In vivo Braak‐staging using 18F‐Flortaucipir‐tau‐PET as a predictive marker for future cognitive decline in Alzheimer's disease.

Background: The development of in vivo amyloid (Aβ) and tau biomarkers has greatly facilitated diagnosing AD (Jack et al., 2018), however, reliable prognosis of AD‐related cognitive decline remains a critical yet unmet challenge. Here, especially tau biomarkers may offer great potential for individualized risk prediction of clinical AD progression, since the progressive spread of tau tracks clinical status more closely than Aβ deposition (Ossenkoppele et al., 2016; Wang et al., 2016). Thus, positron emission tomography (PET) of tau may be a better predictor for future cognitive decline and clinical progression than amyloid‐PET. Here, we examined whether tau‐PET‐based Braak‐staging outperforms amyloid‐PET as a predictor of future cognitive decline and clinical progression in older adults with and without cognitive impairment. Method: We included 396 cognitively normal to AD dementia subjects with Flutemetamol/Florbetapir‐amyloid‐PET, Flortaucipir‐tau‐PET and longitudinal cognitive assessments. Annual change rates in global cognition (i.e. MMSE, ADAS13) and episodic memory (i.e. ADNI‐MEM) were calculated via linear mixed models. We determined global amyloid‐PET (i.e. Centiloid) plus global and Braak‐stage‐specific tau‐PET SUVRs, which were stratified as positive(+)/negative(‐) at pre‐established cut‐offs. Using Braak‐stage‐specific binarized tau‐PET SUVRs, we classified subjects as BraakI+/BraakI‐IV+/BraakI‐VI+/Braakatypical+. In bootstrapped linear regression, we assessed whether global tau‐PET explained more variance in subsequent cognitive decline than Centiloid. Using ANCOVAs, we tested whether advanced Braak‐stage predicted faster future cognitive decline. All models were controlled for age, sex, education, clinical diagnosis, and baseline cognition. Lastly, we determined Braak‐stage‐specific conversion risk to mild cognitive impairment (MCI) or dementia. Result: Higher global tau‐PET SUVR at baseline explained more variance in subsequent cognitive decline across ∼2 years of follow up compared to baseline amyloid‐PET, consistently for MMSE, ADAS13, and memory (all p<0.001; Fig. 1). The association between global tau‐PET and subsequent cognitive changes remained consistent when additionally controlling for baseline amyloid‐PET. Advancing Braak‐stage was associated with gradually worsening future cognitive decline, independent of Centiloid (p<0.001; Fig. 2), and elevated clinical conversion risk to MCI or AD dementia (e.g. 15.38% for BraakI+, 23.81% for BraakI‐IV+, 45.45% for BraakI‐VI+; Fig. 3). Conclusion: Tau‐PET clearly outperforms amyloid‐PET in predicting future cognitive decline, where tau‐PET‐based Braak‐staging may be used for estimating patient‐specific risk for clinical AD progression. [ABSTRACT FROM AUTHOR]