Insights from VOYAGER on statin-induced HDL response

Findings from the VOYAGER meta-analysis have provided insights into the effects of statins on high-density lipoprotein cholesterol (HDL cholesterol). The most important predictor of statin-induced HDL cholesterol response was the pre-treatment level of HDL cholesterol. 

 

Barter PJ, Brandrup-Wognsen G, Palmer M, Nicholls SJ. Effect of statins on HDL: a complex process unrelated to changes in LDL: Analysis of the VOYAGER Database.  J. Lipid Res. published December 2, 2009, doi:10.1194/jlr.P002816

 

The data were reviewed by HDL Forum Editor Professor Philip Barter, Heart Research Institute, Sydney, Australia.

 

While all statins raise HDL cholesterol, the mechanism of this effect and factors responsible, are not clear. To address these questions, the authors evaluated results from an individual patient meta-analysis database of 32,258 patients enrolled in 37 clinical trials who were treated with rosuvastatin, atorvastatin or simvastatin. All trials included in this analysis used fixed doses of these statins, given for at least 4 weeks, and had baseline and on-treatment lipid values recorded for each patient.  All lipid values were measured in the fasting state.

 

While all three statins raised plasma levels of HDL cholesterol, the magnitude of the response varied between different statins (Fig 1.)

 

 

For both rosuvastatin and simvastatin there was a dose-dependent increase in HDL cholesterol, whereas for atorvastatin, the increase in HDL cholesterol was inversely related to dose. Whether these differences in HDL cholesterol response impact on the cardioprotective properties of the individual statins is not definitively known, although there is growing evidence that statin-induced increases in HDL cholesterol impact favourably on atheroma burden (1). Variations in the balance between mechanisms that increase or decrease HDL cholesterol concentration might explain the different pattern of HDL cholesterol response with different statins.

 

With respect to apolipoprotein A-I (apoA-I), the main protein component of HDL, increases were of similar magnitude across the dose range for rosuvastatin (between 6% and 6.5%) or paralleled the HDL cholesterol dose-dependent changes for simvastatin (direct relationship, ranging from 5.2% with 10 mg to 6.4% with 40 mg) or atorvastatin (inverse relationship, ranging from 4.7% with 10 mg to 0.1% with 80 mg).  Across all doses of all statins, statin-induced changes in HDL cholesterol correlated significantly with changes in apoA-I.  In contrast to previous reports, the percentage increase in apoA-I was comparable with that of HDL cholesterol for all three statins.

 

Statin-induced changes in HDL cholesterol were unrelated to pre-treatment levels or changes in low-density lipoprotein (LDL) cholesterol.

 

In contrast, pre-treatment concentrations of HDL cholesterol and triglycerides, as well as the presence of diabetes were predictive of statin-induced HDL cholesterol response. The strongest predictor was the pre-treatment level of HDL cholesterol. Patients with pre-treatment HDL cholesterol levels <39 mg/dL (1.0 mmol/L) showed the greatest increase in HDL cholesterol levels (Fig 2). This could explain much of the observed variability in HDL cholesterol response in different statin studies.

 

 

Additionally, the higher plasma triglycerides were at baseline, the greater the increase in HDL cholesterol at both low and high doses of statin (Fig 3). This relationship was robust even after adjusting for baseline HDL cholesterol levels. Given that reduction in triglyceride-rich lipoproteins is accompanied by a decrease in cholesteryl ester transfer protein (CETP) mediated transfer of cholesteryl ester from HDL to triglyceride-rich lipoproteins, this relationship is probably not surprising.

 

 

However, patients with diabetes had substantially smaller statin-induced HDL cholesterol responses, despite having lower pre-treatment levels of HDL cholesterol than non-diabetics. For example, for the rosuvastatin 10 mg group, the statin-induced HDL cholesterol response was 6.7% for non-diabetics vs. 4.06% for diabetics (p<0.0001, multivariate analysis). This finding clearly warrants further investigation.

 

The authors concluded that the statin-induced HDL cholesterol response varies according to individual statins.  The most important predictor of this response was the pre-treatment level of HDL cholesterol.

 

Reference

1. Nicholls SJ, Tuzcu EM, Sipahi I et al. Statins, high-density lipoprotein cholesterol, and regression of coronary atherosclerosis. JAMA 2007; 297: 499–508.

 


Download Attachment