INS NYC 2024 Program

Paper	Paper Session 10 Program Schedule 02/16/2024 10:15 am - 11:40 am Room: West Side Ballroom - Salon 3 Paper Session 10: Alzheimer's Disease Final Abstract #4 The Mobile Toolbox (MTB) for Assessing Cognition Remotely in Preclinical Alzheimer’s Disease: Associations with In-Clinic Cognitive Assessments and Amyloid and Tau PET. Roos Jutten, Massachusetts General Hospital, Harvard Medical School, Boston, United States Jessa Burling, Massachusetts General Hospital, Harvard Medical School, Boston, United States Michael Properzi, Massachusetts General Hospital, Harvard Medical School, Boston, United States Rebecca Amariglio, Massachusetts General Hospital, Brigham and Women's Hospital, Harvard Medical School, Boston, United States Gad Marshall, Massachusetts General Hospital, Brigham and Women's Hospital, Harvard Medical School, Boston, United States Kathryn Papp, Massachusetts General Hospital, Brigham and Women's Hospital, Harvard Medical School, Boston, United States Keith Johnson, Massachusetts General Hospital, Harvard Medical School, Boston, United States Reisa Sperling, Massachusetts General Hospital, Brigham and Women's Hospital, Harvard Medical School, Boston, United States Dorene Rentz, Massachusetts General Hospital, Brigham and Women's Hospital, Harvard Medical School, Boston, United States Category: Dementia (Alzheimer's Disease) Keyword 1: cognitive functioning Keyword 2: computerized neuropsychological testing Keyword 3: positron emission tomography Objective: There is an urgent need to better understand and characterize the evolution of cognitive changes along the earliest stages of Alzheimer’s disease (AD). However, measuring these very subtle, cognitive changes is challenging using conventional in-clinic paper-pencil cognitive tests. Increasing evidence suggests that digital assessment technologies may improve this by enabling the remote collection of more precise and nuanced information. The Mobile Toolbox (MTB) is a novel smartphone-based application for at-home administration on an individual’s own device. Here, we compared MTB-based scores to established in-clinic cognitive assessments and explored associations between the MTB and AD biomarkers on positron emission tomography (PET) imaging. Participants and Methods: The MTB was administered to cognitively unimpaired (CU) older adults from four well-characterized observational studies. All participants had in-clinic Preclinical Alzheimer’s Cognitive Composite-5 (PACC5) testing available, and a subsample had PiB-PET (n=57) and FTP-PET (n=35). The MTB includes six measures of Fluid Cognition, including episodic memory (Picture Sequence Memory (PSM) and Face-Name Associative Memory Exam (FNAME)), executive functions (EF) (Dimensional Change Card Sort (DCCS), Flanker), working memory (Memory for Sequence (MFS)), and processing speed (Number-Symbol Match (NSM)), and two measures of Crystallized Cognition (Spelling and Vocabulary). We hypothesized that a combination of the MTB measures PSM, FNAME (episodic memory), NSM (processing speed), and Vocabulary (as a general measure of cognition) would mimic the individual PACC5 tests, resulting in an MTB-PACC-like composite. Next, we created multiple MTB-based composites, consecutively adding measures of working memory and EF, resulting in the MTB-composite-3 (PSM, FNAME, MFS); MTB-composite-4 (PSM, FNAME, MFS, NSM); MTB-composite-5A (PSM, FNAME, MFS, NSM, DCCS) and MTB-composite-5B (PSM, FNAME, MFS, NSM, Flanker). Linear regression models correcting for age, sex and years of education compared various MTB-composites to the PACC5, global amyloid burden (PiB DVR) and tau deposition (FTP SUVr, PVC) in the medial temporal lobe. Results: MTB baseline data was available on 78 participants (age 70.9±9.18, 61.5% female, 16.6±2.5 years of education, 22% Aβ+). The theoretically derived MTB-PACC-like composite was associated to PACC5 (corrected β=0.35, 95%CI[0.15,0.55], p=0.001), as were the MTB-composite-3 (corrected β=0.29, 95%CI[0.10,0.48], p=0.003), MTB-composite-4 (corrected β=0.36, 95%CI[0.17,0.56], p=0.001), MTB-composite-5A (corrected β=0.37, 95%CI[0.17,0.57], p<0.001) and MTB-composite-5B (corrected β=0.38, 95%CI[0.17,0.59], p=0.001). No significant associations were observed with MTB composites and global amyloid, although we observed a trend-level association between amyloid and the MTB-composite-3 (corrected β=-0.07, [-0.16,0.30], p=0.156). The following MTB composites were associated with entorhinal tau: MTB-composite-3 (corrected β=-0.23, 95%CI[-0.40,-0.05], p=0.012), MTB-composite-4 (corrected β=-0.21, 95%CI[-0.39,-0.03], p=0.025), MTB-composite 5A (corrected β=-0.20, 95%CI[-0.37,-0.02], p=0.029) and MTB-composite 5B (corrected β=-0.22, 95%CI[-0.42,-0.03], p=0.028). Conclusions: Various combinations of MTB-based composites done at home are similarly associated with standardized in-clinic cognitive assessments. Further, our preliminary results suggest that the most efficacious MTB-composites for detecting cognitive deficits related to AD pathology included measures of episodic memory, processing speed and EF. Overall, these findings suggest that the MTB might provide a promising tool to remotely measure cognition, thereby facilitating the monitoring of individuals at-risk for AD in research as well as real-world settings such as primary care practices.

Paper

02/16/2024
10:15 am - 11:40 am
Room: West Side Ballroom - Salon 3

Paper Session 10: Alzheimer's Disease

Final Abstract #4

The Mobile Toolbox (MTB) for Assessing Cognition Remotely in Preclinical Alzheimer’s Disease: Associations with In-Clinic Cognitive Assessments and Amyloid and Tau PET.

Roos Jutten, Massachusetts General Hospital, Harvard Medical School, Boston, United States
Jessa Burling, Massachusetts General Hospital, Harvard Medical School, Boston, United States
Michael Properzi, Massachusetts General Hospital, Harvard Medical School, Boston, United States
Rebecca Amariglio, Massachusetts General Hospital, Brigham and Women's Hospital, Harvard Medical School, Boston, United States
Gad Marshall, Massachusetts General Hospital, Brigham and Women's Hospital, Harvard Medical School, Boston, United States
Kathryn Papp, Massachusetts General Hospital, Brigham and Women's Hospital, Harvard Medical School, Boston, United States
Keith Johnson, Massachusetts General Hospital, Harvard Medical School, Boston, United States
Reisa Sperling, Massachusetts General Hospital, Brigham and Women's Hospital, Harvard Medical School, Boston, United States
Dorene Rentz, Massachusetts General Hospital, Brigham and Women's Hospital, Harvard Medical School, Boston, United States

Category: Dementia (Alzheimer's Disease)

Keyword 1: cognitive functioning
Keyword 2: computerized neuropsychological testing
Keyword 3: positron emission tomography

Objective:

There is an urgent need to better understand and characterize the evolution of cognitive changes along the earliest stages of Alzheimer’s disease (AD). However, measuring these very subtle, cognitive changes is challenging using conventional in-clinic paper-pencil cognitive tests. Increasing evidence suggests that digital assessment technologies may improve this by enabling the remote collection of more precise and nuanced information. The Mobile Toolbox (MTB) is a novel smartphone-based application for at-home administration on an individual’s own device. Here, we compared MTB-based scores to established in-clinic cognitive assessments and explored associations between the MTB and AD biomarkers on positron emission tomography (PET) imaging.

Participants and Methods:

The MTB was administered to cognitively unimpaired (CU) older adults from four well-characterized observational studies. All participants had in-clinic Preclinical Alzheimer’s Cognitive Composite-5 (PACC5) testing available, and a subsample had PiB-PET (n=57) and FTP-PET (n=35). The MTB includes six measures of Fluid Cognition, including episodic memory (Picture Sequence Memory (PSM) and Face-Name Associative Memory Exam (FNAME)), executive functions (EF) (Dimensional Change Card Sort (DCCS), Flanker), working memory (Memory for Sequence (MFS)), and processing speed (Number-Symbol Match (NSM)), and two measures of Crystallized Cognition (Spelling and Vocabulary). We hypothesized that a combination of the MTB measures PSM, FNAME (episodic memory), NSM (processing speed), and Vocabulary (as a general measure of cognition) would mimic the individual PACC5 tests, resulting in an MTB-PACC-like composite. Next, we created multiple MTB-based composites, consecutively adding measures of working memory and EF, resulting in the MTB-composite-3 (PSM, FNAME, MFS); MTB-composite-4 (PSM, FNAME, MFS, NSM); MTB-composite-5A (PSM, FNAME, MFS, NSM, DCCS) and MTB-composite-5B (PSM, FNAME, MFS, NSM, Flanker). Linear regression models correcting for age, sex and years of education compared various MTB-composites to the PACC5, global amyloid burden (PiB DVR) and tau deposition (FTP SUVr, PVC) in the medial temporal lobe.

Results:

MTB baseline data was available on 78 participants (age 70.9±9.18, 61.5% female, 16.6±2.5 years of education, 22% Aβ+). The theoretically derived MTB-PACC-like composite was associated to PACC5 (corrected β=0.35, 95%CI[0.15,0.55], p=0.001), as were the MTB-composite-3 (corrected β=0.29, 95%CI[0.10,0.48], p=0.003), MTB-composite-4 (corrected β=0.36, 95%CI[0.17,0.56], p=0.001), MTB-composite-5A (corrected β=0.37, 95%CI[0.17,0.57], p<0.001) and MTB-composite-5B (corrected β=0.38, 95%CI[0.17,0.59], p=0.001). No significant associations were observed with MTB composites and global amyloid, although we observed a trend-level association between amyloid and the MTB-composite-3 (corrected β=-0.07, [-0.16,0.30], p=0.156). The following MTB composites were associated with entorhinal tau: MTB-composite-3 (corrected β=-0.23, 95%CI[-0.40,-0.05], p=0.012), MTB-composite-4 (corrected β=-0.21, 95%CI[-0.39,-0.03], p=0.025), MTB-composite 5A (corrected β=-0.20, 95%CI[-0.37,-0.02], p=0.029) and MTB-composite 5B (corrected β=-0.22, 95%CI[-0.42,-0.03], p=0.028).

Conclusions:

Various combinations of MTB-based composites done at home are similarly associated with standardized in-clinic cognitive assessments. Further, our preliminary results suggest that the most efficacious MTB-composites for detecting cognitive deficits related to AD pathology included measures of episodic memory, processing speed and EF. Overall, these findings suggest that the MTB might provide a promising tool to remotely measure cognition, thereby facilitating the monitoring of individuals at-risk for AD in research as well as real-world settings such as primary care practices.