INS NYC 2024 Program

Poster	Poster Session 04 Program Schedule 02/15/2024 12:00 pm - 01:15 pm Room: Majestic Complex (Posters 61-120) Poster Session 04: Neuroimaging \| Neurostimulation/Neuromodulation \| Teleneuropsychology/Technology Final Abstract #74 Using Remote, Digital Cognitive Assessments to Capture Multi-Day Learning and Long-Term Retention in Preclinical Alzheimer’s Disease Cassidy Molinare, Department of Neurology, Brigham and Women's Hospital, Harvard Medical School, Boston, United States Daniel Soberanes, Department of Neurology, Brigham and Women's Hospital, Harvard Medical School, Boston, United States Emma Weizenbaum, Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, United States Stephanie Hsieh, Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, United States Keith Johnson, Department of Neurology, Massachusetts General Hospital, Harvard Medical School and Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, United States Dorene Rentz, Department of Neurology, Brigham and Women's Hospital, Harvard Medical School and Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, United States Reisa Sperling, Department of Neurology, Brigham and Women's Hospital, Harvard Medical School and Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, United States Gad Marshall, Department of Neurology, Brigham and Women's Hospital, Harvard Medical School and Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, United States Rebecca Amariglio, Department of Neurology, Brigham and Women's Hospital, Harvard Medical School and Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, United States Kathryn Papp, Department of Neurology, Brigham and Women's Hospital, Harvard Medical School and Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, United States Roos Jutten, Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, United States Category: Teleneuropsychology/ Technology Keyword 1: technology Keyword 2: cognitive processing Keyword 3: dementia - Alzheimer's disease Objective: Alzheimer’s disease (AD) is presumed to have a long preclinical stage in which amyloid-beta and tau pathology start to accumulate, but overt clinical symptoms remain absent. The medial-temporal lobe (MTL) is particularly vulnerable to early tau deposition in preclinical AD, which is thought to impact the ability to successfully encode and retrieve recently learned material. However, subtle changes in encoding and retention are challenging to detect using traditional single timepoint cognitive assessments. The Boston Remote Assessment for Neurocognitive Health (BRANCH) was designed to mimic learning in everyday life, by exposing individuals to repeated stimuli over frequent short-term intervals (days) as well as by testing the retention of learned material over longer intervals (months). Here, we investigated how learning and retention processes captured using the BRANCH paradigm are associated with MTL pathology. Participants and Methods: We included 139 well-characterized cognitively unimpaired older adults (age=75.1±7.87, 66.2% female, years of education=16.3±2.51, 18.0% Aβ+) who underwent Flortaucipir-PET within 2.16±1.48 years of BRANCH. As part of BRANCH, participants completed the Face-Name Associative Memory Exam (FNAME) assessment for 7 consecutive days at home on a personal device (e.g., smartphone, laptop). After 14.5±4 months, they were prompted to take a FNAME recall task, asking them to identify the correct face-name pairs that they previously learned. We utilized a previously validated multi-day learning curve metric to quantify learning performance across the 7 days. To measure long-term retention (LTR), we calculated a proportion of participants’ accuracy on the recall task to their optimal performance during the initial 7 days. Linear regression models were used to investigate whether FNAME Day 1 performance, multi-day learning curve, and LTR were associated with tau deposition in the entorhinal (ET) cortex. All analyses were corrected for demographics, and analyses with LTR were additionally adjusted for time between the initial and recall assessment. In addition, since time between learning and recall likely impacts LTR performance, we ran a sensitivity analysis after restricting the sample to only include individuals whose time between assessments was below the median (i.e., < 15 months), resulting in a subsample with an average of 11.1±2.8 months between assessments (n=69, age=76±7.80, 59.4% female, years of education=16.3±2.46, 18.8% Aβ+). Results: FNAME Day 1 performance was not associated with ET (β=-0.01,95%CI[-0.12–0.09],p=0.790). However, a lower multi-day learning curve was associated with elevated ET (β=-0.09,95%CI[-0.15–-0.02],p=0.014). Lower LTR was not associated with greater ET (β=-0.09,95%CI[-0.20–0.02],p=0.123) in the total sample. However, in the restricted sample, lower LTR was significantly associated with elevated ET (β=-0.18,95%CI[-0.34–-0.02],p=0.031). Conclusions: We showed that leveraging a multi-day learning paradigm such as BRANCH enables the detection of multiple MTL-related cognitive processes that are not detected using a single timepoint assessment. These findings highlight the advantage of employing a remote, digital tool across multiple timepoints to capture various memory processes that are relevant in preclinical AD. Future work will help us to better understand LTR across various time intervals, and to determine to what extent MTL tau pathology mediates the association between initial encoding and LTR performance.

Poster

02/15/2024
12:00 pm - 01:15 pm
Room: Majestic Complex (Posters 61-120)

Poster Session 04: Neuroimaging | Neurostimulation/Neuromodulation | Teleneuropsychology/Technology

Final Abstract #74

Using Remote, Digital Cognitive Assessments to Capture Multi-Day Learning and Long-Term Retention in Preclinical Alzheimer’s Disease

Cassidy Molinare, Department of Neurology, Brigham and Women's Hospital, Harvard Medical School, Boston, United States
Daniel Soberanes, Department of Neurology, Brigham and Women's Hospital, Harvard Medical School, Boston, United States
Emma Weizenbaum, Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, United States
Stephanie Hsieh, Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, United States
Keith Johnson, Department of Neurology, Massachusetts General Hospital, Harvard Medical School and Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, United States
Dorene Rentz, Department of Neurology, Brigham and Women's Hospital, Harvard Medical School and Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, United States
Reisa Sperling, Department of Neurology, Brigham and Women's Hospital, Harvard Medical School and Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, United States
Gad Marshall, Department of Neurology, Brigham and Women's Hospital, Harvard Medical School and Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, United States
Rebecca Amariglio, Department of Neurology, Brigham and Women's Hospital, Harvard Medical School and Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, United States
Kathryn Papp, Department of Neurology, Brigham and Women's Hospital, Harvard Medical School and Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, United States
Roos Jutten, Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, United States

Category: Teleneuropsychology/ Technology

Keyword 1: technology
Keyword 2: cognitive processing
Keyword 3: dementia - Alzheimer's disease

Objective:

Alzheimer’s disease (AD) is presumed to have a long preclinical stage in which amyloid-beta and tau pathology start to accumulate, but overt clinical symptoms remain absent. The medial-temporal lobe (MTL) is particularly vulnerable to early tau deposition in preclinical AD, which is thought to impact the ability to successfully encode and retrieve recently learned material. However, subtle changes in encoding and retention are challenging to detect using traditional single timepoint cognitive assessments. The Boston Remote Assessment for Neurocognitive Health (BRANCH) was designed to mimic learning in everyday life, by exposing individuals to repeated stimuli over frequent short-term intervals (days) as well as by testing the retention of learned material over longer intervals (months). Here, we investigated how learning and retention processes captured using the BRANCH paradigm are associated with MTL pathology.

Participants and Methods:

We included 139 well-characterized cognitively unimpaired older adults (age=75.1±7.87, 66.2% female, years of education=16.3±2.51, 18.0% Aβ+) who underwent Flortaucipir-PET within 2.16±1.48 years of BRANCH. As part of BRANCH, participants completed the Face-Name Associative Memory Exam (FNAME) assessment for 7 consecutive days at home on a personal device (e.g., smartphone, laptop). After 14.5±4 months, they were prompted to take a FNAME recall task, asking them to identify the correct face-name pairs that they previously learned. We utilized a previously validated multi-day learning curve metric to quantify learning performance across the 7 days. To measure long-term retention (LTR), we calculated a proportion of participants’ accuracy on the recall task to their optimal performance during the initial 7 days. Linear regression models were used to investigate whether FNAME Day 1 performance, multi-day learning curve, and LTR were associated with tau deposition in the entorhinal (ET) cortex. All analyses were corrected for demographics, and analyses with LTR were additionally adjusted for time between the initial and recall assessment. In addition, since time between learning and recall likely impacts LTR performance, we ran a sensitivity analysis after restricting the sample to only include individuals whose time between assessments was below the median (i.e., < 15 months), resulting in a subsample with an average of 11.1±2.8 months between assessments (n=69, age=76±7.80, 59.4% female, years of education=16.3±2.46, 18.8% Aβ+).

Results:

FNAME Day 1 performance was not associated with ET (β=-0.01,95%CI[-0.12–0.09],p=0.790). However, a lower multi-day learning curve was associated with elevated ET (β=-0.09,95%CI[-0.15–-0.02],p=0.014). Lower LTR was not associated with greater ET (β=-0.09,95%CI[-0.20–0.02],p=0.123) in the total sample. However, in the restricted sample, lower LTR was significantly associated with elevated ET (β=-0.18,95%CI[-0.34–-0.02],p=0.031).

Conclusions:

We showed that leveraging a multi-day learning paradigm such as BRANCH enables the detection of multiple MTL-related cognitive processes that are not detected using a single timepoint assessment. These findings highlight the advantage of employing a remote, digital tool across multiple timepoints to capture various memory processes that are relevant in preclinical AD. Future work will help us to better understand LTR across various time intervals, and to determine to what extent MTL tau pathology mediates the association between initial encoding and LTR performance.