INS NYC 2024 Program

Poster	Poster Session 02 Program Schedule 02/15/2024 08:00 am - 09:15 am Room: Shubert Complex (Posters 1-60) Poster Session 02: Aging \| MCI \| Neurodegenerative Disease - PART 1 Final Abstract #23 Data-driven Classification of Cognitively Normal and Mild Cognitive Impairment Subtypes Predicts Progression in the NACC Dataset Emily Edmonds, Banner Alzheimer’s Institute; University of Arizona, Tucson, United States Kelsey Thomas, Veterans Affairs San Diego Healthcare System; University of California San Diego, San Diego, United States Steven Rapcsak, Banner Alzheimer’s Institute; University of Arizona, Tucson, United States Shannon Lindemer, Banner Alzheimer’s Institute, Tucson, United States Lisa Delano-Wood, University of California San Diego; Veterans Affairs San Diego Healthcare System, La Jolla, United States David Salmon, University of California San Diego, La Jolla, United States Mark Bondi, University of California San Diego; Veterans Affairs San Diego Healthcare System, La Jolla, United States Category: MCI (Mild Cognitive Impairment) Keyword 1: mild cognitive impairment Keyword 2: dementia - Alzheimer's disease Objective: Previous work has shown that data-driven methods for classifying MCI based on comprehensive neuropsychological test data can reliably identify MCI subtypes that show stronger associations between cognition and dementia risk factors than do classifications based on conventional diagnostic methods such as the “consensus diagnosis” approach. We aimed to extend this work to the National Alzheimer’s Coordinating Center (NACC) sample. Participants and Methods: Cluster analysis was performed with baseline neuropsychological data from participants age 50 years or older (mean=71.6 years) without dementia in the NACC Uniform Data Set (n=26,255), and repeated with only data from the “normal cognition” subsample (n=16,005). The UDS neuropsychological tests examined included measures of memory (Immediate and Delayed Recall from Logical Memory or Craft Story), attention/working memory (Forward and Backward Digit Span or Number Span), processing speed/executive functioning (Trail Making Test, Parts A and B), and language (Category Fluency [animals, fruits, vegetables], Boston Naming Test or Multilingual Naming Test). Raw scores were converted into demographically-adjusted (age, education, sex) z-scores based on the performance of a robust cognitively normal group. Survival analyses examined progression to MCI or dementia. 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%). Rate of progression to dementia differed across the cluster-derived groups (oCN < tCN < aMCI < mMCI-Mild < mMCI-Severe). Comparison of the two classification methods showed that cluster analysis identified more MCI cases than consensus diagnosis. Within the NACC "normal cognition" subsample, five clusters emerged: High-All Domains (16.7%), Low-Attention/Working Memory (Low-WM; 22.1%), Low-Memory (36.3%), Amnestic MCI (aMCI; 16.7%), and Non-amnestic MCI (naMCI; 8.3%), with differing rates of progression to MCI/dementia (High-All < Low-WM = Low-Memory < aMCI < naMCI). Conclusions: Our data-driven method of classifying participants into MCI subtypes outperformed the consensus diagnostic approach by providing more precise information about risk for future progression, and revealing heterogeneity in cognitive performance and progression risk within the NACC "normal cognition" group. Given that the field is increasingly focused on early detection of subtle cognitive decline in the preclinical phase so treatment or prevention strategies can be implemented early, more nuanced classifications beyond the “normal cognition” category would be valuable. Results have implications for future research by demonstrating a method to identify empirically-derived subtypes of subtle cognitive decline and MCI that optimize prediction of risk for future MCI/dementia.

Poster

02/15/2024
08:00 am - 09:15 am
Room: Shubert Complex (Posters 1-60)

Poster Session 02: Aging | MCI | Neurodegenerative Disease - PART 1

Final Abstract #23

Data-driven Classification of Cognitively Normal and Mild Cognitive Impairment Subtypes Predicts Progression in the NACC Dataset

Emily Edmonds, Banner Alzheimer’s Institute; University of Arizona, Tucson, United States
Kelsey Thomas, Veterans Affairs San Diego Healthcare System; University of California San Diego, San Diego, United States
Steven Rapcsak, Banner Alzheimer’s Institute; University of Arizona, Tucson, United States
Shannon Lindemer, Banner Alzheimer’s Institute, Tucson, United States
Lisa Delano-Wood, University of California San Diego; Veterans Affairs San Diego Healthcare System, La Jolla, United States
David Salmon, University of California San Diego, La Jolla, United States
Mark Bondi, University of California San Diego; Veterans Affairs San Diego Healthcare System, La Jolla, United States

Category: MCI (Mild Cognitive Impairment)

Keyword 1: mild cognitive impairment
Keyword 2: dementia - Alzheimer's disease

Objective:

Previous work has shown that data-driven methods for classifying MCI based on comprehensive neuropsychological test data can reliably identify MCI subtypes that show stronger associations between cognition and dementia risk factors than do classifications based on conventional diagnostic methods such as the “consensus diagnosis” approach. We aimed to extend this work to the National Alzheimer’s Coordinating Center (NACC) sample.

Participants and Methods:

Cluster analysis was performed with baseline neuropsychological data from participants age 50 years or older (mean=71.6 years) without dementia in the NACC Uniform Data Set (n=26,255), and repeated with only data from the “normal cognition” subsample (n=16,005). The UDS neuropsychological tests examined included measures of memory (Immediate and Delayed Recall from Logical Memory or Craft Story), attention/working memory (Forward and Backward Digit Span or Number Span), processing speed/executive functioning (Trail Making Test, Parts A and B), and language (Category Fluency [animals, fruits, vegetables], Boston Naming Test or Multilingual Naming Test). Raw scores were converted into demographically-adjusted (age, education, sex) z-scores based on the performance of a robust cognitively normal group. Survival analyses examined progression to MCI or dementia.

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%). Rate of progression to dementia differed across the cluster-derived groups (oCN < tCN < aMCI < mMCI-Mild < mMCI-Severe). Comparison of the two classification methods showed that cluster analysis identified more MCI cases than consensus diagnosis. Within the NACC "normal cognition" subsample, five clusters emerged: High-All Domains (16.7%), Low-Attention/Working Memory (Low-WM; 22.1%), Low-Memory (36.3%), Amnestic MCI (aMCI; 16.7%), and Non-amnestic MCI (naMCI; 8.3%), with differing rates of progression to MCI/dementia (High-All < Low-WM = Low-Memory < aMCI < naMCI).

Conclusions:

Our data-driven method of classifying participants into MCI subtypes outperformed the consensus diagnostic approach by providing more precise information about risk for future progression, and revealing heterogeneity in cognitive performance and progression risk within the NACC "normal cognition" group. Given that the field is increasingly focused on early detection of subtle cognitive decline in the preclinical phase so treatment or prevention strategies can be implemented early, more nuanced classifications beyond the “normal cognition” category would be valuable. Results have implications for future research by demonstrating a method to identify empirically-derived subtypes of subtle cognitive decline and MCI that optimize prediction of risk for future MCI/dementia.