.5 years; the Lothian sample), upon those derived from major depression disorder (MDD), schizophrenia (SCZ), bipolar disorder (BIP), autism spectral disorder (ASD), Alzheimer's disease (AD), and Parkinson's disease (PD). We studied the causal relations in the two S100B datasets for risk of these six neuropsychiatric disorders. MR suggested increased S100B levels 5-7 days after birth to causally increase the risk of MDD (OR = 1.014; 95%CI = 1.007-1.022; FDR-corrected p = 6.43×10-4). In older adults, MR suggested increased S100B levels to have a causal relation to the risk of BIP (OR = 1.075; 95%CI = 1.026-1.127; FDR-corrected p = 1.35×10-2). No significant causal relations were found for the other five disorders. We did not observe any evidence for reverse causality of these neuropsychiatric or neurological disorders on altered S100B levels. Sensitivity analyses using more stringent SNP-selection criteria and three alternative MR models suggested the results are robust. Altogether, our findings imply a small cause-effect relation for the previously reported associations of S100B and mood disorders. Such findings may provide a novel avenue for the diagnosis and management of disorders.

Tracing the development and lifespan change of population-level structural asymmetry in the cerebral cortex.

Cortical asymmetry is a ubiquitous feature of brain organization that is subtly altered in some neurodevelopmental disorders, yet we lack knowledge of how its development proceeds across life in health. Achieving consensus on the precise cortical asymmetries in humans is necessary to uncover the developmental timing of asymmetry and the extent to which it arises through genetic and later influences in childhood. Here, we delineate population-level asymmetry in cortical thickness and surface area vertex-wise in seven datasets and chart asymmetry trajectories longitudinally across life (4-89 years; observations = 3937; 70% longitudinal). We find replicable asymmetry interrelationships, heritability maps, and test asymmetry associations in large-scale data. Cortical asymmetry was robust across datasets. Whereas areal asymmetry is predominantly stable across life, thickness asymmetry grows in childhood and peaks in early adulthood. Areal asymmetry is low-moderately heritable (max h2SNP ~19%) and correlates phenotypically and genetically in specific regions, indicating coordinated development of asymmetries partly through genes. In contrast, thickness asymmetry is globally interrelated across the cortex in a pattern suggesting highly left-lateralized individuals tend towards left-lateralization also in population-level right-asymmetric regions (and vice versa), and exhibits low or absent heritability. We find less areal asymmetry in the most consistently lateralized region in humans associates with subtly lower cognitive ability, and confirm small handedness and sex effects. Results suggest areal asymmetry is developmentally stable and arises early in life through genetic but mainly subject-specific stochastic effects, whereas childhood developmental growth shapes thickness asymmetry and may lead to directional variability of global thickness lateralization in the population.

Asymmetric thinning of the cerebral cortex across the adult lifespan is accelerated in Alzheimer's disease.

Aging and Alzheimer's disease (AD) are associated with progressive brain disorganization. Although structural asymmetry is an organizing feature of the cerebral cortex it is unknown whether continuous age- and AD-related cortical degradation alters cortical asymmetry. Here, in multiple longitudinal adult lifespan cohorts we show that higher-order cortical regions exhibiting pronounced asymmetry at age ~20 also show progressive asymmetry-loss across the adult lifespan. Hence, accelerated thinning of the (previously) thicker homotopic hemisphere is a feature of aging. This organizational principle showed high consistency across cohorts in the Lifebrain consortium, and both the topological patterns and temporal dynamics of asymmetry-loss were markedly similar across replicating samples. Asymmetry-change was further accelerated in AD. Results suggest a system-wide dedifferentiation of the adaptive asymmetric organization of heteromodal cortex in aging and AD.

Individual variations in 'brain age' relate to early-life factors more than to longitudinal brain change.

Brain age is a widely used index for quantifying individuals' brain health as deviation from a normative brain aging trajectory. Higher-than-expected brain age is thought partially to reflect above-average rate of brain aging. Here, we explicitly tested this assumption in two independent large test datasets (UK Biobank [main] and Lifebrain [replication]; longitudinal observations ≈ 2750 and 4200) by assessing the relationship between cross-sectional and longitudinal estimates of brain age. Brain age models were estimated in two different training datasets (n ≈ 38,000 [main] and 1800 individuals [replication]) based on brain structural features. The results showed no association between cross-sectional brain age and the rate of brain change measured longitudinally. Rather, brain age in adulthood was associated with the congenital factors of birth weight and polygenic scores of brain age, assumed to reflect a constant, lifelong influence on brain structure from early life. The results call for nuanced interpretations of cross-sectional indices of the aging brain and question their validity as markers of ongoing within-person changes of the aging brain. Longitudinal imaging data should be preferred whenever the goal is to understand individual change trajectories of brain and cognition in aging.

Scientists who study the brain and aging are keen to find an effective way to measure brain health, which could help identify people at risk for dementia or memory problems. One popular marker is 'brain age'. This measurement uses a brain scan to estimate a person's chronological age, then compares the estimated brain age to the person's actual age to determine whether their brain is aging faster or slower than expected for their age. However, since brain age relies on one brain scan taken at one point in time, it is not clear whether it really measures brain aging or if it might capture brain differences that have been present throughout the individual's life. Studies comparing individual brain scans over several years would be necessary to know for sure. Now, Vidal-Piñeiro et al. show that the brain-age measurement does not reflect faster brain aging. In the experiments, the researchers compared repeated brain scans of thousands of individuals over 40 years of age. The experiments showed that deviations from normative brain age detected in a single scan reflected early life differences more than changes in the brain over time. For example, people with older-looking brains were more likely to have had a low birth weight or to have a combination of genes associated with having an older looking brain. Vidal-Piñeiro et al. show that brain age mostly reflects a pre-existing brain condition rather than brain aging. The experiments also suggest that genetics and early brain development likely have a strong impact on brain health throughout life. Future studies trying to test or develop brain-aging measurements should use serial measurements to track brain changes over time.

Neural correlates of durable memories across the adult lifespan: brain activity at encoding and retrieval.

Age-related effects on brain activity during encoding and retrieval of episodic memories are well documented. However, research typically tests memory only once, shortly after encoding. Retaining information over extended periods is critical, and there are reasons to expect age-related effects on the neural correlates of durable memories. Here, we tested whether age was associated with the activity elicited by durable memories. One hundred forty-three participants (22-78 years) underwent an episodic memory experiment where item-context relationships were encoded and tested twice. Participants were scanned during encoding and the first test. Memories retained after 90 minutes but later forgotten were classified as transient, whereas memories retained after 5 weeks were classified as durable. Durable memories were associated with greater encoding activity in inferior lateral parietal and posteromedial regions and greater retrieval activity in frontal and insular regions. Older adults exhibited lower posteromedial activity during encoding and higher frontal activity during retrieval, possibly reflecting greater involvement of control processes. This demonstrates that long-lasting memories are supported by specific patterns of cortical activity that are related to age.

The effects of tDCS upon sustained visual attention are dependent on cognitive load.

Transcranial Direct Current Stimulation (tDCS) modulates the excitability of neuronal responses and consequently can affect performance on a variety of cognitive tasks. However, the interaction between cognitive load and the effects of tDCS is currently not well-understood. We recorded the performance accuracy of participants on a bilateral multiple object tracking task while undergoing bilateral stimulation assumed to enhance (anodal) and decrease (cathodal) neuronal excitability. Stimulation was applied to the posterior parietal cortex (PPC), a region inferred to be at the centre of an attentional tracking network that shows load-dependent activation. 34 participants underwent three separate stimulation conditions across three days. Each subject received (1) left cathodal / right anodal PPC tDCS, (2) left anodal / right cathodal PPC tDCS, and (3) sham tDCS. The number of targets-to-be-tracked was also manipulated, giving a low (one target per visual field), medium (two targets per visual field) or high (three targets per visual field) tracking load condition. It was found that tracking performance at high attentional loads was significantly reduced in both stimulation conditions relative to sham, and this was apparent in both visual fields, regardless of the direction of polarity upon the brain's hemispheres. We interpret this as an interaction between cognitive load and tDCS, and suggest that tDCS may degrade attentional performance when cognitive networks become overtaxed and unable to compensate as a result. Systematically varying cognitive load may therefore be a fruitful direction to elucidate the effects of tDCS upon cognitive functions.