P300 Latency with Memory Performance: A Promising Biomarker for Preclinical Stages of Alzheimer's Disease

Abstract

Detecting and tracking the preclinical stages of Alzheimer's disease (AD) is now of particular interest due to the aging of the world's population. AD is the most common cause of dementia, affecting the daily lives of those afflicted. Approaches in development can accelerate the evaluation of the preclinical stages of AD and facilitate early treatment and the prevention of symptom progression. Shifts in P300 amplitude and latency, together with neuropsychological assessments, could serve as biomarkers in the early screening of declines in cognitive abilities. In this study, we investigated the ability of the P300 indices evoked during a visual oddball task to differentiate pre-clinically diagnosed participants from normal healthy adults (HCs). Two preclinical stages, named asymptomatic AD (AAD) and prodromal AD (PAD), were included in this study, and a total of 79 subjects participated, including 35 HCs, 22 AAD patients, and 22 PAD patients. A mixed-design ANOVA test was performed to compare the P300 indices among groups during the processing of the target and non-target stimuli. Additionally, the correlation between these neurophysiological variables and the neuropsychological tests was evaluated. Our results revealed that neither the peak amplitude nor latency of P300 can distinguish AAD from HCs. Conversely, the peak latency of P300 can be used as a biomarker to differentiate PAD from AAD and HCs. The correlation results revealed a significant relationship between the peak latency of P300 and memory domain tasks, showing that less time-demanding neuropsychological assessments can be used. In summary, our findings showed that a combination of P300 latency and memory-requiring tasks can be used as an efficient biomarker to differentiate individuals with AAD from HCs.

Keywords: EPR; P300; asymptomatic AD; early screening; neuropsychology; prodromal AD.

Figure 1

A schematic diagram of the sequence of the oddball task.

Figure 2

Flowchart of the EEG data processing steps.

Figure 3

Mean peak amplitude values (µv) across the 32 channels during target processing in the HC, AAD, and PAD groups.

Figure 4

Mean peak latency values (ms) across the 32 channels during target processing in the HC, AAD, and PAD groups.

Figure 5

Average P300 waveforms recorded at the C3, CZ, and C4 electrodes during the presentation of target stimuli (shown in black) and non-target stimuli (shown in red) for the (A) HC, (B) AAD, and (C) PAD groups. The measuring window spans from 300 to 600 ms (shown in gray shadows).

Figure 6

Correlations between the P300 peak latency and neuropsychological tests in the HC, AAD, and PAD groups and the whole-group correlation during the presentation of the target stimulus (A), and the correlation with frontal/executive tasks and among the cognitive domains (B). The level of significance is shown with (*)/* p < 0.05 and ** p < 0.01 (FDR-corrected).

Figure 6

Correlations between the P300 peak latency and neuropsychological tests in the HC, AAD, and PAD groups and the whole-group correlation during the presentation of the target stimulus (A), and the correlation with frontal/executive tasks and among the cognitive domains (B). The level of significance is shown with (*)/* p < 0.05 and ** p < 0.01 (FDR-corrected).