Amyloid‐, tau‐, and FDG‐PET in amyloid‐negative patients clinically diagnosed with early‐onset Alzheimer’s disease in the LEADS study
Nidhi S. Mundada, David N. Soleimani‐Meigooni, Ehud Zeltzer, Charles Windon, Jeremy A. Tanner, Courtney Lawh Heath, Leonardo Iaccarino, Ranjani Shankar, Alinda Amuiri, Paul S. S Aisen, Ani Eloyan, Robert A. Koeppe, Maria C. Carrillo, Brad C. Dickerson, Liana G. Apostolova, Gil D. Rabinovici, Renaud La Joie- Psychiatry and Mental health
- Cellular and Molecular Neuroscience
- Geriatrics and Gerontology
- Neurology (clinical)
- Developmental Neuroscience
- Health Policy
- Epidemiology
Abstract
Background
Negative AD biomarkers are commonly found in patients with a clinical diagnosis of late‐onset AD, but little is known about biomarker‐negative patients diagnosed with sporadic Early Onset AD (EOAD, <65yo). We explored data from the Longitudinal Early‐onset Alzheimer’s Disease Study (LEADS) to identify such participants and explored amyloid‐, tau‐, and FDG‐PET.
Method
We identified 380 patients who i) met clinical criteria for MCI or dementia due to AD, and ii) completed baseline amyloid‐PET (Florbetaben) and tau‐PET (Flortaucipir). Of these patients, 93 (24.5%) were amyloid‐PET negative based on a combination of visual read and quantification (“EOnonAD”). A subset of EononAD participants completed FDG‐PET (n = 63), 5 had follow‐up amyloid‐PET and 2 had follow‐up tau‐PET. 63 age‐matched cognitively normal controls were also included. Image acquisition and processing followed ADNI protocols.
Result
Compared to amyloid‐positive patients (EOAD), EononAD patients were less frequently female or APOE4 carriers, and less severely impaired (Figure 1). EononAD patients had a mean Centiloid of 5.6 (Figure 2A). At the group level, temporal Flortaucipir‐SUVR was low (1.20.3, not significantly different from controls), although 3 cases had Flortaucipir‐SUVR>2 (Figure 2A). Visual inspection showed these cases had asymmetric AD‐like Flortaucipir‐PET patterns (Figure 2B). Within patients with EononAD, amyloid‐ and tau‐PET were correlated (Spearman’s ρ = 0.29, p = 0.005), even when excluding the 3 high Flortaucipir‐SUVR outliers (Figure‐2C). 5 patients had follow‐up amyloid‐PET: 2 showed major amyloid‐PET increase (+20 CL over 3 years), exceeding the 25 CL threshold at follow‐up, while the other 3 remained amyloid‐negative (Figure 2D). Two patients had follow‐up Flortaucipir‐PET: a low‐Flortaucipir case remained low, and one of the three high‐tau cases showed increased signal over time (Figure 2D). Group‐level FDG‐PET analysis showed frontal, temporal, and/or parietal hypometabolism in EOnonAD (Figure 3A). However, individual maps showed high heterogeneity: while ∼50% showing no clear hypometabolism (normal‐looking FDG), the remaining cases showed heterogeneous patterns suggestive of various underlying etiologies (e.g., FTLD‐like or AD‐like patterns, Figure 3B).
Conclusion
Amyloid‐PET negative patients included in LEADS are heterogeneous. While a minority of cases might have underlying AD with subthreshold amyloid‐ and/or tau‐PET values, FDG‐PET showed heterogeneous patterns suggesting underlying etiologies include various degenerative and non‐degenerative causes. ”