No effects of APOE ε4 on spatial navigation and broader cognition in young adults genetically at risk for Alzheimer's disease

Problem Statement
Alzheimer's disease (AD) is characterized by progressive memory loss and disorientation, with pathological changes in medial temporal lobe regions often preceding clinical symptoms by decades. The apolipoprotein E (APOE) ε4 allele is the major genetic risk factor for sporadic AD. While spatial navigation deficits have been identified in older ε4 carriers and proposed as early biomarkers for AD, it remains unclear whether these deficits are detectable in healthy young adults (18–35 years) who carry the ε4 allele. Previous findings in younger cohorts are mixed, with some studies suggesting neural changes without behavioral deficits, while others report no differences. Furthermore, the cognitive impact of the APOE ε2 allele, often associated with reduced AD risk, remains underexplored in young populations. This study aimed to determine if carrying the APOE ε4 allele is associated with measurable differences in spatial navigation and broader cognitive domains (working memory, executive function, face recognition) in healthy young adults.

Methodology
The study utilized a large-scale sample of 1,000 healthy young adults (18–35 years). Participants provided DNA samples via buccal swabs for APOE genotyping. After excluding individuals with neurological/psychiatric histories, incomplete data, or non-target genotypes, the final analysis focused on 415 participants for spatial navigation (327 ε3ε3 non-carriers and 88 ε3ε4 carriers) and varying sub-samples for other cognitive tasks.

The experimental design involved two sessions:

1. Onsite (Day 1): DNA collection and administration of the Trail Making Test (TMT A & B) to assess visual search, processing speed, and executive function.
2. Online (Day 2): Administration of the app-based navigation game "Sea Hero Quest" (SHQ) to measure spatial navigation (wayfinding distance, duration, distance-to-border, and path integration via flare accuracy). Participants also completed computer-based tasks for working memory (Memory Updating and Spatial Short-Term Memory) and face recognition (Cambridge Face Memory Test), alongside questionnaires on gaming experience and self-reported navigation ability.

Statistical analyses included Wilcoxon rank-sum tests for group comparisons, linear regression and mixed-effects models to control for covariates (sex, age, gaming experience), equivalence testing (TOST) to confirm the absence of meaningful differences, and Bayesian analyses to quantify evidence for the null hypothesis. Exploratory analyses using non-parametric bootstrapping were conducted for homozygous ε4 carriers and ε2 carriers due to small sample sizes.

Key Contributions and Results

• Spatial Navigation (SHQ): Contrary to the hypothesis that ε4 carriers would show deficits, the study found no significant differences between ε3ε3 and ε3ε4 carriers in wayfinding distance, wayfinding duration, or mean distance-to-border. These null results held across different levels of difficulty and after controlling for demographic and cognitive covariates. Equivalence testing and Bayesian analyses (BF01 = 6.34–7.58) provided moderate evidence supporting the null hypothesis, indicating that any true differences are likely negligible.
• Path Integration: No significant differences were found in flare accuracy (a measure of path integration) between ε3ε3 and ε3ε4 carriers, though performance was near ceiling, suggesting the task may lack sensitivity for this specific metric in young adults.
• Broader Cognition: No significant genotype effects were observed for working memory (Memory Updating, Spatial Short-Term Memory), executive function (TMT A & B), or face recognition (CFMT) between ε3ε3 and ε3ε4 groups. Equivalence testing supported the null for most measures, with Bayes factors indicating moderate evidence for no difference.
• Exploratory Findings on Other Genotypes:
  • Homozygous ε4 (ε4ε4): Due to the small sample size (N=11), no significant differences were detected compared to ε3ε3 controls, though the authors note this requires caution.
  • ε2 Carriers: Exploratory bootstrapping suggested potential cognitive advantages in ε2 carriers. Specifically, ε2ε3 carriers navigated closer to environmental borders; ε2ε4 carriers showed better memory updating; and ε2ε2 carriers demonstrated faster processing speed (TMT A) and better face recognition (CFMT) compared to ε3ε3 controls.

Significance and Claims
The paper concludes that in healthy young adults, the APOE ε4 allele does not manifest as detectable behavioral deficits in spatial navigation or broader cognitive domains, even when using sensitive paradigms like SHQ. The authors suggest that while neural changes (such as reduced grid-like codes) may occur in young ε4 carriers, they are likely compensated for at the behavioral level, with deficits potentially emerging only later in life or in the presence of additional risk factors (e.g., vascular burden, family history).

The study highlights the challenge of detecting early cognitive signatures of genetic dementia risk in young populations using standard behavioral tasks. It also points to the underexplored role of the APOE ε2 allele, suggesting it may confer cognitive benefits even in young adulthood, consistent with its known protective effects in older age. The authors emphasize that future research should employ paradigms sensitive to very subtle changes and consider the interplay of multiple risk factors to understand the preclinical trajectory of AD.