Inference by interference: Task-dependent pupil responses as an early detection method for Alzheimer's disease related brainstem functional change

Massachusetts General Hospital, Boston MA

Investigators

Linked publications & trials

Abstract

ABSTRACT Trials aimed at changing the hallmark proteinopathies of Alzheimer Disease (AD), Aβ and tau, have been disappointing in halting or reducing cognitive decline. As these trials are often performed in symptomatic individuals, where brain changes are irreversible, shifting interventions to asymptomatic individuals may be more effective. The conundrum is to identify people who are more likely to be at risk of AD and who will benefit from these interventions. While Aβ starts in the neocortex about 15 years before clinical symptoms, tau pathology starts to accumulate from age 20 in the locus coeruleus (LC) before propagating towards the transenthorhinal cortex, the limbic system and subsequently to the cortical regions. However, it is important to note that even though by age 50, tau is omnipresent in the LC, not everyone will develop AD. Given that accumulation of tau in the LC is associated with loss of projections, we posit that functional measures of the LC-Norepinephrine system (LC-NE) may be more sensitive to distinguish individuals at-risk from those who are likely to maintain cognitive health. The LC-NE system plays an important modulatory role in behavioral and cognitive processes including ‘executive’ functions like attention and working memory. The overall goal of the proposed project is to detect the earliest AD related changes through read-outs related to LC function, which will ultimately contribute to improved identification of the target population for interventions and increase the efficacy of these interventions. LC neuronal function is characterized by tonic (that we can elicit by increasing CO2) and phasic firing patterns (that can be elicited by novelty). Changes in pupil dilation receive contributions from the LC-NE system, and by providing specific cognitive task loads, phasic and tonic LC function can be derived. Wehypothesize that the pupil response in a task reacting to new tones after a short breath hold vs no breath hold can be a biomarker of very early AD-related processes. We will test these hypotheses in a group of clinically normal individuals (age range 55-80 years, n=70, 50% female) leveraging to an ongoing deeply phenotyped cohort (SerialMK) that already collects Aβ and tau-PET imaging. The results of this proposed project will contribute to improved detection of individuals at-risk by determining (Aim 1): To investigate the tonic-phasic balance in pupil responses in different task settings (oddball, and oddball after breath hold) and their relation to age and sex. (Aim 2): Relating pupillary responses in these different task settings to AD pathology. The proposed research is innovative as it aims to identify at-risk populations in a non- invasive, functionally relevant manner, using a readout of one of the earliest affected brain systems and thereby opening up the opportunity to benefit prevention trials at a time in life when pathology is not yet extensive.