Biomarkers

Abstract

Background: Cognitively unimpaired (CU) older adults with abnormal levels of β-amyloid (Aβ) deposition are considered in the preclinical stage of Alzheimer's disease (AD) and, when combined with the subjective cognitive decline (SCD), are proposed as the stage 2 AD in the NIA-AA framework. Here, we aim to investigate whether neuropathologic deterioration increases in early stages, particularly in stage 2 AD.

Method: We included 341 CU participants over 50 years of age from the Sino Longitudinal Study on Cognitive Decline (SILCODE) study. SCD was defined based on baseline cognitive complaints with associated concern/worry, whereas normal control (NC) did not express any concern/worry. Aβ status was defined based on amyloid-PET or, if not available, on plasma Aβ42/40 ratio. All selected participants had available plasma phospho-tau181 (ptau181), and were divided into four groups according to SCD combined with baseline Aβ positivity. General linear models and linear mixed-effect models were performed to determine whether ptau181 levels differed between groups (i.e., NC.Aβ-, SCD.Aβ-, NC.Aβ+, SCD.Aβ+) at baseline and over time.

Result: At baseline, the four groups were significantly different in terms of ptau181 levels (p=0.001), with the SCD.Aβ+ group having higher levels than NC.Aβ- (p=0.004) and SCD.Aβ- (p=0.003), but not different from NC.Aβ+ (p=0.63). In the smallest sample of 81 CU with available follow-up for ptau181 (i.e., 15 NC.Aβ-, 35 SCD.Aβ-, 8 NC.Aβ+, 23 SCD.Aβ+), we replicated the baseline differences (p=0.037). Longitudinal analysis showed that ptau181 levels changed differently between groups over time (p=0.016), with SCD.Aβ+ showing a faster increase over time than SCD.Aβ- (p=0.011).

Conclusion: Our results showed that participants with stage 2 AD (SCD.Aβ+) have a higher ptau181 burden at baseline and increasing plasma ptau181 levels over time in a Chinese sample. This suggests that this group is at higher risk for neuropathologic changes, although no differences were found with stage 1 (NC.Aβ+). It should be noted that sample sizes for longitudinal analyses were small, especially for NC, which may partly explain this lack of differences between stages 1 and 2 AD. The SILCODE follow-up is ongoing, and we plan to test this hypothesis at a later stage.