Introduction
The main controversy in the area of lipid management today is related to the usefulness of 'non-statin’ agents to maximize cardiovascular risk reduction in particular patient types, such as those with diabetes and metabolic syndrome. Fibrate drugs, such as gemfibrozil (Lopid) and fenofibrate (TriCor), are undoubtedly the best tools to address the condition characterized by high triglycerides, low high-density lipoprotein (HDL), and small dense low-density lipoprotein (LDL) (atherogenic dyslipidemia), but the most recent guidelines from both the American Diabetes Association (ADA) and the National Cholesterol Education Panel continue to focus on LDL control as the target of therapy. However, statin-based interventions are unlikely to correct problems of triglycerides and HDL, whereas the use of fibrates has a stronger effect on the atherogenic dyslipidemia and might even produce significant LDL reduction in some patients. Thus, physicians dealing with the common phenotype of insulin resistance are presented with a practical problem - the use of statins primarily to control LDL, or the use of fibrates primarily to control triglycerides and HDL. Well-designed outcome studies published in the last few years have shown the value of fibrate therapy in patients with obesity, the metabolic syndrome, and diabetes, particularly when LDL levels are below 130mg/dl and when triglycerides and HDL levels are only modestly abnormal.
Mechanism of Action of Fibrates
Gemfibrozil and fenofibrate are the fibrates currently approved for use in the US and, in addition to these, bezafibrate and ciprofibrate are also available in Europe. The fibrates have been in use since the late 1960s, and for about 25 years their mode of action was not known. The relatively recent understanding of the molecular mechanisms of fibrate action1 represents one of the biggest breakthroughs in cardiovascular pharmacology. Fibrates are now known to alter the transcription of several genes involved in lipoprotein metabolism and other pathways.2 Fibrates are able to activate gene transcription because they are synthetic ligands for peroxisome proliferator-activated receptor (PPARα), a ligand-activated transcription factor and member of the nuclear hormone receptor superfamily. PPARα transmits signals from lipid-soluble factors, such as fatty acids, eicosanoids, hormones and vitamins, to genes in the nucleus by binding to DNA within specific response elements (PPRE).1
Effects of Fibrates on Plasma Lipids
A number of factors influence the response of plasma lipid levels to treatment with fibrates, including the baseline lipoprotein profile, the genetic and environmental factors underlying the hyperlipidemia, and the drug used. Fibrates are clearly the drug of choice for treatment of severe hypertriglyceridemia (TG >500mg/dl) or chylomicronemia syndrome (TG >1,000mg/dl), conditions associated with an increased risk of pancreatitis.3 Patients with severe hypertriglyceridemia frequently have low levels of LDL cholesterol, and treatment with fibrates may raise their LDL cholesterol levels by increasing intravascular lypolysis of very low-density lipoproteins (VLDL) through lipoprotein lipase, with resultant accumulation of newly formed LDL ('beta-shift’ phenomenon). However, in the setting of more moderate hypertriglyceridemia (TG<500mg/dl), fibrates produce 30% to 50% reductions in serum triglycerides, 15% to 25% increases in HDL cholesterol, and have variable effects on LDL depending on the molecule used. Gemfibrozil has neutral LDL effects, whereas fenofibrate may produce LDL reductions ranging from 5% to 35%.4 An important aspect of LDL metabolism in patients with triglyceride and HDL problems is the accumulation of small, dense LDL. Fenofibrate decreases small, dense LDL particles, which are more susceptible to oxidation and more 'atherogenic’ than larger, more buoyant LDL particles.5,6 Several non-traditional risk factors are also influenced by fibrates. Fenofibrate decreases plasma levels of Lp(a) by 7% to 23%,4,7 reduces fibrinogen,8 and lowers levels of serum uric acid to the point that it may have therapeutic effects on gout.9
Effects of Fibrates on the Vessel Wall
The pleiotropic effects of fibrates may result in direct anti-atherogenic effects in the artery wall. PPARα is expressed by all of the major cell types in atherosclerotic lesions, including macrophages, endothelial cells, and vascular smooth muscle cells (SMC).10 PPARα may act as a negative regulator of the vascular inflammatory gene response by antagonizing the activity of the transcription factors nuclear factor-κb (NFκB) and activating protein- 1 (AP-1).11,12 Fibrates have been reported to inhibit tumor necrosis factor (TNFα)-induced expression of vascular cell adhesion molecule-1 (VCAM-1) by endothelial cells.13 Therefore, fibrates may decrease the recruitment and adhesion of mononuclear cells to endothelial cells, a crucial step in the initiation of atherosclerotic lesions.
Activation of PPARα by fibrates has also been reported to influence cholesterol homeostasis inside the macrophage by inducing expression of ATP-binding cassette transporter (ABCA1) and scavenger receptor - class B (SR-B1), drivers of cholesterol efflux out of the cell.7 Furthermore, fenofibrate inhibits activation of vascular smooth muscle cells14 and lowers C-reactive protein (CRP) levels to a degree similar to that previously reported for statin therapy.15,16
Combination Therapy of Fibrates with Statins
The risk of clinically important myositis and rhabdomyolysis (defined as muscle pain with creatinine phosphokinase (CPK) levels >10 times the upper limits of normal) is the major concern in lipid-lowering therapies including the use of statins as monotherapy or in combination regimens. Recent evidence indicates that gemfibrozil causes increased levels of statins in the blood.17,18 In contrast, fenofibrate appears to have little effect on the pharmacokinetic properties of simvastatin, atorvastatin, cerivastatin, or rosuvastatin, which may explain why there are fewer reports of significant interactions between fenofibrate and statins.18 As a consequence, the ADA has issued new recommendations that give preference to fenofibrate over gemfibrozil in combination with statins. Support to this position was given by the LDS trial, stopped in 2001 because of the withdrawal of cerivastatin from the market, which showed the absence of toxicity from fenofibrate taken in combination with the most dangerous of statins.
Fibrates for Primary Prevention of CHD
In the Helsinki Heart Study,19,20 more than 4,000 men with moderate coronary heart disease (CHD) risk were enrolled to receive 1,200mg of gemfibrozil or placebo for five years. Gemfibrozil reduced triglycerides by 35% and LDL by 11%, while raising HDL by 11%.The effect on outcomes was a very significant 34% reduction in CHD events. Interestingly, intervention with gemfibrozil was more beneficial in subjects with high triglycerides and low HDL than in the treated group as a whole (71% versus 34% CHD risk reduction). Also, diabetic subjects benefited more than the normoglycemic individuals from treatment with gemfibrozil (68% versus 34% CHD risk reduction).
In line with these results, overweight subjects (body mass index (BMI) >26) experienced more risk reduction from gemfibrozil compared with lean subjects (BMI <26), whereas the most striking risk reduction was produced by overweight or obese subjects (BMI >30) with high triglycerides and low HDL.21 These data support the notion that the patient type most amenable to cardiovascular (CV) risk reduction by fibrate therapy is an overweight patient, with metabolic syndrome or diabetes, and the atherogenic dyslipidemia.
Fibrates for Secondary Prevention of CHD
The Veterans Administration HDL intervention trial (VA-HIT) evaluated the effect of gemfibrozil in CHD patients with type 2 diabetes or the metabolic syndrome and low HDL.22 The baseline lipids included: LDL 111mg/dl, HDL 32mg/dl, and triglycerides 164mg/dl. Five years of therapy with gemfibrozil resulted in an average 8% increase in HDL, a 24% reduction in triglycerides, and no changes in LDL levels.This was accompanied with a very significant reduction in coronary and cerebrovascular events (22% and 31%, respectively).23 The intervention produced a number needed to treat (NNT) of about 20 for coronary events, which compares favorably with the results of the statin trials. In addition, the effect of fibrate therapy on CHD rates among the nearly 700 diabetic subjects enrolled in this study was particularly large and apparently superior to the effects of statins in the same patient type (see Figure 1). Among the non-diabetic subgroup of this study, the best predictor of CHD risk reduction afforded by fibrate therapy was fasting plasma insulin level, with subjects in the lowest quartile experiencing no benefits, and those in the highest quartile experiencing the most significant benefits.24 These data support the value of fibrate therapy in the metabolic syndrome patient.
However, the enthusiastic endorsement of fibrate therapy that could be derived from the VA-HIT results is partly tempered by the recent results of the Bezafibrate Infarction Prevention (BIP) Trial. In this study, 3,122 CHD subjects with LDL of 150mg/dl and HDL of 34mg/dl were treated with bezafibrate or placebo for five years and evaluated for lipid changes and CHD event rates. Despite significant changes in triglycerides and HDL, bezafibrate failed to produce significant CHD reduction in the overall population.25 However, bezafibrate was very beneficial for the subjects who had hypertriglyceridemia, and LDL levels approximately 130mg/dl, at baseline.
In this group, risk reduction started during the first year of the study and reached an impressive 40% after five years.25 The Diabetes Atherosclerosis Intervention Study (DAIS) has tested the hypothesis that fibrate therapy can help diabetic patients with a baseline LDL around 130mg/dl. In this three-year study, fenofibrate treatment was accompanied by a 40% reduction in progression of focal coronary artery disease (CAD) and a 23% reduction in the rate of CV events, including myocardial infarctions (MIs), compared with placebo.26 Interestingly, these effects appeared to be explained not only by the changes in HDL, triglyceride, and LDL levels,but also by the significant increase in LDL particle size induced by fenofibrate.27
Fibrates Versus Statins in the Treatment of the Atherogenic Dyslipidemia
The new guidelines of the National Cholesterol Education Program (NCEP) and the ADA highlight the importance of LDL reduction in high-risk patients, but at the same time encourage physicians to position all diabetic and insulin-resistant patients in the high cardiovascular risk category.28-29 Because these patients are commonly affected by a dyslipidemia characterized by hypertriglyceridemia with low HDL, it could be argued that the best lipid intervention in these cases would be one directed at the primary metabolic abnormality.
The importance of the atherogenic dyslipidemia can also be indirectly inferred by the results of the major statin trials. For example, the most impressive statin effects on CHD risk reduction in a high-risk population were seen in the 4S trial, where patients had high LDL but normal triglycerides and HDL.31 The equivalent pravastatin studies, Cholesterol and Recurrent Events (CARE) and Long-term Intervention with Pravastatin in Ischaemic Disease (LIPID), where subjects were enrolled with triglycerides as high as 350mg/dl and a baseline LDL between 140 and 15mg/dl, showed more modestly positive results, which suggests that, in a population with different forms of dyslipidemia, the exclusive attention to LDL may produce fewer benefits compared with a more comprehensive or targeted lipid management approach.32,33 It is also important to note that in the CARE study, subjects with starting triglycerides above the median value did not experience significant risk reduction despite a significant LDL-lowering effect.32
The recently published Heart Protection Study (HPS) investigated the risk reduction potential afforded by simvastatin in a population of 20,000 subjects with high risk (75% CHD, 25% diabetes or multiple risk factors) and 'normal’ lipid levels (51% of subjects had an LDL below 130mg/dl).34 The most impressive benefit in this trial was experienced by people who had low baseline HDL.The effect of treatment in subjects with a baseline HDL lower than 35mg/dl was a 4.4% absolute risk reduction and NNT of 24, compared with a 2.7% absolute risk reduction and NNT of 37 in the subjects with the highest baseline LDL (above 130mg/dl).34 Even though these data seem to support the use of statins in low HDL patients, it is important to note that the residual risk in HPS patients with low HDL after treatment was higher than that in any other HPS lipid sub-category, suggesting that greater risk reduction could have been achieved in this population through aggressive control of HDL (and perhaps triglycerides, i.e. through fibrate therapy).
Table 1 shows the effect of statins or gemfibrozil in diabetic patients in 4S, HPS, CARE, and VA-HIT. It is evident that an efficient risk reduction is not simply related to drug choice (simvastatin produced very different effects in 4S versus HPS), but is mostly the result of matching the right drug with the right patient type (i.e. a statin for the hypercholesterolemic patients of 4S, a fibrate for the hypertriglyceridemic/low HDL patients of VA-HIT). Whether the diabetic patients of CARE and HPS would have been better served by fibrate therapy remains a matter of speculation.
The question arises, of course, of whether combination therapy with statin and fibrate would produce additive or synergistic effects and greater risk reductions compared with monotherapy with either drug in high-risk patients. Such a question is being investigated in the NIH-sponsored ACCORD trial, designed to compare the effect of simvastatin monotherapy with the combination with fenofibrate in diabetic patients (www.accordtrial.org). The first results from this study should be expected in 2010.
Conclusions
Fibrates meet the criteria of excellent drugs for cardiovascular prevention and are appropriate for the long-term management of dyslipidemia in patients with diabetes and the metabolic syndrome. In addition, the knowledge that fibrates reduce the risk of CV events 3 without lowering cholesterol levels suggests the possibility for pleiotropic effects of PPARα activation on the vasculature, and thus proposes an evolution of the therapeutic potential of this class of drugs remarkably parallel to that of the HMG-CoA reductase inhibitors.
These observations argue against the use of any one class of drugs in the management of high-risk dyslipidemia, and support a practice stance of matching a specific diagnosis of lipid abnormality with the therapeutic agent most likely to correct it.< p/>
The importance of such a position is evident for the common subjects with very high CVD risk and moderate combined dyslipidemia, i.e. the obese, metabolic syndrome, and diabetic patients. The current guidelines encourage aggressive lipid lowering in patients with the metabolic syndrome and diabetes, but one should keep in mind that these patients present with different forms and degrees of dyslipidemia. There are obvious scenarios where the drug choice is limited. For example, a diabetic patient with an LDL of 150mg/dl or higher will need statin therapy irrespective of triglyceride or HDL levels. Likewise, a diabetic subject with triglycerides higher than 500mg/dl will need fibrate coverage irrespective of LDL levels. However, a large number of these patients fall into the gray zone of moderate combined dyslipidemia, with triglycerides below 500mg/dl, HDL in the 30s, and LDL ranging between 100 and 130mg/dl. In these subjects, a more targeted approach to therapy may be warranted based on the specific lipid abnormality.