HAPTOGLOBIN PHENOTYPE, HDL-CHOLESTEROL RAISING THERAPY (FIBRATE), AND RISK OF CARDIOVASCULAR DISEASE EVENTS WITHIN THE ACTION TO CONTROL CARDIOVASCULAR RISK IN DIABETES (ACCORD) LIPID TRIAL

Abstract

Background: The failure of many clinical trials aimed at preventing cardiovascular disease (CVD) by pharmacologically raising high-density lipoprotein (HDL) cholesterol has cast doubt on the idea that more HDL-cholesterol is better and that the function of HDL may be of more importance. The haptoglobin (Hp)2-2 phenotype (~40% of people) is associated with dysfunctional HDL that can become heavily oxidized in hyperglycemia. The effect of raising potentially pro-atherogenic HDL-cholesterol on risk of CVD in the Hp2-2 phenotype in hyperglycemia is unknown.

Objectives: 1) To determine whether the effect of adding fenofibrate therapy to simvastatin therapy on risk of coronary heart disease (CHD) and CVD events depends on Hp phenotype, and 2) to determine whether the association between HDL-cholesterol and cardiovascular events (CHD and CVD) depend on Hp phenotype, in type 2 diabetes mellitus (T2DM). These objectives were also examined in men and women and in primary and secondary prevention patients separately.

Methods: Haptoglobin phenotype was determined using a validated assay in 4,996 men and women with T2DM who participated in the ACCORD lipid trial with a mean follow-up of 4.7 years for CVD and 4.6 for CHD. In an intention-to-treat analysis (objective 1), multivariable Cox proportional hazards regression was used to determine the effect of fenofibrate therapy on CHD and CVD events for the two Hp phenotype groups separately. In a biomarker analysis with repeated measures over time (objective 2), multivariable Cox proportional hazards regression was used to determine the association between a 1-mg/dL increase in HDL-cholesterol and CHD and CVD events, stratifying by Hp phenotype.

Results: Fenofibrate with background simvastatin, compared to simvastatin alone, reduced the risk of CHD in Hp1 allele carriers but not in the Hp2-2 phenotype (p-value for interaction=0.009). The effects also differed by sex with a reduced risk of CHD in men who were Hp1 allele carriers and an increased risk of CHD in women with the Hp2-2 phenotype. There was a significant, inverse association between HDL-cholesterol and cardiovascular outcomes in Hp1 allele carriers but there was no significant association between HDL-cholesterol and cardiovascular outcomes in patients with the Hp2-2 phenotype, although the interaction was not significant.

Conclusion: The effect of adding fenofibrate to simvastatin on CHD and CVD risk may depend on Hp phenotype and sex. Further research is needed to understand both the relationship between fenofibrate therapy and cardiovascular events, and between HDL function and cardiovascular events in different Hp phenotypes.