Journal of Atherosclerosis and Thrombosis
Online ISSN : 1880-3873
Print ISSN : 1340-3478
ISSN-L : 1340-3478
Original Article
Apolipoprotein E-Containing High-Density Lipoprotein is Independently Associated with Atherosclerotic Plaque Progression
Pinfei Ni, Jiangtao Li, Youling Duan, Piaopiao Hu, Qiuju Deng, Yongchen Hao, Zhao Yang, Lizhen Han, Yue Qi, Jing Liu
著者情報
キーワード: Apolipoprotein E, High-density lipoprotein, Carotid atherosclerosis, Cohort study
ジャーナル オープンアクセス HTML

2026 年 33 巻 1 号 p. 40-54

Browse “Advance online publication” version
詳細
PDFをダウンロード (977K)
メタデータをダウンロード RIS形式

(EndNote、Reference Manager、ProCite、RefWorksとの互換性あり)

BIB TEX形式

(BibDesk、LaTeXとの互換性あり)

テキスト
メタデータのダウンロード方法
発行機関連絡先
Abstract

Aim: Mounting evidence suggests apolipoprotein E-containing high-density lipoprotein cholesterol (APOE-HDLC) as an indicator of the anti-atherogenic function of HDLC, but data are lacking on whether or not APOE-HDLC is involved in the development of atherosclerosis in humans. This study was performed to explore whether or not APOE-HDLC is associated with atherosclerotic plaque progression in humans.

Methods: Among 823 participants 45 to 74 years old who were free of cardiovascular disease, we assessed nuclear magnetic resonance spectroscopy-measured HDL particle concentrations, APOE-HDLC levels and HDLC levels at baseline, and performed carotid ultrasound measurements in surveys conducted in 2002 and again in 2007 after a 5-year interval. The ratio of APOE-HDLC to total HDLC (APOE-HDLC/HDLC ratio) was calculated to assess the relative proportion of APOE-HDLC in total HDLC, given the strong correlation between them.

Results: The baseline APOE-HDLC/HDLC ratio was significantly associated with the risk of 5-year plaque progression (relative risk [RR] = 0.71; 95% confidence interval [CI] = 0.53-0.95), which is independent of the ratio of HDLC to the HDL particle number (HDLC/P ratio). In particular, participants with an HDLC/P ratio ≥ 44.8 (denoted very high level of cholesterol content per HDLP, a marker of dysfunctional HDL) had a 36% reduced 5-year plaque progression risk (RR = 0.64; 95% CI = 0.43-0.97) if combined with the highest APOE-HDLC/HDLC ratio, as compared with the lowest APOE-HDLC/HDLC ratio.

Conclusions: These results highlight the potential utility of APOE-containing HDL as a candidate emerging biomarker for the anti-atherosclerotic function of HDL particles.

Pinfei Ni and Jiangtao Li contributed equally to this manuscript.

See editorial vol. 33: 24-25

Abbreviations: APOE-HDLC, APOE-containing HDL cholesterol; CETP, cholesteryl ester transfer protein; HDLC, HDL cholesterol; HDLP, HDL particle; HDLC/P ratio, ratio of HDL cholesterol to HDL particle number; hs-CRP, high-sensitivity C-reactive protein; RR, relative risk

Introduction

The Framingham Heart Study in 1986, as well as a large body of subsequent epidemiologic evidence, demonstrated an inverse relationship between high-density lipoprotein cholesterol (HDLC) and cardiovascular disease (CVD) events1), yet this has not translated into benefits in reducing cardiovascular risk as a result of raising HDLC levels, partly because HDLC does not reflect the HDL function2-7). Thus, a recent major shift in HDL research has been toward identifying new parameters beyond HDLC as potential therapeutic targets8).

As the most heterogeneous particle in lipoproteins, HDL carries a variety of proteins associated with a range of biological functions9) including reverse cholesterol transport, lipid metabolism, and other mechanisms. Apolipoprotein E (APOE), an important mediator of lipoprotein metabolism in circulation, has been reported to play a pivotal role in atherosclerosis10). In our previous community-based cohort study, we reported that elevated levels of APOE-containing HDLC (APOE-HDLC) were associated with a decreased risk of 10-year coronary heart disease incidence11) and even attenuated the impact of extreme high level of cholesterol content per HDL particle, which was associated with an increased risk of atherosclerosis progression. Although the underlying mechanisms driving APOE-containing HDL-mediated cardioprotective effects remain uncertain, previous experimental studies have shown that APOE-containing HDL might have the potential to regulate reverse cholesterol transport as well as anti-inflammation and antioxidant capacity, which are key pathophysiological processes in atherosclerosis 12, 13). A recent study showed an association between APOE-HDLC levels and the presence of non-calcified plaque in patients with coronary artery disease14). However, data on whether or not circulating APOE-HDLC levels influence the development of atherosclerosis are scarce.

Aim

In the present study, we measured levels of APOE-HDLC and calculated the ratio of HDLC to HDL particle number (HDLC/P ratio) to estimate the cholesterol content per HDL particle in a CVD-free population. A high HDLC/P ratio is indicative of a very high level of cholesterol content per HDLP. Therefore, we tested the hypothesis that APOE-HDLC was associated with the progression of atherosclerotic plaque in a community-based cohort study and whether or not the atherogenic impact of a very high level of cholesterol content per HDLP can be attenuated.

Methods

Study Participants

Study participants were recruited from the Chinese Multi-provincial Cohort Study - Beijing Project, a prospective population-based cohort study focused on investigating the progression and determinants of carotid atherosclerosis15, 16). A total of 1324 participants aged 45 to 74 years old completed examinations involving demographic characteristics, traditional cardiovascular risk factors, and carotid ultrasound examination at baseline in 2002. After excluding participants with clinical CVD and those without blood samples at baseline, 1117 participants were invited for a re-examination of risk factors and carotid ultrasound measurements in 2007. We excluded participants for the following reasons: death (n = 16), lost to follow-up (n = 58), and no re-examination of carotid ultrasound (n = 220), leaving 823 participants with complete data who were eligible for the current study (Supplemental Fig.1).

Supplemental Fig.1.

Flowchart for selection of study participants

Written informed consent was obtained from all participants. The protocol was approved by the Ethics Committee of Beijing Anzhen Hospital, Capital Medical University. This study abides by the Declaration of Helsinki principles.

Risk Factor Survey

This study complied with the protocol of the World Health Organization-Monitoring of Trends and Determinants in Cardiovascular Disease. Demographic characteristics, personal medical history, and medical therapy information were collected using a standardized questionnaire. Height, weight, and blood pressure were measured during physical examinations. The body mass index was calculated as weight in kilograms divided by height in meters squared. Hypertension was defined as a systolic blood pressure ≥ 140 mmHg and/or diastolic blood pressure ≥ 90 mmHg, or the use of anti-hypertensive medications during the past 2 weeks. Type 2 diabetes was considered with fasting blood glucose levels ≥ 7.0 mmol/L or with a clinical diagnosis. Details of survey methods and definitions of risk factors have been described previously15).

Laboratory Measurements

Peripheral blood from enrolled participants was collected after an overnight fast for laboratory measurements. Total cholesterol, low-density lipoprotein cholesterol, HDLC, fasting blood glucose, triglyceride, and high-sensitivity C-reactive protein (hs-CRP) were measured using fresh samples taken on the same day of collection in 2002, according to previous reports11, 17). The remaining samples were stored at −80℃ without being exposed to repeated freeze-thaw cycles to minimize any degradation. Total cholesterol, triglyceride, and fasting blood glucose levels were determined using enzymatic methods (Human Diagnostics, Wiesbaden, Germany). Low-density lipoprotein cholesterol and HDLC levels were measured by a homogeneous assay (Daiichi, Tokyo, Japan). In brief, low-density lipoprotein cholesterol was assayed by a two-reagent system containing two specific surfactants. HDLC was measured using a synthetic polymer together with a polyanion to block the non-HDL lipoproteins, and a detergent then exposed only cholesterol in HDL to the enzymes, giving specificity for HDLC18). Plasma lipoprotein particle numbers were ascertained in 2013 using the nuclear magnetic resonance (NMR) spectroscopy technique provided by LipoScience, Inc. (Raleigh, NC, USA), following a previously outlined protocol19). Samples destined for lipoprotein particle evaluation via proton NMR spectroscopy were thawed, separated into 500-mL aliquots, re-frozen, and then transported on dry ice to LipoScience. This technique relies on the distinctive signals emitted by lipoprotein subfractions of varying sizes as the fundamental basis for quantification.

Cholesterol carried by APOE-containing HDL was measured using an automated homogenous assay (Denka Co., Ltd. Tokyo, Japan) as described previously20), and analyzed on a Hitachi 7180 automatic analyzer in 2015. In brief, this method consisted of a primary reaction to remove non-HDLC, a secondary reaction to measure APOE-deficient HDLC, and a tertiary reaction to measure APOE-containing HDLC (Denka Seiken Co. Ltd., Tokyo, Japan). The coefficient of variation (CV) was 1.33% for low-range controls and 1.53% for high-range controls. Apolipoprotein A-I and APOE concentrations were both determined using an immunoturbidimetric method (Sekisui Medical Co, Ltd, Tokyo, Japan) in 2015. With regard to apolipoprotein A-I, the CV was 2.08% for low-range controls and 1.58% for high-range controls. For APOE, the CV was 2.94% for low-range controls and 3.23% for high-range controls.

Carotid Atherosclerosis Examinations

A standard measurement protocol was used for carotid B-mode ultrasound at baseline and the re-examination. In brief, carotid ultrasonography was performed in six carotid segments (far and near walls of bilateral common carotid arteries, bifurcations, and internal carotid arteries). The presence of plaques was defined as a focal region with intimal-medial thickness ≥ 1.5 mm or a focal structure that encroached into the arterial lumen measuring at least 0.5 mm or 50% of the surrounding intimal-medial thickness21). Progression of carotid plaque was defined as the appearance of at least one plaque at re-examination in a previously plaque-free arterial segment22). The incidence of carotid plaque was defined as the appearance of at least one plaque at the re-examination among participants with no plaque in all segments of the carotid artery at baseline. The results of reproducibility at the baseline and re-examination were described previously15). In brief, the kappa value for intra-observer and inter-observer agreement for the presence of plaque was more than 0.80 at baseline and re-examination. Furthermore, the kappa value of the inter-observer agreement was 0.79 between the 2 survey examinations.

Study Power Estimation

Limited studies have investigated the association between baseline APOE-HDLC levels and carotid atherosclerosis progression. Based on our previous study, the relative risk of coronary heart disease incidence associated with a higher ratio of APOE-HDLC to total HDLC (≥ 9.8%) was 0.16 11). In the current study, the 5-year progression rate of carotid plaque was 44.6%. The R-squared value for the ratio of APOE-HDLC to total HDLC with other known cardiovascular risk factors was 0.440. The prevalence of the ratio of APOE-HDLC to total HDLC ≥ 9.8% was 36.8%. The estimated sample size was 140, assuming an alpha (probability of type I error) of 0.05 and a beta (probability of type II error) of 0.10. The actual sample size of 823 yielded sufficient statistical power.

Data Management and Statistical Analyses

Continuous variables are expressed as the mean (±standard deviation [SD]) in the case of a normal distribution or as median (interquartile range) and were compared using an unpaired Student’s t-test or Mann-Whitney U test, as appropriate. Categorical variables are expressed as the number (percentage) and were compared using the χ2 test. Correlations were estimated using a partial correlation analysis adjusted for the age and sex.

An estimate of cholesterol molecules per HDL particle (HDLC/P ratio) was obtained by calculating the ratio of HDLC to HDL particle number (units of HDLC were transformed to mmol/L from the original mg/dL unit when calculating the HDLC/P ratio). HDLC/P ratios were categorized into 3 groups (＜44.8, 44.8-52.2, or ≥ 52.3) in order to maximize statistical power, with the cutoff point set using the upper third (52.3) and a fixed decrement of 7.5. Given that APOE-HDLC was strongly correlated with HDLC (partial-r = 0.98), a proportion of APOE-HDLC in total HDLC (APOE-HDLC/HDLC ratio) was obtained by calculating the ratio of APOE-HDLC to total HDLC. To ensure sufficient statistical power and maintain the consistency of data across categories, the APOE-HDLC/HDLC ratio was also categorized into 3 groups of ＜9.1%, 9.1%-10.0%, and ≥ 10.1%, with the cutoff point set using the upper third (10.1%) and a fixed decrement of 1%.

The relative risk (RR) and 95% confidence interval (CI) for the 5-year progression of carotid atherosclerosis within subgroups of APOE-HDLC/HDLC ratio and the HDLC/P ratio were calculated using the modified Poisson regression model17) after adjusting for known cardiovascular risk factors, including the age, sex, body mass index, systolic blood pressure, diabetes, current smoking, low-density lipoprotein cholesterol levels, triglycerides, hs-CRP levels, HDL particle size, anti-hypertensive medications, and lipid-lowering medications. Participants in the lowest category of HDLC/P ratio or APOE-HDLC/HDLC ratio were defined as the reference groups, respectively. Furthermore, the HDLC/P ratio and APOE-HDLC/HDLC ratio were categorized and cross-combined into the following 6 groups: low APOE-HDLC/HDLC ratio (＜9.1%) and low HDLC/P ratio (＜44.8); low APOE-HDLC/HDLC ratio (＜9.1%) and high HDLC/P ratio (≥ 44.8); medium APOE-HDLC/HDLC ratio (9.1%-10.0%) and low HDLC/P ratio (＜44.8); medium APOE-HDLC/HDLC ratio (9.1%-10.0%) and high HDLC/P ratio (≥ 44.8); high APOE-HDLC/HDLC ratio (≥ 10.1%) and low HDLC/P ratio (＜44.8); high APOE-HDLC/HDLC ratio (≥ 10.1%) and high HDLC/P ratio (≥ 44.8). The association of these combined subgroups with the 5-year progression of carotid atherosclerosis was analyzed after adjusting for all of the risk factors mentioned above. The subgroup with a low APOE-HDLC/HDLC ratio (＜9.1%) and low HDLC/P ratio (＜44.8) was defined as the reference group.

To test whether missing data yielded potential bias, comparisons were made between participants eligible for the final analysis and those unavailable for a re-examination. No significant differences were observed in lipid-related biomarkers (Supplemental Table 1).

Supplemental Table 1.Differences between study participants who were eligible and unavailable for re-examination

Baseline characteristics

Eligible participants

(n = 823)

Participants unavailable

for re-examination(n = 294)

P value
Age (years) 59.1±7.8 59.9±7.6 0.147
Male, n (%) 366 (44.5) 146 (49.7) 0.125
Current smoker, n (%) 83 (10.1) 27 (9.2) 0.656
Body mass index (kg/m2) 25.0±3.2 24.7±3.1 0.251
Systolic blood pressure (mmHg) 128.9±18.2 126.2±17.3 0.031
Diastolic blood pressure (mmHg) 80.7±9.9 80.2±9.9 0.526
Fasting blood glucose (mmol/L) 4.66 (4.38 - 5.05) 4.72 (4.44 - 5.06) 0.398
Type 2 diabetes, n (%) 62 (7.5) 20 (6.8) 0.680
Hypertension, n (%) 412 (50.1) 149 (50.7) 0.855
Anti-hypertensive medications, n (%) 234 (28.4) 74 (25.2) 0.283
Lipid-lowering medications, n (%) 98 (11.9) 31 (10.5) 0.530
Total cholesterol (mg/dL) 215.3±39.8 214.1±39.3 0.640
LDL cholesterol (mg/dL) 135.7±35.9 129.9±32.2 0.663
HDL cholesterol (mg/dL) 58.5±13.4 53.0±11.8 0.297
Triglyceride (mg/dL) 116.0 (83.0 - 166.0) 117.0 (84.0 - 171.0) 0.831

Abbreviation: HDL, high-density lipoprotein; LDL, low-density lipoprotein.

Data are expressed as number (percent) for categorical variables and as mean (standard deviation) for continuous variables in case of normal distributions and median (inter-quartile range) otherwise.

P values were compared between the study participants who were eligible and unavailable for re-examination. Unpaired Student’s t-test (when satisfying a normal distribution) or Mann-Whitney U test (when not satisfying a normal distribution) for quantitative variables and χ2 test for categorical variables.

The statistical analyses were performed using the R software program, version 4.2.0 (The R Foundation for Statistical Computing, Vienna, Austria). A value of P＜0.05 in the 2-sided test was considered to indicate statistical significance. Sample size estimation was calculated using the PASS software program, version 11.0 (NCSS, Kaysville, UT, USA).

Results

Baseline Characteristics of Study Participants

Among 823 participants with a mean (±SD) age of 59.1 (±7.8) years old, the baseline mean APOE-HDLC levels and APOE-HDLC/HDLC ratio were 5.59±1.62 (range, 0.8-12.3) mg/dL and 9.4%±0.9% (range, 4.2%-11.8%), respectively. Baseline characteristics of study participants are shown in Table 1. Participants with plaque progression over a five-year interval were more likely to have lower APOE-HDLC, APOA-I, HDLC levels, and HDL particle numbers, and higher levels of lipid carried by atherogenic lipoproteins than participants with no progression (Table 1). The correlations of APOE-HDLC with other risk factors were similar to those with total HDLC (Fig.1). Given the strong correlation between APOE-HDLC and HDLC (partial-r = 0.98), the APOE-HDLC/HDLC ratio was calculated as a new indicator of APOE-HDL. As shown in Fig.1, the APOE-HDLC/HDLC ratio was moderately correlated with HDL particle numbers (partial-r = 0.61).

Table 1.Baseline characteristics of study participants

Baseline characteristics

Total participants

(n = 823)

Participants with

plaque progression

(n = 367)

Participants with

non-plaque progression

(n = 456)

P value
Age (years) 59.1±7.8 61.0±7.6 57.5±7.7 ＜0.001
Male, n (%) 366 (44.5) 187 (51.0) 179 (39.3) ＜0.001
Current smoker, n (%) 83 (10.1) 45 (12.3) 38 (8.3) 0.034
Body mass index (kg/m2) 25.0±3.2 25.3±3.3 24.7±3.2 0.003
Systolic blood pressure (mmHg) 128.9±18.2 132.5±18.4 126.0±17.6 ＜0.001
Diastolic blood pressure (mmHg) 80.7±9.9 81.5±10.1 80.0±9.7 0.025
Fasting blood glucose (mmol/L) 4.66 (4.38 - 5.05) 4.72 (4.38 - 5.05) 4.66 (4.33 - 5.05) 0.070
Hs-CRP (mg/L) 0.81 (0.36 - 1.76) 0.88 (0.41 - 1.76) 0.72 (0.34 - 1.64) 0.026
Type 2 diabetes, n (%) 62 (7.5) 29 (7.9) 33 (7.2) 0.719
Hypertension, n (%) 412 (50.1) 213 (58.0) 199 (43.6) ＜0.001
Glucose-lowering medications, n (%) 29 (3.5) 14 (3.8) 15 (3.3) 0.762
Anti-hypertensive medications, n (%) 234 (28.4) 116 (31.6) 118 (25.9) 0.053
Lipid-lowering medications, n (%) 98 (11.9) 47 (12.8) 51 (11.2) 0.200
Total cholesterol (mg/dL) 215.3±39.8 222.0±40.3 209.9±38.6 ＜0.001
LDL cholesterol (mg/dL) 135.7±35.9 144.2±35.6 128.9±34.7 ＜0.001
HDL cholesterol (mg/dL) 58.5±13.4 56.9±12.1 59.8±14.2 0.001
Triglyceride (mg/dL) 116.0 (83.0 - 166.0) 126.5 (91.0 - 187.3) 98.0 (73.0 - 131.5) 0.003
APOE-HDLC (mg/dL) 5.59±1.62 5.44±1.49 5.71±1.71 0.018
APOE-HDLC/HDLC ratio (%) 9.4±0.9 9.5±0.9 9.4±0.9 0.499
Apolipoprotein AⅠ(mg/dL) 143.0±20.3 140.6±19.9 145.0±20.4 0.002
Apolipoprotein E (mg/dL) 4.58±1.59 4.68±1.63 4.51±1.55 0.126
HDL particle number (umol/L) 30.35±4.58 29.70±4.44 30.88±4.64 ＜0.001
HDL particle size (nm) 9.05±0.44 9.00±0.42 9.10±0.45 0.001
HDLC/P ratio 46.3±9.4 45.6±8.8 46.8±9.8 0.073

Abbreviations: APOE-HDLC, apolipoprotein E-containing high-density lipoprotein cholesterol; APOE-HDLC/HDLC ratio, ratio of apolipoprotein E containing high-density lipoprotein cholesterol to total high-density lipoprotein cholesterol; HDLC/P ratio, ratio of HDL cholesterol to HDL particle number, giving the content of cholesterol per HDL particle; Hs-CRP, high-sensitivity C-reactive protein; HDL, high-density lipoprotein; LDL, low-density lipoprotein.

Data are expressed as number (percent) for categorical variables and as mean (standard deviation) for continuous variables in case of normal distributions and median (inter-quartile range) otherwise.

P values were compared between the study participants with plaque progression and non-plaque progression. Unpaired Student’s t-test (when satisfying a normal distribution) or Mann-Whitney U test (when not satisfying a normal distribution) for quantitative variables and χ2 test for categorical variables.

Fig.1. Correlation of APOE-HDLC and other cardiovascular disease risk factors

Circle colors indicate the direction of correlation: blue for positive and red for negative, with darker shades representing stronger correlations. Neutral gray denotes no significant correlation. Circle sizes correspond to the absolute magnitude of the correlation coefficient. APOE-HDLC, apolipoprotein E-containing high-density lipoprotein cholesterol; APOE-HDLC/HDLC ratio, ratio of apolipoprotein E-containing high-density lipoprotein cholesterol to total high-density lipoprotein cholesterol; HDLC/P ratio, ratio of HDLC to HDL particle number; hs-CRP, high-sensitivity C-reactive protein; LDL, low-density lipoprotein.

Association between the APOE-HDLC/HDLC Ratio and Risk of 5-Year Progression in Carotid Atherosclerosis

A total of 367 (44.6%) participants developed at least 1 new plaque in a previously plaque-free carotid segment over a 5-year interval. A modified Poisson regression analysis showed that the RR for plaque progression of the highest APOE-HDLC/HDLC ratio was 0.75 (95% CI: 0.56-1.00) relative to the lowest level, after adjusting for the age, BMI, systolic blood pressure, triglycerides, and LDL cholesterol (Table 2). The relationship between the APOE-HDLC/HDLC ratio and plaque progression was preserved after additional adjustment for the HDLC/P ratio. Of 660 participants who had no baseline plaque in all segments of carotid artery, a total of 282 (42.7%) individuals had new-onset plaque. A modified Poisson regression analysis also showed that the APOE-HDLC/HDLC ratio was associated with the incidence of new-onset plaque (Supplemental Table 2). These associations were robust after excluding participants with lipid-lowering treatment (Supplemental Table 3). Similar results were found for Apo A-I and HDLC (Supplemental Tables 4, 5, 6, 7).

Table 2.Association of APOE-HDLC/HDLC ratio and HDLC/P ratio with progression of carotid atherosclerosis

Variables Model 1 Model 2
RR (95%CI) P value RR (95%CI) P value
APOE-HDLC/HDLC ratio (%)
＜9.1 Ref Ref
9.1-10.0 0.88 (0.72 - 1.08) n.s. 0.86 (0.71 - 1.06) n.s.
≥ 10.1 0.75 (0.56 - 1.00) 0.050 0.71 (0.53 - 0.95) 0.021
HDLC/P ratio
＜44.8 Ref
44.8-52.2 1.27 (1.04 - 1.54) 0.017
≥ 52.3 1.37 (1.02 - 1.85) 0.039
Age (per 5 years) 1.16 (1.10 - 1.22) ＜0.001 1.15 (1.09 - 1.22) ＜0.001
Body mass index (1 kg/m2) 1.03 (1.00 - 1.06) 0.032 1.03 (1.00 - 1.06) 0.025
Systolic blood pressure (per 10 mmHg) 1.05 (1.01 - 1.10) 0.017 1.06 (1.01 - 1.10) 0.010
Log-transformed triglyceride 1.21 (0.83 - 1.75) n.s. 1.78 (1.08 - 2.95) 0.024
LDL cholesterol (per 20 mg/dL) 1.15 (1.09 - 1.20) ＜0.001 1.13 (1.08 - 1.19) ＜0.001

Model 1: Adjusted for age, sex, body mass index, diabetes, smoking, systolic blood pressure, antihypertensive and lipid-lowering medications, hs-CRP, triglycerides, LDL cholesterol, and HDL particle size.

Model 2: Further adjusted for HDLC/P ratio in addition to all variables in Model 1.

Abbreviations: APOE-HDLC, apolipoprotein E-containing high-density lipoprotein cholesterol; APOE-HDLC/HDLC ratio, ratio of apolipoprotein E containing high-density lipoprotein cholesterol to total high-density lipoprotein cholesterol; CI, confidence interval; HDLC/P ratio, ratio of HDL cholesterol to HDL particle number, giving the content of cholesterol per HDL particle; HDL, high-density lipoprotein; LDL, low-density lipoprotein; RR, relative risk.

RRs (95%CI) were calculated by modified Poisson regression analysis, and all substantial models were adjusted for age (per 5 years), sex, BMI, diabetes, smoking, systolic blood pressure (per 10mmHg), anti-hypertensive medications, lipid-lowering medications, log-transformed high-sensitivity C-reactive protein levels, log-transformed triglyceride, low-density lipoprotein cholesterol (per 20mg/dL) and HDL particle size (per 0.4 nm), where the significant variables are shown in the figure.

Supplemental Table 2.Modified Poisson regression analysis of APOE-HDLC/HDLC ratio and HDL-C/P ratio in 5-year risk of plaque progression and plaque incidence, respectively

Variables Plaque progression Plaque incidence
RR (95%CI) P value RR (95%CI) P value
APOE-HDLC/HDLC ratio (%)
＜9.1 Ref Ref
9.1-10.0 0.86 (0.71 - 1.06) 0.152 0.86 (0.68 - 1.08) 0.195
≥ 10.1 0.71 (0.53 - 0.95) 0.021 0.72 (0.51 - 1.01) 0.056
HDL-C/P ratio
＜44.8 Ref Ref
44.8-52.2 1.27 (1.04 - 1.54) 0.017 1.19 (0.95 - 1.50) 0.138
≥ 52.3 1.37 (1.02 - 1.85) 0.039 1.34 (0.95 - 1.89) 0.096
Age (per 5 years) 1.15 (1.09 - 1.22) ＜0.001 1.14 (1.07 - 1.21) ＜0.001
Male 1.07 (0.89 - 1.27) 0.470 1.06 (0.86 - 1.30) 0.609
Type 2 diabetes 1.03 (0.78 - 1.36) 0.820 0.90 (0.62 - 1.30) 0.571
Body mass index (1 kg/m2) 1.03 (1.00 - 1.06) 0.025 1.02 (0.99 - 1.05) 0.141
Current smoker 1.20 (0.95 - 1.51) 0.124 1.25 (0.95 - 1.63) 0.109
Systolic blood pressure (per 10 mmHg) 1.06 (1.01 - 1.10) 0.010 1.06 (1.01 - 1.12) 0.016
Anti-hypertensive medications 0.94 (0.80 - 1.11) 0.485 1.00 (0.83 - 1.22) 0.963
Lipid-lowering medications 0.89 (0.71 - 1.11) 0.314 0.82 (0.61 - 1.10) 0.188
Log-transformed hs-CRP 0.96 (0.82 - 1.14) 0.660 0.98 (0.81 - 1.20) 0.875
Log-transformed triglyceride 1.78 (1.08 - 2.95) 0.024 1.48 (0.81 - 2.73) 0.204
LDL cholesterol (per 20 mg/dL) 1.13 (1.08 - 1.19) ＜0.001 1.16 (1.09 - 1.23) ＜0.001
HDL particle size (per 0.4 nm) 0.96 (0.88 - 1.05) 0.393 0.95 (0.85 - 1.06) 0.328

Abbreviation: APOE-HDLC, apolipoprotein E-containing high-density lipoprotein cholesterol; APOE-HDLC/HDLC ratio, ratio of apolipoprotein E containing high-density lipoprotein cholesterol to total high-density lipoprotein cholesterol; CI, confidence interval; HDL-C/P ratio, ratio of HDL cholesterol to HDL particle number, giving the content of cholesterol per HDL particle; HDL, high-density lipoprotein; Hs-CRP, high-sensitivity C-reactive protein; LDL, low-density lipoprotein; RR, relative risk.

Supplemental Table 3.Modified Poisson regression analysis of APOE-HDLC/HDLC ratio and HDL-C/P ratio in 5-year risk of plaque progression and plaque incidence in participants without lipid-lowering medications, respectively

Variables Plaque progression Plaque incidence
RR (95%CI) P value RR (95%CI) P value
APOE-HDLC/HDLC ratio (%)
＜9.1 Ref Ref
9.1-10.0 0.80 (0.65 - 0.99) 0.039 0.77 (0.61 - 0.99) 0.038
≥ 10.1 0.70 (0.52 - 0.96) 0.027 0.67 (0.47 - 0.95) 0.024
HDL-C/P ratio
＜44.8 Ref Ref
44.8-52.2 1.25 (1.01 - 1.54) 0.038 1.21 (0.95 - 1.54) 0.124
≥ 52.3 1.43 (1.03 – 1.98) 0.031 1.44 (1.00 - 2.08) 0.048

Abbreviation: APOE-HDLC/HDLC ratio, ratio of apolipoprotein E containing high-density lipoprotein cholesterol to total high-density lipoprotein cholesterol; CI, confidence interval; HDL-C/P ratio, ratio of HDL cholesterol to HDL particle number, giving the content of cholesterol per HDL particle; RR, relative risk.

All substantial models are additionally adjusted for age (5 years), sex, diabetes, body mass index (1kg/m2), smoking, systolic blood pressure (10 mm Hg), anti-hypertensive medications, log-transformed high-sensitivity C-reactive protein levels, log-transformed triglyceride levels, low-density lipoprotein cholesterol (20 mg/dL), and HDL particle size (0.4 nm).

Supplemental Table 4.Modified Poisson regression analysis of APOA-I and HDL-C/P ratio in 5-year risk of plaque progression and plaque incidence, respectively

Plaque Progression Plaque Incidence
RR (95% CI) P value RR (95% CI) P value
APOA-I (mg/dL)
＜134.7 Ref Ref
134.7 - 150.4 0.99 (0.83 - 1.18) 0.892 1.01 (0.83 - 1.24) 0.908
≥ 150.5 0.73 (0.58 - 0.93) 0.011 0.72 (0.55 - 0.94) 0.018
HDL-C/P ratio
＜44.8 Ref Ref
44.8 - 52.2 1.29 (1.06 - 1.57) 0.012 1.22 (0.97 - 1.54) 0.090
≥ 52.3 1.40 (1.04 - 1.88) 0.026 1.39 (0.99 - 1.96) 0.056

RRs (95% CI) were calculated by modified Poisson regression analysis, and all substantial models were adjusted for age (per 5 years), sex, BMI, diabetes, smoking, systolic blood pressure (per 10mmHg), anti-hypertensive medications, lipid-lowering medications, log-transformed high-sensitivity C-reactive protein levels, log-transformed triglyceride, low-density lipoprotein cholesterol (per 20mg/dL), and HDL particle size (per 0.4 nm).

Supplemental Table 5.Modified Poisson regression analysis of APOA-I and HDL-C/P ratio in 5-year risk of plaque progression and plaque incidence in participants without lipid-lowering medications, respectively

Plaque Progression Plaque Incidence
RR (95% CI) P value RR (95% CI) P value
APOA-I (mg/dL)
＜134.7 Ref Ref
134.7 - 150.4 1.09 (0.87 - 1.38) 0.452 1.20 (0.90 - 1.62) 0.216
≥ 150.5 0.99 (0.68 - 1.45) 0.957 1.10 (0.69 - 1.77) 0.686
HDL-C/P ratio
＜44.8 Ref Ref
44.8 - 52.2 1.28 (1.03 - 1.58) 0.024 1.24 (0.97 - 1.59) 0.081
≥ 52.3 1.52 (1.09 - 2.12) 0.013 1.51 (1.05 - 2.17) 0.024

RRs (95% CI) were calculated by modified Poisson regression analysis, and all substantial models were adjusted for age (per 5 years), sex, BMI, diabetes, smoking, systolic blood pressure (per 10mmHg), anti-hypertensive medications, lipid-lowering medications, log-transformed high-sensitivity C-reactive protein levels, log-transformed triglyceride, low-density lipoprotein cholesterol (per 20mg/dL), and HDL particle size (per 0.4 nm).

Supplemental Table 6.Modified Poisson regression analysis of HDL-C and HDL-C/P ratio in 5-year risk of plaque progression and plaque incidence, respectively

Plaque Progression Plaque Incidence
RR (95% CI) P value RR (95% CI) P value
HDL-C (mg/dL)
＜51.8 Ref Ref
51.8 - 62.7 0.88 (0.73 - 1.07) 0.194 0.87 (0.70 - 1.09) 0.219
≥ 62.8 0.77 (0.59 - 1.02) 0.069 0.69 (0.50 - 0.96) 0.026
HDL-C/P ratio
＜44.8 Ref Ref
44.8 - 52.2 1.29 (1.07 - 1.58) 0.012 1.24 (0.98 - 1.57) 0.076
≥ 52.3 1.42 (1.04 - 1.93) 0.027 1.44 (1.01 - 2.04) 0.042

RRs (95% CI) were calculated by modified Poisson regression analysis, and all substantial models were adjusted for age (per 5 years), sex, BMI, diabetes, smoking, systolic blood pressure (per 10mmHg), anti-hypertensive medications, lipid-lowering medications, log-transformed high-sensitivity C-reactive protein levels, log-transformed triglyceride, low-density lipoprotein cholesterol (per 20mg/dL), and HDL particle size (per 0.4 nm).

Supplemental Table 7.Modified Poisson regression analysis of HDL-C and HDL-C/P ratio in 5-year risk of plaque progression and plaque incidence in participants without lipid-lowering medications, respectively

Plaque Progression Plaque Incidence
RR (95% CI) P value RR (95% CI) P value
HDL-C (mg/dL)
＜51.8 Ref Ref
51.8 - 62.7 1.13 (0.87 - 1.47) 0.360 1.12 (0.84 - 1.49) 0.459
≥ 62.8 1.28 (0.82 - 2.00) 0.276 1.10 (0.70 - 1.74) 0.674
HDL-C/P ratio
＜44.8 Ref Ref
44.8 - 52.2 1.25 (1.01- 1.55) 0.037 1.22 (0.95 - 1.57) 0.113
≥ 52.3 1.50 (1.08 - 2.10) 0.016 1.50 (1.04 - 2.17) 0.031

RRs (95% CI) were calculated by modified Poisson regression analysis, and all substantial models were adjusted for age (per 5 years), sex, BMI, diabetes, smoking, systolic blood pressure (per 10mmHg), anti-hypertensive medications, lipid-lowering medications, log-transformed high-sensitivity C-reactive protein levels, log-transformed triglyceride, low-density lipoprotein cholesterol (per 20mg/dL), and HDL particle size (per 0.4 nm).

Impact of the APOE-HDLC/HDLC Ratio on Plaque Progression Associated with a High HDLC/P Ratio

The APOE-HDLC/HDLC ratio was negatively associated with plaque progression at both low and high HDLC/P ratios, despite no hint of significance for the low-level group. Especially among participants with an HDLC/P ratio ≥ 44.8, the RR for plaque progression was 0.64 (95% CI: 0.43-0.97) at the highest APOE-HDLC/HDLC ratio compared with the lowest level (Supplemental Table 8). In addition, a joint analysis of the APOE-HDLC/HDLC ratio and HDLC/P ratio showed that the RR for plaque progression was 1.39 (95% CI: 1.04-1.88) in participants with a high HDLC/P ratio but low APOE-HDLC/HDLC ratio and 0.74 (95% CI: 0.51-1.08) in those with a high HDLC/P ratio and high APOE-HDLC/HDLC ratio, relative to those with a low HDLC/P ratio and low APOE-HDLC/HDLC ratio after multivariable adjustment (Fig.2).

Supplemental Table 8.Modified Poisson regression analysis of the APOE-HDLC/HDLC ratio with progression of carotid atherosclerosis across HDL-C/P ratio subgroups＊

APOE-HDLC/HDLC ratio (%) HDL-C/P ratio ＜44.8 HDL-C/P ratio ≥ 44.8
RR (95%CI) P value RR (95%CI); P value
＜9.1 Ref Ref
9.1-10.0 0.93 (0.71 - 1.22) 0.621 0.76 (0.57 - 1.03) 0.075
≥ 10.1 0.76 (0.50 - 1.14) 0.185 0.64 (0.43 - 0.97) 0.033

Abbreviations: APOE-HDLC/HDLC ratio, ratio of apolipoprotein E-containing high-density lipoprotein cholesterol to total high-density lipoprotein cholesterol; HDL-C/P ratio, ratio of high-density lipoprotein cholesterol to high-density lipoprotein particle numbers; RR, relative risk; CI, confidence interval.

＊All substantial models are adjusted for age (per 5 years), sex, BMI, diabetes, smoking, systolic blood pressure (per 10mmHg), anti-hypertensive medications, lipid-lowering medications, log-transformed high-sensitivity C-reactive protein levels, log-transformed triglyceride, low-density lipoprotein cholesterol (per 20mg/dL) and HDL particle size (per 0.4 nm).

Fig.2. Relative risk of plaque progression associated with combined subgroups of APOE-HDLC/HDLC ratio and HDLC/P ratio

APOE-HDLC/HDLC ratio, ratio of apolipoprotein E-containing high-density lipoprotein cholesterol to total high-density lipoprotein cholesterol; CI, confidence interval; HDLC/P ratio, ratio of high-density lipoprotein cholesterol to high-density lipoprotein particle number; LDL, low-density lipoprotein; RR, relative risk. Modified Poisson models adjusted for age (per 5 years), sex, body mass index, diabetes, smoking, systolic blood pressure (per 10 mmHg), anti-hypertensive medications, lipid-lowering medications, log-transformed high-sensitivity C-reactive protein levels, log-transformed triglyceride, low-density lipoprotein cholesterol (per 20 mg/dL), and HDL particle size (per 0.4 nm); significant variables are shown in the figure. The subgroup with a low APOE-HDLC/HDLC ratio and low HDLC/P ratio was defined as the reference group. The APOE-HDLC/HDLC ratio and HDLC/P ratio were categorized and cross-combined into six groups. The APOE-HDLC/HDLC ratio was categorized into low (＜9.1%), medium (9.1%-10.0%), high (≥ 10.1%) groups. HDLC/P ratios were categorized into low (＜44.8) and high (≥ 44.8) groups.

Discussion

In this community-based cohort study, we report for the first time that an elevated APOE-HDLC/HDLC ratio was significantly and independently associated with a lower risk of atherosclerotic plaque progression. Furthermore, an elevated APOE-HDLC/HDLC ratio can modify the impact of a very high cholesterol content per HDLP on increased progression of carotid atherosclerosis. These results suggested that APOE-containing HDL may serve as a candidate emerging biomarker for the anti-atherosclerotic function of HDL.

Recent major shifts in the research area of HDL have been toward identifying new parameters to better reflect HDL function. The contribution of APOE-HDL has attracted attention owing to the key role of APOE in modulating cholesterol efflux and in affecting the metabolism of apolipoprotein B-containing lipoproteins. Our previous study demonstrated an association of elevated APOE-HDLC with a decreased risk of 10-year coronary heart disease incidence in a community-based cohort study11). Morton et al. also reported that a subspecies of HDL that contains APOE but not Apolipoprotein CIII was found to be associated with a lower risk of coronary heart disease23). Furthermore, a recent observational study has shown that low levels of APOE-HDLC are strongly associated with the presence of non-calcified coronary plaque assessed using quantitative coronary computed tomography angiography14). However, limited studies have explored the impact of circulating APOE-HDLC levels on the development of atherosclerosis. To our knowledge, this is the first large-scale study conducted in asymptomatic individuals who were free of CVD to assess this association. The progression of atherosclerosis is influenced by several traditional risk factors, including age, hypertension, obesity, and dyslipidemia. After accounting for these traditional risk factors, we found a decreased progression of carotid atherosclerosis associated with a high APOE-HDLC/HDLC ratio as compared with a low APOE-HDLC/HDLC ratio. These results highlight the potential of APOE-containing HDL as a candidate emerging biomarker for the anti-atherosclerotic function of HDL particles. Furthermore, our previous findings suggested that very high levels of cholesterol molecules per HDL particle (HDLC/P ratio), defined as a very high cholesterol content per HDLP, increase the progression of carotid atherosclerosis and might represent an indicator of dysfunctional HDL17). Previous studies have further shown that high levels of very high level of cholesterol content per HDLP promote the formation of APOE-containing HDL in patients with cholesteryl ester transfer protein (CETP) deficiency24, 25) and those treated with CETP inhibitors26, 27). In addition, high levels of APOE promoted by paraoxonase-1 have been shown to ameliorate HDL dysfunction by promoting HDL maturation and macrophage cholesterol leakage, thereby alleviating the development of atherosclerosis28). These findings suggest that APOE-HDL may be necessary to correct the dysfunction of very high level of cholesterol content per HDLP.

The present study identified the modifiable impact of an elevated APOE-HDLC/HDLC ratio on reducing progression of carotid atherosclerosis associated with very high level of cholesterol content per HDLP. These findings provide a mechanistic explanation for the association between elevated APOE-HDL and the decreased risk of cardiovascular disease. Based on these findings, we propose APOE-HDL as a novel indicator beyond HDLC as a potential therapeutic target.

Numerous pre-clinical and clinical studies have demonstrated the pleiotropic anti-atherogenic properties of APOE29-31), which may be endowed with the HDL that carries it. APOE serves as a ligand for various liver receptors that mediate the elimination of atherosclerotic lipoproteins, including low-density lipoprotein receptors32), low-density lipoprotein receptor-associated proteins33), and heparin/heparin sulfate proteoglycan34-37). Moreover, Murphy et al. showed that expression of APOE by hematopoietic stem cells promotes cholesterol efflux via ATP-binding cassette transporter A1 (ABCA1)/ATP-binding cassette subfamily G1 and decreases cell proliferation and infiltration into the intima38). More interestingly, early work by Shinohata et al. showed that an increase in APOE-containing HDL induced by a high-fat/high-cholesterol diet may be involved in cholesterol efflux from the liver through increased ABCA1-mediated free-cholesterol efflux39). In addition, several other biological functions of APOE may contribute to the anti-atherogenic role of HDL, such as anti-inflammatory and anti-oxidative protection40, 41). In clinical trials, APOE-derived mimetic peptides have attracted much research attention as candidates for anti-atherogenic therapy42). The chimeric protein Ac-hE18A-NH2 has been proven to reduce plasma cholesterol, promote macrophage cholesterol efflux, and to have anti-inflammatory and anti-oxidative properties43, 44). Therefore, APOE-HDL may be a multifunctional protein that serves as a potential marker for anti-atherosclerotic function of HDL through multiple potential pathways. However, the exact mechanisms underlying the anti-atherogenic role of APOE-containing HDL remain uncertain and require further study.

Several strengths of the current study are worth noting. To our knowledge, the present study is the first large-scale study conducted in a general population to investigate the relationship between APOE-HDLC related parameters and five-year progression of carotid atherosclerosis. In this study, we found that an increased APOE-HDLC/HDLC ratio was independently associated with a slower progression of carotid atherosclerosis and could attenuate the atherogenic impact of a very high cholesterol content per HDLP. However, despite these advantages, several potential study limitations should be pointed out. First, while many confounders were adjusted for in the present study, some residual confounding could not be ruled out, including dietary factors, types of lipid-lowering drugs, and APOE present on non-HDL APOB-containing lipoproteins. Despite being a common limitation in a nonrandomized study, this mandates further investigation. Second, the sample size in the present study was sizeable, but it only reflected a fraction of the original cohort comprising a largely Chinese population. Although the missing data did not generate substantial bias because of non-significant differences between study participants who were eligible and unavailable for a re-examination, further confirmation in larger studies with more diverse populations is needed. Third, our study included a large cohort of prospectively enrolled participants who underwent repeat ultrasound for the assessment of carotid plaque progression among asymptomatic individuals with no history of CVD in primary prevention. Future studies should be conducted among patients with advanced, established arterial lesions in secondary prevention. We also assessed the impact of APOE-HDLC on plaque progression, but whether or not APOE-HDLC exerts different effects on the progression of plaque composition requires further study. Fourth, our study did not consider the relationships among APOE isoforms (E2/E3/E4) or the impact of APOE distribution on HDL, which would be desirable to clarify in future. Finally, more mechanism studies are needed to investigate the specific and complex biological effects of APOE-HDL on the development of atherosclerosis.

Conclusion

Our study findings suggested that a high APOE-HDLC/HDLC ratio is significantly and independently associated with a decreased risk of five-year carotid atherosclerosis progression, and an elevated APOE-HDLC/HDLC ratio could decrease the progression of carotid atherosclerosis and further modify the impact of very high level of cholesterol content per HDLP. The findings suggest that APOE-containing HDL may serve as a measurable marker for identifying the anti-atherogenic function of HDL and correcting the role of dysfunctional HDL particles. This possibility should be further investigated in large prospective epidemiological studies to evaluate their clinical utility in the prevention of CVD.

Conflict of Interest

The authors declare that they have no conflict of interest.

Funding

This work was supported by Capital Funds for health Improvement and Research (CFH 2024-2G-1052); Beijing municipal medical research institutes pilot reform project [grant number 2023-9]; National Key Research and Development Program of China [grant numbers 2022YFC3602501 and 2021YFC2500601]; National Natural Science Foundation of China [grant numbers 82470465, 82073635, 82103962, and 12226005]; and Beijing municipal medical research institutes financial project (11000024T000002831102).

Author Contributions

Y.Q. and J.L. contributed to the conception or design of the work. P.F.N., J.T.L., Y.L.D., P.P.H., Q.J.D., Y.C.H., Z.Y., and L.Z.H. contributed to the acquisition, analysis, or interpretation of data for the work. P.F.N. and J.T.L. drafted the manuscript. Y.Q. and J.L. critically revised the manuscript. All gave final approval and agree to be accountable for all aspects of work ensuring integrity and accuracy.

Acknowledgments

We gratefully acknowledge the contribution of all the investigators from participating centers in the Chinese Multi-provincial Cohort Study for data collection.

References
Related Articles
 

This article is licensed under a Creative Commons [Attribution-NonCommercial-ShareAlike 4.0 International] license.
https://creativecommons.org/licenses/by-nc-sa/4.0/
feedback
 Top