Biomarkers.

Area of Science:

• Neuroimaging
• Biomarker Discovery
• Artificial Intelligence

Background:

• The PREDICTOM project, funded by the EU, aims to develop novel biomarkers for early identification of individuals at high risk for Alzheimer's Disease (AD).
• This study focuses on detailing the Magnetic Resonance (MR) imaging protocol to be implemented across seven European clinical centers.

Purpose of the Study:

• To describe the rationale and design of the PREDICTOM MR protocol for a multicenter study.
• To establish a harmonized imaging protocol for consistent data acquisition across diverse clinical sites and scanner types.

Main Methods:

• The protocol involves collecting MR data from 615 healthy subjects over 50 years old, with a target ratio of 80/20% for higher/lower AD risk.
• Data acquisition will occur at seven European centers using 3T MR scanners from two vendors and six different models, ensuring compatibility with modern clinical scanners.
• The comprehensive protocol includes eight datatypes, designed to extract anatomical and functional features relevant to AD, with a total exam time not exceeding one hour.

Main Results:

• The PREDICTOM MRI protocol incorporates Quantitative Susceptibility Mapping (QSM) for microbleed detection and Magnetic Resonance Spectroscopy (MRS) for identifying early AD-related metabolic alterations.
• Adaptations for scanners with lower gradient performance include potential use of partial Fourier in resting-state fMRI (rsfMRI) and a reduced 2-shell diffusion protocol.
• The protocol has undergone initial testing on four scanners (2 vendors, 3 models), with ongoing harmonization efforts.

Conclusions:

• The PREDICTOM MR protocol is designed for the exploration of existing and novel MRI biomarkers.
• The protocol aims to facilitate the early identification of individuals at high risk of developing Alzheimer's Disease.
• Ongoing harmonization is crucial for reducing inter-center and inter-scanner bias in this multicenter study.