Latent classes of mild cognitive impairment are associated with clinical outcomes and neuropathology: Analysis of data from the National Alzheimer's Coordinating Center

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.

Latent Class Proportional Hazards Regression with Heterogeneous Survival Data

Heterogeneous survival data are commonly present in chronic disease studies. Delineating meaningful disease subtypes directly linked to a survival outcome can generate useful scientific implications. In this work, we develop a latent class proportional hazards (PH) regression framework to address such an interest. We propose mixture proportional hazards modeling, which flexibly accommodates class-specific covariate effects while allowing for the baseline hazard function to vary across latent classes. Adapting the strategy of nonparametric maximum likelihood estimation, we derive an Expectation-Maximization (E-M) algorithm to estimate the proposed model. We establish the theoretical properties of the resulting estimators. Extensive simulation studies are conducted, demonstrating satisfactory finite-sample performance of the proposed method as well as the predictive benefit from accounting for the heterogeneity across latent classes. We further illustrate the practical utility of the proposed method through an application to a mild cognitive impairment (MCI) cohort in the Uniform Data Set.

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.

Evaluating the association between latent classes and competing risks outcomes with multiphenotype data

Latent class analysis is an intuitive tool to characterize disease phenotype heterogeneity. With data more frequently collected on multiple phenotypes in chronic disease studies, it is of rising interest to investigate how the latent classes embedded in one phenotype are related to another phenotype. Motivated by a cohort with mild cognitive impairment (MCI) from the Uniform Data Set (UDS), we propose and study a time-dependent structural model to evaluate the association between latent classes and competing risk outcomes that are subject to missing failure types. We develop a two-step estimation procedure which circumvents latent class membership assignment and is rigorously justified in terms of accounting for the uncertainty in classifying latent classes. The new method also properly addresses the realistic complications for competing risks outcomes, including random censoring and missing failure types. The asymptotic properties of the resulting estimator are established. Given that the standard bootstrapping inference is not feasible in the current problem setting, we develop analytical inference procedures, which are easy to implement. Our simulation studies demonstrate the advantages of the proposed method over benchmark approaches. We present an application to the MCI data from UDS, which uncovers a detailed picture of the neuropathological relevance of the baseline MCI subgroups.

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.

An Exploration of Subgroups of Neuropsychiatric Symptoms in Mild Cognitive Impairment and Their Risks of Conversion to Dementia or Death

OBJECTIVE

To explore the heterogeneity of neuropsychiatric symptom (NPS) complexes in individuals with mild cognitive impairment (MCI) and assess the relative risks of converting to dementia or dying.

DESIGN

Latent class analysis using 7,971 participants with MCI.

SETTING

Participants in the Uniform Data Set (UDS) from 39 NIH Alzheimer's Disease Centers.

PARTICIPANTS

Persons with a diagnosis of MCI at initial visit from each center and with either a Mini-Mental State Examination (MMSE) score of 22 or greater or an equivalent education-adjusted Montreal Cognitive Assessment (MoCA) score of 16 or greater.

MEASUREMENTS

Neuropsychiatric Inventory Questionnaire (NPI-Q) administered at initial visit.

RESULTS

In addition to a subgroup with mild or no NPS (relative frequency, 50%), three empirically-based subgroups of NPS were identified: 1) an "affect" or "negative mood" subgroup (27%) with depression, anxiety, apathy, nighttime disturbance, and change in appetite; 2) a "hyperactive" subgroup (14%) with agitation, irritability, and disinhibition; and 3) a "psychotic with additional severe NPS" subgroup (9%) with the highest risk of delusions and hallucinations, as well as highest risk of all other NPS. Each of these three subgroups had significantly higher risk of converting to dementia than the "mild NPS" class, with the "psychotic with additional severe NPS" subgroup possessing a 64% greater risk. The subgroups did not differ in their risks of death without dementia.

CONCLUSION

Our findings of three NPS subgroups in MCI characterized by affect, hyperactive, or psychotic features are largely consistent with a previous 3-factor model of NPS found in a demented population. The consistency of these findings across studies and samples, coupled with our results on the associated risks of converting to dementia, suggests that the NPS structure is robust, and warrants further consideration in classification models of MCI.

Diagnosing Mild Cognitive Impairment Among Racially Diverse Older Adults: Comparison of Consensus, Actuarial, and Statistical Methods

BACKGROUND

Actuarial and statistical methods have been proposed as alternatives to conventional methods of diagnosing mild cognitive impairment (MCI), with the aim of enhancing diagnostic and prognostic validity, but have not been compared in racially diverse samples.

OBJECTIVE

We compared the agreement of consensus, actuarial, and statistical MCI diagnostic methods, and their relationship to race and prognostic indicators among diverse older adults.

METHODS

Participants (N = 354; M age = 71; 68% White, 29% Black) were diagnosed with MCI or normal cognition (NC) according to clinical consensus, actuarial neuropsychological criteria (Jak/Bondi), and latent class analysis (LCA). We examined associations with race/ethnicity, longitudinal cognitive and functional change, and incident dementia.

RESULTS

MCI rates by consensus, actuarial criteria, and LCA were 44%, 53%, and 41%, respectively. LCA identified three MCI subtypes (memory; memory/language; memory/executive) and two NC classes (low normal; high normal). Diagnostic agreement was substantial, but agreement of the actuarial method with consensus and LCA was weaker than the agreement between consensus and LCA. Among cases classified as MCI by actuarial criteria only, Black participants were over-represented, and outcomes were generally similar to those of NC participants. Consensus diagnoses best predicted longitudinal outcomes overall, whereas actuarial diagnoses best predicted longitudinal functional change among Black participants.

CONCLUSION

Consensus diagnoses optimize specificity in predicting dementia, but among Black older adults, actuarial diagnoses may be more sensitive to early signs of decline. Results highlight the need for cross-cultural validity in MCI diagnosis and should be explored in community- and population-based samples.

Traditional Thai exercise (Ruesi Dadton) for improving motor and cognitive functions in mild cognitive impairment: a randomized controlled trial

This study determined the effectiveness of a 12-week cycle of Ruesi Dadton (RSD) among older adults with mild cognitive impairment (MCI), for improving cognitive and physical performance. Seventy-six participants were included and were divided equally into two groups. A group performed RSD exercise for 60 min, 3 times/wk for 12 weeks, and the control group did not perform RSD exercise. The primary endpoint was cognitive function, as assessed by the Montreal cognitive assessment (MoCA), Mini-Mental State Examination, verbal fluency (VF) test, and trail making test parts A and B (TMT-A and TMT-B). The secondary endpoints were the Timed Up and Go (TUG) test, handgrip, and gait speed results, which were used to evaluate the physical function. There were significant differences in the TMT-B and handgrip scores (P<0.05) between the two groups. Both groups had improved MoCA scores (P<0.05) and normal walking speeds (P<0.01). Additionally, the RSD group showed improved VF test (P<0.01), TMT-B (P<0.01), and TUG test (P<0.05); a negative correlation was found between MoCA and TUG test (P<0.05). However, high walking speed and handgrip (P<0.05) worsened in the control group. RSD exercise resulted in relevant improvements in the cognitive and physical functions in MCI.

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

BACKGROUND AND OBJECTIVES

Given prior work demonstrating that mild cognitive impairment (MCI) can be empirically differentiated into meaningful cognitive subtypes, we applied actuarial methods to comprehensive neuropsychological data from the University of California San Diego Alzheimer's Disease Research Center (ADRC) in order to identify cognitive subgroups within ADRC participants without dementia and to examine cognitive, biomarker, and neuropathologic trajectories.

METHODS

Cluster analysis was performed on baseline neuropsychological data (n = 738; mean age 71.8). Survival analysis examined progression to dementia (mean follow-up 5.9 years). CSF Alzheimer disease (AD) biomarker status and neuropathologic findings at follow-up were examined in a subset with available data.

RESULTS

Five clusters were identified: optimal cognitively normal (CN; n = 130) with above-average cognition, typical CN (n = 204) with average cognition, nonamnestic MCI (naMCI; n = 104), amnestic MCI (aMCI; n = 216), and mixed MCI (mMCI; n = 84). Progression to dementia differed across MCI subtypes (mMCI > aMCI > naMCI), with the mMCI group demonstrating the highest rate of CSF biomarker positivity and AD pathology at autopsy. Actuarial methods classified 29.5% more of the sample with MCI and outperformed consensus diagnoses in capturing those who had abnormal biomarkers, progressed to dementia, or had AD pathology at autopsy.

DISCUSSION

We identified subtypes of MCI and CN with differing cognitive profiles, clinical outcomes, CSF AD biomarkers, and neuropathologic findings over more than 10 years of follow-up. Results demonstrate that actuarial methods produce reliable cognitive phenotypes, with data from a subset suggesting unique biological and neuropathologic signatures. Findings indicate that data-driven algorithms enhance diagnostic sensitivity relative to consensus diagnosis for identifying older adults at risk for cognitive decline.

Interaction of neuropsychiatric symptoms with APOE ε4 and conversion to dementia in MCI patients in a Memory Clinic

To date, very few studies have been focused on the impact of the convergence of neuropsychiatric symptoms (NPS) and APOE ε4 on the conversion to dementia in patients with Mild Cognitive Impairment patients (MCI), and none has been based in a clinical setting. The objective of the study is to determine the predictive value of additive and multiplicative interactions of NPS and APOE ε4 status on the prediction of incident dementia among MCI patients monitored in a Memory Clinic. 1512 patients (aged 60 and older) with prevalent MCI were followed for a mean of 2 years. Neuropsychiatric symptoms were assessed at baseline using the Neuropsychiatric Inventory Questionnaire. Cox proportional hazards models were calculated. Additive interactions for depression, apathy, anxiety, agitation, appetite, or irritability and a positive ε4 carrier status were obtained, significantly increasing the hazard ratios of incident dementia (HR range 1.3-2.03). Synergistic interactions between NPS and APOE ε4 are identified among MCI patients when predicting incident dementia. The combination of the behavioral status and the genetic trait could be considered a useful strategy to identify the most vulnerable MCI patients to dementia conversion in a Memory Clinic.

Recognition Memory Performance as a Cognitive Marker of Prodromal Alzheimer's Disease

BACKGROUND

The utility of recognition memory for identifying persons with biomarker evidence of Alzheimer's disease (AD) is unclear since prior studies of mild cognitive impairment (MCI) relied only on clinical diagnosis and did not include simultaneous measures of central amyloidosis and tauopathy.

OBJECTIVE

We evaluated whether recognition memory and associated indices, including discriminability and response bias from signal detection theory, differentiate persons with amnestic MCI (aMCI) due to prodromal AD from non-prodromal AD.

METHOD

Sixty older adults with aMCI were classified as prodromal AD (n = 28) or non-prodromal AD (n = 32) based upon cerebrospinal fluid levels of amyloid-β and tau. Memory was assessed using the Hopkins Verbal Learning Test-Revised which includes free recall and recognition.

RESULTS

ANCOVAs adjusting for age indicated comparable (all p > 0.05) performances between prodromal and non-prodromal MCI groups respectively on traditional HVLT-R recognition measures of hits (mean±SD: 9.5±3.0 versus 10.9±1.7), false alarms (1.8±1.8 versus 1.5±1.5), and hits minus false alarms (7.7±3.0 versus 9.2±2.6). In contrast, discriminability (d'), which reflects how easily targets and distractors are distinguished, was significantly (p = 0.009) poorer in the prodromal versus non-prodromal groups (3.1±1.9 versus 4.8±2.0, effect size = 0.87). In addition, only d' significantly predicted group membership (OR = 0.66, CI = 0.48-0.92, p = 0.04). Response bias, the tendency to report that a target did or did not appear, was comparable between groups (0.08±1.1 versus -0.04±1.3).

CONCLUSION

Recognition discriminability is significantly poorer in aMCI with biomarker evidence of prodromal AD. In contrast to traditional recognition indices, discriminability from signal detection theory may be superior in identifying aMCI due to AD versus non-AD etiologies.

Identification of Patterns of Cognitive Impairment for Early Detection of Dementia

Early detection of dementia is crucial to devise effective interventions. Comprehensive cognitive tests, while being the most accurate means of diagnosis, are long and tedious, thus limiting their applicability to a large population, especially when periodic assessments are needed. The problem is compounded by the fact that people have differing patterns of cognitive impairment as they progress to different forms of dementia. This paper presents a novel scheme by which individual-specific patterns of impairment can be identified and used to devise personalized tests for periodic follow-up. Patterns of cognitive impairment are initially learned from a population cluster of combined normals and cognitively impaired subjects, using a set of standardized cognitive tests. Impairment patterns in the population are identified using a 2-step procedure involving an ensemble wrapper feature selection followed by cluster identification and analysis. These patterns have been shown to correspond to clinically accepted variants of Mild Cognitive Impairment (MCI), a prodrome of dementia. The learned clusters of patterns can subsequently be used to identify the most likely route of cognitive impairment, even for pre-symptomatic and apparently normal people. Baseline data of 24,000 subjects from the NACC database was used for the study.

Scalable and robust latent trajectory class analysis using artificial likelihood

Latent trajectory class analysis is a powerful technique to elucidate the structure underlying population heterogeneity. The standard approach relies on fully parametric modeling and is computationally impractical when the data include a large collection of non-Gaussian longitudinal features. We introduce a new approach, the first based on artificial likelihood concepts, that avoids undue modeling assumptions and is computationally tractable. We show that this new method provides reliable estimates of the underlying population structure and is from 20 to 200 times faster than conventional methods when the longitudinal features are non-Gaussian. We apply the approach to explore subgroups among research participants in the early stages of neurodegeneration.

Efficacy and Safety of Cholinesterase Inhibitors for Mild Cognitive Impairment:A Systematic Review and Meta-Analysis

BACKGROUND

The clinical benefit of cholinesterase inhibitors (ChEIs) for mild cognitive impairment (MCI) remains inconclusive.

OBJECTIVE

We performed a systematic review and meta-analysis of the efficacy/safety of ChEIs on subjects with MCI.

METHODS

We included randomized controlled trials (RCTs) of ChEIs in subjects with MCI, using cognitive function scores as a primary outcome measure.

RESULTS

Fourteen RCTs (six using donepezil, four using galantamine, and four using rivastigmine) with 5,278 subjects were included. We found no significant difference in cognitive function scores between the ChEIs and placebo groups [standardized mean difference (SMD) = -0.06, p = 0.38, I2 = 76% ]. However, in the secondary outcomes, ChEIs were associated with a lower incidence of progression to dementia compared with placebo (risk ratio = 0.76, the number needed to treat = 20). For safety outcomes, ChEIs were associated with a lower prevalence of fall than placebo. On the other hand, compared with placebo, ChEIs were associated with a higher incidence of discontinuation due to all causes, discontinuation due to adverse events, at least one adverse event, abnormal dreams, diarrhea, dizziness, headache, insomnia, loose stools, muscle cramps, nausea, vomiting, and weight loss.

CONCLUSIONS

Although ChEIs have a slight efficacy in the treatment of MCI, there are many safety issues. Therefore, ChEIs are difficult to recommend for MCI. However, the efficacy and safety of ChEIs on MCI with a biomarker-based diagnosis is unclear. Further RCTs are needed to confirm the efficacy and safety of ChEIs when used for individual neuropathological classifications of MCI.

Neuropsychological profile of amyloid-positive versus amyloid-negative amnestic Mild Cognitive Impairment

INTRODUCTION

Patients diagnosed with amnestic mild cognitive impairment (aMCI) are at high risk of progressing to dementia. It became possible, through the use of biomarkers, to diagnose those patients with aMCI who have Alzheimer's disease. However, it is presently unfeasible that all patients undergo biomarker testing. Since neuropsychological testing is required to make a formal diagnosis of aMCI, it would be interesting if it could be used to predict the amyloid status of patients with aMCI.

METHODS

Participants with aMCI, known amyloid status (Aβ+ or Aβ-) and a comprehensive neuropsychological evaluation, were selected from the Cognitive Complaints Cohort database for this study. Neuropsychological tests were compared in Aβ+ and Aβ- aMCI patients. A binary logistic regression analysis was conducted to model the probability of being amyloid positive.

RESULTS

Of the 216 aMCI patients studied, 117 were Aβ+ and 99 were Aβ-. Aβ+ aMCI patients performed worse on several memory tests, namely Word Total Recall, Logical Memory Immediate and Delayed Free Recall, and Verbal Paired Associate Learning, as well as on Trail Making Test B, an executive function test. In a binary logistic regression model, only Logical Memory Delayed Free Recall retained significance, so that for each additional score point in this test, the probability of being amyloid positive decreased by 30.6%. The resulting model correctly classified 64.6% of the aMCI cases regarding their amyloid status.

CONCLUSIONS

The neuropsychological assessment remains an essential step to diagnose and characterize patients with aMCI; however, neuropsychological tests have limited value to distinguish the aMCI patients who have amyloid pathology from those who might suffer from other clinical conditions.

Does Empirically Derived Classification of Individuals with Subjective Cognitive Complaints Predict Dementia?

(1) Background: Early identification of mild cognitive impairment (MCI) in people reporting subjective cognitive complaints (SCC) and the study of progression of cognitive decline are important issues in dementia research. This paper examines whether empirically derived procedures predict progression from MCI to dementia. (2) Methods: At baseline, 192 participants with SCC were diagnosed according to clinical criteria as cognitively unimpaired (70), single-domain amnestic MCI (65), multiple-domain amnestic MCI (33) and multiple-domain non-amnestic MCI (24). A two-stage hierarchical cluster analysis was performed for empirical classification. Categorical regression analysis was then used to assess the predictive value of the clusters obtained. Participants were re-assessed after 36 months. (3) Results: Participants were grouped into four empirically derived clusters: Cluster 1, similar to multiple-domain amnestic MCI; Cluster 2, characterized by subjective cognitive decline (SCD) but with low scores in language and working memory; Cluster 3, with specific deterioration in episodic memory, similar to single-domain amnestic MCI; and Cluster 4, with SCD but with scores above the mean in all domains. The majority of participants who progressed to dementia were included in Cluster 1. (4) Conclusions: Cluster analysis differentiated between MCI and SCD in a sample of people with SCC and empirical criteria were more closely associated with progression to dementia than standard criteria.