Alteration of prefrontal functional connectivity in preclinical Alzheimer's disease: an fNIRS study

Abstract

Background: Early detection of Alzheimer's disease (AD) is vital for delaying its progression through timely intervention. The preclinical stage, the longest phase of AD, often goes undetected due to a lack of noticeable symptoms. Developing an accessible and quantitative screening method for AD is essential for enabling appropriate interventions during this stage.

Methods: Functional near-infrared spectroscopy was used to investigate prefrontal functional connectivity in preclinical AD subjects. A total of 99 participants, including healthy controls and preclinical subjects who were amyloid beta (Aβ) positive (n = 45), were recruited. We designed a mixed phonemic and semantic verbal fluency task for the experimental protocol. Functional connectivity was then analyzed as z-values in the left, right, and interhemispheric prefrontal regions during a verbal fluency task. Finally, we assessed the correlation between the participants' z-values and clinical indices.

Results: The preclinical AD group exhibited increased interhemispheric functional connectivity derived from oxygenated and deoxygenated hemoglobin during verbal tasks involving the first phonemic letter. Additionally, significant right and left functional connectivity differences were observed in the healthy control group during verbal tasks with the letter and categories, but not in the preclinical AD group. Lastly, the difference in interhemispheric functional connectivity of oxygenated hemoglobin between the first and second verbal trials was significantly greater in the preclinical AD group. These interhemispheric functional connectivity values were significantly correlated with Aβ results from positron emission tomography.

Conclusion: The initial increase and subsequent reduction of interhemispheric functional connectivity in the preclinical AD group across task repetitions suggests that task-related prefrontal network alterations may occur during the preclinical phase of AD and shows its potential as a biomarker for screening preclinical AD.

Keywords: Alzheimer's disease; functional connectivity; functional near-infrared spectroscopy; older adult; preclinical stage; prefrontal cortex; verbal fluency task.

Figure 1

Experimental paradigm of the verbal fluency task.

Figure 2

Schematic representation of fNIRS channel locations. Channels 1 and 2 cover the right prefrontal area, while channels 3 and 4 cover the left prefrontal area.

Figure 3

Representative fNIRS signals. Changes in oxygenated and deoxygenated hemoglobin concentrations were averaged for channels and each group. The solid line indicates the preclinical group, and the dotted line indicates the HC group and the pink and blue shaded areas surrounding the solid and dotted lines represent the standard errors.

Figure 4

Functional connectivity graph: (A) Oxygenated hemoglobin, (B) Deoxygenated hemoglobin. Functional connectivity values are displayed in a 4 × 4 matrix. Numbers 1 to 4 indicate the channel numbers. Asterisks indicate statistically significant differences between the healthy control and preclinical groups (FDR corrected, *p < 0.05 and **p < 0.01).

Figure 5

Functional connectivity graph derived from the change in oxygenated hemoglobin. The color indicates the t-score, showing only statistically significant connectivity during the first phonemic verbal fluency task (P1) (FDR corrected, p < 0.05). The darker blue indicates greater significance between the HC and preclinical AD groups.

Figure 6

Interhemispheric functional connectivity difference between phonemic (P) and semantic (S) verbal fluency tasks derived from HbO: (A) Average of inter2 FC during VFT, (B) Inter FC difference between phonemic and semantic VFT. The scatter plots on the left show the mean and standard error, while the boxplots on the right display the interquartile range, with whiskers extending to the lower and upper bound. The asterisk indicates FDR corrected p < 0.01.