[18F]FDG PET may differentiate cerebral amyloid angiopathy from Alzheimer’s disease

• Published in: European journal of neurology Vol. 28; no. 5; pp. 1511 - 1519
• Main Authors: Bergeret, Sébastien, Queneau, Mathieu, Rodallec, Mathieu, Curis, Emmanuel, Dumurgier, Julien, Hugon, Jacques, Paquet, Claire, Farid, Karim, Baron, Jean‐Claude
• Format: Journal Article
• Language: English
• Published: England John Wiley & Sons, Inc 01.05.2021
• Subjects: [18F]FDG; Alzheimer's disease; Biomarkers; Cerebral amyloid angiopathy; Cerebrospinal fluid; Cognitive ability; Computed tomography; Criteria; Diagnosis; Fluorine isotopes; Hemorrhage; Magnetic resonance imaging; Neurodegenerative diseases; Older people; Perfusion; PET; Pons; Positron emission; Positron emission tomography; Tomography; [18F]FDG; cerebral amyloid angiopathy; Alzheimer's disease; PET
• ISSN: 1351-5101; 1468-1331
• DOI: 10.1111/ene.14743

Background Cerebral amyloid angiopathy (CAA) is a frequent cause of both intracerebral hemorrhage (ICH) and cognitive impairment in the elderly. Diagnosis relies on the Boston criteria, which use magnetic resonance imaging markers including ≥2 exclusively lobar cerebral microbleeds (lCMBs). Although amyloid positron emission tomography (PET) may provide molecular diagnosis, its specificity relative to Alzheimer's disease (AD) is limited due to the prevalence of positive amyloid PET in cognitively normal elderly. Using early‐phase 11C‐Pittsburgh compound B as surrogate for tissue perfusion, a significantly lower occipital/posterior cingulate (O/PC) tracer uptake ratio in probable CAA relative to AD was recently reported, consistent with histopathological lesion distribution. We tested whether this finding could be reproduced using [18F]fluorodeoxyglucose (FDG)‐PET, a widely available modality that correlates well with early‐phase amyloid PET in both healthy subjects and AD. Methods From a large memory clinic database, we retrospectively included 14 patients with probable CAA (Boston criteria) and 21 patients with no lCMB fulfilling AD criteria including cerebrospinal fluid biomarkers. In all, [18F]FDG‐PET/computed tomography (CT) was available as part of routine care. No subject had a clinical history of ICH. Regional standardized [18F]FDG uptake values normalized to the pons (standard uptake value ratio [SUVr]) were obtained, and the O/PC ratio was calculated. Results The SUVr O/PC ratio was significantly lower in CAA versus AD (1.02 ± 0.14 vs. 1.19 ± 0.18, respectively; p = 0.024). Conclusions Despite the small sample, our findings are consistent with the previous early‐phase amyloid PET study. Thus, [18F]FDG‐PET may help differentiate CAA from AD, particularly in cases of amyloid PET positivity. Larger prospective studies, including in CAA‐related ICH, are however warranted. Cerebral amyloid angiopathy (CAA) is a frequent cause of both intracerebral hemorrhage (ICH) and cognitive impairment in the elderly. Although amyloid positron emission tomography (PET) is positive in most patients with CAA, differentiation from Alzheimer’s disease (AD) may be difficult. We show here that the occipital/posterior cingulate tracer uptake ratio using fluorodeoxyglucose (FDG) PET is significantly lower in probable CAA patients than in AD patients without magnetic resonance imaging markers of CAA, consistent with the known distribution of pathological and FDG markers in both conditions. Larger prospective studies, including in CAA‐related ICH, are warranted to confirm these findings and assess their clinical utility.