New IDEAS

New IDEAS (Imaging Dementia - Evidence for Amyloid Scanning, NCT04426539) is the follow-up to the original IDEAS study. It is a US coverage-with-evidence-development study in which Medicare paid for one clinical amyloid PET scan per participant with mild cognitive impairment (MCI) or dementia. Unlike its predecessor, it set enrollment targets for Black and Latinx participants and grouped patients by typical (progressive amnestic) or atypical presentation. The American College of Radiology sponsored and ran it, and the Alzheimer's Association directed it. De-identified amyloid PET images are distributed by LONI through the Image and Data Archive under a data use agreement. Clinical data are shared through the Global Alzheimer's Association Interactive Network (GAAIN).

Composition

Of 6,023 people registered between December 2020 and March 2024, 5,757 had complete pre-PET data and form the analyzed cohort: 1,248 Black, 1,166 Latinx and 3,343 of all other races and ethnicities. 3,606 had MCI and 2,151 dementia, 56% were women, and the median age was 75 (range 35 to 98). Participants came from 151 clinical practices with 270 dementia specialists.

Acquisition

Amyloid PET was done as routine clinical imaging at 107 PET facilities with one of the three FDA-approved tracers: florbetapir, florbetaben or flutemetamol. Scans took place within 60 days of the pre-PET visit. Archiving of the images was an optional part of the study with its own consent. A brain MRI or CT within 24 months before enrollment was an entry criterion but was not part of the study protocol, and the data sharing policy states that only amyloid PET images are available.

Annotations

Each scan has a binary visual read (positive or negative) by a trained local imaging specialist, following the approved criteria for the tracer. The study recorded the diagnosis and intended management before the scan and the implemented management about 90 days after it, including changes in Alzheimer's medication, counseling and lecanemab use.

Known limitations

The number of archived scans and the content of the LONI release are not documented publicly, so the counts here describe the analyzed study cohort. Protocols and scanners vary across sites, and reads were local, not central. Amyloid status is missing for about 16% of the analyzed participants. Medicare claims outcomes require separate permission from CMS.