Association of the polymorphisms of the cholesteryl ester transfer protein gene with coronary artery disease: a meta-analysis

Aims This meta-analysis aimed to assess the association of the polymorphisms of cholesterol ester transfer protein (CETP) rs708272 (G>A), rs5882 (G>A), rs1800775 (C>A), rs4783961 (G>A), rs247616 (C>T), rs5883 (C>T), rs1800776 (C>A), and rs1532624 (C>A) with coronary artery disease (CAD) and the related underlying mechanisms. Methods A comprehensive search was performed using five databases such as PubMed, EMBASE, Web of Science, Cochrane Library and Scopus to obtain the appropriate articles. The quality of the included studies was assessed by the Newcastle-Ottawa Scale. The statistical analysis of the data was performed using STATA 17.0 software. The association between CETP gene polymorphisms and risk of CAD was estimated using the pooled odds ratio (OR) and 95% confidence interval (95% CI). The association of CETP gene polymorphisms with lipids and with CETP levels was assessed using the pooled standardized mean difference and corresponding 95% CI. P < 0.05 was considered statistically significant. Results A total of 70 case-control studies with 30,619 cases and 31,836 controls from 46 articles were included. The results showed the CETP rs708272 polymorphism was significantly associated with a reduced risk of CAD under the allele model (OR = 0.846, P < 0.001), the dominant model (OR = 0.838, P < 0.001) and the recessive model (OR = 0.758, P < 0.001). AA genotype and GA genotype corresponded to higher high-density lipoprotein cholesterol (HDL-C) concentrations in the blood compared with GG genotype across the studied groups (all P < 0.05). The CETP rs5882 and rs1800775 polymorphisms were not significantly associated with CAD under the allele model (P = 0.802, P = 0.392), the dominant model (P = 0.556, P = 0.183) and the recessive model (P = 0.429, P = 0.551). Similarly, the other mentioned gene polymorphisms were not significantly associated with CAD under the three genetic models. Conclusions The CETP rs708272 polymorphism shows a significant association with CAD, and the carriers of the allele A are associated with a lower risk of CAD and higher HDL-C concentrations in the blood compared to the non-carriers. The CETP rs5882, rs1800775, rs4783961, rs247616, rs5883, rs1800776, and rs1532624 are not significantly associated with CAD. Systematic Review Registration https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42023432865, identifier: CRD42023432865.


Introduction
Characterized by the narrowing of coronary arteries, coronary artery disease (CAD) is the leading cause of morbidity and mortality worldwide (1).The etiology and progression of CAD involve complex interactions between genetic and environmental factors (2), the latter includes age, gender, hypertension, smoking, and dyslipidemia (3).Dyslipidemia refers to lipid abnormalities that include high levels of low-density lipoprotein cholesterol (LDL-C), triglycerides (TG) and total cholesterol (TC), as well as low levels of high-density lipoprotein cholesterol (HDL-C).Substantial advances in understanding the genetic basis of dyslipidemia have recently suggested that the cholesteryl ester transfer protein (CETP) is involved in the pathogenesis of CAD (4).CETP is a plasma glycoprotein that facilitates the exchange of triglycerides for cholesterol ester from HDL to apolipoprotein B-containing lipoproteins, reducing the concentration of HDL-C (5).CETP is encoded by the homonymous gene on chromosome 16q13, and its polymorphisms influence protein activity and plasma lipid profiles, thus affecting CAD development and progression (6).Many studies evaluated the association between the polymorphisms of the CETP gene and the risk of CAD to find the underlying mechanisms and potential clinical implications.Most of the studies focused on three polymorphisms in the CETP gene, such as rs708272, rs5882, and rs180075.The first (also known as TaqIB) is one of the most common polymorphisms, consisting of a G-to-A substitution in the 279th nucleotide in the first intron of the gene (7).The second polymorphism is characterized by a single nucleotide polymorphism (SNP) that leads to the substitution of an isoleucine (I) with a valine (V) at the position 405 of the CETP protein sequence (8).The third polymorphism is characterized by an SNP at position 629 of the CETP gene, which results in the substitution of C-to-A (9).Although a review already in 2008 reported that CETP rs708272, rs5882, and rs180075 polymorphisms are significantly associated with the CAD risk (10), some recent studies published contradictory results.This meta-analysis aims to assess the association between CAD and common CETP gene polymorphisms represented by three SNPs (rs708272, rs5882, rs1800775), as well as five uncommon SNPs (rs4783961, rs247616, rs5883, rs1800776, rs1532624).Figure 1 illustrates the CETP gene structure and mutation locations.

Methods
This meta-analysis was performed in accordance with the guidelines outlined by the statement of the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) (11).

Search strategy
We conducted a comprehensive search using the following databases: PubMed, EMBASE, Web of Science, Cochrane Library, and Scopus.The search spanned from January 1, 1988, to May 9, 2023.We used a comprehensive search strategy with the following keywords: "Cholesterol Ester Transfer Protein" or "CETP" and "Myocardial Infarction" or "Cardiovascular Stroke" or "Heart Attack" or "Coronary Disease" or "CAD" or "CHD" and "Polymorphism, Single Nucleotide" or "SNP" or "Genotype" or "mutant" or "variant".

Inclusion and exclusion criteria
Clinical case-control studies that assessed the associations between CETP gene polymorphisms and CAD were included if they provided sufficient information on genotype counts for CAD patients and controls, allowing us to calculate the odds ratio (OR) and 95% confidence interval (95% CI).Additionally, all cases had to meet the diagnostic criteria for CAD.
We excluded duplicates, as well as articles related to animal experiments, reviews, meta-analyses, case reports, meeting abstracts, letters, and editorial comments.Non-English-language publications and those with insufficient statistical results were also excluded.myocardial infarction), genotyping methods, case/control sources, age distributions of cases/controls, gender distributions of cases/ controls, sample sizes of cases/controls, genotype counts in cases/ controls, and p-values for Hardy-Weinberg Equilibrium (HWE).Additionally, if available, data on the associations of CETP gene polymorphisms with serum lipid concentrations and CETP levels, as well as mean and standard deviation values for CETP level, HDL-C, LDL-C, TG, and TC concentrations across genotypes, were extracted to elucidate underlying mechanisms.

Statistical analysis
Statistical analysis was performed using the STATA software (Version 17.0, Stata Corporation, College Station, TX, USA).We estimated the association between CETP gene polymorphisms and CAD risk using the pooled OR and its 95% CI under three genetic models: the allele model (M vs. W), the dominant model The flowchart shows the procedure for selecting studies.The statistical significance of the pooled OR and pooled SMD was evaluated using the Z test.The significant threshold P < 0.05 was selected.Cochran's Q-statistic and I 2 tests were used to assess potential heterogeneity among the included studies.If the Q-test resulted in a P < 0.05 or I 2 > 50% indicating a significant heterogeneity, we employed a random-effects model.Otherwise, a fixed-effects model was used (13).The genotype count in the control was measured for HWE using a chi-square test.Sensitivity analysis was performed to assess the impact of individual studies on the overall pooled OR and SMD by sequentially excluding one study at a time and examining the effect.We conducted sensitivity analysis by systematically excluding one study at a time and examining its impact on the overall pooled OR and SMD.To evaluate publication bias, we used funnel plots, Begg's test, and Egger's test (14).A significance threshold of P < 0.05 was selected for determining the statistical significance of Begg's test and Egger's test.Additionally, we applied the trim-and-fill method to estimate the potential number of missing studies and their outcomes, which could contribute to publication bias.

Flowchart and initial selection
Figure 2 shows the flowchart of the details of the selection procedure.Initially, 908 articles were collected from five databases.A total of 763 publications were excluded based on their titles and abstracts, the full texts were evaluated, and additional 99 articles were removed.Finally, 46 articles meeting the inclusion criteria were included .

Characteristics of included studies
Seventy case-control studies, meeting the inclusion criteria, were included in this meta-analysis, comprising a total of 30,619 cases and 31,836 controls from 46 articles.The main characteristics of the selected studies and the genotypic distributions of CETP gene polymorphisms are shown in Tables 1, 2.

Genotyping methods
Genotyping methods varied among the included studies.Polymerase chain reaction-restriction fragment length polymorphisms (PCR-RFLP) is a molecular biology technology that analyzes genetic polymorphisms and sequence variations in DNA samples by

Gender and age
Gender and age are known to be influential factors in the association between SNPs and the occurrence of CAD.Gender ratio and mean age varied across the studies.
The NOS scores of the included articles were ≥5 and the average was 6.8.

Association of the three common CETP gene polymorphisms with CAD
The details of the overall and subgroup analyses of the association of the CETP rs708272, rs5882 and rs180075 polymorphisms with CAD are listed in Table 3.
After the exclusion of the studies deviating from HWE in the control, the statistical significance of the analysis did not substantially change.Subgroup analysis based on ethnicity identified a significance of association of the CETP rs708272 polymorphism with CAD both in Caucasians and Asians.The allele rs708272-A based on CAD subtypes significantly reduced the risk of myocardial infarction under the allele model (OR=0.849,P = 0.032) and recessive model (OR = 0.874, P = 0.005), but not under the dominant model (OR = 0.965, P = 0.361).Substantial heterogeneity in the coronary stenosis subgroup under the three genetic models (I 2 = 62.0%,I 2 = 60.2%,I 2 = 56.5%)was not observed in the myocardial infarction subgroup (I 2 = 0.0% for all).Table 3 also shows significant associations in NON-RFLP, RFLP, HB, and PB subgroups (all P < 0.05).Based on the patient sample size, the risk estimation in studies with large samples (size ≥ 500) appeared relatively conserved under the allele model (OR = 0.907, P = 0.006), dominant model (OR = 0.883, P = 0.02) and recessive model (OR = 0.831, P = 0.001), while the risk estimation in studies with small samples (size < 500) was slightly overestimated.Based on the size of the control, the heterogeneity of both studies with large samples (size ≥ 500) and small samples (size < 500) decreased under the allele model (I 2 = 49.7%,I 2 = 57.9%)and dominant model (I 2 = 20.5%,I 2 = 51.4%),suggesting that the heterogeneity might originate from this factor.Based on the gender ratio, the CETP rs708272 polymorphism was found to have a significant association with CAD under the three genetic models (all P < 0.05) in the cases and controls where males constituted the majority.However, in the cases where females predominated, this SNP was not significantly associated with CAD under the allele model (OR = 0.786, P = 0.091), dominant model (OR = 0.784, P = 0.360), or recessive model (OR = 0.668, P = 0.006).This non-association was also found in the femaledominated controls under the dominant model (OR = 0.818 P = 0.075).Based on the mean age of cases/controls, the significant association of this SNP with CAD was observed in all age groups under the three genetic model (all P < 0.05), and this finding was consistent with the overall analysis.
The overall analysis indicated that the CETP rs5882 and rs180075 polymorphisms were not significantly associated with both large samples (size ≥ 500) and small samples (size < 500) under the three genetic models, suggesting that the heterogeneity might be attributed to variations in sample size.

Association of the five uncommon CETP gene polymorphisms with CAD
Table 4 shows no significant associations of CETP rs4783961, rs247616, rs5883, rs1800776, and rs1532624 polymorphisms with CAD under the three genetic models.The results might be more susceptible to random factors due to the limited sample size, which might contribute to an increased heterogeneity.Subgroup analysis was not performed due to the lack of a sufficient number of studies.

Association of the CETP rs708272 polymorphism with CETP level and lipid serum concentrations
A meta-analysis was performed after the extraction of the data of CETP and lipid levels to find the association.Table 5 shows which information should be extracted using HDL-C as an example.Table 6 shows the association of the CETP rs708272 polymorphism with CETP level and HDL-C, LDL-C, TG, and TC concentrations in the patients, control, and all subjects under the homozygote model and the heterozygote model.AA genotype had higher HDL-C concentrations compared with the GG genotype in the case group (SMD = 0.459, P < 0.001) and control group (SMD = 0.279, P < 0.001) without significant heterogeneity (I 2 = 0.0%, I 2 = 40.6%)(Figure 6).Similarly, the GA genotype had higher HDL-C concentrations than the GG genotype in the patient group (SMD = 0.216, P < 0.001) and control group (SMD = 0.197, P < 0.001) without significant heterogeneity (I 2 = 20.6%,I 2 = 0.0%).The overall analysis indicated that the AA genotype and GA genotype were statistically significant compared with the GG genotype, with considerable heterogeneity (I 2 = 83.0%,I 2 = 74.0%)which might result from the types of groups.In conclusion, the carriers of allele A had higher HDL-C concentrations than the non-carriers.HDL particles play a pivotal role in removing excess cholesterol from peripheral tissues, reducing inflammation, protecting the endothelium, and preventing oxidative damage-all of which contribute to their anti-atherogenic properties.Higher levels of HDL in the blood are generally associated with a lower risk of atherosclerosis and cardiovascular disease.No significant correlation was found between the CETP rs708272 polymorphism and TC, TG, and LDL-C across the studied groups under two genetic models.In the patient group, the AA genotype had a lower CETP level (SMD = −0.485,P = 0.010) compared with the GG genotype.However, this conclusion is not reliable due to the limited sample size and the presence of only one study.

Meta-regression analysis
The subgroup analysis of rs708272 and rs1800775 indicated that the heterogeneity might be due to the sample size which was equivalent to the sum of each genotype count.Univariate and multivariate meta-regression analyses, including three types of genotype counts of case/control, were conducted to find potential sources of heterogeneity in the studies on the association of CETP gene polymorphism with CAD.As regards rs708272, the heterogeneity could be explained by GG genotype counts of the patients and AA and GG genotype counts of the control under the allele model (P = 0.027, P = 0.004, P = 0.041).AA and GA genotype counts of the patients and AG genotype counts of the control could explain the source of heterogeneity under the recessive model (P = 0.002, P = 0.001, P = 0.002).AA and GA genotype counts of the case/control could explain the source of heterogeneity under the dominant model (P = 0.003, P = 0.025, P = 0.001, P = 0.049).As regards rs5882, the heterogeneity could be explained by the GG genotype counts of the control under the allele model (P = 0.012).The meta-regression analysis failed to find the source of heterogeneity under the dominant model and the recessive model.As regards rs1800775, the source of heterogeneity was explained by the AA and CA genotype counts of the patients under the recessive model (P = 0.007, P = 0.019).The source of heterogeneity was explained by the AA and CA genotype counts of the patients and CC genotype counts of the control under the allele model (P = 0.008, P = 0.032, P = 0.010).The source of heterogeneity was explained by the CC genotype counts of the control under the allele model (P = 0.008).

Sensitivity analysis and publication bias
The sensitivity analysis of the association between the CETP gene polymorphism and CAD indicated that no single study affected the overall OR and the statistical significance under the three genetic models.As regards the rs708272 polymorphism, the funnel plot showed an imbalance between the number of data points on the left and right sides, resulting in an asymmetrical appearance (Figure 7).The Egger's test confirmed a remarkable publication bias under the allele model (P = 0.014), dominant model (P = 0.035), and recessive model (P = 0.02).The Begg's test did not detect any publication bias.The trim-and-fill method did not trim any studies for imputation.The Egger's test revealed the disappearance of publication bias under the allele model (P = 0.190) and the dominant model (P = 0.504), but not under the recessive model (P = 0.045) after excluding studies with significant departure from HWE.As regards the rs5882 polymorphism and rs1800775 polymorphism, the shapes of the funnel plots did not show any evident asymmetry (Figures 8, 9).Egger's test confirmed no publication bias under the allele model (P = 0.191, P = 0.173), the dominant model (P = 0.298, P = 0.400), and the recessive model (P = 0.097, P = 0.138).
No individual studies substantially influenced the overall SMD when associated with HDL-C.The funnel plots were almost symmetrical (Figure 10).Egger's test and Begg's test did not find any publication bias.The inclusion of studies investigating the association of the CETP rs708272 polymorphism with other lipid serum concentrations did not have any significant impact on the overall SMD.The funnel plots showed nearly symmetrical distribution of the studies.Egger's test and Begg's test did not detect any evidence of publication bias.

Discussion
In the present meta-analysis, our findings suggested that carriers of allele A of the CETP rs708272 polymorphism had higher HDL-C concentrations, lower CETP levels, and lower risk of CAD than no-carries.However, the reliability of the association between the CETP rs708272 polymorphism and CETP level was questionable considering the restricted sample size and the absence of additional studies.HDL particles exhibit a diverse range of atheroprotective activities including the efflux of cellular cholesterol, reduction of the inflammatory responses, and protection against pathological oxidation (61).Therefore, the potential mechanism underlying the observed association between the CETP rs708272 polymorphism and a reduced risk of CAD might be due to the increase of HDL-C concentrations attributed to the rs708272 G-to-A mutation.Based on the findings mentioned above, our speculation was that CETP inhibitors might prevent the development of CAD.Although some studies showed promising effects, including a modest increase in HDL cholesterol levels and a potential reduction in LDL cholesterol and triglyceride levels, the translation of these lipid-modifying effects into significant improvements in cardiovascular outcomes such as the decrease in the incidence of CAD has been inconsistent (62)(63)(64)(65).It is possible that HDL-C concentrations may not fully reflect the functionality and diversity of HDL particles, since they primarily served as a general measure of cholesterol carried by HDL.Subgroup analyses was evaluated to examine the potential effects of CAD sub-type, genotyping techniques, source of control, sample size, gender, and age on the connection between genetic polymorphisms and CAD.The findings of the subgroup analysis revealed a significant association between the CETP rs708272 polymorphism and a decreased risk of CAD in almost all subgroups.It's worth noting that in case and control groups predominantly composed of females, no significant association was found.However, due to the limited number of studies and  small sample sizes, these conclusions are not reliable.During the analysis of articles, we observed that the majority of studies include case and control groups with a significantly higher number of males than females, and gender differences were not further analyzed.This suggests that future researchers should pay more attention to the impact of gender.However, no significant association of the CETP rs5882 and rs1800775 polymorphisms with the risk of CAD was found, which was in contradiction with the finding reported by the previous review.Subgroup analysis revealed no significant association between these two SNPs and CAD in most subgroups.It was worth noting that rs1800775 polymorphism was associated with a reduced risk of CAD in studies using large samples, while it was associated with an increased risk of CAD in studies using small samples.More accurate results might be obtained including more studies with larger sample sizes.
No significant correlation was identified between the five uncommon CETP gene polymorphisms and CAD.However, the reliability of these results was limited due to the inclusion of only 2-3 studies for each SNP.
SNPs are closely associated with the occurrence and progression of many diseases (66).The association between SNPs and specific diseases might help identifying the role of genetic factors in disease development, providing a basis for an early   Forest plot of the meta-analysis between the CETP rs708272 polymorphism and HDL-C concentrations under the homozygote model (AAvs.GG).
Funnel plot of the meta-analysis between the CETP rs5882 polymorphisms and CAD under the allele model (Avs.G).
Funnel plot of the meta-analysis between the CETP rs708272 polymorphism and CAD under the allele model (Avs.G).diagnosis, prevention, and personalized treatment (67).This metaanalysis revealed a significant association between the CETP rs708272 polymorphism and CAD.Furthermore, it is worth considering the potential for early intervention strategies aimed at individuals carrying the CETP rs708272 risk allele, with the goal of preventing CAD.Our results might help in the determination of disease etiology, as well as improve early intervention to delay or prevent the onset of CAD.
Our current meta-analysis has certain potential limitations.Firstly, studies published in English were the only included, resulting in an insufficient sample size and a limited number of studies for some SNPs.Secondly, a significant heterogeneity was found in our study, potentially compromising the robustness of the findings.Thirdly, only the relationship between each SNP and CAD was investigated due to the lack of sufficient raw data, without examining the interactions between SNPs.While these limitations are acknowledged, it is also prudent to consider the potential impact of publication bias.Publication bias, whereby studies with significant or positive findings are more likely to be published, could introduce a bias in our meta-analysis results.Although we made efforts to minimize this bias through a comprehensive literature search, it remains a potential limitation that should be recognized.

Conclusion
The CETP rs708272 polymorphism is significantly associated with a lower risk of CAD and higher HDL-C concentration.However, the CETP rs5882 and rs1800775, rs4783961, rs247616, rs5883, rs1800776, and rs1532624 do not show any significant association with CAD.

FIGURE 4 Forest
FIGURE 4Forest plot of the meta-analysis between the CETP rs5882 polymorphism and CAD under the allele model (Avs.G).

FIGURE 5 Forest
FIGURE 5Forest plot of the meta-analysis between the CETP rs1800775 polymorphism and CAD under the allele model (Avs.C).

TABLE 1
Main characteristics of the selected studies of CETP gene polymorphisms.

TABLE 4
Overall analyses of the association of the three uncommon CETP gene polymorphisms with CAD.

TABLE 5
Selected studies including the data of the HDL-C concentrations across the CETP rs708272 polymorphism genotypes.

TABLE 6
Overall and subgroup analyses of the association of the CETP rs708272 polymorphisms with CETP and lipids levels.