Skip to main content

Effects of APOE2 and APOE4 on brain microstructure in older adults: modification by age, sex, and cognitive status



APOE4 is the strongest genetic risk factor for sporadic Alzheimer’s disease (AD), whereas APOE2 confers protection. However, effects of APOE on neurodegeneration in cognitively intact individuals, and how these associations evolve with cognitive decline, are unclear. Furthermore, few studies have evaluated whether effects of APOE on neurodegenerative changes are modified by other AD key risk factors including age and sex.


Participants included older adults (57% women; 77 ± 7 years) from the Rancho Bernardo Study of Health Aging and the University of California San Diego Alzheimer’s Disease Research Center, including 192 cognitively normal (CN) individuals and 33 with mild cognitive impairment. Participants underwent diffusion MRI, and multicompartment restriction spectrum imaging (RSI) metrics were computed in white matter, gray matter, and subcortical regions of interest. Participants were classified as APOE4 carriers, APOE2 carriers, and APOE3 homozygotes. Analysis of covariance among CN (adjusting for age, sex, and scanner) assessed differences in brain microstructure by APOE, as well as interactions between APOE and sex. Analyses across all participants examined interactions between APOE4 and cognitive status. Linear regressions assessed APOE by age interactions.


Among CN, APOE4 carriers showed lower entorhinal cortex neurite density than non-carriers, whereas APOE2 carriers showed lower cingulum neurite density than non-carriers. Differences in entorhinal microstructure by APOE4 and in entorhinal and cingulum microstructure by APOE2 were present for women only. Age correlated with lower entorhinal restricted isotropic diffusion among APOE4 non-carriers, whereas age correlated with lower putamen restricted isotropic diffusion among APOE4 carriers. Differences in microstructure between cognitively normal and impaired participants were stronger for APOE4-carriers in medial temporal regions, thalamus, and global gray matter, but stronger for non-carriers in caudate.


The entorhinal cortex may be an early target of neurodegenerative changes associated with APOE4 in presymptomatic individuals, whereas APOE2 may support beneficial white matter and entorhinal microstructure, with potential sex differences that warrant further investigation. APOE modifies microstructural patterns associated with aging and cognitive impairment, which may advance the development of biomarkers to distinguish microstructural changes characteristic of normal brain aging, APOE-dependent pathways, and non-AD etiologies.


Apolipoprotein E (APOE) is a polymorphic lipoprotein with three predominant isoforms that modify risk for sporadic Alzheimer’s disease (AD). Compared to ε3 homozygotes, individuals with one or two ε4 alleles harbor approximately 3 and 15 times increased risk, respectively, whereas carriers of the protective ε2 allele have an estimated 1.7 times decreased risk [1]. APOE is multifunctional, with roles in lipid and glucose metabolism, neuroinflammation, amyloid and tau accumulation, and blood-brain barrier dysfunction [2]. However, the downstream regulation of neurodegeneration or neuroprotection by APOE-dependent pathways remains elusive, due in part to the challenges of in vivo approaches that are typically limited in their measurement of subtle cytoarchitectural changes. Improved characterization of APOE effects on brain microstructure in non-demented older adults may be dually instrumental in optimizing preclinical AD biomarkers and therapeutic strategies targeting APOE-dependent pathways.

Diffusion MRI studies have identified microstructural white matter (WM) abnormalities in cognitively normal older APOE4 carriers, including increased mean diffusivity and reduced fractional anisotropy or restricted diffusion [3,4,5], whereas others observed null [6, 7] or inconsistent [8] effects. Prior discrepancies may be attributable to cohort differences in factors such as age range, health, or socioeconomic status, to methodological differences in diffusion MRI techniques, or inclusion of APOE2 carriers in the reference group. Whereas previous studies have predominantly focused on WM integrity, more recent investigations have interrogated cortical or subcortical gray matter with inconclusive findings [9, 10]. Further investigation of APOE effects on regions that are early targets of AD pathology, including the entorhinal cortex and hippocampus, is warranted.

Uncertainty persists regarding the potential neuroprotective effects of APOE2 on brain aging, due largely to its low population prevalence of only 10%, making it understudied relative to APOE4. Although lower amyloid burden has been observed among APOE2 carriers, differences in tau or atrophy have been inconclusive [11, 12], with some reports of slower hippocampal atrophy among APOE2 carriers [13]. The few diffusion MRI studies evaluating APOE2 have similarly generated mixed findings, with one study observing higher WM fractional anisotropy among APOE2 carriers [14], yet others reporting null effects [7, 15]. Thus, despite the established protective effect of APOE2 against AD, it remains unclear whether higher brain reserve in the form of preserved cytoarchitectural integrity, underlies this resistance to AD.

APOE is differentially associated with AD risk across populations [16], such that key AD risk factors, including older age and female sex, may obscure associations of APOE with neurodegenerative trajectories. Consistent with accelerating APOE4-dependent risk for AD and cognitive decline with older age [17], some studies have reported stronger effects of APOE4 on WM microstructure with age [3, 15, 18], whereas others observed no such interaction [8]. Although sex differences in effects of APOE on microstructure have been scarcely examined, studies reporting sex-specific associations of APOE4 with WM integrity [19] and stronger APOE-related differences in brain metabolism and cortical atrophy for women than men [20, 21] provide grounds for deeper exploration of sex differences in underlying cytoarchitecture.

Considering the prolonged preclinical period preceding AD onset, identifying the earliest APOE-dependent neurodegenerative changes may provide a window of opportunity for timely detection at critical points along the disease course, warranting focus on asymptomatic individuals. However, given the heterogeneous nature of mild cognitive impairment (MCI), characterizing neurodegenerative patterns that manifest with AD-specific risk may help to develop biomarkers that aid in more accurate differential diagnosis. While prior diffusion MRI studies have focused on cognitively normal individuals or those with AD dementia, limited evidence suggests that APOE4 may drive accelerated hippocampal atrophy and network disruption with MCI [22], though others report no modifying effect of APOE4 on WM microstructure in MCI [23]. Thus, further research interrogating differential markers of cytoarchitectural injury according to APOE across the preclinical to prodromal AD continuum may help to inform the probable etiology of nascent cognitive decline.

Thus, further research is needed to characterize APOE-related patterns of microstructural brain aging while considering key AD risk modifiers, to improve precision diagnostic approaches for targeted early disease detection. In this study, we employed restriction spectrum imaging (RSI) [24], a multicompartment diffusion MRI model that offers more comprehensive characterization of tissue cytoarchitecture than morphometric MRI or diffusion tensor imaging (DTI), to examine effects of APOE2 and APOE4 on brain microstructure in older adults across the continuum from cognitively normal to mildly impaired. Building upon our prior work demonstrating excellent sensitivity of RSI metrics to microstructural brain injury in MCI [25, 26], and to more subtle cytoarchitectural changes that manifest with normal aging [27], herein we further probe effects of APOE on brain microstructure and their modification by age and sex.



Eligible participants included predominantly non-Hispanic White community-dwelling participants of the Rancho Bernardo Study (RBS) of Healthy Aging and the UC San Diego Shiley-Marcos Alzheimer’s Disease Research Center (ADRC) longitudinal study who completed a diffusion MRI scan and had available genetic data. MRI data were acquired for RBS participants between 2014 and 2016 and for ADRC participants from 2013 to 2022. ADRC participants underwent consensus diagnosis by two senior neurologists, and those with diagnoses of cognitively normal (CN), MCI [28], or mild AD [29], were eligible for study inclusion. RBS participants did not undergo clinical evaluation but completed the Modified Mini-Mental State Exam (3MS), a cognitive screening tool for the assessment of dementia. Cognitively impaired (CI) individuals included ADRC participants with diagnoses of MCI or mild AD and RBS participants with a 3MS score < 78 [30]. Exclusion criteria included history of head injury, stroke, dementia, neurological disease, treatment for an alcohol use disorder, or safety contraindication for MRI. After excluding six participants due to poor MRI data quality and three participants with APOE2/4 genotype due to conflicting risk effects of the ε2 and ε4 alleles, the final sample included 225 participants (139 RBS, 86 ADRC; 192 CN, 33 CI; 57% women; age at MRI: mean ± SD 76.5 ± 7.0, range 56–97 years).

Standard protocol approvals and participant consents

Study procedures were approved by the University of California, San Diego Human Research Protections Program Board and participants provided informed written consent prior to participation.

Demographic and health assessment

Education level was acquired at enrollment and converted to years of education. History of medical conditions was obtained from standard questionnaires. Blood pressure was measured in seated, resting participants. Participant height and weight were measured to compute body mass index (BMI, kg/m2).

APOE genotyping

For RBS participants, DNA was extracted by Sequana Therapeutics (La Jolla, CA) using standard techniques (Puregene; Gentra, Minneapolis, MN) and genotyping was conducted by Diagnomics, Inc. Genotyping for ADRC participants used a commercially available Illumina BeadChip array. Participants were classified as APOE2/3, APOE4 carrier (APOE3/4 or APOE4/4), or APOE3/3 (there were no APOE2 homozygotes and APOE2/4 participants were excluded from analysis). One participant with an inconclusive genotype of APOE2/3 versus APOE3/3 was excluded from APOE2 analyses.

Amyloid and tau measurement

A subset of ADRC participants (N = 70) underwent lumber puncture with standardized procedures, preanalytical preparation, and storage of cerebrospinal fluid in accordance with the recommended best practices [31]. Briefly, lumbar puncture was conducted early in the morning after overnight fasting to collect 15–25 mL cerebrospinal fluid. Samples were processed, aliquoted into 500 μL fractions in polypropylene microtubes, snap-frozen, and stored at -80 °C until assayed. Samples were analyzed with the automated Lumipulse platform using assays developed with established monoclonal antibodies (Fujirebio Inc.) to measure beta-amyloid (Aβ)-42 and 40, total tau (t-tau), and phosphorylated tau (p-tau). T-tau/Aβ42 > 0.54 was considered positive for AD pathology [32]. Because only a subset of participants (31% of the full sample) underwent lumbar puncture, cerebrospinal fluid measures are reported for AD biomarker characterization and were not incorporated into analyses.

Imaging data acquisition

Imaging data were acquired on two 3.0 Tesla Discovery 750 scanners (GE Healthcare, Milwaukee, WI, USA) at the University of California, San Diego Center for Functional MRI (100% of RBS participants and 58% of ADRC participants) and the Altman Clinical and Translational Research Institute (42% of ADRC participants), using an eight-channel phased array head coil. MRI sequences common to the RBS and ADRC protocols included a three-plane localizer; a sagittal 3D fast spoiled gradient echo T1-weighted structural scan optimized for maximum tissue contrast (TE = 3.2 ms, TR = 8.1 ms, inversion time = 600 ms, flip angle = 8°, FOV = 256 × 256 mm, matrix = 256 × 192, slice thickness = 1.2 mm, resampled to a 1 × 1 × 1.2 mm resolution, scan time 8:27); and an axial 2D single-shot pulsed-field gradient spin-echo echo-planar diffusion-weighted sequence (45 gradient directions, b-values = 0, 500, 1500, 4000 s/mm2, one b = 0 volume and 15 gradient directions for each non-zero b-value; TE = 80.6 ms, TR = 7 s, FOV = 240 × 240 mm, matrix = 96 × 96, slice thickness = 2.5 mm, resampled to a 1.875 × 1.875 × 2.5 mm resolution, scan time 6:34).

Data processing

All raw and processed MR images were visually inspected for artifacts and processed using an automated image processing pipeline that integrates FreeSurfer ( with in-house software [33]. Cortical gray matter, WM, and CSF boundaries were reconstructed from T1-weighted structural images using FreeSurfer (version 5.3) and subcortical regions were automatically segmented according to a subcortical atlas [34]. Using previously detailed methods [33], diffusion MRI data underwent eddy current correction in the phase-encode direction with displacements modeled as a function of spatial location, gradient orientation, and gradient strength, and correction for head motion with rigid-body registration. Spatial and intensity distortions caused by B0 field inhomogeneity were corrected by aligning b = 0 images acquired in opposite phase encoding direction using nonlinear registration and correcting subsequent images using the displacement volume. The b = 0 images were registered to T1 images using mutual information after coarse pre-alignment to atlas brains, and diffusion images were aligned with a fixed rotation and translation relative to the T1 image. For cortical surface-based analyses, RSI metrics were sampled with linear interpolation from 0.8 to 2.0 mm from the gray/white matter boundary and, to minimize partial volume effects, were computed using a weighted average based on the proportion of gray matter in each voxel [35]. RSI cortical surface maps were registered to common space and smoothed with a FWHM 10 mm kernel. WM fiber tracts were labeled using AtlasTrack, a WM fiber atlas based on prior probability and orientation information [36], and voxels containing primarily gray matter or CSF were excluded from WM [34]. T1 images were used to nonlinearly coregister brains to common space and diffusion orientation estimates were compared to atlas orientations to refine voxel-wise probabilities of belonging to a given fiber. All raw and processed structural and diffusion images underwent visual quality control. Manual editing of the cortical surface reconstruction, including adding white matter control points or removing mislabeled non-brain voxels, was conducted when applicable; the majority of scans underwent at least minimal editing.

Computation of RSI metrics

For primary analyses, RSI measures were computed in regions of interest (averaged between left and right hemispheres), identified based upon their early involvement in AD, including the entorhinal cortex and hippocampus, which are targets of early AD neuropathology and neurodegeneration, and eight WM limbic and association fibers (cingulum, corpus callosum, fornix, inferior fronto-occipital fasciculus, inferior longitudinal fasciculus, parahippocampal cingulum, superior longitudinal fasciculus, and uncinate). Exploratory analyses also examined RSI measures in global white and cortical gray matter, across cortical gray matter surface vertices, and in four additional subcortical regions (thalamus, putamen, caudate, and amygdala). Global white and gray matter measures were calculated as the mean across all WM fibers or all cortical gray matter, respectively. Computed metrics included restricted isotropic diffusion (RI), a measure of highly restricted, non-oriented diffusion that likely corresponds with the intracellular fraction; neurite density (ND), a measure of oriented restricted diffusion that accounts for multiple diffusion orientations, presumably reflecting cellular processes such as axons, dendrites, or glial processes; hindered isotropic diffusion (HI), a measure of non-restricted diffusion that is hindered by cellular barriers and consistent with diffusion within large cell bodies or the extracellular space; and isotropic free water (IF), a measure of cerebrospinal fluid (Additional file 1: Table S1) [24, 27]. All measures were computed in cortical and subcortical gray matter, while HI was not examined in WM fibers due its poor representation in WM [24].

Statistical analysis

Differences in demographic and health factors by APOE genotype were examined using analyses of variance for continuous variables or chi-squared tests for categorical variables. BMI was adjusted for sex.

Analyses of RSI measures included covariates of age, sex, and scanner. To examine effects of APOE4 on brain microstructure in asymptomatic individuals, analyses of covariance (ANCOVA) were conducted among CN with APOE4 (carrier versus non-carrier) as the independent variable and RSI metric as the dependent variable. To evaluate effects of APOE2, models were repeated with APOE4 non-carriers further classified as APOE2/3 or APOE3/3.

To evaluate sex differences in associations between APOE4 or APOE2 and brain microstructure, models were repeated with a term for the interaction between sex and APOE status.

To assess modification of age-microstructure associations by APOE, linear regressions were conducted with factors of APOE, age, and the interaction between APOE and age. Secondary models included additional factors of age2 and the interaction between APOE and age2 to assess possible nonlinear age effects. Interaction models used mean-centered variables to reduce multicollinearity. For any RSI measure demonstrating significant interactions, APOE-stratified analyses were performed.

Finally, to probe whether APOE4 modified effects of cognitive impairment on microstructure, ANCOVAs were conducted with factors of APOE4 and cognitive status (CN versus CI) and their interaction. Significant interactions were followed by analysis of cognitive status stratified by APOE4.

Region-of-interest analyses were conducted in SPSS version 28.0 (IBM Corp, Armonk, NY, USA) and cortical surface analyses were performed in FreeSurfer version 6.0. Significance was set to p < 0.05. Uncorrected p-values are reported for transparency, but to account for multiple comparisons across eight fibers and four subcortical regions, significance was assessed using Bonferroni corrected thresholds set to p < 0.006 for WM fibers and p < 0.012 for subcortical regions. Cortical surface general linear models were corrected with the false discovery rate (FDR) method.


Participant characteristics

Participant characteristics by APOE4 (74 carriers, 151 non-carriers) for the full sample, and stratified by cognitive status, are shown in Additional file 1: Table S2. APOE4 carriers (28%) were more likely to be cognitively impaired (p < 0.001) than non-carriers (8%) and had higher diastolic blood pressure (p = 0.009). Further analysis of diastolic blood pressure identified a sex by APOE4 interaction among CN (F(1,182) = 5.87, p = 0.02), such that the difference was only present among men (p < 0.001) but not women (p = 0.31) (full sample: men p = 0.03, women p = 0.09). Participant age, sex, years of education, systolic blood pressure, BMI, or diabetes did not differ by APOE4 (p > 0.05). CI participants were more likely to be male and more highly educated than CN participants (p < 0.001; Additional file 1: Table S3). APOE2/3 participants did not differ from APOE2 non-carriers on any demographic or health factor (Table 1). The subset of 70 participants who underwent lumbar puncture included 31% CI, 47% APOE4 carriers, and 54% women and had a mean age of 75.5 ± 5.6 years and mean time between MRI and lumbar puncture of 2.4 ± 2.2 years. APOE4 carriers and CI participants were more likely to be positive for AD pathology (tau/Aβ42, p < 0.05), and to have lower Aβ42/40 and higher p-tau than APOE4 non-carriers and CN participants (p < 0.01) (Additional file 1: Table S4). Participant characteristics by cohort are presented in Additional file 1: Table S5.

Table 1 Participant characteristics (mean ± SD or N(%)) by APOE genotype for all subjects and cognitively normal subjects

Effects of APOE4 on brain microstructure

When brain microstructure was examined among CN according to APOE4 status, entorhinal cortex ND was lower for APOE4 carriers compared to non-carriers (F(1,187) = 4.46, p = 0.04; Additional file 1: Table S6; Fig. 1A). This difference was unchanged by adjustment for entorhinal cortex thickness (F(1,185) = 4.39, p = 0.04). Sex-stratified analyses revealed that entorhinal ND differed by APOE4 only for women (F(1,115) = 4.90, p = 0.03) but not for men (F(1,69) = 0.25, p = 0.62) (Additional file 1: Figure S1A). However, sex did not interact with APOE4 for entorhinal ND (p = 0.42) or any other measure. Because male APOE4 carriers had higher diastolic blood pressure than non-carriers, analyses were further adjusted for diastolic blood pressure. Differences in entorhinal ND were moderately attenuated for the full sample (F(1,186) = 2.23, p = 0.14) and trivially changed among women (F(1,114) = 3.60, p = 0.06), for whom blood pressure did not differ by APOE4. Age interacted with APOE4 for entorhinal cortex RI (p = 0.02), with correlations between older age and lower RI for APOE4 non-carriers (p < 0.001) but not carriers (p = 0.88) (Fig. 1B). This interaction remained after adjustment for entorhinal cortex thickness (p = 0.02) and diastolic blood pressure (p = 0.04). Exploratory analyses revealed no further differences by APOE4 nor interactions with age or sex, across the cortical gray matter surface (p < 0.05 FDR corrected), in global white or cortical gray matter, or within subcortical regions (Bonferroni corrected).

Fig. 1
figure 1

Differences in entorhinal cortex microstructure by APOE4 among cognitively normal participants. Entorhinal cortex neurite density was lower for cognitively normal APOE4 carriers than non-carriers (A). Entorhinal restricted isotropic diffusion correlated with age only for APOE4 non-carriers (B). Values are residuals, adjusted for age, sex, and scanner

Effects of APOE2 on brain microstructure

To probe differences by APOE2, APOE4 non-carriers were further classifed as APOE2/3 (there were no APOE2 homozogotes) and APOE3 homozygotes. Across all CN participants, cingulum ND differed by APOE2 (F(2,185) = 5.42, p = 0.005), with lower ND for APOE2/3 than APOE3/3 and APOE4 carriers (Additional file 1: Table S7; Fig. 2A). In sex-stratified analyses, this difference was present for women (F(2,114) = 6.33, p = 0.002) but not for men (F(2,67) = 0.70, p = 0.50) (Additional file 1: Figure S1C). Similarly, there was an effect of APOE2 on entorhinal HI for women (F(2,114) = 4.43, p = 0.01) but not men (F(2,67) = 0.10, p = 0.90), reflecting lower HI for APOE2/3 than APOE3/3 and APOE4 women (Additional file 1: Figure S1B). The difference for women was strengthened after adjustment for entorhinal cortex thickness (F(2,112) = 7.53, p < 0.001). However, sex by APOE interactions did not reach significance for cingulum ND (p = 0.28), entorhinal HI (p = 0.056), or any other measure. Exploratory analyses revealed an interaction between APOE and age for putamen RI (p = 0.003), with age-related decline present only for APOE4 carriers (Fig. 2B).

Fig. 2
figure 2

Effects of APOE2 on brain microstructure among cognitively normal participants. Cingulum neurite density was lower for cognitively normal APOE2/3 than for APOE2 non-carriers (A). APOE interacted with age on putamen restricted isotropic diffusion, with correlations present for APOE4 carriers only (B). Values are residuals, adjusted for age, sex, and scanner

Interactions between APOE4 and cognitive status on brain microstructure

When models were conducted across the entire sample (both CN and CI) including cognitive status, APOE4, and their interaction, pronounced differences by cognitive status were observed, with additional differences by APOE4. As we previously reported in an overlapping sample [25], CI demonstrated widespread microstructural abnormalilties compared to CN (Additional file 1: Figure S2A), with the strongest differences in entorhinal cortex IF and HI, and hippocampal RI (p < 0.001); CI showed higher IF, and lower RI and HI, than CN. APOE4 carriers exhibited lower entorhinal cortex HI and global gray matter ND and HI, as well as higher thalamic RI and ND, and global gray matter IF, than non-carriers (Additional file 1: Figure S2B).

APOE4 interacted with cognitive status for hippocampal IF and HI (p < 0.05), and for entorhinal HI (p = 0.004) (Additional file 1: Table S8, Additional file 1: Figure S2C). As shown in Fig. 3, CI showed higher hippocampal IF and lower entorhinal cortex HI than CN among both APOE4 carriers and non-carriers, with more pronounced differences for carriers. CI showed lower hippocampal HI than CN among APOE4 carriers only. After adjustment for cortical thickness or volume respectively, the interactions for entorhinal cortex HI (p = 0.14) and hippocampal IF (p = 0.16) and HI (p = 0.17) were attenuated, driven by greater atrophy with cognitive impairment among APOE4 carriers. Exploratory analyses also revealed significant interactions for global gray matter RI, HI, and IF (p < 0.01), thalamus RI (p < 0.001) and ND (p = 0.009), and caudate RI (p = 0.003), although whole-brain analyses revealed no signifcant localized interactions within cortical gray matter. Among APOE4 carriers, CI exhibited lower gray matter RI and HI and higher IF, yet higher thalamic RI and ND, than CN. Among APOE4 non-carriers, CI showed lower caudate RI than CN. Results were unchanged by further adjustment for diastolic blood pressure. Because education levels were higher among CI than CN (Additional file 1: Table S3), models were repeated with further adjustment for education, which did not alter results.

Fig. 3
figure 3

Interactions between APOE4 and cognitive status on brain microstructure. RSI measures demonstrating significant interactions between APOE4 and cognitive status are shown. Values are residuals, adjusted for age, sex, and scanner

To evaluate region-specific associations among microstructural metrics, correlations among RSI measures within any region demonstrating significant effects of APOE are presented in Additional file 1: Table S9.


Leveraging the improved characterization of cellular microstructure by RSI, this study extends prior evidence of APOE-associated brain atrophy and diffusion abnormalities to further delineate modifying effects of APOE4 at key stages along the cognitive aging spectrum, and to help illuminate the lesser understood role of APOE2 in brain aging. Results revealed effects of APOE4 on microstructure localized to the entorhinal cortex, which is among the earliest cortical targets of tau neuropathology, in contrast to effects of APOE2 on cingulum microstructure. Sex stratification revealed that effects of APOE2 and APOE4 on brain microstructure were present only among women, extending prior evidence that APOE more profoundly modifies AD risk for women than men [16]. Findings implicate differential regulation of disease- and aging-related cytoarchitectural changes by APOE genotype, such that APOE4 carriers undergo entorhinal neurodegeneration in preclinical and prodromal AD, with age-related striatal changes, whereas APOE4 non-carriers present with age-related entorhinal cortex microstructural abnormalities and striatal injury accompanying cognitive impairment.

Among cognitively intact individuals, entorhinal neurite density was lower for APOE4 carriers, which could reflect loss of dendritic density or complexity, or axonal damage occurring early along the AD trajectory. The regional specificity of these findings aligns with evidence that entorhinal cortex is a site of preclinical neuropathological tau deposition that is elevated for APOE4 carriers [37] and predicts cognitive decline [38], and which undergoes accelerated atrophy in preclinical AD [39]. However, longitudinal examination is needed to distinguish preclinical neurodegeneration from pre-existing morphometric vulnerability that may lower brain reserve, as APOE4 carriers demonstrate lower entorhinal cortex thickness even in youth [40]. In contrast, entorhinal RI declined with age among APOE4 non-carriers, which may reflect aging-related neuronal loss or dystrophy among those with reduced genetic risk for AD, consistent with age-related entorhinal atrophy observed in the general population [41]. Given the challenge of distinguishing neurodegenerative from age-related structural brain changes, this dissociation of microstructural features affected by APOE4 (ND) and aging (RI) within a region highlights the value of multicompartment diffusion MRI for developing more discriminative cytoarchitectural biomarkers. Notably, associations of entorhinal microstructural with APOE and age were independent of cortical thinning, pointing to more subtle preclinical cytoarchitectural changes among individuals with elevated AD risk that may precede atrophy. Because of our hypothesis-driven focus on the entorhinal cortex and hippocampus in a priori analyses, these analyses did not adjust for multiple comparisons across regions. However, whole-brain exploratory cortical surface-based analyses identified no significant differences by APOE, warranting caution when interpreting the regional specificity of the entorhinal cortex effects.

Cingulum ND was reduced for APOE2 carriers relative to non-carriers, an unexpected observation indicating a lower fraction of oriented restricted diffusion in a major association fiber among those with low probability of converting to AD. WM ND is influenced by various tissue properties, which may be reduced with lower axonal density or myelination, which presents a significant barrier to water diffusion. Consistent with our finding, Westlye and colleagues [8] observed higher WM mean and radial diffusivity for APOE2 carriers than APOE3 homozygotes. However, others reported no difference in WM integrity for APOE2 carriers [7, 15] and one DTI study identified higher fractional anisotropy for APOE2 carriers than APOE3 homozygotes within a small cluster of the posterior cingulum [14]. Further investigation is needed to reconcile these incongruent findings, which may be attributed to the inability of DTI to account for crossing fibers, the sample size of APOE3 homozygotes in [14] or regional variability in WM organization within the cingulum. However, the functional implications of our finding warrant further investigation and may indicate WM vulnerability among APOE2-carriers, perhaps related to their heightened cerebrovascular risk. Alternatively, APOE2-related reductions in WM neurite density may reflect axonal organization supporting more efficient structural connectivity and restriction of neuropathological spread. Indeed, “large world” connectivity of distal networks including projections of the cingulate with frontal and temporal cortex [42] predominates in AD, accompanied by degraded “rich hubs,” localized networks of high connectivity [43]. Higher brain network segregation may also attenuate cortical tau propagation [44]. Thus, reduced structural connectivity of long-range association fibers such as the cingulum may optimize “small world” organization over distal connectivity, which may reinforce circuits supporting cognitive functions that decline with age.

Sex-stratified analyses suggested that observed effects of both APOE2 and APOE4 on entorhinal (ND and HI) and cingulum (ND) microstructure were limited to women, highlighting the importance of sex stratification in investigations of APOE. Women are more vulnerable to effects of APOE4 on memory decline [45] as well as risk for AD dementia [16] and neuropathology [46], and more limited data suggest that APOE2 also confers greater protection against AD risk for women than men [16]. Our findings extend this evidence to identify entorhinal and white microstructure as novel neurodegenerative markers of this sex disparity. Given that interactions between sex and APOE did not reach significance, further study is warranted to replicate our findings in larger datasets and to identify mechanisms by which sex-specific hormonal, sociocultural, or other factors interact with APOE.

Despite previous reports of WM compromise among APOE4-carriers [3,4,5], our data revealed no robust difference in WM microstructure by APOE4 status. However, widespread WM differences by cognitive status were observed, consistent with our prior findings in an overlapping sample [25], that were not modified by APOE4. These findings would be consistent with a pathological cascade that originates with medial temporal neurodegeneration, followed by WM degeneration as neuropathology and associated cellular injury spreads throughout the cortex with disease progression. Further replication and longitudinal investigation are needed to clarify the temporal sequence of neurodegeneration and underlying mechanisms.

Our findings extend widely reported microstructural abnormalities in individuals with MCI [47] to demonstrate that APOE4 modifies the microstructural signature of MCI, perhaps reflecting unique etiologies within this notably heterogeneous condition. While entorhinal hindered diffusion was lower and hippocampal free water was higher for cognitively impaired participants across the full sample, these differences were accentuated among APOE4 carriers, consistent with their role as preclinical targets of AD neuropathology and neurodegeneration. These effects were partially attributable to atrophy with cognitive impairment, as has been widely reported previously, in contrast to the atrophy-independent effects of APOE on entorhinal microstructure observed among cognitively normal individuals. APOE4 carriers also exhibited abnormal cortical gray matter and thalamic microstructure with MCI, patterns that were not present among non-carriers. Reduced gray matter RI and HI, which may represent cell death or dystrophy, and increased free water in cognitively impaired APOE4 carriers are consistent with cortical atrophy that propagates throughout the neocortex with AD progression. The role of thalamic structure and function in AD appears complex and perhaps nonlinear, as thalamic atrophy in MCI [48] and hypometabolism in APOE4 carriers [49] have been observed, whereas others have reported increased thalamic volume [50] and hypermetabolism with amyloid pathology [51] among APOE4 carriers. Elevated thalamic RI and ND for cognitively impaired APOE4 carriers could reflect morphological changes occurring with gliosis including somatic swelling of activated microglia, increased density and length of processes of activated astrocytes, or proliferation of either glial cell type, consistent with the interpretation that increased thalamic volume reflects inflammatory signaling involving microglial activation [50]. APOE4 non-carriers demonstrated a distinct pattern of reduced striatal intracellular integrity (caudate RI) with cognitive impairment, which may point to AD-independent striatal contributions to clinical symptoms, as various conditions including Parkinson’s disease and psychiatric disorders are accompanied by striatal dysfunction [52, 53]. In contrast, age-related decline in striatal intracellular integrity (putamen RI) occurred only for APOE4 carriers. Together, these findings support a model in which APOE4 carriers are particularly susceptible to preclinical neurodegenerative changes in medial temporal regions that progress with cognitive decline, whereas among APOE4 non-carriers, microstructural changes in these regions are more closely tied to aging than cognitive impairment. Conversely, striatal intracellular injury may be a marker of APOE4-independent cognitive impairment that demonstrates more benign age-related changes in APOE4 carriers.

Diastolic blood pressure was elevated for male APOE4 carriers relative to non-carriers, consistent with a documented susceptibility of APOE4 carriers to cardiovascular risk factors [54], including higher diastolic blood pressure observed in some cohorts [55,56,57]. Concordantly, adjustment for diastolic blood pressure slightly attenuated the difference in entorhinal ND by APOE4 in the sex-combined sample, but minimally affected this difference within women, who showed no difference in blood pressure by APOE, and did not alter any other APOE4 effects. Though sex differences in vascular effects of APOE remain poorly understood, these preliminary findings suggest that associations of APOE4 with blood pressure may be stronger in men, in contrast to more pronounced associations between APOE and microstructural injury in women, such that blood pressure does not meaningfully account for the observed effects of APOE4. Nevertheless, considering the still elusive role of APOE-dependent vascular contributions to AD, further research is needed to better characterize the extent to which vascular dysfunction mediates effects of APOE4 on neurodegeneration, and how these pathways differ by sex.

A limitation of this study is lack of AD biomarker characterization on the full sample, precluding consideration of amyloid and tau in statistical analyses, which could aid in informing potential associations between AD neuropathology and APOE-dependent mechanisms underlying microstructural differences. Nevertheless, available biomarker data demonstrated more AD-like profiles (lower CSF amyloid and higher tau) for APOE4 carriers and CI participants, supporting the interpretation that these individuals were more likely to be in preclinical AD stages. Furthermore, the cross-sectional nature of this study precludes tracking patterns of change or nonlinear dynamics of diffusion markers over time. Because the sample almost entirely comprised non-Hispanic White participants and we were unable to account for racial/ethnic or ancestry differences, findings may not generalize to non-white participants, considering effects of APOE are not uniform across racial groups. Given the small number of APOE4 homozygotes and lack of APOE2 homozygotes, further investigation is needed to probe allele dose effects. Finally, it is important to highlight that RSI identifies diffusion patterns consistent with distinct tissue compartments, but cannot directly image cell morphometry or inform physiological mechanisms underlying microstructural properties. As for all biophysical models, accuracy of diffusion MRI measures relies on the validity of assumptions regarding cell architecture, as well as selection of protocol parameters, which may complicate interpretability.


Together, our findings indicate that microstructural signatures of APOE genotype emerge prior to cognitive impairment in patterns that may differ by sex, with entorhinal neurite loss as a potential preclinical AD biomarker, and WM microstructure associated with reduced AD risk. APOE genotype may shape regionally specific patterns of cellular changes occurring with aging and neurodegenerative disease, with effects of normal aging predominating in the absence of pathogenic drivers. Patterns of microstructural abnormalities associated with mild cognitive impairment differed according to APOE4, serving as potential differential indicators of underlying etiology.

Availability of data and materials

The datasets used in this study are available from the corresponding author on reasonable request. Data for the RBS are available at:





Alzheimer’s disease


Alzheimer’s disease research center


Apolipoprotein E


Body mass index


Cognitively impaired


Cognitively normal


Diffusion tensor imaging


False discovery rate


Hindered isotropic


Isotropic free water


Mild cognitive impairment


Neurite density


Rancho Bernardo Study


Restricted isotropic


Restriction spectrum imaging


White matter


  1. Farrer LA, Cupples LA, Haines JL, Hyman B, Kukull WA, Mayeux R, Myers RH, Pericak-Vance MA, Risch N, van Duijn CM. Effects of age, sex, and ethnicity on the association between apolipoprotein E genotype and Alzheimer disease. A meta-analysis. APOE and Alzheimer Disease Meta Analysis Consortium. JAMA. 1997;278(16):1349–56.

    Article  CAS  PubMed  Google Scholar 

  2. Raulin A-C, Doss SV, Trottier ZA, Ikezu TC, Bu G, Liu C-C. ApoE in Alzheimer’s disease: pathophysiology and therapeutic strategies. Mol Neurodegener. 2022;17(1):72.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  3. Adluru N, Destiche DJ, Lu SY-F, Doran ST, Birdsill AC, Melah KE, Okonkwo OC, Alexander AL, Dowling NM, Johnson SC, et al. White matter microstructure in late middle-age: effects of apolipoprotein E4 and parental family history of Alzheimer’s disease. NeuroImage Clin. 2014;4:730–42.

    Article  PubMed  PubMed Central  Google Scholar 

  4. Cavedo E, Lista S, Rojkova K, Chiesa PA, Houot M, Brueggen K, Blautzik J, Bokde ALW, Dubois B, Barkhof F, et al. Disrupted white matter structural networks in healthy older adult APOE ε4 carriers – An international multicenter DTI study. Neuroscience. 2017;357:119–33.

    Article  CAS  PubMed  Google Scholar 

  5. Kljajevic V, Meyer P, Holzmann C, Dyrba M, Kasper E, Bokde ALW, Fellgiebel A, Meindl T, Hampel H, Teipel S. The ε4 genotype of apolipoprotein E and white matter integrity in Alzheimer’s disease. Alzheimers Dement. 2014;10(3):401–4.

    Article  PubMed  Google Scholar 

  6. Nyberg L, Salami A. The APOE ε4 allele in relation to brain white-matter microstructure in adulthood and aging. Scand J Psychol. 2014;55(3):263–7.

    Article  PubMed  Google Scholar 

  7. Lacey C, Gawryluk J. White matter microstructure and cognitive functioning across healthy older adults with different APOE alleles. Alzheimers Dement. 2021;17(S1):e051190.

    Article  Google Scholar 

  8. Westlye LT, Reinvang I, Rootwelt H, Espeseth T. Effects of APOE on brain white matter microstructure in healthy adults. Neurology. 2012;79(19):1961–9.

    Article  CAS  PubMed  Google Scholar 

  9. Mole JP, Fasano F, Evans J, Sims R, Kidd E, Aggleton JP, Metzler-Baddeley C. APOE-ε4-related differences in left thalamic microstructure in cognitively healthy adults. Sci Rep. 2020;10(1):19787.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Nir TM, Zhu AH, Gari IB, Dixon D, Islam T, Villalon-Reina JE, Medland SE, Thompson PM, Jahanshad N. Effects of ApoE4 and ApoE2 genotypes on subcortical magnetic susceptibility and microstructure in 27,535 participants from the UK Biobank. In: Biocomputing. 2022. edn.: 121–132.

  11. Groot C. Multimodal imaging correlates of the protective APOE2 allele. Alzheimers Dement. 2020;16(S3):e039421.

    Article  Google Scholar 

  12. Grothe MJ, Salvadó G, Groot C, Moscoso A, Gispert JD, Schöll M, Ossenkoppele R. Differential effects of APOE2 and APOE4 alleles on PET-measured amyloid-β and tau deposition in older individuals without dementia. Alzheimers Dement. 2020;16(S4):e040440.

    Article  Google Scholar 

  13. Chiang GC, Insel PS, Tosun D, Schuff N, Truran-Sacrey D, Raptentsetsang ST, Jack CR Jr, Aisen PS, Petersen RC, Weiner MW, et al. Hippocampal atrophy rates and CSF biomarkers in elderly APOE 2 normal subjects. Neurology. 2010;75(22):1976–81.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Chiang GC, Zhan W, Schuff N, Weiner MW. White matter alterations in cognitively normal apoE epsilon2 carriers: insight into Alzheimer resistance? AJNR Am J Neuroradiol. 2012;33(7):1392–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Nir TM, Zhu AH, Gari IB, Dixon D, Islam T, Villalon-Reina JE, Medland SE, Thompson PM, Jahanshad N. Effects of ApoE4 and ApoE2 genotypes on subcortical magnetic susceptibility and microstructure in 27,535 participants from the UK Biobank. Pac Symp Biocomput. 2022;27:121–32.

    PubMed  PubMed Central  Google Scholar 

  16. Neu SC, Pa J, Kukull W, Beekly D, Kuzma A, Gangadharan P, Wang L-S, Romero K, Arneric SP, Redolfi A, et al. Apolipoprotein E genotype and sex risk factors for Alzheimer disease: a meta-analysis. JAMA Neurol. 2017;74(10):1178–89.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Genin E, Hannequin D, Wallon D, Sleegers K, Hiltunen M, Combarros O, Bullido MJ, Engelborghs S, De Deyn P, Berr C, et al. APOE and Alzheimer disease: a major gene with semi-dominant inheritance. Mol Psychiatry. 2011;16(9):903–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Operto G, Molinuevo JL, Cacciaglia R, Falcon C, Brugulat-Serrat A, Suárez-Calvet M, Grau-Rivera O, Bargalló N, Morán S, Esteller M, et al. Interactive effect of age and APOE-ε4 allele load on white matter myelin content in cognitively normal middle-aged subjects. NeuroImage Clin. 2019;24:101983.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Horgusluoglu-Moloch E, Wang M, Zhou X, Zeng Q, Zhang B. Sex and APOE genotype-specific white matter changes in Alzheimer’s disease. Alzheimers Dement. 2021;17(S1):e055755.

    Article  Google Scholar 

  20. Sampedro F, Vilaplana E, de Leon MJ, Alcolea D, Pegueroles J, Montal V, Carmona-Iragui M, Sala I, Sánchez-Saudinos MB, Antón-Aguirre S, et al. APOE-by-sex interactions on brain structure and metabolism in healthy elderly controls. Oncotarget. 2015;6(29):26663–74.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Hobel Z, Isenberg AL, Raghupathy D, Mack W, Pa J. for the Alzheimer’s Disease Neuroimaging I, the Australian Imaging B, Lifestyle flagship study of a: APOE ɛ4 gene dose and sex effects on Alzheimer’s disease MRI biomarkers in older adults with mild cognitive impairment. J Alzheimers Dis. 2019;71:647–58.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Ma C, Wang J, Zhang J, Chen K, Li X, Shu N, Chen Y, Liu Z, Zhang Z. Disrupted brain structural connectivity: pathological interactions between genetic APOE ε4 status and developed MCI condition. Mol Neurobiol. 2017;54(9):6999–7007.

    Article  CAS  PubMed  Google Scholar 

  23. Wang Z, Dai Z, Shu H, Liu D, Guo Q, He Y, Zhang Z. Cortical thickness and microstructural white matter changes detect amnestic mild cognitive impairment. J Alzheimers Dis. 2017;56:415–28.

    Article  PubMed  Google Scholar 

  24. White NS, Leergaard TB, D’Arceuil H, Bjaalie JG, Dale AM. Probing tissue microstructure with restriction spectrum imaging: histological and theoretical validation. Hum Brain Mapp. 2013;34(2):327–46.

    Article  PubMed  Google Scholar 

  25. Reas ET, Hagler DJ Jr, White NS, Kuperman JM, Bartsch H, Cross K, Loi RQ, Balachandra AR, Meloy MJ, Wierenga CE, et al. Sensitivity of restriction spectrum imaging to memory and neuropathology in Alzheimer’s disease. Alzheimers Res Ther. 2017;9(1):55.

    Article  PubMed  PubMed Central  Google Scholar 

  26. Reas ET, Hagler DJ Jr, White NS, Kuperman JM, Bartsch H, Wierenga CE, Galasko D, Brewer JB, Dale AM, McEvoy LK. Microstructural brain changes track cognitive decline in mild cognitive impairment. NeuroImage Clin. 2018;20:883–91.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Reas ET, Hagler DJ Jr, Andrews MJ, Lee RR, Dale AM, McEvoy LK. Associations between age and brain microstructure in older community-dwelling men and women: the Rancho Bernardo Study. Neurobiol Aging. 2020;95:94–103.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E. Mild cognitive impairment: clinical characterization and outcome. Arch Neurol. 1999;56(3):303–8.

    Article  CAS  PubMed  Google Scholar 

  29. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology. 1984;34(7):939–44.

    Article  CAS  PubMed  Google Scholar 

  30. McDowell I, Kristjansson B, Hill GB, Hébert R. Community screening for dementia: the Mini Mental State Exam (MMSE) and modified Mini-Mental State Exam (3MS) compared. J Clin Epidemiol. 1997;50(4):377–83.

    Article  CAS  PubMed  Google Scholar 

  31. Vanderstichele H, Bibl M, Engelborghs S, Le Bastard N, Lewczuk P, Molinuevo JL, Parnetti L, Perret-Liaudet A, Shaw LM, Teunissen C, et al. Standardization of preanalytical aspects of cerebrospinal fluid biomarker testing for Alzheimer’s disease diagnosis: a consensus paper from the Alzheimer’s Biomarkers Standardization Initiative. Alzheimers Dement. 2012;8(1):65–73.

    Article  CAS  PubMed  Google Scholar 

  32. Kaplow J, Vandijck M, Gray J, Kanekiyo M, Huyck E, Traynham CJ, Esquivel R, Fagan AM, Luthman J. Concordance of Lumipulse cerebrospinal fluid t-tau/Abeta42 ratio with amyloid PET status. Alzheimers Dement. 2020;16(1):144–52.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Hagler DJ Jr, Hatton S, Cornejo MD, Makowski C, Fair DA, Dick AS, Sutherland MT, Casey B, Barch DM, Harms MP. Image processing and analysis methods for the Adolescent Brain Cognitive Development Study. Neuroimage. 2019;202:116091.

    Article  CAS  PubMed  Google Scholar 

  34. Fischl B, Salat DH, Busa E, Albert M, Dieterich M, Haselgrove C, van der Kouwe A, Killiany R, Kennedy D, Klaveness S, et al. Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain. Neuron. 2002;33(3):341–55.

    Article  CAS  PubMed  Google Scholar 

  35. Elman JA, Panizzon MS, Hagler DJ, Fennema-Notestine C, Eyler LT, Gillespie NA, Neale MC, Lyons MJ, Franz CE, McEvoy LK, et al. Genetic and environmental influences on cortical mean diffusivity. Neuroimage. 2017;146:90–9.

    Article  PubMed  Google Scholar 

  36. Hagler DJ Jr, Ahmadi ME, Kuperman J, Holland D, McDonald CR, Halgren E, Dale AM. Automated white-matter tractography using a probabilistic diffusion tensor atlas: application to temporal lobe epilepsy. Hum Brain Mapp. 2009;30(5):1535–47.

    Article  PubMed  Google Scholar 

  37. Young CB, Johns E, Kennedy G, Belloy ME, Insel PS, Greicius MD, Sperling RA, Johnson KA, Poston KL, Mormino EC, et al. APOE effects on regional tau in preclinical Alzheimer’s disease. Mol Neurodegener. 2023;18(1):1.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Sperling RA, Mormino EC, Schultz AP, Betensky RA, Papp KV, Amariglio RE, Hanseeuw BJ, Buckley R, Chhatwal J, Hedden T, et al. The impact of amyloid-beta and tau on prospective cognitive decline in older individuals. Ann Neurol. 2019;85(2):181–93.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Miller MI, Younes L, Ratnanather JT, Brown T, Trinh H, Postell E, Lee DS, Wang M-C, Mori S, O’Brien R, et al. The diffeomorphometry of temporal lobe structures in preclinical Alzheimer’s disease. NeuroImage Clin. 2013;3:352–60.

    Article  PubMed  PubMed Central  Google Scholar 

  40. Shaw P, Lerch JP, Pruessner JC, Taylor KN, Rose AB, Greenstein D, Clasen L, Evans A, Rapoport JL, Giedd JN. Cortical morphology in children and adolescents with different apolipoprotein E gene polymorphisms: an observational study. Lancet Neurol. 2007;6(6):494–500.

    Article  CAS  PubMed  Google Scholar 

  41. Fjell AM, Walhovd KB, Fennema-Notestine C, McEvoy LK, Hagler DJ, Holland D, Brewer JB, Dale AM. One-year brain atrophy evident in healthy aging. J Neurosci. 2009;29(48):15223–31.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. He Y, Chen Z, Evans A. Structural insights into aberrant topological patterns of large-scale cortical networks in Alzheimer’s disease. J Neurosci. 2008;28(18):4756–66.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Drenthen GS, Backes WH, Freeze WM, Jacobs HIL, Verheggen ICM, van Boxtel MPJ, Hoff EI, Verhey FR, Jansen JFA. Rich-club connectivity of the structural covariance network relates to memory processes in mild cognitive impairment and Alzheimer’s disease. J Alzheimers Dis. 2022;89:209–17.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  44. Steward A, Biel D, Brendel M, Dewenter A, Roemer S, Rubinski A, Luan Y, Dichgans M, Ewers M, Franzmeier N, et al. Functional network segregation is associated with attenuated tau spreading in Alzheimer’s disease. Alzheimers Dement. 2023;19(5):2034–46.

    Article  PubMed  Google Scholar 

  45. Wang X, Zhou W, Ye T, Lin X, Zhang J. for Alzheimer’s Disease Neuroimaging I: sex difference in the association of APOE4 with memory decline in mild cognitive impairment. J Alzheimers Dis. 2019;69:1161–9.

    Article  CAS  PubMed  Google Scholar 

  46. Altmann A, Tian L, Henderson VW, Greicius MD. Alzheimer’s Disease Neuroimaging Initiative I: sex modifies the APOE-related risk of developing Alzheimer disease. Ann Neurol. 2014;75(4):563–73.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  47. Sexton CE, Kalu UG, Filippini N, Mackay CE, Ebmeier KP. A meta-analysis of diffusion tensor imaging in mild cognitive impairment and Alzheimer’s disease. Neurobiol Aging. 2011;32(12):2322 e2325-2318.

    Article  Google Scholar 

  48. Chételat G, Landeau B, Eustache F, Mézenge F, Viader F, de la Sayette V, Desgranges B, Baron JC. Using voxel-based morphometry to map the structural changes associated with rapid conversion in MCI: a longitudinal MRI study. Neuroimage. 2005;27(4):934–46.

    Article  PubMed  Google Scholar 

  49. Langbaum JBS, Chen K, Caselli RJ, Lee W, Reschke C, Bandy D, Alexander GE, Burns CM, Kaszniak AW, Reeder SA, et al. Hypometabolism in Alzheimer-affected brain regions in cognitively healthy Latino individuals carrying the apolipoprotein E ε4 allele. Arch Neurol. 2010;67(4):462–8.

    Article  PubMed  PubMed Central  Google Scholar 

  50. Cacciaglia R, Molinuevo JL, Falcón C, Brugulat-Serrat A, Sánchez-Benavides G, Gramunt N, Esteller M, Morán S, Minguillón C, Fauria K, et al. Effects of APOE-ε4 allele load on brain morphology in a cohort of middle-aged healthy individuals with enriched genetic risk for Alzheimer’s disease. Alzheimers Dement. 2018;14(7):902–12.

    Article  PubMed  Google Scholar 

  51. Johnson SC, Christian BT, Okonkwo OC, Oh JM, Harding S, Xu G, Hillmer AT, Wooten DW, Murali D, Barnhart TE, et al. Amyloid burden and neural function in people at risk for Alzheimer’s disease. Neurobiol Aging. 2014;35(3):576–84.

    Article  CAS  PubMed  Google Scholar 

  52. Zhai S, Shen W, Graves SM, Surmeier DJ. Dopaminergic modulation of striatal function and Parkinson’s disease. J Neural Transm. 2019;126(4):411–22.

    Article  CAS  PubMed  Google Scholar 

  53. Karcher NR, Rogers BP, Woodward ND. Functional connectivity of the striatum in Schizophrenia and psychotic bipolar disorder. Biol Psychiatry Cogn Neurosci Neuroimaging. 2019;4(11):956–65.

    PubMed  PubMed Central  Google Scholar 

  54. Song Y, Stampfer MJ, Liu S. Meta-analysis: apolipoprotein E genotypes and risk for coronary heart disease. Ann Intern Med. 2004;141(2):137–47.

    Article  PubMed  Google Scholar 

  55. Hestad KA, Engedal K, Whist JE, Farup PG. The effect of ApoE e4 on blood pressure in patients with and without depression. Neuropsychiatr Dis Treat. 2016;12:1365–70.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  56. Stewart KC, Subramanian D, Neal UJ, Hanson AJ. APOE genotype influences postprandial blood pressure after high fat feeding in older adults. J Nutr Health Aging. 2019;23(5):408–13.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  57. Sapkota B, Subramanian A, Priamvada G, Finely H, Blackett PR, Aston CE, Sanghera DK. Association of APOE polymorphisms with diabetes and cardiometabolic risk factors and the role of APOE genotypes in response to anti-diabetic therapy: results from the AIDHS/SDS on a South Asian population. J Diabetes Complications. 2015;29(8):1191–7.

    Article  PubMed  PubMed Central  Google Scholar 

Download references


Not applicable.


This work was supported by the National Institute on Aging (R01 AG062483, R00 AG057797), National Institute on Alcohol Abuse and Alcoholism (R01 AA021187), and National Institute on Drug Abuse (1U24DA041123-01).

Author information

Authors and Affiliations



ETR conceptualized the study design, conducted statistical analysis, and drafted the manuscript. CT conducted statistical analysis and revised the manuscript. SJB revised the manuscript. LKM conceptualized the study design and revised the manuscript.

Corresponding author

Correspondence to Emilie T. Reas.

Ethics declarations

Ethics approval and consent to participate

Study procedures were approved by the University of California, San Diego Human Research Protections Program Board and participants provided informed written consent prior to participation.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Table S1.

Description of restriction spectrum imaging metrics. Table S2. Participant characteristics (mean ± SD or N(%)) by APOE4 for the full sample and stratified by cognitive status. Table S3. Participant characteristics (mean ± SD or N(%)) by cognitive status. Table S4. Amyloid and tau measures (mean ± SD or N(%)) by APOE4 genotype and cognitive status for the subset of participants who underwent lumbar puncture. Table S5. Participant characteristics (mean ± SD or N(%)) by cohort and cognitive status. Table S6. Effect sizes (F-values) and p-values for differences in RSI metrics within all regions examined, by APOE4 among cognitively normal participants. Table S7. Effect sizes (F-values) and p-values for differences in RSI metrics within all regions examined, by APOE (APOE2/3 vs APOE3/3 vs APOE4-carrier) among cognitively normal participants. Table S8. Interaction between APOE4 and cognitive impairment on brain microstructure (mean ± SD, adjusted for age, sex, and scanner). Table S9. Correlations (Pearson’s r) between RSI metrics within selected regions of interest demonstrating significant microstructural differences by APOE. Figure S1. Sex-specific differences in brain microstructure by APOE in cognitively normal participants. Entorhinal cortex neurite density was lower for APOE4 carriers than for non-carriers among women only (A). Entorhinal cortex hindered isotropic diffusion (B) and cingulum neurite density (C) were lower for APOE2/3 than for APOE3/3 and APOE4 carriers among women only. Values are residuals, adjusted for age and scanner. Pairwise comparisons in B and C are Bonferroni corrected for multiple comparisons. Figure S2. Effects of cognitive status and APOE4, and their interaction, on brain microstructure. F-values are illustrated for main effects of cognitive status (A), APOE4 (B), and their interaction (C) on RSI metrics in entorhinal cortex, subcortical regions of interest, fiber tracts of interest, and global gray and white matter. Effect sizes for main effects are plotted such that positive values indicate means for cognitively normal (CN) > cognitively impaired (CI) and APOE4 non-carriers > carriers, whereas negative values indicate means for CI > CN and APOE4 carriers > non-carriers. Effects reaching significance after correction for multiple comparisons are highlighted with a black border.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Reas, E.T., Triebswetter, C., Banks, S.J. et al. Effects of APOE2 and APOE4 on brain microstructure in older adults: modification by age, sex, and cognitive status. Alz Res Therapy 16, 7 (2024).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: