Localizing function in the brain has been an elusive long-term goal in the study of cognition. Prior studies have utilized four reference abilities (RAs) to capture cognition (Salthouse, 2009). Full-brain cortical networks have been tied to these abilities using common multi-voxel patterns across subjects in distinct age groups. Using voxel searchlights the current study explores purely local cortical representations of cognition, less commonly explored. This work analyzes 240 subjects’ responses to cognitive tasks from the four RAs. The current study further employs representational similarity analysis (RSA, Kriegeskorte, Mur, & Bandettini, 2008) to the similarity of brain activities from tasks within the same RA; RSA can capture representational consistencies within each subject even when exact voxel pattern may vary across subjects. We found distinct topographical localizations for each RA that were mostly consistent across age and suggested refinements of broader functional divisions of the brain from prior literature.

Background

The ideal participants for Alzheimer's disease (AD) clinical trials would show cognitive decline in the absence of treatment (i.e., placebo arm) and would also respond to the therapeutic intervention.

Objective

To investigate if predictive models can be an effective tool for identifying and excluding people unlikely to show cognitive decline as an enrichment strategy in AD trials.

Method

We used data from the placebo arms of two phase 3, double-blind trials, EXPEDITION and EXPEDITION2. Patients had 18 months of follow-up. Based on the longitudinal data from the placebo arm, we classified participants into two groups: one showed cognitive decline (any negative slope) and the other showed no cognitive decline (slope is zero or positive) on the Alzheimer's Disease Assessment Scale-Cognitive subscale (ADAS-cog). We used baseline data for EXPEDITION to train regression-based classifiers and machine learning classifiers to estimate probability of cognitive decline. Models were applied to EXPEDITION2 data to assess predicted performance in an independent sample. Features used in predictive models included baseline demographics, apolipoprotein E ε4 genotype, neuropsychological scores, functional scores, and volumetric magnetic resonance imaging.

Result

In EXPEDITION, 46.3% of placebo-treated patients showed no cognitive decline and the proportion was similar in EXPEDITION2 (45.6%). Models had high sensitivity and modest specificity in both the training (EXPEDITION) and replication samples (EXPEDITION2) for detecting the stable group. Positive predictive value of models was higher than the base prevalence of cognitive decline, and negative predictive value of models were higher than the base rate of participants who had stable cognition.

Conclusion

Excluding persons with AD unlikely to decline from the active and placebo arms of clinical trials using predictive models may boost the power of AD trials through selective inclusion of participants expected to decline.

Background

Predicting clinical course of cognitive decline can boost clinical trials' power and improve our clinical decision-making. Machine learning (ML) algorithms are specifically designed for the purpose of prediction; however. identifying optimal features or algorithms is still a challenge.

Objective

To investigate the accuracy of different ML methods and different features to classify cognitively normal (CN) individuals from Alzheimer's disease (AD) and to predict longitudinal outcome in participants with mild cognitive impairment (MCI).

Methods

A total of 1,329 participants from the Alzheimer's Disease Neuroimaging Initiative (ADNI) were included: 424 CN, 656 MCI, and 249 AD individuals. Four feature-sets at baseline (hippocampal volume and volume of 47 cortical and subcortical regions with and without demographics and APOE4) and six machine learning methods (decision trees, support vector machines, K-nearest neighbor, ensemble linear discriminant, boosted trees, and random forests) were used to classify participants with normal cognition from participants with AD. Subsequently the model with best classification performance was used for predicting clinical outcome of MCI participants.

Results

Ensemble linear discriminant models using demographics and all volumetric magnetic resonance imaging measures as feature-set showed the best performance in classification of CN versus AD participants (accuracy = 92.8%, sensitivity = 95.8%, and specificity = 88.3%). Prediction accuracy of future conversion from MCI to AD for this ensemble linear discriminant at 6, 12, 24, 36, and 48 months was 63.8% (sensitivity = 74.4, specificity = 63.1), 68.9% (sensitivity = 75.9, specificity = 67.8), 74.9% (sensitivity = 71.5, specificity = 76.3), 75.3%, (sensitivity = 65.2, specificity = 79.7), and 77.0% (sensitivity = 59.6, specificity = 86.1), respectively.

Conclusions

Machine learning models trained for classification of CN versus AD can improve our prediction ability of MCI conversion to AD.

Background

Amyloid-β positivity (Aβ+) based on PET imaging is part of the enrollment criteria for many of the clinical trials of Alzheimer's disease (AD), particularly in trials for amyloid-targeted therapy. Predicting Aβ positivity prior to PET imaging can decrease unnecessary patient burden and costs of running these trials.

Objective

The aim of this study was to evaluate the performance of a machine learning model in estimating the individual risk of Aβ+ based on gold-standard of PET imaging.

Methods

We used data from an amnestic mild cognitive impairment (aMCI) subset of the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort to develop and validate the models. The predictors of Aβ status included demographic and ApoE4 status in all models plus a combination of neuropsychological tests (NP), MRI volumetrics, and cerebrospinal fluid (CSF) biomarkers.

Results

The models that included NP and MRI measures separately showed an area under the receiver operating characteristics (AUC) of 0.74 and 0.72, respectively. However, using NP and MRI measures jointly in the model did not improve prediction. The models including CSF biomarkers significantly outperformed other models with AUCs between 0.89 to 0.92.

Conclusions

Predictive models can be effectively used to identify persons with aMCI likely to be amyloid positive on a subsequent PET scan.

The genetic basis of resilience, defined as better cognitive functioning than predicted based on neuroimaging or neuropathology, is not well understood. Our objective was to identify genetic variation associated with executive functioning resilience. We computed residuals from regression models of executive functioning, adjusting for age, sex, education, Hachinski score, and MRI findings (lacunes, cortical thickness, volumes of white matter hyperintensities and hippocampus). We estimated heritability and analyzed these residuals in models for each SNP. We further evaluated our most promising SNP result by evaluating cis-associations with brain levels of nearby (±100 kb) genes from a companion data set, and comparing expression levels in cortex and cerebellum from decedents with AD with those from other non-AD diseases. Complete data were available for 750 ADNI participants of European descent. Executive functioning resilience was highly heritable (H² = 0.76; S.E. = 0.44). rs3748348 on chromosome 14 in the region of RNASE13 was associated with executive functioning resilience (p-value = 4.31 × 10⁻⁷). rs3748348 is in strong linkage disequilibrium (D' of 1.00 and 0.96) with SNPs that map to TPPP2, a member of the α-synuclein family of proteins. We identified nominally significant associations between rs3748348 and expression levels of three genes (FLJ10357, RNASE2, and NDRG2). The strongest association was for FLJ10357 in cortex, which also had the most significant difference in expression between AD and non-AD brains, with greater expression in cortex of decedents with AD (p-value = 7 × 10⁻⁷). Further research is warranted to determine whether this signal can be replicated and whether other loci may be associated with cognitive resilience.

Mild cognitive impairment (MCI) and Alzheimer's disease (AD) are associated with a progressive loss of cognitive abilities. In the present report, we assessed the relationship of memory and executive function with brain structure in a sample of 810 Alzheimer's Disease Neuroimaging Initiative (ADNI) participants, including 188 AD, 396 MCI, and 226 healthy older adults (HC). Composite scores of memory (ADNI-Mem) and executive function (ADNI-Exec) were generated by applying modern psychometric theory to item-level data from ADNI's neuropsychological battery. We performed voxel-based morphometry (VBM) and surface-based association (SurfStat) analyses to evaluate relationships of ADNI-Mem and ADNI-Exec with grey matter (GM) density and cortical thickness across the whole brain in the combined sample and within diagnostic groups. We observed strong associations between ADNI-Mem and medial and lateral temporal lobe atrophy. Lower ADNI-Exec scores were associated with advanced GM and cortical atrophy across broadly distributed regions, most impressively in the bilateral parietal and temporal lobes. We also evaluated ADNI-Exec adjusted for ADNI-Mem, and found associations with GM density and cortical thickness primarily in the bilateral parietal, temporal, and frontal lobes. Within-group analyses suggest these associations are strongest in patients with MCI and AD. The present study provides insight into the spatially unbiased associations between brain atrophy and memory and executive function, and underscores the importance of structural brain changes in early cognitive decline.

Differences in brain metabolism as measured by FDG-PET in prodromal and early Alzheimer's disease (AD) have been consistently observed, with a characteristic parietotemporal hypometabolic pattern. However, exploration of brain metabolic correlates of more nuanced measures of cognitive function has been rare, particularly in larger samples. We analyzed the relationship between resting brain metabolism and memory and executive functioning within diagnostic group on a voxel-wise basis in 86 people with AD, 185 people with mild cognitive impairment (MCI), and 86 healthy controls (HC) from the Alzheimer's Disease Neuroimaging Initiative (ADNI). We found positive associations within AD and MCI but not in HC. For MCI and AD, impaired executive functioning was associated with reduced parietotemporal metabolism, suggesting a pattern consistent with known AD-related hypometabolism. These associations suggest that decreased metabolic activity in the parietal and temporal lobes may underlie the executive function deficits in AD and MCI. For memory, hypometabolism in similar regions of the parietal and temporal lobes were significantly associated with reduced performance in the MCI group. However, for the AD group, memory performance was significantly associated with metabolism in frontal and orbitofrontal areas, suggesting the possibility of compensatory metabolic activity in these areas. Overall, the associations between brain metabolism and cognition in this study suggest the importance of parietal and temporal lobar regions in memory and executive function in the early stages of disease and an increased importance of frontal regions for memory with increasing impairment.