Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 314 - Statistical Advancements in Neurodegeneration Trial Designs and Analyses
Type: Topic Contributed
Date/Time: Wednesday, August 5, 2020 : 10:00 AM to 11:50 PM
Sponsor: Section for Statistical Programmers and Analysts
Abstract #313858
Title: TRCPAD: Accelerating Participant Recruitment in Alzheimer’s Disease Clinical Trials Using Machine Learning Techniques
Author(s): Oliver Langford* and Michael C Donohue and Gustavo Jimenez-Maggiora and Reisa Sperling and Jeff Cummings and Paul S Aisen and Rema C Raman
Companies: University of Southern California and University of Southern California and University of Southern California and Harvard Medical School and Cleveland Clinic Lou Ruvo Center for Brain Health and University of Southern California and University of Southern California
Keywords: Machine Learning; webstudy; Alzheimers

Trial Ready Cohort for Preclinical/Prodromal Alzheimer's Dementia (TRCPAD) aims to develop a large, biomarker-confirmed, trial-ready cohort to facilitate rapid enrollment into AD prevention trials. Preclinical Alzheimer’s studies to date have had more than a 2/3rd amyloid screen fail rate, resulting in prolonged and expensive recruitment. One of our aims is to optimize an innovative, adaptive risk algorithm to efficiently identify the most appropriate trial participants. Participants are enrolling in our web-based registry where they complete a number of questionnaires and cognitive assessments. With these data, we assess their eligibility for in-clinic assessments via a multi-stage algorithm and use Machine Learning techniques to predict amyloid status. Participants who are confirmed amyloid positive are invited into clinical trials of anti-amyloid interventions. In this presentation, we will take a detailed look at the Machine Learning methods chosen and our motivations for these choices. We will also see how the addition of APOE genotyping significantly improves the predictive performance, AUC (~0.6 to ~0.7), when estimating Amyloid status.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2020 program