Improved longitudinal [(18)F]-AV45 amyloid PET by white matter reference and VOI-based partial volume effect...

Area of Science:

• Neuroimaging
• Nuclear Medicine
• Alzheimer's Disease Research

Background:

• Amyloid Positron-Emission Tomography (PET) is crucial for Alzheimer's disease (AD) research and diagnosis.
• Traditional methods often use the cerebellum as a reference region and neglect partial volume effects (PVE), leading to variability in longitudinal studies.
• Previous longitudinal studies using cerebellum reference tissue showed minimal amyloid binding increases and high inter-subject variability.

Purpose of the Study:

• To evaluate the impact of different reference regions and partial volume correction (PVEC) on the diagnostic discriminatory power of amyloid PET.
• To assess how these quantitative methods affect the longitudinal performance of amyloid PET in tracking disease progression.
• To optimize amyloid PET analysis for improved accuracy in Alzheimer's disease detection and monitoring.

Main Methods:

• Analysis of [(18)F]-AV45 PET and T1-weighted MRI data from 962 subjects at baseline and 258 at two-year follow-up.
• Calculation of cortical composite volume-of-interest (VOI) values with and without VOI-based PVEC.
• Standardized Uptake Value Ratios (SUVR) computed using cerebellum, brainstem, or white matter as reference regions.
• Receiver Operating Characteristic (ROC) analysis to determine diagnostic discrimination between healthy controls (HC), mild cognitive impairment (MCI), and AD.
• Comparison of longitudinal changes in amyloid load across diagnostic groups under different quantification methods.

Main Results:

• White matter demonstrated the most consistent Standard Uptake Value (SUV) across diagnostic groups.
• PVEC decreased SUV in healthy controls (-18%) and MCI (-10%) but increased it in AD (+7%).
• Highest diagnostic discrimination was achieved with PVEC and white matter scaling for HC/AD (AUC=0.907) and PVEC with brainstem scaling for HC/MCI (AUC=0.658).
• PVEC enhanced longitudinal amyloid increases across all groups, with the white matter reference yielding the lowest inter-subject variability.

Conclusions:

• VOI-based PVEC combined with white matter or brainstem reference regions significantly improves the discriminatory power of [(18)F]-AV45 PET for Alzheimer's disease.
• These optimized quantification methods enhance the detection of longitudinal amyloid accumulation, crucial for monitoring disease progression.
• The findings suggest that adjusting for PVE and selecting appropriate reference tissues are vital for robust amyloid PET imaging in AD research and clinical practice.