Amyloid-β PET and CSF in an autopsy-confirmed cohort

Accumulation of amyloid-β is among the earliest changes in Alzheimer's disease (AD). Amyloid-β positron emission tomography (PET) and Aβ₄₂ in cerebrospinal fluid (CSF) both assess amyloid-β pathology in-vivo, but 10-20% of cases show discordant (CSF+/PET- or CSF-/PET+) results. The neuropathological correspondence with amyloid-β CSF/PET discordance is unknown.

We included 21 patients from our tertiary memory clinic who had undergone both CSF Aβ₄₂ analysis and amyloid-β PET, and had neuropathological data available. Amyloid-β PET and CSF results were compared with neuropathological ABC scores (comprising of Thal (A), Braak (B), and CERAD (C) stage, all ranging from 0 [low] to 3 [high]) and neuropathological diagnosis.

Neuropathological diagnosis was AD in 11 (52%) patients. Amyloid-β PET was positive in all A3, C2, and C3 cases and in one of the two A2 cases. CSF Aβ₄₂ was positive in 92% of ≥A2 and 90% of ≥C2 cases. PET and CSF were discordant in three of 21 (14%) cases: CSF+/PET- in a patient with granulomatosis with polyangiitis (A0B0C0), CSF+/PET- in a patient with FTLD-TDP type B (A2B1C1), and CSF-/PET+ in a patient with AD (A3B3C3). Two CSF+/PET+ cases had a non-AD neuropathological diagnosis, that is FTLD-TDP type E (A3B1C1) and adult-onset leukoencephalopathy with axonal spheroids (A1B1C0).

Our study demonstrates neuropathological underpinnings of amyloid-β CSF/PET discordance. Furthermore, amyloid-β biomarker positivity on both PET and CSF did not invariably result in an AD diagnosis at autopsy, illustrating the importance of considering relevant comorbidities when evaluating amyloid-β biomarker results.