HAI Book 2025 - Flipbook - Page 154
Cody, Karly
22
The relative prevalence of tau severity stages is related to amyloid
burden, clinical diagnosis and age
Karly Cody1, Andrzej Sokolowski1, Mackenzie Carlson1, Lucah Medina1, Joe Winer1, Christina
Young1, Kyan Younes1, Alaina Durant2, Logan Dumitrescu2, Derek Archer2, Duygu TosunTurgut3, Sterling Johnson4, Theresa Harrison5, Timothy Hohman2, Elizabeth Mormino1
1
Department of Neurology and Neurological Sciences, Stanford University, Stanford, CA, US
Vanderbilt Memory and Alzheimer’s Center, Vanderbilt Genetics Institute, Department of Neurology, Vanderbilt
University Medical Center, Nashville, TN, US
3
Department of Radiology, University of California-San Francisco, San Francisco, CA, US
4
Department of Medicine, University of Wisconsin-Madison, Madison, WI, US
5
Neuroscience Department, University of California-Berkeley, Berkeley, CA, US
2
Objectives: This study leverages multi-site harmonized PET data from the AD Sequencing Project-Phenotype
Harmonization Consortium to characterize the prevalence of tau PET stages along the amyloid continuum.
Methods: Participants (n=2,153; Table 1) from A4, ADNI, SCAN-Prospective, SCAN-Mixed Protocol, and WRAP who
underwent amyloid and tau PET imaging were included in this study. PET data corresponding to ligand-specific
post-injection windows were processed and harmonized using an MRI-free pipeline. Amyloid PET SUVRs were
translated to centiloids (CLs) and regional tau positivity (T+/-) was derived within ligand using gaussian mixture
modeling. A hierarchical staging schema was used to classify tau severity (e.g. T-, T12+, T34+, T56+). The
cumulative probability of hierarchical tau stages was estimated according to CLs, age, and clinical stage (e.g. CU,
MCI, Dementia) using ordinal logistic regression models.
Results: The frequency and spatial extent of T+ generally increased with CL level and clinical stage (Figure 1).
Analyses estimating the cumulative prevalence of tau stages indicated significant CL×clinical stage, CL×age, and
age×clinical stage interactions, suggesting that the age- and CL-related probability of tau stages differed by
clinical stage and that CL-related probability differed by age. Overall, we found that the prevalence of tau stages
reflected the expected trend that both increasing CLs and increasing clinical severity corresponded to a higher
likelihood of tau PET severity (Figure 2A). Additionally, we observed that the effect of CL and clinical stage was
moderated by age (Figure 2B). Among individuals with impairment, the probability of late-stage tau (T56+) was
highest among those with younger ages and higher CLs. By contrast, the age effect was diminished in CUs at high
CLs as well as at lower CLs in MCI and dementia.
Discussion: Heterogeneity in tau PET severity has important implications for clinical trial design and for inferring
contributions of the AD pathway to clinical impairment.
HAI2025 - 154
