Data-Driven vs Consensus Diagnosis of MCI: Enhanced Sensitivity for Detection of Clinical, Biomarker, and Neuropathologic Outcomes

Comparing neuropsychological, typical, and ADNI criteria for the diagnosis of mild cognitive impairment in Vietnam-era veterans

OBJECTIVE

Neuropsychological criteria for mild cognitive impairment (MCI) more accurately predict progression to Alzheimer's disease (AD) and are more strongly associated with AD biomarkers and neuroimaging profiles than ADNI criteria. However, research to date has been conducted in relatively healthy samples with few comorbidities. Given that history of traumatic brain injury (TBI) and post-traumatic stress disorder (PTSD) are risk factors for AD and common in Veterans, we compared neuropsychological, typical (Petersen/Winblad), and ADNI criteria for MCI in Vietnam-era Veterans with histories of TBI or PTSD.

METHOD

267 Veterans (mean age = 69.8) from the DOD-ADNI study were evaluated for MCI using neuropsychological, typical, and ADNI criteria. Linear regressions adjusting for age and education assessed associations between MCI status and AD biomarker levels (cerebrospinal fluid [CSF] p-tau181, t-tau, and Aβ42) by diagnostic criteria. Logistic regressions adjusting for age and education assessed the effects of TBI severity and PTSD symptom severity simultaneously on MCI classification by each criteria.

RESULTS

Agreement between criteria was poor. Neuropsychological criteria identified more Veterans with MCI than typical or ADNI criteria, and were associated with higher CSF p-tau181 and t-tau. Typical and ADNI criteria were not associated with CSF biomarkers. PTSD symptom severity predicted MCI diagnosis by neuropsychological and ADNI criteria. History of moderate/severe TBI predicted MCI by typical and ADNI criteria.

CONCLUSIONS

MCI diagnosis using sensitive neuropsychological criteria is more strongly associated with AD biomarkers than conventional diagnostic methods. MCI diagnostics in Veterans would benefit from incorporation of comprehensive neuropsychological methods and consideration of the impact of PTSD.

Data-driven classification of cognitively normal and mild cognitive impairment subtypes predicts progression in the NACC dataset

INTRODUCTION

Data-driven neuropsychological methods can identify mild cognitive impairment (MCI) subtypes with stronger associations to dementia risk factors than conventional diagnostic methods.

METHODS

Cluster analysis used neuropsychological data from participants without dementia (mean age = 71.6 years) in the National Alzheimer's Coordinating Center (NACC) Uniform Data Set (n = 26,255) and the "normal cognition" subsample (n = 16,005). Survival analyses examined MCI or dementia progression.

RESULTS

Five clusters were identified: "Optimal" cognitively normal (oCN; 13.2%), "Typical" CN (tCN; 28.0%), Amnestic MCI (aMCI; 25.3%), Mixed MCI-Mild (mMCI-Mild; 20.4%), and Mixed MCI-Severe (mMCI-Severe; 13.0%). Progression to dementia differed across clusters (oCN < tCN < aMCI < mMCI-Mild < mMCI-Severe). Cluster analysis identified more MCI cases than consensus diagnosis. In the "normal cognition" subsample, five clusters emerged: High-All Domains (High-All; 16.7%), Low-Attention/Working Memory (Low-WM; 22.1%), Low-Memory (36.3%), Amnestic MCI (16.7%), and Non-amnestic MCI (naMCI; 8.3%), with differing progression rates (High-All < Low-WM = Low-Memory < aMCI < naMCI).

DISCUSSION

Our data-driven methods outperformed consensus diagnosis by providing more precise information about progression risk and revealing heterogeneity in cognition and progression risk within the NACC "normal cognition" group.

Empirically derived psychosocial-behavioral phenotypes in Black/African American and Hispanic/Latino older adults enrolled in HABS-HD: Associations with AD biomarkers and cognitive outcomes

INTRODUCTION

Identification of psychosocial-behavioral phenotypes to understand within-group heterogeneity in risk and resiliency to Alzheimer's disease (AD) within Black/African American and Hispanic/Latino older adults is essential for the implementation of precision health approaches.

METHODS

A cluster analysis was performed on baseline measures of socioeconomic resources (annual income, social support, occupational complexity) and psychiatric distress (chronic stress, depression, anxiety) for 1220 racially/ethnically minoritized adults enrolled in the Health and Aging Brain Study-Health Disparities (HABS-HD). Analyses of covariance adjusting for sociodemographic factors examined phenotype differences in cognition and plasma AD biomarkers.

RESULTS

The cluster analysis identified (1) Low Resource/High Distress (n = 256); (2) High Resource/Low Distress (n = 485); and (3) Low Resource/Low Distress (n = 479) phenotypes. The Low Resource/High Distress phenotype displayed poorer cognition and higher plasma neurofilament light chain; differences between the High Resource/Low Distress and Low Resource/Low Distress phenotypes were minimal.

DISCUSSION

The identification of psychosocial-behavioral phenotypes within racially/ethnically minoritized older adults is crucial to the development of targeted AD prevention and intervention efforts.

Cognition and Amyloid-β in Older Veterans: Characterization and Longitudinal Outcomes of Data-Derived Phenotypes

Background

Within older Veterans, multiple factors may contribute to cognitive difficulties. Beyond Alzheimer's disease (AD), psychiatric (e.g., PTSD) and health comorbidities (e.g., TBI) may also impact cognition.

Objective

This study aimed to derive subgroups based on objective cognition, subjective cognitive decline (SCD), and amyloid burden, and then compare subgroups on clinical characteristics, biomarkers, and longitudinal change in functioning and global cognition.

Methods

Cluster analysis of neuropsychological measures, SCD, and amyloid PET was conducted on 228 predominately male Vietnam-Era Veterans from the Department of Defense-Alzheimer's Disease Neuroimaging Initiative. Cluster-derived subgroups were compared on baseline characteristics as well as 1-year changes in everyday functioning and global cognition.

Results

The cluster analysis identified 3 groups. Group 1 (n = 128) had average-to-above average cognition with low amyloid burden. Group 2 (n = 72) had the lowest memory and language, highest SCD, and average amyloid burden; they also had the most severe PTSD, pain, and worst sleep quality. Group 3 (n = 28) had the lowest attention/executive functioning, slightly low memory and language, elevated amyloid and the worst AD biomarkers, and the fastest rate of everyday functioning and cognitive decline.

CONCLUSIONS

Psychiatric and health factors likely contributed to Group 2's low memory and language performance. Group 3 was most consistent with biological AD, yet attention/executive function was the lowest score. The complexity of older Veterans' co-morbid conditions may interact with AD pathology to show attention/executive dysfunction (rather than memory) as a prominent early symptom. These results could have important implications for the implementation of AD-modifying drugs in older Veterans.

Grey matter volume and CSF biomarkers predict neuropsychological subtypes of MCI

There is increasing evidence of different subtypes of individuals with mild cognitive impairment (MCI). An important line of research is whether neuropsychologically-defined subtypes have distinct patterns of neurodegeneration and cerebrospinal fluid (CSF) biomarker composition. In our study, we demonstrated that MCI participants of the ADNI database (N = 640) can be discriminated into 3 coherent neuropsychological subgroups. Our clustering approach revealed amnestic MCI, mixed MCI, and cluster-derived normal subgroups. Furthermore, classification modeling revealed that specific predictive features can be used to differentiate amnestic and mixed MCI from cognitively normal (CN) controls: CSF Aβ142 concentration for the former and CSF Aβ1-42 concentration, tau concentration as well as grey matter atrophy (especially in the temporal and occipital lobes) for the latter. In contrast, participants from the cluster-derived normal subgroup exhibited an identical profile to CN controls in terms of cognitive performance, brain structure, and CSF biomarker levels. Our comprehensive data analytics strategy provides further evidence that multimodal neuropsychological subtyping is both clinically and neurobiologically meaningful.

Brain Structure-Function Fusing Representation Learning Using Adversarial Decomposed-VAE for Analyzing MCI

Integrating the brain structural and functional connectivity features is of great significance in both exploring brain science and analyzing cognitive impairment clinically. However, it remains a challenge to effectively fuse structural and functional features in exploring the complex brain network. In this paper, a novel brain structure-function fusing-representation learning (BSFL) model is proposed to effectively learn fused representation from diffusion tensor imaging (DTI) and resting-state functional magnetic resonance imaging (fMRI) for mild cognitive impairment (MCI) analysis. Specifically, the decomposition-fusion framework is developed to first decompose the feature space into the union of the uniform and unique spaces for each modality, and then adaptively fuse the decomposed features to learn MCI-related representation. Moreover, a knowledge-aware transformer module is designed to automatically capture local and global connectivity features throughout the brain. Also, a uniform-unique contrastive loss is further devised to make the decomposition more effective and enhance the complementarity of structural and functional features. The extensive experiments demonstrate that the proposed model achieves better performance than other competitive methods in predicting and analyzing MCI. More importantly, the proposed model could be a potential tool for reconstructing unified brain networks and predicting abnormal connections during the degenerative processes in MCI.

A graph theoretic approach to neurodegeneration: five data-driven neuropsychological subtypes in mild cognitive impairment

Mild cognitive Impairment (MCI) is notoriously heterogenous in terms of clinical presentation, neuroimaging correlates, and subsequent progression. Predicting who will progress to dementia, which type of dementia, and over what timeframe is challenging. Previous work has attempted to identify MCI subtypes using neuropsychological measures in an effort to address this challenge; however, there is no consensus on approach, which may account for some of the variability. Using a hierarchical community detection approach, we examined cognitive subtypes within an MCI sample (from the Alzheimer's Disease Neuroimaging Initiative [ADNI] study). We then examined whether these subtypes were related to biomarkers (e.g., cortical volumes, fluorodeoxyglucose (FDG)-positron emission tomography (PET) hypometabolism) or clinical progression. We identified five communities (i.e., cognitive subtypes) within the MCI sample: 1) predominantly memory impairment, 2) predominantly language impairment, 3) cognitively normal, 4) multidomain, with notable executive dysfunction, 5) multidomain, with notable processing speed impairment. Community membership was significantly associated with 1) cortical volume in the hippocampus, entorhinal cortex, and fusiform cortex; 2) FDG PET hypometabolism in the posterior cingulate, angular gyrus, and inferior/middle temporal gyrus; and 3) conversion to dementia at follow up. Overall, community detection as an approach appears a viable method for identifying unique cognitive subtypes in a neurodegenerative sample that were linked to several meaningful biomarkers and modestly with progression at one year follow up.

Uncovering heterogeneous cognitive trajectories in mild cognitive impairment: a data-driven approach

BACKGROUND

Given the complex and progressive nature of mild cognitive impairment (MCI), the ability to delineate and understand the heterogeneous cognitive trajectories is crucial for developing personalized medicine and informing trial design. The primary goals of this study were to examine whether different cognitive trajectories can be identified within subjects with MCI and, if present, to characterize each trajectory in relation to changes in all major Alzheimer's disease (AD) biomarkers over time.

METHODS

Individuals with a diagnosis of MCI at the first visit and ≥ 1 follow-up cognitive assessment were selected from the Alzheimer's Disease Neuroimaging Initiative database (n = 936; age 73 ± 8; 40% female; 16 ± 3 years of education; 50% APOE4 carriers). Based on the Alzheimer's Disease Assessment Scale-Cognitive Subscale-13 (ADAS-Cog-13) total scores from baseline up to 5 years follow-up, a non-parametric k-means longitudinal clustering method was performed to obtain clusters of individuals with similar patterns of cognitive decline. We further conducted a series of linear mixed-effects models to study the associations of cluster membership with longitudinal changes in other cognitive measures, neurodegeneration, and in vivo AD pathologies.

RESULTS

Four distinct cognitive trajectories emerged. Cluster 1 consisted of 255 individuals (27%) with a nearly non-existent rate of change in the ADAS-Cog-13 over 5 years of follow-up and a healthy-looking biomarker profile. Individuals in the cluster 2 (n = 336, 35%) and 3 (n = 240, 26%) groups showed relatively mild and moderate cognitive decline trajectories, respectively. Cluster 4, comprising about 11% of our study sample (n = 105), exhibited an aggressive cognitive decline trajectory and was characterized by a pronouncedly abnormal biomarker profile.

CONCLUSIONS

Individuals with MCI show substantial heterogeneity in cognitive decline. Our findings may potentially contribute to improved trial design and patient stratification.

Mild Cognitive Impairment is Associated with Poorer Everyday Decision Making

BACKGROUND

Older adults are faced with many unique and highly consequential decisions such as those related to finances, healthcare, and everyday functioning (e.g., driving cessation). Given the significant impact of these decisions on independence, wellbeing, and safety, an understanding of how cognitive impairment may impact decision making in older age is important.

OBJECTIVE

To examine the impact of mild cognitive impairment (MCI) on responses to a modified version of the Short Portable Assessment of Capacity for Everyday Decision making (SPACED).

METHODS

Participants were community-dwelling, actively driving older adults (N = 301; M age = 77.1 years, SD = 5.1; 69.4% with a college degree or higher; 51.2% female; 95.3% White) enrolled in the Advancing Understanding of Transportation Options (AUTO) study. A generalized linear model adjusted for age, education, sex, randomization group, cognitive assessment method, and study site was used to examine the relationship between MCI status and decision making.

RESULTS

MCI status was associated with poorer decision making; participants with MCI missed an average of 2.17 times more points on the SPACED than those without MCI (adjusted mean ratio: 2.17, 95% CI: 1.02, 4.61, p = 0.044).

CONCLUSION

This finding supports the idea that older adults with MCI exhibit poorer decision-making abilities than cognitively normal older adults. It also suggests that older adults with MCI may exhibit poorer decision making across a wide range of decision contexts.

Midlife Neuropsychological Profiles and Associated Vascular Risk: The Bogalusa Heart Study

BACKGROUND

Individuals with Alzheimer's disease (AD) often present with coexisting vascular pathology that is expressed to different degrees and can lead to clinical heterogeneity.

OBJECTIVE

To examine the utility of unsupervised statistical clustering approaches in identifying neuropsychological (NP) test performance subtypes that closely correlate with carotid intima-media thickness (cIMT) in midlife.

METHODS

A hierarchical agglomerative and k-means clustering analysis based on NP scores (standardized for age, sex, and race) was conducted among 1,203 participants (age 48±5.3 years) from the Bogalusa Heart Study. Regression models assessed the association between cIMT ≥50th percentile and NP profiles, and global cognitive score (GCS) tertiles for sensitivity analysis.

RESULTS

Three NP profiles were identified: Mixed-low performance [16%, n = 192], scores ≥1 SD below the mean on immediate, delayed free recall, recognition verbal memory, and information processing; Average [59%, n = 704]; and Optimal [26%, n = 307] NP performance. Participants with greater cIMT were more likely to have a Mixed-low profile [OR = 3.10, 95% CI (2.13, 4.53), p < 0.001] compared to Optimal. After adjusting for education and cardiovascular (CV) risks, results remained. The association with GCS tertiles was more attenuated [lowest (34%, n = 407) versus highest (33%, n = 403) tertile: adjusted OR = 1.66, 95% CI (1.07, 2.60), p = 0.024].

CONCLUSION

As early as midlife, individuals with higher subclinical atherosclerosis were more likely to be in the Mixed-low profile, underscoring the potential malignancy of CV risk as related to NP test performance, suggesting that classification approaches may aid in identifying those at risk for AD/vascular dementia spectrum illness.

Decreased Electroencephalography Global Field Synchronization in Slow-Frequency Bands Characterizes Synaptic Dysfunction in Amnestic Subtypes of Mild Cognitive Impairment

Background

Mild cognitive impairment (MCI) is highly prevalent in a memory clinic setting and is heterogeneous regarding its clinical presentation, underlying pathophysiology, and prognosis. The most prevalent subtypes are single-domain amnestic MCI (sd-aMCI), considered to be a prodromal phase of Alzheimer’s disease (AD), and multidomain amnestic MCI (md-aMCI), which is associated with multiple etiologies. Since synaptic loss and dysfunction are the closest pathoanatomical correlates of AD-related cognitive impairment, we aimed to characterize it in patients with sd-aMCI and md-aMCI by means of resting-state electroencephalography (EEG) global field power (GFP), global field synchronization (GFS), and novel cerebrospinal fluid (CSF) synaptic biomarkers.

Methods

We included 52 patients with sd-aMCI (66.9 ± 7.3 years, 52% women) and 30 with md-aMCI (63.1 ± 7.1 years, 53% women). All patients underwent a detailed clinical assessment, resting-state EEG recordings and quantitative analysis (GFP and GFS in delta, theta, alpha, and beta bands), and analysis of CSF biomarkers of synaptic dysfunction, neurodegeneration, and AD-related pathology. Cognitive subtyping was based on a comprehensive neuropsychological examination. The Mini-Mental State Examination (MMSE) was used as an estimation of global cognitive performance. EEG and CSF biomarkers were included in a multivariate model together with MMSE and demographic variables, to investigate differences between sd-aMCI and md-aMCI.

Results

Patients with sd-aMCI had higher CSF phosphorylated tau, total tau and neurogranin levels, and lower values in GFS delta and theta. No differences were observed in GFP. The multivariate model showed that the most important synaptic measures for group separation were GFS theta, followed by GFS delta, GFP theta, CSF neurogranin, and GFP beta.

Conclusion

Patients with sd-aMCI when compared with those with md-aMCI have a neurophysiological and biochemical profile of synaptic damage, neurodegeneration, and amyloid pathology closer to that described in patients with AD. The most prominent signature in sd-aMCI was a decreased global synchronization in slow-frequency bands indicating that functional connectivity in slow frequencies is more specifically related to early effects of AD-specific molecular pathology.

Cognitive Heterogeneity and Risk of Progression in Data-Driven Subtle Cognitive Decline Phenotypes

BACKGROUND

There is increasing recognition of cognitive and pathological heterogeneity in early-stage Alzheimer's disease and other dementias. Data-driven approaches have demonstrated cognitive heterogeneity in those with mild cognitive impairment (MCI), but few studies have examined this heterogeneity and its association with progression to MCI/dementia in cognitively unimpaired (CU) older adults.

OBJECTIVE

We identified cluster-derived subgroups of CU participants based on comprehensive neuropsychological data and compared baseline characteristics and rates of progression to MCI/dementia or a Dementia Rating Scale (DRS) of ≤129 across subgroups.

METHODS

Hierarchical cluster analysis was conducted on individual baseline neuropsychological test scores from 365 CU participants in the UCSD Shiley-Marcos Alzheimer's Disease Research Center longitudinal cohort. Cox regressions examined the risk of progression to consensus diagnosis of MCI or dementia, or to DRS score ≤129, by cluster group.

RESULTS

Cluster analysis identified 5 groups: All-Average (n = 139), Low-Visuospatial (n = 46), Low-Executive (n = 51), Low-Memory/Language (n = 83), and Low-All Domains (n = 46). Subgroups had unique demographic and clinical characteristics. Rates of progression to MCI/dementia or to DRS ≤129 were faster for all subgroups (Low-All Domains progressed the fastest > Low Memory/Language≥Low-Visuospatial and Low-Executive) relative to the All-Average subgroup.

CONCLUSION

Faster progression in the Low-Visuospatial, Low-Executive, and Low-Memory/Language groups compared to the All-Average group suggests that there are multiple pathways and/or unique subtle cognitive decline profiles that ultimately lead to a diagnosis of MCI/dementia. Use of comprehensive neuropsychological test batteries that assess several domains may be a key first step toward an individualized approach to early detection and fewer missed opportunities for early intervention.

Mapping Actuarial Criteria for Parkinson’s Disease-Mild Cognitive Impairment onto Data-Driven Cognitive Phenotypes

Prevalence rates for mild cognitive impairment in Parkinson’s disease (PD-MCI) remain variable, obscuring the diagnosis’ predictive utility of greater dementia risk. A primary factor of this variability is inconsistent operationalization of normative cutoffs for cognitive impairment. We aimed to determine which cutoff was optimal for classifying individuals as PD-MCI by comparing classifications against data-driven PD cognitive phenotypes. Participants with idiopathic PD (n = 494; mean age 64.7 ± 9) completed comprehensive neuropsychological testing. Cluster analyses (K-means, Hierarchical) identified cognitive phenotypes using domain-specific composites. PD-MCI criteria were assessed using separate cutoffs (−1, −1.5, −2 SD) on ≥2 tests in a domain. Cutoffs were compared using PD-MCI prevalence rates, MCI subtype frequencies (single/multi-domain, executive function (EF)/non-EF impairment), and validity against the cluster-derived cognitive phenotypes (using chi-square tests/binary logistic regressions). Cluster analyses resulted in similar three-cluster solutions: Cognitively Average (n = 154), Low EF (n = 227), and Prominent EF/Memory Impairment (n = 113). The −1.5 SD cutoff produced the best model of cluster membership (PD-MCI classification accuracy = 87.9%) and resulted in the best alignment between PD-MCI classification and the empirical cognitive profile containing impairments associated with greater dementia risk. Similar to previous Alzheimer’s work, these findings highlight the utility of comparing empirical and actuarial approaches to establish concurrent validity of cognitive impairment in PD.