PET-guided Assessment of Amyloid Clearance and Outcomes in a Real-World Cohort of Patients with Alzheimer Disease undergoing Anti-Amyloid Therapy

BACKGROUND AND PURPOSE: Beta-amyloid (Aβ)-PET is central to confirming Alzheimer disease (AD) before treatment with anti-amyloid monoclonal antibody therapies (AAT), however its role in treatment response monitoring in routine clinical practice remains unclear. This study aimed to evaluate longitudinal Aβ-PET changes following AAT and their association with clinical and safety outcomes in a real-world cohort. MATERIALS AND METHODS: We conducted a retrospective single-center study of patients with mild cognitive impairment (MCI) or mild dementia due to AD who underwent Aβ-PET before and after treatment with AAT (lecanemab or donanemab). Aβ-PET scans were assessed using visual interpretation and quantitative measures including Centiloid level (CL) and regional Z-scores. Treatment-related amyloid clearance (TRAC) was determined based on magnitude of CL changes (ΔCL). Associations between Aβ-PET changes and baseline Fazekas score, amyloid-related imaging abnormalities (ARIA) and APOE-ε4 carrier status were examined. Associations between ΔCL and cognitive performance from baseline to post-AAT as assessed per Mini-Mental State Examination (ΔMMSE) were examined using multivariable linear regression models incorporating baseline CL and adjusting APOE-ε4 status and occurrence of ARIA. Region-specific multivariable analyses evaluated associations between regional Z-score changes and ΔMMSE. RESULTS: Thirty-two patients met inclusion criteria (lecanemab, N=15; donanemab, N=17). Significant Aβ reduction was observed across the cohort (median ΔCL 59.11; p < 0.001) as well as within each treatment group. Most patients meeting TRAC criteria achieved full or partial TRAC (26/29, 89.7%). Baseline Aβ burden, as well as cognitive outcomes, did not differ significantly across TRAC categories (p = 0.25). ΔCL was not significantly associated with APOE-ε4 status or ARIA occurrence. In adjusted analyses, greater global CL reduction was associated with greater MMSE improvement, particularly at lower baseline burden. Region-specific analyses demonstrated marked, highly significant decrease in Z-scores across all regions. CONCLUSIONS: Longitudinal Aβ-PET demonstrated substantial Aβ clearance following AAT in routine clinical practice. Aβ reduction was not significantly associated with baseline Aβ burden, ARIA, APOE-ε4 status. In adjusted analyses, greater Aβ reduction was associated with greater MMSE improvement. Our findings support Aβ-PET as a sensitive biomarker of biological treatment effect, while highlighting the complexity of linking Aβ clearance to short-term cognitive outcomes.