Cognitive indices correlate with [18F]GTP1 tau PET signal and white matter hyperintensities in prodromal‐to‐mild Alzheimer's disease: Baseline data from the Tauriel Study

Abstract

AbstractBackgroundIn prior cross‐sectional studies using [18F]GTP1 or other tau PET tracers across a spectrum of Alzheimer’s disease (AD) severity, tau PET signal has independently correlated with cognition, even after adjustments for β‐amyloid (Aβ) load and atrophy. However, it remains uncertain how robust these correlations are in more homogeneous patient populations (e.g., clinical trial participants) and after adjustment for MRI white‐matter hyperintensities (WMH). We sought to address this question by examining baseline data from the Phase 2 Tauriel study (NCT03289143), which is investigating the safety and efficacy of semorinemab (an anti‐tau antibody) in prodromal‐to‐mild AD.MethodWe analyzed data from a subset of Tauriel participants who underwent baseline [18F]GTP1 tau PET imaging and met criteria for probable AD dementia (n=244) or mild cognitive impairment (MCI; n=137), with MMSE of 20‐30, global Clinical Dementia Rating (CDR) of 0.5 or 1, significant Aβ pathology (PET or CSF), and significant episodic memory impairment on the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS). Other screening/baseline assessments included the 13‐item version of the Alzheimer’s Disease Assessment Scale‐Cognitive Subscale (ADAS‐Cog13) and MRI (WMH scored with Fazekas scale). Relationships between cognitive indices, whole cortical gray (WCG) [18F]GTP1 SUVR, WMH, and age were assessed using Pearson correlations and multiple linear regression.ResultUnivariate analyses revealed significant correlations between all cognitive indices [MMSE, CDR Sum of Boxes, ADAS‐Cog13, and RBANS] and [18F]GTP1 SUVR across the whole cortical gray (WCG; p’s<0.001). Significant univariate correlations were also seen between WMH and ADAS‐Cog13 scores (p=0.035) and between age and RBANS scores (p<0.001). Multivariate linear regression analyses that included all three factors consistently identified both WCG [18F]GTP1 SUVR and WMH as independent predictors of cognitive performance across all indices (p’s<0.05), with a much larger proportion of variance attributable to WCG [18F]GTP1 SUVR.ConclusionAmongst participants with prodromal‐to‐mild AD enrolled in the Tauriel study, [18F]GTP1 tau PET signal was more strongly associated with baseline cognitive performance than the extent of WMH. These data suggest that even in AD participants with co‐morbid cerebral microvascular ischemic disease, therapeutic interventions targeting tau have the potential to ameliorate subsequent cognitive decline.