Differential Structural Brain Development in Adolescents with Different Apolipoprotein Eε Genotypes

1219 

Abstract Submission 

Linda Chang¹, Mahsa Mayeli¹, Amal Isaiah¹, shuo chen², Thomas Ernst¹

¹University of Maryland School of Medicine, Baltimore, MD, ²university of maryland, baltimore, baltimore, MD

Linda Chang  
University of Maryland School of Medicine
Baltimore, MD

Mahsa Mayeli, M.D.  
University of Maryland School of Medicine
Baltimore, MD

Amal Isaiah  
University of Maryland School of Medicine
Baltimore, MD

shuo chen  
university of maryland, baltimore
baltimore, MD

Thomas Ernst  
University of Maryland School of Medicine
Baltimore, MD

Apolipoprotein E epsilon (APOEε) allele has three isoforms, yielding six genotypes (ε2ε2, ε2ε3, ε2ε4, ε3ε3, ε3ε4, ε4ε4). Children with different APOEε genotypes showed differential brain maturation in a cross-sectional study (Chang 2016). Specifically, children with ε2ε4 had smallest hippocampi and young children (<10 years) with ε4ε4 had the lowest hippocampal fractional anisotropy relative to those with other APOEε genotypes. Furthermore, young children with ε2ε4 genotype performed worst on attention tasks while those with ε4ε4 had lowest scores on executive function and working memory. The ε4 allele is associated with greater risks for Alzheimer's disease (Fernández-Calle, 2022), or poorer outcomes from brain injuries (Chang, 2011; Kassam 2016). Understanding the effects of ε4 allele on brain development may provide useful early biomarkers. The current study aims to validate prior findings using the large longitudinal dataset of typically developing adolescents from the Adolescents Brain Cognitive Development (ABCD) study.

Data release 4.0 included data from 11,875 healthy children (www.abcdstudy.org). We extracted the genomics data and identified six APOEε genotypes for the associated single nucleotide polymorphisms (rs429358 and rs7412) using Plink 2.0. We included 2,476 participants with complete datasets from baseline (ages 9-10 years), 2-year and 4-year follow-up visits. We analyzed structural MRI [processed with Freesurfer v5.3+ using regions of interests (ROIs) from the Desikan-Killiany atlas (Halger, 2019)] and cognitive scores from the NIH Toolbox-Cognitive Battery (NIHTB-CB). APOEε2/ε2 were excluded due to the small sample size. We investigated APOEε genotype differences and APOEε genotype-by-Age interactions in thickness, surface area, and volume for cortical ROIs, and in subcortical ROI volumes. Longitudinal changes in these measures were assessed using a linear mixed-effects model, incorporating APOEε group, age, sex at birth, family income, study site, scanner ID, and intracranial volume (for volume measurements only) as fixed effects, and participant as random effect. The false discovery rate (FDR) was used to correct for multiple comparisons. All analyses were conducted in R 4.3.1.

Participant characteristics are summarized (Figure 1). APOEε genotype effects or APOEε*Age interactions were significant in multiple ROIs for cortical thickness, surface area or volumes, and for subcortical volumes. Only six cortical measures remained significant after FDR-correction: cortical thickness of bilateral Cuneus, bilateral Pericalcarine cortices, left Lingual gyrus and left postcentral gyrus (Figure. 1). However, all subcortical gray matter volumes remained significant after FDR-correction for APOEε-main effects (p<0.001) or APOEε*Age interactions (p<0.0001), marginal effect size R2=0.19-0.45 (Figure 2). Specifically, relative to the other genotypes, the ε2ε4 group (red lines) had the smallest hippocampi and amygdalae, but largest pallidum, caudate and putamen bilaterally. Conversely, the ε4ε4 group (green lines) had the smallest putamen and largest amygdalae. In addition, the volumes of the corpus callosum showed group differences (Posterior: Age*APOEε-corrected-p=0.005, APOEε-corrected-p<0.001, Middle and Central: APOEε-corrected-p<0.001). Furthermore, global volumes of cerebral white matter, cerebellar cortex and white matter all showed APOEε and APOEε*Age effects (corrected-p<0.001). However, no significant effects for APOEε or APOEε*Age were found for any of the NIH-TB-CB test scores.

Adolescents with different APOEε genotypes show significant variations in the growth trajectories of their cortical thickness and subcortical gray matter volumes, especially those with ε2ε4 and ε4ε4, validating prior observations (Chang, 2016). Whether these early morphometric differences persist into adulthood, leading to greater vulnerability for brain injury or lower repair capacity at later life, requires further follow-up studies.

Neurodevelopmental/ Early Life (eg. ADHD, autism)

Genetic Association Studies ²

Early life, Adolescence, Aging ¹

Development

MRI

NORMAL HUMAN

Other - Apolipoprotein E