Imaging Dementia-Evidence for Amyloid Scanning (IDEAS) Study

The IDEAS Study is an observational, open-label, longitudinal cohort study designed to assess the impact of amyloid PET on patient-oriented outcomes in Medicare beneficiaries with mild cognitive impairment (MCI) or dementia of uncertain etiology. The study falls under the Centers for Medicare \& Medicaid Services (CMS) Coverage with Evidence Development (CED) policy. A total of 18,488 Medicare beneficiaries meeting Appropriate Use Criteria (AUC) for amyloid PET will be enrolled over 24 months at sites throughout the United States. Dementia specialists will team with PET facilities able to perform amyloid PET and with trained radiologists/nuclear medicine physicians, all of whom will consent to completing the data requirements and timelines for the study. Amyloid PET will be performed and interpreted at each facility with results provided to the ordering physician for support in further clinical decision making, which will be captured for the study.

Our over-arching hypothesis is that, in diagnostically uncertain cases, knowledge of amyloid status as determined by amyloid PET will lead to significant changes in patient management, and that this will translate into improved long-term outcomes. We will pursue two specific aims:

Aim 1 investigates the impact of amyloid PET on short-term patient management, by comparing pre-PET intended management (ascertained in a case report form \[CRF\] prior to PET) to post-PET actual management 90-days post-PET). The primary objective will be to test whether amyloid PET leads to a ≥ 30% change between intended and actual patient management within 90 days in a cumulative endpoint consisting of: Alzheimer's disease (AD) drug therapy, other drug therapy, and counseling about safety and future planning. Secondary objectives will assess the impact of amyloid PET results on clinical diagnosis and prevention of unnecessary diagnostic procedures and treatments.

Aim 2 utilizes Medicare claims data to compare medical outcomes at 12 months for patients enrolled in the longitudinal cohort (amyloid PET-known) with those for a matched control cohort of patients who have never undergone amyloid PET imaging (amyloid PET-naïve). The primary objective will be to determine if amyloid PET in the amyloid PET-known cohort of patients is associated with a ≥ 10% reduction in hospitalizations and emergency room visits in comparison to the matched amyloid PET-naïve patients. Secondary objectives will examine whether knowledge of amyloid PET status reduces hospitalizations related to ambulatory-sensitive conditions, whether the association between amyloid PET knowledge and health outcomes varies by baseline cognitive status (MCI versus dementia) and amyloid status (amyloid positive versus negative). The amyloid PET-naïve cohort will be identified via a matching algorithm where each individual in the amyloid PET-known cohort will be matched to one individual with similar dementia diagnosis, pre-scan dementia-related resource utilization, age, race, gender, ethnicity, geographic location, and comorbid chronic conditions likely to impact cognition or the outcomes of interest seen at the same time as the amyloid PET-known patient (concurrent control).

In pursuing these Aims, we will generate valuable observational data on clinical utility that will inform future use of this technology in diagnostic algorithms, and develop a cohort of patients who undergo amyloid PET and can serve as a foundation to address future research questions.