Entorhinal-based path integration selectively predicts midlife risk of Alzheimer's disease

Abstract

Introduction: Entorhinal cortex (EC) is the first cortical region to exhibit neurodegeneration in Alzheimer's disease (AD), associated with EC grid cell dysfunction. Given the role of grid cells in path integration (PI)-based spatial behaviors, we predicted that PI impairment would represent the first behavioral change in adults at risk of AD.

Methods: We compared immersive virtual reality (VR) PI ability to other cognitive domains in 100 asymptomatic midlife adults stratified by hereditary and physiological AD risk factors. In some participants, behavioral data were compared to 7T magnetic resonance imaging (MRI) measures of brain structure and function.

Results: Midlife PI impairments predicted both hereditary and physiological AD risk, with no corresponding multi-risk impairment in episodic memory or other spatial behaviors. Impairments associated with altered functional MRI signal in the posterior-medial EC.

Discussion: Altered PI may represent the transition point from at-risk state to disease manifestation in AD, prior to impairment in other cognitive domains.

Keywords: Alzheimer's disease; entorhinal cortex; path integration; preclinical; virtual reality.

FIGURE 1

Hereditary and physiological risk factor PI impairments. A, The PI task schematic and three return condition types. B, Higher CAIDE dementia risk score significantly correlated with decline in performance (viz. increase in location error) on no distal cues condition relative to baseline in both sexes. C, Decline in performance on no distal cues relative to baseline was largest in FH+/APOE ε4+. D, FH+ and APOE ε4+ performance decline on no distal cues relative to baseline was specific to males (combined for display). E, FH+/APOE ε4+ or CAIDE (not pictured) decline in performance on no optic flow condition relative to baseline was not significant. ** P _Tukey < 0.01. APOE, apolipoprotein E; CAIDE, Cardiovascular Risk Factors, Aging and Dementia Study; FH, family history; PI, path integration

FIGURE 2

Angular rather than distance errors contributed to risk factor–associated impairment of PI after removal of distal orientation cues. A, Schematic of angular error (left) and distance error (right) calculations. B, CAIDE dementia risk score correlated with change in angular but not distance error from baseline to no distal cues. C, Multi‐hereditary risk interacted with sex for angular but not distance error changes in PI. *P_Tukey  < 0.05. APOE, apolipoprotein E; CAIDE, Cardiovascular Risk Factors, Aging and Dementia Study; DRS, dementia risk score; FH, family history; PI, path integration

FIGURE 3

Right posterior–medial entorhinal fMRI correlates of decline in path integration performance. A, Image of object memory location fMRI task. B, Greater change in location error from baseline to “no distal cues” return condition associated with lower 6‐fold grid‐like activity, which reflects behavioral findings of showing a stronger effect in males with hereditary risk (C) or individuals with a higher CAIDE score (D). High risk is defined as FH+/APOE ε4+ and low risk as individual or no FH/APOE ε4 risk factors. E, The negative 6‐fold grid magnitudes in (B,C) might relate to a unidirectional head direction‐like signal, which was stronger in individuals that showed worse PI in the absence of orientation cues. Again, this appeared specific to males with hereditary risk (F) in an exploratory analysis. APOE, apolipoprotein E; CAIDE, Cardiovascular Risk Factors, Aging and Dementia Study; FH, family history; fMRI, functional magnetic resonance imaging; PI, path integration