Mean concentration of non–high-density lipoprotein (HDL) apoE and apoE in HDL containing or lacking apoC3 at screening visit by APOE genotype in 995 older individuals selected as a random subcohort of the Ginkgo Evaluation of Memory study.17 APOE indicates apolipoprotein E gene.
eTable 1. ApoE and ApoE Subspecies at Screening Visit by APOE Genotype in the Random Subcohort of the Ginkgo Evaluation of Memory Study (N=995)
eTable 2. Hazard Ratios (HRs) for Risk of Dementia or Difference in Cognitive Scores at Baseline and 95% Confidence Intervals (95% CI) per SD Higher Plasma Concentrations of ApoE and ApoE Subspecies at Screening Visit in the Ginkgo Evaluation of Memory Case-cohort, N=1351
eFigure 1. Flowchart Summarizing Case-cohort Sampling
eFigure 2. Measurement of Plasma Concentration of ApoE in Different Lipoprotein Subspecies by ELISA
eFigure 3. Average Trajectories of ADAS-cog Scores by Tertiles of Whole Plasma ApoE (A) and ApoE in HDL (B) at Screening Visit in the Ginkgo Evaluation of Memory Case-cohort
eFigure 4. HDL and its Apolipoprotein-defined Subspecies and Other Apolipoproteins at Study Entry (2000 to 2002) and Follow-up
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Koch M, DeKosky ST, Goodman M, et al. Association of Apolipoprotein E in Lipoprotein Subspecies With Risk of Dementia. JAMA Netw Open. 2020;3(7):e209250. doi:10.1001/jamanetworkopen.2020.9250
What is the association of apolipoprotein E (apoE) protein levels in different lipoproteins with cognitive function and risk of dementia?
In this case-cohort study including 1351 community-dwelling participants 74 years and older, the presence of apoE in high-density lipoproteins that lack apoC3 was associated with better cognitive function and decreased risk of dementia. In contrast, the presence of apoE in high-density lipoproteins that contain apoC3 was unrelated to cognitive function and risk of dementia.
The findings of this study extend the beneficial associations of the novel apoE–positive, apoC3–negative lipoprotein from cardiovascular disease to dementia.
The ε4 allele of the apolipoprotein E (APOE) gene and lower apolipoprotein E (apoE) protein levels in plasma are risk factors for Alzheimer disease, but the underlying biological mechanisms are not fully understood. Half of plasma apoE circulates on high-density lipoproteins (HDLs). Higher apoE levels in plasma HDL were previously found to be associated with lower coronary heart disease risk, but the coexistence of another apolipoprotein, apoC3, modified this lower risk.
To investigate associations between the presence of apoE in different lipoproteins with cognitive function, particularly the risk of dementia.
Design, Setting, and Participants
This prospective case-cohort study embedded in the Ginkgo Evaluation of Memory Study (2000-2008) analyzed data from 1351 community-dwelling participants 74 years and older. Of this group, 995 participants were free of dementia at baseline (recruited from September 2000 to June 2002) and 521 participants were diagnosed with incident dementia during follow-up until 2008. Data analysis was performed from January 2018 to December 2019.
Enzyme-linked immunosorbent assay–measured concentration of apoE in whole plasma, HDL-depleted plasma (non-HDL), HDL, and HDL subspecies that contain or lack apoC3 or apoJ.
Main Outcomes and Measures
Adjusted hazard ratios for risk of dementia and Alzheimer disease during follow-up and adjusted differences (β coefficients) in Alzheimer Disease Assessment–Cognitive Subscale (ADAS-cog) and Modified Mini-Mental State Examination scores at baseline.
Among 1351 participants, the median (interquartile range) age was 78 (76-81) years; 639 (47.3%) were women. The median (interquartile range) follow-up time was 5.9 (3.7-6.5) years. Higher whole plasma apoE levels and higher apoE levels in HDL were associated with better cognitive function assessed by ADAS-cog (whole plasma, β coefficient, −0.15; 95% CI, −0.24 to −0.06; HDL, β coefficient, −0.20; 95% CI, −0.30 to −0.10) but were unassociated with dementia or Alzheimer disease risk. When separated by apoC3, a higher apoE level in HDL that lacks apoC3 was associated with better cognitive function (ADAS-cog per SD: β coefficient, 0.17; 95% CI, −0.27 to −0.07; Modified Mini-Mental State Examination score per SD: β coefficient, 0.25; 95% CI, 0.07 to 0.42) and lower risk of dementia (hazard ratio per SD, 0.86; 95% CI, 0.76 to 0.99). In contrast, apoE levels in HDL that contains apoC3 were unassociated with any of these outcomes.
Conclusions and Relevance
In a prospective cohort of older adults with rigorous follow-up of dementia, the apoE level in HDL that lacked apoC3 was associated with better cognitive function and lower dementia risk. This finding suggests that the cardioprotective associations of this novel lipoprotein extend to dementia.
The ε4 allele of the apolipoprotein E (APOE) gene is the most important known single genetic risk factor for late-onset Alzheimer disease (AD),1,2 although the mechanism by which it increases AD risk remains incompletely understood. The APOE genotype influences plasma apolipoprotein E (apoE) concentrations. Plasma concentrations of apoE themselves are associated with lower risk of dementia and AD, even after accounting for the APOE genotype.3
Although apoE is a multifunctional protein, research on apoE has focused primarily on its role in lipid binding, transport, and metabolism. It has a major effect on the clearance of plasma lipids by mediating the binding of apoE–containing lipoproteins and lipid complexes to cell-surface lipoprotein receptors.4-6 In plasma, most apoE is derived from the liver and a constituent of diverse lipoproteins, including very low-density lipoproteins, intermediate-density lipoproteins, chylomicrons, and high-density lipoproteins (HDLs). Given the distinct metabolic roles of these lipoproteins, the association of apoE with dementia might depend on the lipoprotein where it is located. Thus far, investigations of total apoE levels in relation to dementia risk have not accounted for the distribution of apoE among a heterogeneous mix of lipoproteins.
In individuals with normal lipid levels, approximately half of total apoE is found on HDL.7,8 Beyond apoE, HDL particles also contain dozens of other proteins.9 Thus, HDL can be separated into subfractions based on its protein components.10 Accumulating evidence suggests that these subfractions are differentially associated with coronary heart disease (CHD)11 and diabetes,12 2 risk factors of dementia.13,14 For instance, among generally healthy middle-aged adults free of CHD, higher apoE in HDL is inversely associated with the risk of acute CHD but only in the absence of apoC3 on HDL particles.15 These findings support the hypothesis that the properties of plasma lipoproteins and their role in health and disease depend on apolipoprotein composition. Another apolipoprotein found on HDL, apoJ, has been implicated in the pathophysiology of dementia.16 The presence of specific apolipoproteins may identify apoE subspecies that are more or less involved in certain disease processes. To address the association of apoE (overall and its subspecies in the fraction of plasma without HDL [non-HDL], in HDL, and in HDL that contains or lacks apoC3 or apoJ) with cognitive function, incidence of dementia, and AD, this study measured the apoE concentration in these lipoprotein subspecies in a prospective, well-phenotyped population of older adults. We hypothesized that higher apoE levels in HDL are inversely associated with risk of dementia but only in the absence of apoC3.
The Ginkgo Evaluation of Memory Study (GEMS)17 enrolled 3069 community-dwelling participants 74 years and older with normal cognition or mild cognitive impairment recruited from October 2000 to May 2002 at 4 field centers in the US. In the trial,18 an intervention of 240 mg of Ginkgo biloba daily was found to be ineffective in reducing the incidence of all-cause dementia, but the trial provided an extraordinary resource for secondary analyses because of its dedication of resources to neurologist-adjudicated risk of all-cause dementia and AD.19 For the present analysis, we used the case-cohort design, as described Koch et al.20 From the 3069 GEMS participants, we included a random subcohort of 1000 participants (32.6%) free of dementia at baseline and an additional 523 (17.0%) participants diagnosed with dementia during follow-up, of whom 166 overlapped (a feature of the case-cohort design) (eFigure 1 in the Supplement). Thus, a total of 1357 participants (44.2% of all GEMS participants) were included in the case-cohort. We excluded 2 participants with no available plasma samples and 4 participants with missing apoE measurements, leaving 1351 participants in the final analysis. Institutional review boards at each investigational center approved the study, and participants and their proxies provided written informed consent. The study meets the criteria for exemption by the Harvard T.H. Chan School of Public Health Office of Human Research Administration per the regulations found at 45 CFR 46.104(d) (4). This study follows the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
The nutritional biomarker laboratory at the Harvard T.H. Chan School of Public Health, Boston, Massachusetts, measured the levels of plasma apoE concentrations in different lipoproteins (eFigure 2 in the Supplement) in all participants of this case-cohort study at study entry between September 2000 and June 2002 and for a subset of participants during follow-up (806 participants at 3-year follow-up, 103 participants at 4-year follow-up). To minimize between-plate variation, plasma samples collected from the subset of participants at baseline and follow-up were incubated separately but on the same plates. A total of 307 of the 1351 case-cohort participants (22.7%) self-reported fasting for at least 4 hours at the time of blood sample collection.
We first quantified the concentrations of apoE in whole plasma using a sandwich enzyme-linked immunosorbent assay (ELISA) (Academy Bio-Medical Co). Then, apo B–containing lipoproteins were removed through precipitation using magnesium chloride and dextran sulfate. The concentration of apoE was measured in the remaining fraction (ie, HDL). The concentration of apoE in HDL was subtracted from the concentration of apoE in whole plasma to obtain the concentration of apoE in non-HDL. The measurement of the level of apoE in HDL that contained or lacked apoC3 used a patented and modified sandwich ELISA-based procedure.9,21 Antibodies specific for apoC3 (Academy Bio-Medical Co) were precoated onto ELISA plates, and diluted HDL samples were added. After incubation, plates were washed and a tween-containing diluent (1× phosphate-buffered saline with 2% and bovine serum albumin with 0.05% Tween 20) was added to release the lipoproteins that bound to the anti–apoC3 antibodies. Next, these lipoproteins were transferred to a plate precoated with anti–apoE antibodies to quantify the concentration of apoE in HDL that contained apoC3. The concentration of apoE in HDL that contained apoC3 was subtracted from the concentration of apoE in HDL to calculate the concentration of apoE in HDL lacking apoC3. The concentration of apoE in HDL containing or lacking apoJ was measured using the same protocol using apoJ antibodies (R&D Systems). Enzymatic assays were used to measure plasma triglycerides (Thermo Fisher Scientific).
From 2000 to 2008, participants were assessed semiannually for incident dementia using criteria of the Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition). To qualify for study entry, participants underwent a detailed neuropsychological battery of 10 tests capturing the cognitive domains of construction, memory, language, executive function, attention and psychomotor speed, and premorbid intellectual functioning. Participants completed the Clinical Dementia Rating scale and the Modified Mini-Mental State Examination (3MSE) semiannually. From September 2000 to August 2004, the cognitive subscale of the Alzheimer Disease Assessment Scale (ADAS-cog) was administered semiannually, and the results of the 3 cognitive tests were used to mandate readministration of the neuropsychological battery when the scores declined by a prespecified number.19 From August 2004 to April 2008, the ADAS-cog and the neuropsychological battery were both administered annually.22
Participants considered to potentially have cognitive impairment were referred for a full neurological evaluation and brain magnetic resonance imaging. After this evaluation, a panel reviewed these data and applied a validated protocol for dementia diagnosis (using criteria from the National Institute of Neurological Disorders and Stroke–Association Internationale pour la Recherche et l’Enseignement en Neurosciences, the National Institute of Neurological and Communication Disorders and Stroke, Alzheimer Disease and Related Disorders Association, and the Alzheimer Disease Diagnostic and Treatment Centers).23-25 Using these criteria, the panel assigned participants with dementia to 1 of the following dementia subtypes: vascular dementia, AD, mixed dementia, or other dementia.19 Based on the criteria from the International Working Group on Mild Cognitive Impairment,26 participants were considered to have mild cognitive impairment if participants scored at or below the 10th percentile for education-adjusted and age-adjusted norms on 2 or more of 10 selected neuropsychological test scores from each cognitive domain (using the Cardiovascular Health Study population as the norm) and if participants had a score of 0.5 on the Clinical Dementia Rating scale.27
At study entry, trained technicians collected data on age, sex, educational attainment, smoking status, medical history, and race/ethnicity in interviews, and measured blood pressure, height, and weight. Participants brought prescription drugs and over-the-counter medications to the study visit for entry into the database. Participants were screened for depression using the Center for Epidemiologic Studies-Depression scale. One missing value on this scale was replaced with the population’s median value of 3.
We recalibrated apoE measures using procedures proposed by Rosner et al28 to account for batch-to-batch variation. Among participants with repeated apoE measurements, we assessed the correlation of apoE measured at baseline and follow-up, controlling for sex and age at study entry. The Wilcoxon signed rank test was used to test for equality of apoE and apoE subspecies assessed at study entry and follow-up. To account for the oversampling of participants with dementia in association analyses, participants without dementia were assigned a weight inversely proportional to the sampling probability (3069/1000).
For incident dementia (the primary outcome) and AD, we fit inverse-sampling probability-weighted Cox proportional hazards models with robust estimate of variance, with study time as the underlying time axis censoring at death, drop-out, or dementia diagnosis, whichever occurred first. We tested the proportional hazards assumption based on Schoenfeld residuals.
For cognitive test scores, inverse-sampling probability-weighted linear regression models were used to assess the association of apolipoproteins with the ADAS-cog and 3MSE scores. Because ADAS-cog scores decreased (ie, improved, suggesting learning effects) during the first half of follow-up and showed a nonlinear increase thereafter (eFigure 3 in the Supplement) and because cognitive testing was discontinued in participants diagnosed with dementia during follow-up, we restricted analyses of cognitive scores to baseline values.
The concentration of apoE in whole plasma, HDL-depleted plasma (non-HDL), HDL, and HDL subspecies that contain or lack apoC3 or apoJ were modeled continuously (per 1 standard deviation increment). We adjusted for age, sex, race/ethnicity (as white or nonwhite), clinic site, fasting status (ie, <4 hours or ≥4 hours), education, the weekly number of alcoholic drinks (none, 0.1-0.9, 1.0-7.0, 7.1-14.0, >14, or missing), smoking status (never, former, current, or missing), body mass index (calculated as weight in kilograms divided by height in meters squared; <20, 20-24.9, 25-29.9, ≥30, or missing), lipid-lowering medication use, history of cardiovascular disease, history of diabetes, Center for Epidemiologic Studies-Depression score, treatment assignment from the original trial (placebo or Ginkgo biloba), and APO E ε4 carrier status (carrier, noncarrier, or missing). Complementary apoE subspecies (eg, apoE in HDL that contained or lacked apoC3) were modeled simultaneously, and likelihood ratio tests were used to assess heterogeneity of the slopes. We conducted sensitivity analyses without adjustment for APO E ε4 status and in a subset after exclusion of participants carrying at least 1 APO E ε4 allele. Furthermore, in sensitivity analyses, we evaluated the association of the concentration of apoE in HDL that contained or lacked apoC3 with dementia risk after excluding the first 2 years of follow-up. Analyses were performed from January 2018 to December 2019, and used Stata version 12.1 (Stata Corp). Statistical significance was set at 2-sided P < .05.
Among 1351 participants, the median (interquartile range) age was 78 (76-81) years; 639 (47.3%) were women; and 521 participants (38.6%) were diagnosed with dementia, including 352 participants (26.1%) with AD. The median (interquartile range) follow-up time was 5.9 (3.7-6.5) years. Compared with subcohort members, more participants who subsequently developed dementia during follow-up had MCI and carried the APOE ε4 allele (dementia cases during follow-up: 197 of 521 participants [38%]; random subcohort: 154 of 995 participants [15.5%]) (Table 1). ApoE and apoE subspecies at baseline and follow-up were highly correlated (all r ≥ 0.55) (eFigure 4 in the Supplement). The concentration of apoE in whole plasma and subspecies of apoE in different lipoproteins (apoE concentration in the non-HDL fraction of plasma, in HDL, and in HDL that contains or lacks apoC3 or apoJ) by APO E genotype is shown in eTable 1 in the Supplement. There were inverse associations of both apoE in whole plasma and apoE in HDL with APOE genotype. Compared with APOE ε2/2 and APO E ε2/3 carriers, the remaining genotypes had a lower proportion of their total apoE in the HDL fraction of plasma (eg, APOE ε2/2, 61% [5th-95th percentile, 35%-95%]; APOE ε2/3, 62% [40%-92%]; APOE ε3/3, 55% [35%-92%]). Across all genotypes, the HDL fractions containing or lacking apoC3 each contained similar amounts of apoE (Figure). For example, among APOE ε2/3 carriers, the concentration of apoE in HDL that contains apoC3 was 5.0 mg/dL (to convert measures apolipoprotein concentration to g/L, multiply mg/dL by 0.01) and the concentration of apoE in HDL that lacks apoC3 was 4.9 mg/dL.
Higher apoE levels in whole plasma, in non-HDL, and in HDL were not significantly associated with risk of dementia or AD in fully adjusted models (Table 2; eTable 2 in the Supplement). We assessed associations of lipoprotein species with cognition at baseline (Table 2). Higher apoE in whole plasma and higher apoE levels in HDL (but not in non-HDL) were associated with lower ADAS-cog scores at baseline, indicating better cognitive function (whole plasma, β coefficient, −0.15; 95% CI, −0.24 to −0.06; HDL, β coefficient, −0.20; 95% CI, −0.30 to −0.10). The concentration of apoE in whole plasma, non-HDL, or HDL were not statistically associated with 3MSE scores.
In contrast with total apoE, a higher apoE level in HDL that lacked apoC3 was significantly associated with lower dementia risk (Table 3). The hazard ratios (HRs) for apoE level in HDL that contained or lacked apoC3 with dementia risk were statistically significantly different (HR per 1-SD higher apoE in HDL that contained apoC3: 1.07 [95% CI, 0.95-1.19] vs 0.86 [95% CI, 0.76-0.99]; P for heterogeneity = .03). ApoE subspecies had similar associations with AD as with all-cause dementia. In sensitivity analyses excluding the first 2 years of follow-up, the results were not materially different. The HR for dementia per 1-SD higher concentration of apoE in HDL that contained apoC3 was 1.11 (95% CI, 0.98-1.25) and was 0.87 (95% CI, 0.76-1.00) for apoE in HDL that lacked apoC3.
A higher apoE level in HDL that lacked apoC3 was also associated with lower ADAS-cog scores (ie, better cognitive function) at baseline, while a higher apoE level in HDL that contained apoC3 was not associated with ADAS-cog scores (difference in ADAS-cog score per 1-SD higher apoE in HDL that contains apoC3: −0.08 [95% CI, −0.17 to 0.01]; difference in ADAS-cog score per 1-SD higher apoE in HDL that lacks apoC3: −0.17 [95% CI, −0.27 to −0.07]) (Table 3). Furthermore, the concentration of apoE in HDL containing or lacking apoJ was differentially associated with ADAS-cog scores (difference in ADAS-cog score per 1-SD higher apoE in HDL that contains apoJ, 0.05 [95% CI, −0.06 to −0.16]; difference in ADAS-cog score per 1-SD higher apoE in HDL that lacks apoJ, −0.24 [95% CI, −0.36 to −0.12]; P for heterogeneity = .005) but not with dementia (Table 4). Consistent with the ADAS-cog results, apoE levels in HDL containing or lacking apoC3 as well as apoE levels in HDL containing or lacking apoJ were differentially associated with 3MSE scores. A higher apoE level in HDL that lacked apoC3 was associated with higher 3MSE scores (ie, better cognitive function) at baseline (difference in 3MSE score per 1-SD higher apoE in HDL that contains apoC3, −0.12 [95% CI, −0.28 to 0.04]; difference in 3MSE score per 1-SD higher apoE in HDL that lacks apoC3, 0.25 [95% CI, 0.07 to 0.42]; P for heterogeneity = .005), while a higher apoE level in HDL that contained apoC3 was unassociated with 3MSE scores. Higher apoE levels in HDL containing apoJ were associated with lower 3MSE scores (difference in 3MSE score per 1-SD higher apoE in HDL that contains apoJ, −0.22 [95% CI, −0.41 to −0.02]; P for heterogeneity = .01). In contrast, higher apoE levels in HDL lacking apoJ were associated with higher 3MSE scores (difference in 3MSE score per 1-SD higher apoE in HDL that lacks apoJ, 0.24 [95% CI, 0.03 to 0.44]).
In this study analyzing the association of apoE level in HDL with cognition and dementia, substantial differences existed based on the presence or absence of other apolipoproteins in the HDL fractions. Specifically, the presence of apoC3 in HDL appeared to modulate the association of apoE in HDL with risk of dementia and AD. While the concentration of apoE in HDL that contained apoC3 was not associated with either end point, higher concentrations of apoE in HDL that lacked apoC3 were associated with a lower risk of both total dementia and AD.
Large-scale studies without information on lipoprotein subspecies have demonstrated that higher plasma apoE is associated with less cognitive decline and a lower risk of subsequent AD and other dementias.29,30 Consistent with these studies, we found an inverse association between apoE in HDL and dementia and AD risk in basic adjusted models. However, the association did not persist after adjustment for lifestyle factors and clinical characteristics. It is possible that apoE levels are less strongly associated with dementia risk in older populations like that studied in the GEMS.17 On the other hand, the association observed in other studies may reflect an unrecognized association restricted to apoE in HDL without apoC3.
Although apoE is expressed in the brain, the major source of apoE in plasma is the liver, and plasma and brain pools are distinct31 but correlated in some but not all studies.32,33 So far, most studies have focused on neuropathological effects of the apoE ε4 isoform within the central nervous system, while mechanisms by which circulatory apoE might contribute to cognitive decline and onset of AD and AD pathology are poorly understood.34 The observation that apoE deficiency in mice is associated with blood–brain barrier dysfunction has led to the hypothesis that apoE is involved in cerebrovascular integrity.35
With regard to apoE subspecies, in the absence of apoC3, higher apoE in the HDL fraction of plasma was associated with better cognitive function and a lower risk of dementia and AD. The latter findings parallel observations from the Danish Diet, Cancer, and Health study,15 in which higher apoE in plasma HDL was associated with lower CHD risk,15 but as this study found, only in the absence of apoC3. Lower CHD risk is itself a risk factor for dementia. Although it is not known if apoE in HDL without apoC3 is causally associated with lower risk of dementia or cardiovascular disease, metabolic studies have shown that while HDL that contains apoE is cleared from the circulation more than 7-fold faster than HDL that lacks apoE, the clearance of HDL is attenuated if apoE coexists with apoC3.15 Thus, at least for clearance, apoC3 effectively counteracts the effect of apoE—similar to what we have now observed for both CHD and dementia.
Despite the strong genetic association of APOJ with dementia, population-based studies linking plasma apoJ levels to AD risk have provided inconsistent results.36,37 The high affinity of apoJ to bind amyloid β in cerebrospinal fluid has led to suggestions that apoJ has a role in AD pathology.38 In vitro studies demonstrate that amyloid β–apoJ complexes bind the low-density lipoprotein receptor–related protein 2, also called megalin, resulting in amyloid β clearance from the brain.39 Although apoE can also bind to megalin, the functional role of this interaction is incompletely understood.40 The A variant of the megalin promotor polymorphism rs3755166 leads to lower transcript levels of the megalin protein and has been associated with higher risk of AD and greater cognitive decline in prior studies.41-43
Studies addressing potential interactions of plasma apolipoproteins on HDL are sparse. Apo A1 is the main structural apolipoprotein found in all circulatory HDL. In 1 of our previous investigations among GEMS participants,21 the amount of apoA1 in HDL was not associated with cognitive function or risk of dementia, regardless of the presence of apoC3, apoJ, or apoE in HDL. However, in a substudy of GEMS participants who remained free of clinical dementia, the amount of apoE in subspecies of HDL that contained or lacked apoC3 or apoJ was uniformly inversely associated with amyloid β deposition on positron emission tomographic scans.34 Thus, the amount of apoE, rather than the amount of apoA1, might be particularly strongly associated with dementia risk. Further studies are needed to clarify if the amount of apoE in HDL containing or lacking apoC3 or apoJ is associated with dementia via brain-specific or systemic pathways.
Important limitations warrant mention. Our study population was limited to older participants, and the underlying biological processes leading to incident dementia during follow-up were presumably already underway, thus making generalization to younger populations difficult. Cognitive testing was systematic but was truncated when dementia (the primary end point of GEMS17) occurred because it served primarily as a method to trigger formal dementia ascertainment and adjudication. Mild cognitive impairment was not a trial outcome and could not be classified retrospectively because, up until 2004, the full neuropsychological battery was administered only to participants with scores of the brief battery declining by a prespecified number.
This study also had strengths. To our knowledge, this study is the first examination of novel apoE subspecies. Other major strengths of this study are the detailed, frequent, and prospective evaluation of neurological status, the substantial number of participants with dementia, and the comprehensive data available for adjustment, including by APOE genotype.
In this study, higher apoE levels in HDL lacking apoC3 in an older population were associated with better cognitive function and a lower risk of dementia or AD. In contrast, apoE in HDL that contained apoC3 was unassociated with cognitive function and risk of dementia or AD. The study suggests that previously identified cardioprotective associations of this novel apolipoprotein extend to dementia and highlights the need for studies that assess the relevance of apoE in HDL that lacks apoC3 as new targets for modifiable behavioral or therapeutic interventions and risk stratification.
Accepted for Publication: April 22, 2020.
Published: July 10, 2020. doi:10.1001/jamanetworkopen.2020.9250
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Koch M et al. JAMA Network Open.
Corresponding Author: Manja Koch, PhD, Department of Nutrition, Harvard T.H. Chan School of Public Health, 655 Huntington Ave, Boston, MA 02115 (firstname.lastname@example.org).
Author Contributions: Dr Koch had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Koch, Goodman, Jensen.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Koch.
Critical revision of the manuscript for important intellectual content: DeKosky, Goodman, Sun, Furtado, Fitzpatrick, Mackey, Cai, Lopez, Kuller, Mukamal, Jensen.
Statistical analysis: Koch, Goodman, Sun, Cai, Kuller.
Obtained funding: DeKosky, Fitzpatrick, Lopez, Jensen.
Administrative, technical, or material support: Furtado, Fitzpatrick, Mukamal.
Supervision: DeKosky, Cai, Kuller, Mukamal, Jensen.
Conflict of Interest Disclosures: Dr DeKosky reported receiving research funding from the National Institutes of Health during the conduct of this study. Dr Furtado reported having patent No. US009494606B2 issued. Dr Fitzpatrick reported receiving grants from the University of Washington during the conduct of this study. Dr Lopez reported receiving a grant from the National Institute on Aging outside the submitted work. Dr Jensen reported receiving grants from the National Institute of Neurological Disorders and Stroke during the conduct of the study, and reported having patent No. US88463212B issued. No other disclosures were reported.
Funding/Support: This study was supported by grant U01AT000162 from the National Center for Complementary and Alternative Medicine; grant R01NS089638 from the National Institute on Neurological Disorders and Stroke; the Office of Dietary Supplements of the National Institute on Aging; the National Heart, Lung, and Blood Institute; grant P50AG05133 from University of Pittsburgh Alzheimer’s Disease Research Center; the Roena Kulynych Center for Memory and Cognition Research; and Wake Forest University School of Medicine. Plasma samples from the National Cell Repository for Alzheimer’s Disease, which receives government support under cooperative agreement grant U24 AG21886 awarded by the National Institute on Aging, were used in this study.
Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Additional Contributions: We thank the study participants and their families, whose help and participation made this work possible.