Abstract
There are several hypotheses regarding the mechanisms underlying high residual risk in statin-treated diabetic patients. The observation that high plasma concentrations of triglycerides (TGs) and low high-density lipoprotein cholesterol (HDL-C) predict high residual risk, even when LDL-C concentrations reach values below 70 mg/dL, has suggested that insufficient control of TGs and HDL-C in statin-treated diabetics accounts for much of the residual risk. Following this logic, the clinical trend has been to add lipid-lowering agents known to improve plasma TGs and HDL-C concentrations to statins therapy. However, evidence from randomized clinical trials for effectiveness and safety of such approach has been lacking.
The recently published results of the ACCORD LIPID trial 6 provide us with some information that fills the knowledge gap. This trial was supported by the National Institutes of Health/National Heart, Lung and Blood Institute (NIH/NHLBI) to compare combination therapy of fenofibrate plus simvastatin with simvastatin monotherapy in over 5,000 patients with T2DM. During an average follow up of 4.7 years, the following was observed: (1) The combination therapy was safe but did not significantly reduce first occurrence of CVD death, nonfatal myocardial infarction (MI), or nonfatal stroke as compared to simvastatin monotherapy. (2) A reduction in primary end point was found with the combination therapy in the subgroup of patients with high TGs and low HDL-C (both for males and females). (3) There was heterogeneity in treatment according to gender, with a trend toward increased events with combination therapy in the women. These three points deserve careful consideration in drawing our conclusions for clinical implications of the ACCORD LIPID trial.
The results for the overall population are clear. Adding fenofibrate to statins in T2DM patients who have median LDL-C 96 mg/dL, HDL-C 38 mg/dL, and TGs 164 mg/dL (baseline values for the fenofibrate group) is of no benefit for CVD outcome. The changes in lipid profile induced by fenofibrate combination as compared to statin monotherapy were small. The largest numeric difference was found for the plasma TG concentrations (22 mg/dL lower median in the fenofibrate combination arm). Although statistically significant, median HDL-C was only 1 mg/dL higher and the total cholesterol 4 mg/dL lower in the fenofibrate combination arm as compared to statin monotherapy at the end of study. Larger lipid differences were noted only within the first 4–12 months of the study. A more significant difference in TG and HDL-C concentrations was found between simvastatin monotherapy and simvastatin + fenofibrate in patients with combined hyperlipidemia in the SAFARI trial. 7
SAFARI was an efficacy trial that also demonstrated larger reduction in concentrations of both apolipoprotein B (ApoB) and non-HDL-C (the two parameters were very closely related in this study) when fenofibrate is combined to statin as compared to statin only. 8 It is known that in patients with hypertriglyceridemia, non-HDL-C, or ApoB are better predictors of risk than LDL-C concentration. Also, lowering of non-HDL-C (or ApoB) is associated with improved outcome in randomized clinical trials. Therefore, the question is: Could higher response to fenofibrate in the hypertriglyceridemic subgroup have been associated with better outcome in the ACCORD LIPID trial? The observation that a benefit was detected in the subgroup of patients with the highest tertile of plasma TGs concentrations and the lowest tertile of HDL-C concentrations (>204 and <34 mg/dL, respectively) together with the results from the FIELD trial in the subgroup of patients with the metabolic syndrome 9 would support this view.
The question is then: Why did the study not exclude patients with low TG and high HDL-C. The inclusion criteria were specific for TG (<750 mg/dL if no treatment, <400 mg/dL if subjects were receiving lipid-lowering therapy) and HDL-C concentrations (<55 for women and blacks and <50 mg/dL for all other groups), but did not include criteria to adequately enrich the study population with patients who have high TG and low HDL-C. Only 15–20% of the overall T2DM population has TG >204 and HDL <34 mg/dL. 10 The authors believed that difficulty in enrollment and lack of applicability of the results for the whole T2DM population would have compromised the validity of the trial. The decision was therefore made not to exclude patients with relatively normal TG and HDL-C.
Taken together, the results of the ACCORD LIPID trial have important clinical implications. The results of this trial reaffirm that the majority of patients with T2DM should be treated with statins to decrease plasma concentrations of ApoB-containing particles. Statins are known to also improve TG and HDL-C concentrations. If abnormalities in these lipid fractions remain, combination with fenofibrate appears to be a safe approach that could determine additional reduction in cardiovascular risk. However, the strength of this effect awaits more definitive confirmation. In the meantime, the decision of whether to use combination therapy will have to be tailored to the specific patients and to first assure that compliance is not compromised. If combination therapy means potential reduction of compliance in patients who are already taking a large number of expensive and potentially interacting medications, the risk-to-benefit ratio for addition of fenofibrate is not acceptable on the basis of the available data. Reduction of excessive residual risk in statin-treated patients with T2DM will have to be addressed through coordinated management of coexisting risk factors. Most importantly, better control of CVD risk in patients with T2DM could be achieved by earlier identification of diabetes and earlier institution of comprehensive cardiovascular risk reduction. The approach of intensifying therapeutic goals is ineffective to further reduce risk and has the potential to promote reduced compliance with long-term therapy by increasing costs and drug interaction–related adverse events. Future research will have to define the benefit of early intervention with less intensive but comprehensive treatment goals. The combined results of the ACCORD and STENO II trials 11 have paved the way to this change in focus for CVD prevention strategies in the growing population of patients with T2DM.