Detection of PART in vivo using tau‐PET: preliminary findings using [18F]RO948

Abstract

AbstractBackgroundPrimary age‐related tauopathy (PART) is a neuropathological entity describing tau pathology in the entorhinal cortex and hippocampus in the absence of amyloid‐β (Aβ) plaques. Though cross‐sectional work using [18F]flortaucipir tau‐PET has shown an association between medial temporal lobe signal and worse cognitive performance, longitudinal studies are needed to examine whether medial temporal lobe tau is associated with cognitive decline and cortical tau accumulation in cognitively unimpaired (CU) individuals. We here aimed to investigate whether CU individuals who are Aβ‐negative and tau‐PET positive in the medial temporal lobe show cognitive decline and increases in CSF phosphorylated tau (P‐tau) or tau PET over time.MethodsUsing BioFINDER‐2, PART was defined as i) CU individuals (not fulfilling criteria for mild cognitive impairment of dementia), ii) age ≥ 60 years, iii) Aβ‐negative (Centiloids < 20 and negative CSF Aβ42/40), and iv) an [18F]RO948 tau‐PET standardized uptake value ratio (SUVR) >1.26 (cut‐off based on Gaussian mixture modelling in BioFINDER‐2, n = 1754) in a volume‐weighted ROI encompassing the entorhinal cortex and hippocampus. [18F]RO948 data was collected 70 to 90 min post‐injection with longitudinal follow‐up collected after ∼2.5 years (average 2.73 ± 1.14 years). Longitudinal P‐tau217 was measured using immunoprecipitation‐mass spectrometry (average 2.67 ± 1.12 years). Longitudinal cognitive measures (average 2.75 ± 0.94 years) included the MMSE and modified PACC. For both longitudinal [18F]RO948, CSF P‐tau217 and cognition, change was assessed using slopes derived from linear mixed models.Results38 individuals fulfilled inclusion criteria (Table 1). These individuals showed significant increases in [18F]RO948 SUVR in the entorhinal cortex/hippocampus (p<0.05, t‐test vs 0)—but not in the temporal cortex—and in CSF P‐tau217 (p<0.01, t‐test vs 0) over time (Figure 1A). This group also showed cognitive decline over time using the PACC (p<0.01, t‐test vs 0) but not the MMSE (Figure 1B). Additional analyses are ongoing using two additional tau‐PET tracers and additional CSF‐based P‐tau measures.ConclusionsThese preliminary findings using [18F]RO948 suggest that PART can be detected in vivo using tau‐PET and that Aβ‐negative CU individuals who are tau‐PET positive in the medial temporal lobe are at risk for focal tau accumulation and cognitive decline.