HAI Book 2025 - Flipbook - Page 376
30
Harvard Medical School, Department of Radiology, Boston, MA, US
Gordon Center for Medical Imaging, Massachusetts General Hospital, Boston, MA, US
32
Amsterdam University Medical Center, Neuroscience Campus Amsterdam, Alzheimercenter, Amsterdam, NL
33
Laboratory of Neuro Imaging, USC Stevens Neuroimaging and Informatics Institute, Keck School of Medicine of
USC, University of Southern California, Los Angeles, CA, US
34
Department of Neurology, Mayo Clinic, Rochester, MI, US
35
Department of Neurology, Memory and Aging Center, Weill Institute for Neurosciences, University of California, San
Francisco, CA, US
36
Department of Radiology & Biomedical Imaging, University of California, San Francisco, CA, US
37
Translational Neuroimaging Laboratory, Department of Neurology and Neurosurgery, Faculty of Medicine, The
McGill University Research Centre for Studies in Aging, McGill University, Montreal, QC, CA
38
Montreal Neurological Institute, McGill University, Montreal, QC, CA
39
Department of Neurology, Skåne University Hospital, Lund, SE
40
Alzheimer’s Association, Chicago, IL, US
41
Senior advisor to CPAD Consortium, Critical Path Institute, Tucson, AZ, US
42
The Australian Dementia Network (ADNeT), The University of Melbourne, Melbourne, AU
43
Memory Clinic, Skåne University Hospital, Malmö, SE
44
Health and Biosecurity Flagship, The Australian eHealth Research Centre, CSIRO, Melbourne, AU
31
Objectives: The CenTauR scale has been recently proposed for between-tracer harmonization of tau-PET images.
Here, we studied how CenTauRs enhance the comparability of tau-PET quantification by comparing the
prevalence of CenTauR-defined tau-PET positivity across two matched cohorts scanned with different tau-PET
tracers.
Methods: The Joint Propagation Model (JPM) is a statistical model developed within the Critical Path for
Alzheimer9s Consortium (CPAD) led Tau PET Harmonization Working Group (Leuzy et al., Alzheimers Dement. 2024;
Figure 1A). The JPM simultaneously models the relationships of data from anchor point subjects and data from
subjects in head-to-head studies with the CenTauR scale, providing conversion equations for multiple tracers. We
applied these JPM equations to convert [18F]flortaucipir SUVR (Meta-temporal ROI) data from 561 cognitively
impaired participants from the ADNI and OASIS-3 studies into CenTauRs. We then used ROC analysis to find the
CenTauR cut-off for positivity (T+) that best discriminated between FDA-approved positive/negative visual reads..
The prevalence of T+ was compared across two matched cohorts consisting of A´-positive cognitively impaired
(MCI and AD dementia) individuals scanned with either [18F]flortaucipir (ADNI) or [18F]MK-6240 (AIBL). The cohorts
were matched at the individual level based on age, MMSE, and Centiloids.
Results: A cut-off of 18 CenTauRs was found to optimally classify [18F]flortaucipir PET visual reads (Figure 1B). The
matching procedure identified 46 cases in total (23 [18F]flortaucipir, 23 [18F]MK-6240). The prevalence of T+ using
the 18 CenTauRs cut-off for the Meta-temporal ROI was found to be highly comparable across tracers
([18F]flortaucipir, n=19; 83%; [18F]MK-6240, n=17; 74%). Further analyses involving other ROIs, tracers and cohorts
along with SUVR-to-CenTauR conversion equations, will be presented.
Conclusions: Implementation of standardized methodologies is key to harmonizing Tau-PET results across
distinct AD research and interventional clinical trials. These preliminary analyses suggest that JPM-derived
CenTauR equations are suitable for harmonization of tau-PET outcome data.
HAI2025 - 376
