The CETP TaqIB Polymorphism Is Associated With the Risk of Sudden Death in Type 2 Diabetic Patients

The design and results of the DIABHYCAR study have been reported (17–19). Briefly, DIABHYCAR was a multicentric, random, double-blind, parallel group trial to compare the cardiovascular and renal outcomes of patients taking ramipril (1.25 mg/day) and those taking placebo, in addition to their usual treatment (both groups). The participants were men or women with type 2 diabetes, aged ≥50 years, with serum creatinine ≤150 μmol/l, and elevated urinary albumin excretion (UAE) (≥20 mg/l, two times consecutively). The investigators examined the participants every 6 months for at least 3 years. The mean duration of follow-up was 4 years. The low dose of ramipril was found to be ineffective (19).

For logistical reasons, only French participants (3,124 of the 4,912 participants) were included in this study. Incident myocardial infarction was defined as the first occurrence of a fatal or nonfatal myocardial infarction after the baseline examination. Sudden death was defined as death occurring instantaneously or within 1 h after the onset of new cardiac symptoms (arrhythmia or other cardiovascular causes) or nonwitnessed death, where the body of the deceased was found, and no cause could be discovered. Fatal stroke and myocardial infarction were not included in this group. CHD was defined as the combination of myocardial infarction and sudden death. However, as some people with myocardial infarction died from sudden death, the number of CHD patients is not the sum of myocardial infarction and sudden death groups. An independent adjudication committee without access to the genotyping data evaluated the events (18). The study design was approved by the Angers University Ethics Committee. All participants provided written informed consent.

The CETP TaqIB SNP was genotyped using a PCR molecular beacon technique (20). The PCR was performed in a 96-well microtitration plate. A total of 200 ng of DNA was amplified in a total volume of 25 μl containing 20 pmol of 5′ and 3′ primers, 0.2 mmol/l dNTPs, 4 mmol/l MgCl₂, 10 mmol/l Tris-HCl (pH 8.3), 5 pmol of each allele-specific molecular beacon, and 1 unit of Taq polymerase (gold Taq; Perkin-Elmer, Paris, France). DNA denaturation and Taq activation were performed at 95°C for 10 min in a thermocycler (PTC-200; MJ Research, Watertown, MA) followed by 40 cycles of 20 s at 55°C, 10 s at 72°C, and 10 s at 95°C. After a final denaturation at 95°C for 2 min, hybridization with the probes was performed at 60°C for 1 min. Fluorescence emission was recorded with a plate fluorometer (FLUOstar; BMG Labtech, Offenburg, Germany) using two wavelength systems: 480–520 nm for fluorescein and 520–590 nm for tetramethylrhodamine. The amplifiers and allele-specific probes were synthesized by Eurogentec (Seraing, Belgium). All of the sequences of primers and probes are available from the authors on request.

Quantitative variables are described by means ± SD or geometric means (95% CI) if the distribution was skewed. The association of CETP TaqIB genotypes with baseline characteristics, not including lipids, was tested by ANOVA (continuous variables) or a χ² test (categorical variables) in univariate analysis. The association of CETP TaqIB genotypes with baseline lipid levels was tested by ANCOVA, adjusting for potential confounding variables (age, sex, BMI, A1C, smoking, and alcohol). We defined survival time as the period from the date of entry into the study to the date of the first event (myocardial infarction or sudden death) or the end of the study. Cox regression was used to estimate the hazard ratio; we adjusted for age, sex, prior myocardial infarction, systolic blood pressure, A1C, C-reactive protein (CRP), UAE, serum creatinine, total and HDL cholesterol, smoking, diabetes duration, BMI, and triglyceride levels and for the use of drugs at baseline: different antidiabetes and lipid-lowering treatments, treatments for hypertension, and platelet antiaggregants. We also adjusted for insulin and ACE inhibitor treatments, which could only be introduced during the follow-up (treatments by insulin and/or an ACE inhibitor at entry were exclusion criteria). The distribution of the use of the different drugs did not differ among genotypes. There was no interaction effect between genotype and the use of drugs. No interaction effect was found between sex and genotype, either on lipid levels or on CHD events. Therefore, data are not presented separately according to sex. All calculations were performed using SYSTAT 11 for Windows.

Table 1 provides a summary of features and clinical and lipid profiles at baseline of the DIABHYCAR population as well as the stratified data according to combined CHD incidence (myocardial infarction or cardiovascular death). Subjects with incident CHD during follow-up differed in age; total, LDL, and HDL cholesterol and triglyceride levels; blood pressure; known duration of diabetes; prevalence of prior coronary antecedents (myocardial infarction and angina pectoris); serum CRP; serum creatinine; and UAE. The sex ratio was similar in the two groups. Genotype and allele distributions of the TaqIB SNP are shown in Table 2. The genotype frequencies were in Hardy-Weinberg equilibrium. None of the variables (distribution or levels), including prior coronary events, differed according to CETP Taq1B genotypes (data not shown) except lipid levels (Table 2). The CETP TaqI B1 allele was associated with low levels of HDL cholesterol (P < 0.001), with a gene dose effect (Table 2).

The CHD incidence (myocardial infarction and/or sudden death) according to the CETP genotype is shown in Table 3. There was a higher percentage of B1B1 subjects in the CHD group than in the group without CHD. Cox survival analysis indicated that B1B1 homozygotes had more risk than carriers of the B2 allele, independent of other risk factors, including many variables listed in Table 1 (age, A1C, systolic blood pressure, diabetes duration, prior myocardial infarction, total plasma cholesterol, triglycerides, serum CRP, serum creatinine, and UAE). There was no interaction between genotype and the presence of prior coronary events on the incidence of CHD. Nevertheless, the association was still significant when we excluded subjects with prior myocardial infarction.

After further adjustment for HDL cholesterol levels, the association remained significant and only slightly lower than that without the adjustment. Sudden death was a major factor in this association (Table 3). The 4-year overall and sudden death–free survival rates were 91 and 95% for B1B1 and B2 carriers, respectively. The B1B1 genotype was also more frequent in the “incident myocardial infarction” group (Table 3), but the association was not significant. This could be due to a lack of power of the study because there were so few cases of incident myocardial infarction.

We report a prospective study based on a large cohort of French type 2 diabetic patients with micro- or macroalbuminuria and found, for the first time, a strong negative association between the B2 allele and CHD incidence and particularly sudden death. Although the CETP TaqIB polymorphism was associated with HDL cholesterol in these patients, the association with the occurrence of sudden death remained highly significant after adjustment for most potential confounders, including age and HDL cholesterol.

The association of the CETP TaqIB polymorphism and HDL cholesterol levels is widely recognized in many different populations, including type 2 diabetic patients. The B1 allele has also been associated with a higher prevalence of CHD (11–13). Nevertheless, there is a lack of consistent correlation between the CETP genotype and the clinical outcome. This association may depend on gene-gene or gene-environment interactions (21). A recent meta-analysis of large studies (11) showed that CETP TaqIB was associated with cardiovascular risk and that this relationship was mediated by lower HDL cholesterol levels. In contrast, we show that HDL cholesterol modified the risk associated with the TaqIB genotype only slightly. One reason could be the interaction with the diabetic phenotype. Studies of type 2 diabetic populations mostly show an association between CETP and cardiovascular disease (9,12,22,23). HDL cholesterol levels do not greatly modify the relationship between the CETP genotype and the clinical outcome. This finding could be specific to type 2 diabetic subjects in whom HDL cholesterol levels are already lower than those in nondiabetic subjects.

The most striking result from our study is the association with sudden death. There have been only a few prospective studies of CETP TaqIB polymorphisms. Blankenberg et al. (24) investigated the association between the risk of fatal cardiovascular events in 1,211 patients with coronary artery disease and −629C>A, an SNP in near complete positive linkage disequilibrium with the TaqIB polymorphism (10). Mortality was lower for carriers of the minor A allele (4–4.6%) (the one linked to TaqIB2 allele) than for CC homozygotes (10.8%); this was not the case for other cardiovascular outcomes. This allele was also associated with higher HDL cholesterol levels and lower CETP activity, but, similarly to our results, the strong protective effect on future cardiovascular mortality was independent of its effect on HDL cholesterol levels. The TaqIB polymorphism in cases of micro- and macroalbuminuria has been associated with atrial fibrillation (25). This association is very relevant to our findings for type 2 diabetic patients with micro- or macroalbuminuria as atrial fibrillation is associated with an increase in cardiovascular mortality (26,27).

The B2 allele has been associated with lower levels and activity of CETP, which leads to higher HDL cholesterol levels. However, there is evidence that CETP inhibition goes beyond raising HDL cholesterol levels alone for the purpose of prevention of cardiovascular risk (28,29), and our study supports this hypothesis. Nevertheless, the Investigation of Lipid Level Management to Understand Its Impact in Atherosclerotic Events (ILLUMINATE) trial with a CETP inhibitor, torcetrapib, was interrupted because of a higher mortality rate in the torcetrapib and atorvastatin group than in the atorvastatin group (30). In another trial (Investigation of Lipid Level Management Using Coronary Ultrasound to Assess Reduction of Atherosclerosis by CETP Inhibition and HDL Elevation [ILLUSTRATE]), there was no significant difference in the atherosclerotic plaque burden in the two study groups (atorvastatin with or without torcetrapib) (31). In both of these trials, the use of torcetrapib was associated with increased blood pressure. In our study, the TaqIB polymorphism was not associated with blood pressure (data not shown). Therefore, our results indirectly suggest that the problem may have been due to an adverse effect of the drug, unrelated to the CETP inhibition, as already suggested (32).

We did not observe associations between the CETP TaqIB SNP and prior myocardial infarction. We think that the prospective design of the DIABHYCAR study allowed us to avoid the biases of cross-sectional/case-control studies, which can lower the power to detect associations, but such biases could explain why we observed only an effect on the CHD incidence but not on the prevalence at entry.

In summary, we have demonstrated that the TaqIB polymorphism of the CETP gene is a good genetic predictor of CHD complications of type 2 diabetes and especially cardiac mortality. This association is not fully explained by the effect of TaqIB SNP on HDL cholesterol levels.

The DIABHYCAR study group included the following: Principal Investigator: Prof. M. Marre (Bichat Hospital, Paris, France); Steering Committee: F. Alhenc-Gelas, J.P. Boissel, F. Cambien, S. Etienne, A. Girault-Louvel, P. Gueret, M. Lièvre, J. Mann (Vice-Chairman), M. Marre, J. Ménard, P. Passa (Chairman), P.F. Plouin, D. Vasmant, L. Vaur (Secretary), G.C. Viberti, and C. Weisselberg; Central Co-ordinating Center: J.P. Boissel and M. Lièvre; Executive Committee: J.P. Boissel, V. Bost, M. Cambien, Y. Gallois, N. Genes, J. Gillet, M. Hervé, M. Lièvre, M. Marre, L. Martin, A. Perret-Hantzperg, P.F. Plouin, and L. Vaur; Biological Committee: F. Alhenc-Gelas (Chairman), F. Cambien, A. Girault-Louvel (Vice-Chairman), M. Lièvre, M. Marre, and J. Ménard; Central End Point Committee: E. Bonnefoy, G. Chatellier (Chairman), T. Moreau, and L. Pinède; and Independent Data and Safety Committee: E. Eschwege, C.E. Mogensen, N. Victor, and S. Weber.

This work was supported by a grant from the Association Française des Diabétiques.

Parts of this study were presented as an oral communication at the 66th annual meeting of the American Diabetes Association, Washington, DC, 9–13 June 2006.