Reports

This report is based on medical evidence presented at sanctioned medical congress, from peer reviewed literature or opinion provided by a qualified healthcare practitioner. The consumption of the information contained within this report is intended for qualified Canadian healthcare practitioners only.

JOURNAL CLUB

May 2008

Editorial Overview:

Daniel Gaudet, MD, PhD

Professor of Medicine Université de Montréal Montreal, Quebec

The recently published ENHANCE study defied expectations. The trial was conducted in patients with familial hypercholesterolemia, a relatively rare genetic predisposition for this disorder. Results demonstrated that additional reductions in plasma levels with the cholesterol absorption inhibitor ezetimibe relative to a statin alone did not produce additional regression in the intima media thickness of the carotid and femoral arteries. The clinical significance of this finding is unclear; however, this was a study of a surrogate marker of lipid-lowering benefits and should not produce any changes in current efforts to bring patients to LDL-C goals now firmly established in multiple event-driven studies conducted in the last few decades.

The ENHANCE (Ezetimibe and Simvastatin in Hypercholesterolemia Enhances Atherosclerosis Regression) study was not designed for nor does it have the power to change clinical management of lipid disorders in Canada. The trial evaluated a surrogate marker in a subgroup of high-risk patients with a familial predisposition to hyperlipidemia, the majority of whom were undergoing aggressive lipid-lowering therapy for several years prior to entering the study. From a clinical standpoint, this is an encouraging step, as significant efforts have been made to aggressively decrease LDL-C in patients with familial hypercholesterolemia in the last 30 years using different clinical strategies, alone or in combination. These include rigorous diet and lifestyle changes, ileocecal bypass, resins, fibrates, statins, plasmapheresis and/or cholesterol absorption inhibitors. As a consequence, compared to the situation in the 1970s and 1980s, patients with familial hypercholesterolemia now live significantly longer and better lives.

It should also be considered that the mean baseline intima media thickness (IMT) of the carotid arteries among ENHANCE participants was lower than expected and than what is usually observed in other studies involving patients with familial hypercholesterolemia. This finding might have had an influence on the results of the study but it also represents a surrogate illustration of the clinical benefits of previous treatment. Thus, although there has been a relationship between the degree of LDL-C lowering and change in a variety of surrogate markers of atherosclerosis progression in past studies, it cannot be assumed that lack of regression signifies an absence of clinical effect. For example, previous studies have established that plaque stability is a far more important predictor of thrombotic events than plaque size. Surrogate end points are often misleading, underscoring the critical importance of measuring events. Our experience with patients with homozygous familial hypercholesterolemia, an extremely severe form of this disorder (and more acute than the one affecting ENHANCE participants), teaches us that even in the presence of very severe IMT and atherosclerotic plaque, it is possible to improve survival and decrease the incidence of coronary events through a sustained effort in decreasing LDL-C. Although the event-driven trial testing ezetimibe is underway, there are large data sets supporting the current LDL-C goals, which remain unaffected by ENHANCE.

It would be relevant to explore ENHANCE results in greater depth and to compare them with those from other regression studies involving patients with familial hypercholesterolemia. In ENHANCE, 720 patients with this disorder were followed with B-mode ultrasonography for changes in carotid IMT (cIMT) after being randomized to simvastatin 80 mg monotherapy or combination simvastatin 80 mg/ezetimibe 10 mg (Kastelein et al. N Engl J Med 2008;358(14):1431-43). As predicted by previous studies with ezetimibe, which inhibits absorption of dietary cholesterol from the small intestine, LDL levels were reduced by a median of 16.5% on the combination relative to the statin alone (3.65 mmol/L vs. 4.98 mmol/L; P<0.01) (Figure 1). In addition, the combination was associated with a 25.7% (P<0.01) greater relative reduction in highly-sensitive C-reactive protein (hs-CRP) and a 6.6% (P<0.01) greater reduction in triglycerides than the statin alone (Figure 2).

Such a substantial relative reduction in LDL would have been expected to correspond with a regression in IMT of major arteries. A similar study by the same group, also conducted in patients with familial hypercholesterolemia, had previously associated greater LDL reductions with reduction in cIMT (Smilde et al. Lancet 2001;357(9256):577-81). In the ASAP (Atorvastatin vs. Simvastatin on Atherosclerosis Progression) study, IMT decreased by a median 0.031 mm from baseline (P=0.0017) in those who received atorvastatin 80 mg vs. a 0.036 mm median increase from baseline (P=0.0005) in those who received simvastatin 40 mg. However, this study also demonstrated a diminishing effect over time. In an extension study, the regression per year on the more aggressive lipid-lowering dose fell to 0.005 mm from the rate of approximately 0.015 mm observed over the first two years (van Wissen et al. Am J Cardiol 2005;95(2):264-6).

In ENHANCE, more than 80% of the enrolled patients had been treated previously with statins. There was a placebo run-in period during which all lipid-lowering treatments were discontinued. Following the run-in, baseline LDL-C levels were more than 8.0 mmol/L, confirming that patients had hypercholesterolemia off treatment. The primary outcome, which was the difference in cIMT, showed no significant difference (P=0.29) between the two groups despite the greater lipid lowering with the combination. Although this defied the primary hypothesis that greater lipid lowering with ezetimibe would produce a favourable change in IMT, it is notable that neither treatment strategy produced regression. This differs from the ASAP study, also involving patients with familial hypercholesterolemia, in which the more aggressive lipid-lowering treatment arm did produce regression.

Figure 1. ENHANCE: LDL-C Reductions

Since ENHANCE was not an outcome trial, the relevance of these results to preventing clinical events is unknown. There is substantial evidence to suggest that stabilization of plaques is more important than halting plaque growth. While there has been a correlation between doses of lipid-lowering strategies that have been shown to produce stabilization or regression of atherosclerotic plaques in studies with B-mode or intracoronary ultrasound and benefit in clinical event trials, the relationship may be more complex than expected. For example, regression, although an initial measure of plaque stabilization, may lose relevance in more aggressively controlled patients when components of the plaque, such as the fibrous cap, are less lipid-related and less susceptible to regression through lipid lowering. As suggested by the ASAP extension and other studies, the potential for plaque regression may be limited.

Findings from ENHANCE, ASAP, RADIANCE-1 (Rating Atherosclerotic Disease Change by Imaging with a New CETP Inhibitor trial) and other studies using imaging to evaluate the effect of lipid-lowering treatment on atherosclerosis in familial hypercholesterolemia are hypothesis-generating. They provide insight into the mechanisms of disease but cannot substitute for trials comparing treatment strategies for effect on outcome. It is on the basis of such outcome studies that evidence-based treatment guidelines depend. From this perspective, there are more than 25 years of data from large studies that have redefined LDL-C targets for both secondary and primary prevention. In Canada, the current guidelines recommend an LDL-C goal of <2.0 mmol/L in patients at high risk, defined as those who have atherosclerosis (or diabetes) or a 10-year risk of coronary artery disease greater than 20%.

Neither the goals nor the strategies for reaching these goals have been affected by results of ENHANCE. In Canada, statins have been and remain the first-line pharmacologic therapy in patients who cannot reach treatment goals on lifestyle changes alone. In those who cannot reach goals on maximally tolerated statins, adjunctive therapy is appropriate, including use of ezetimibe. Although we do not yet have evidence to confirm that ezetimibe reduces events when added to statins to further reduce LDL-C, we do have abundant evidence to support the LDL-C goals. Indeed, the level at which there is no further clinical benefit from LDL-C lowering in high-risk patients has not yet been defined but appears to be lower than the current target.

Importantly, IMPROVE-IT (Improved Reduction of Outcomes: Vytorin Efficacy International Trial) is designed to generate the data needed to confirm that ezetimibe does provide clinical benefit commensurate with predictions from its additional reductions in LDL-C. In this multicentre study, 18,000 high-risk patients are being randomized to ezetimibe 10 mg plus simvastatin 40 mg or simvastatin 40 mg alone. More than half of the enrolment has been completed. The primary composite end point is death, myocardial infarction (MI), rehospitalization for acute coronary syndrome or revascularization (occurring 30 days or more after the initial event).
Reductions in hs-CRP

<img1725|center>

Until IMPROVE-IT is completed, the preponderance of data on the benefits of LDL-C on risk reduction supports the use of ezetimibe as an adjunctive therapy in those patients who cannot reach current targets on statins alone. Ezetimibe has been effective in both lowering LDL and reducing hs-CRP, an inflammatory marker which is currently being evaluated as a potentially treatable CV risk factor. Perhaps most importantly, the cholesterol absorption inhibitor is capable of increasing the proportion of patients who reach goal relative to statins alone. In addition, it is well tolerated, a critical variable in an equation weighing potential benefit against potential risk. This means that in the context of the strongly supported lipid hypothesis, the unproven but substantial potential benefit of ezetimibe is not significantly undermined by tolerability issues.

While ENHANCE findings have no immediate clinical implications in Canada, where guidelines have been clear about the importance of maximizing statin therapies before turning to adjunctive agents to reach LDL targets, they have had a profound impact in the US, where clinicians have frequently employed ezetimibe to avoid maximal doses of statins. In a study of patterns of ezetimibe use, the population-based rate of prescribing statins with ezetimibe in 2006 was almost five times higher in the US relative to Canada (15.2% vs. 3.4%) (Jackevicius et al. N Engl J Med 2008;358(17):1819-28). This difference is likely due to the availability of a single pill in the US that encouraged use of a relatively low dose of a statin with ezetimibe. This pill was not made available in Canada.

The unpredictability of clinical trials underscores the importance of evidence-based medicine. Although the results of the ENHANCE study were unexpected, they should not be overinterpreted. A relatively small study conducted in a very specific subset of individuals with hypercholesterolemia, ENHANCE does not refute the overwhelming data that confirm the importance of aggressive lipid lowering in high-risk patients. The ability of ezetimibe to improve outcome in this population must be proven, but ENHANCE does not alter treatment goals.

Summary

As a surrogate end point, IMT regression has provided valuable support for the activity of lipid-lowering therapies. In ENHANCE, the failure of a more aggressive lipid-lowering therapy to alter this surrogate end point relative to a less aggressive regimen has provided potentially important information into the dynamics of atherosclerotic disease, but its relevance to clinical outcomes is unknown. Since the late 1970s, a growing database of corroborative results has shown the benefits of LDL-C reduction on risk management and clinical outcomes. It is not reasonable to conclude that the additional lipid lowering observed in this study is not providing clinical benefit by the absence of change in this surrogate. Clinical end point trials are needed. ENHANCE results have no effect on current treatment guidelines or strategies.

Questions and Answers

Panel

David Fritchett, MD, FRCPC Division of Cardiology, St. Michael's Hospital

Jean Grégoire, MD, CSPQ, FRCPC Coordinator, Ambulatory Care Clinic Montreal Heart Institute

Eva M. Lonn, MD, FRCPC, FACC Division of Preventive Cardiology Hamilton Health Sciences Centre

G.B. John Mancini, MD, FRCPC, FACC Division of Cardiology Vancouver Hospital & Health Sciences Centre

Is cIMT an appropriate outcome for trials? Is it a valid proxy on which to judge a compound’s clinical efficacy?

Dr. Grégoire: The cIMT studies have allowed us to understand the mechanism of lipid-lowering drugs, but they cannot be a substitute for clinical trials measuring effects on outcomes. The same team that performed ENHANCE published a similar study called ASAP several years ago and although that did show a reduction in cIMT with intensive lipid-lowering therapy, it should not be considered any more clinically relevant. The cIMT studies provide the basis to explain clinical results or perhaps to justify a clinical trial, but they are not meaningful for clinical decisions on their own.

Dr. Mancini: The cIMT studies demonstrating that progression of atherosclerosis can be modified with drug therapy has been very reassuring, but this is a tool for evaluating the mechanism of drug benefit. It is does not provide incontrovertible evidence of clinical benefit. Although it has been a useful surrogate, it is still only a surrogate and not a substitute for change in outcome.

Dr. Fitchett: There is a relationship between cIMT, LDL-C and BP levels. Clinical trials of LDL-C lowering have shown either a regression or a decreased progression of cIMT compared to either placebo or less vigorous LDL-C lowering. However, no clinical trial has directly linked cIMT reduction to a reduction of clinical events. It is important to recognize that cIMT is a surrogate measure for atherosclerosis. As a consequence, studies of therapies on cIMT should not affect our use of drug therapy to alter important clinical outcomes.

Dr. Lonn: cIMT is an intermediate end point. Although it has been extensively validated, its role is to test the biologic effects of new therapies. Nothing can replace clinical end points for confirming benefit in the management of patients. It cannot be considered a substitute for efficacy studies in which the goal is to determine whether the drug is useful for preventing events.

Why do you think the ENHANCE trial did not show a significant difference in cIMT from baseline and between treatment arms? What are the possible explanations for the different results observed in the ASAP and METEOR trials compared to those observed in the ENHANCE trial?

Dr. Fitchett: There are three possible explanations for the failure of the ENHANCE study to show a reduction in cIMT with ezetimibe despite the additional LDL reduction. One is that the methodology was not sufficiently sensitive to detect small changes in cIMT. However, the standard deviation of the digital imaging in ENHANCE was 25% of that in previous trials, such as ASAP, so this explanation is unlikely. Another possible explanation was that the addition of ezetimibe was not anti-atherogenic. A third potential explanation is that the vigorous statin treatment received by patients prior to entering ENHANCE already achieved maximum cIMT reductions. In ASAP, which also evaluated patients with familial hypercholesterolemia, the median baseline cIMT was 0.93 mm vs. 0.68 mm in ENHANCE. In METEOR (Measuring Effects on Intima-Media Thickness: An Evaluation of Rosuvastatin), patients were required to have a baseline cIMT >1.2 mm. It is a possibility that maximal cIMT reduction was already achieved among ENHANCE patients before they entered the study. This is supported by the extension study of ASAP, which found no further regression from aggressive lipid lowering after the first two years.

Dr. Mancini: The lack of a reduction was probably due to a combination of factors, although I do not think, as some have suggested, that there was any problem with the laboratory technique of cIMT measurement. Rather, I think it is probably important that the majority of the patients in ENHANCE had received a high-dose statin prior to entering the study. This may have reduced the potential for further regression. Also, although ezetimibe produced a 16.5% additional reduction in LDL, which is what would have been predicted on the dose used, this is a great deal smaller than the 40% or greater reductions that can be achieved on initial high-dose statin, so we would not expect to see the degree of improvement observed in regression studies with high doses of statins employed in treatment-naïve patients.

Dr. Lonn: The most likely explanation for the results of ENHANCE is that 80% of the patients had been pre-treated with statins. We do not really know what their real baseline cholesterol levels were. The publication provides cholesterol levels after a short washout period. This short washout was sufficient to briefly raise lipid levels but not to affect the vascular wall. This explains why the baseline cIMT was lower than expected. It would take two years or longer of washout for the cIMT to return to a pre-statin level. Due to the relatively modest cIMT abnormality at baseline, the study was underpowered to show a difference. This does not rule out the possibility that ezetimibe was ineffective for achieving regression despite the reduction in LDL, but we cannot judge the effects of ezetimibe from the ENHANCE study.

Dr. Grégoire: I think it is very appropriate to compare ENHANCE to ASAP, because the studies were conducted by the same group of investigators in the same special patient population. The most important difference was that the cIMT was not very abnormal at baseline in ENHANCE relative to ASAP. It is very difficult to show regression from this starting point. METEOR, for example, which required a greater baseline cIMT, provided a greater opportunity to achieve regression.

The recent halt of the JUPITER trial gives the first clue that lowering hs-CRP is also important. Given the hs-CRP lowering associated with ezetimibe, does this provide reassurance for potential benefit of intensive lipid lowering with a combination that includes ezetimibe?

Dr. Mancini: Based on the data so far, hs-CRP is an extremely interesting target. The JUPITER (Justification for the Use of Statins in Primary Prevention: An Intervention Trial Evaluating Rosuvastatin) appears to provide evidence that therapy that produces a reduction in hs-CRP will also reduce the risk of clinical events, but the study is not yet published so there is not much we can conclude. The evidence that ezetimibe lowers hs-CRP may suggest another pathway of clinical benefit, but hs-CRP is a surrogate marker of inflammation. We still have some distance to go before we can validate that a reduction in hs-CRP translates directly into a reduction in clinical risk.

Dr. Lonn: It would be premature to speculate heavily about the JUPITER trial until the results are published. So far, with the possible exception of acute coronary syndromes, we do not have much evidence that reducing hs-CRP predicts a reduction in clinical events. JUPITER may change this, but we need to see the complete results.

Dr. Fitchett: The JUPITER trial selected patients at intermediate risk for CV disease events who had elevated levels of hs-CRP and treated them with either rosuvastatin therapy or placebo. The study was stopped prematurely because of the benefit of treatment. We await the publication of the study before further comments can be made. For patients receiving treatment, a recent meta-analysis indicated that hs-CRP reduction relates to the amount LDL-C is reduced, rather than the result of any pleiotropic effect of statin therapy. The ENHANCE study supports this conclusion by showing that drugs other than statins that lower LDL will also lower hs-CRP. In ENHANCE, the combination of ezetimibe and simvastatin reduced hs-CRP by 25.7% compared to patients receiving the statin alone.

Dr. Grégoire: Because hs-CRP is a non-specific marker of inflammation, we need to be very cautious about how an effect on hs-CRP will alter likelihood of clinical events. We need more data to separate effect on hs-CRP from an effect on LDL. Although the JUPITER trial appears to have provided a positive result, it will be important to wait until the study is published to better understand the role of hs-CRP in primary prevention. It is exciting to consider the possibility that drugs that reduce hs-CRP and LDL will provide better protection than drugs that reduce LDL alone, but I do not think we have proof of that yet.

Do the ENHANCE results put into question the importance of lowering LDL-C? Should we still be trying to achieve a target of <2.0 mmol/L for high-risk patients? Do you believe that it matters how you lower LDL-C?

Dr. Lonn: The ENHANCE study has absolutely no relevance to the LDL hypothesis. The ENHANCE study involved measuring an intermediate end point in 720 patients with familial hypercholesterolemia. The LDL hypothesis is built on thousands of patients in multiple trials dating back more than 10 years. We know that lowering LDL is very important for risk reduction. The result should not have any influence on clinicians who should stick with established targets.

Dr. Grégoire: There is a reason that we are treating to established targets. These are based on very strong evidence-based medicine. ENHANCE should not have an effect on how we interpret the LDL hypothesis or on our current strategies to reach targets, including the use of ezetimibe if patients do not reach targets on maximally tolerated doses of statins.

Dr. Mancini: My fear is that clinicians will interpret the results of ENHANCE as being a reason not to treat to goal. First of all, this study was conducted in patients with familial hyperlipidemia, which represents less than 1% of all patients with dyslipidemia. It is, in fact, an unusual population. These results should not change anyone’s mind about the importance of reaching targets, which are well established by many studies. It is important to start with statins, but if the target is not reached, then additional therapies, including ezetimibe, should be considered for getting patients to the goals established in current guidelines.

Dr. Fitchett: The ENHANCE trial does not alter the importance of lowering LDL-C in patients at risk for atherosclerotic CV disease. Multiple clinical trials have shown that reducing LDL-C reduces CV adverse outcomes. A linear relationship between the reduction of LDL-C and the reduction of CV events has been established, irrespective of the method of LDL-C lowering, by studies with a variety of treatments, including bile acid sequestrants and ileal surgical bypass. Recent clinical trials suggest relative improvements in outcome for LDL reductions to levels <1.0 mmol/L. Studies of coronary atherosclerosis with intracoronary ultrasound have shown coronary atherosclerosis progression can be arrested when LDL-C is reduced 50% and levels of <2.0 mmol/L are achieved. Hence the LDL-C hypothesis is still very much alive and is not affected in any way by the results of the ENHANCE study. We should still be attempting to achieve both a target LDL-C of < 2.0mmol/L and a 50% reduction of LDL-C in patients at high risk for CV disease.

Based on the ENHANCE results, do you still feel that ezetimibe is an appropriate add-on strategy to statin therapy for patients not at LDL-C goal?

Dr. Mancini: Ezetimibe continues to be a valuable adjunctive agent for patients who do not reach goals on a maximum dose of statins. Other agents may also be useful, but ezetimibe achieves dependable LDL lowering when added to a statin and it is well tolerated. Based on the importance of reaching LDL targets, ezetimibe should be used if needed.

Dr. Fitchett: The American College of Cardiology (ACC) recommends that major clinical decisions not be made on the basis of the ENHANCE study alone. Furthermore, the US National Lipid Association stated, “Given that other studies are currently in progress to examine CV end points, extrapolating ENHANCE results to different populations with regard to clinical outcomes is premature.” Consequently, the ENHANCE study does not change our guideline recommendations to use add-on therapy with agents including ezetimibe for patients not at LDL-C goal. Other agents, such as bile acid sequestrants or niacin, are also effective and should be considered, but ezetimibe is the only agent that is well tolerated and associated with high levels of patient adherence.

Dr. Grégoire: It is clear that statins, even at optimal doses, are not sufficient to get all patients to treatment targets. So what are the options? Resins and nicotinic acid will improve LDL but they are both associated with tolerability issues that will pose problems with adherence except in very motivated patients. Fibrates are good for lowering triglycerides but not very effective for lowering LDL. For many patients, the only practical option is to add ezetimibe. We know that this is effective in lowering LDL and it is very well tolerated.

Dr. Lonn: The evidence to support the guidelines is overwhelming. This means that if you cannot get patients to the targets on statins alone, you should use adjunctive therapy. I think that ezetimibe continues to be a reasonable choice. It is well tolerated and it is effective for LDL lowering. It is important to eventually have the results from the IMPROVE-IT study to confirm that this method of lipid lowering (ezetimibe added to a statin) reduces risk of clinical events. At present, until these data become available, ezetimibe should be used for lipid lowering, which we know to be beneficial.

Are there important safety and tolerability concerns about adding ezetimibe to statin therapy?

Dr. Fitchett: The multiple clinical trials which have evaluated the efficacy of ezetimibe with statins, including the ENHANCE trial, have shown that abnormalities of liver function and increments of creatine kinase are no more common in patients treated with ezetimibe and a statin than with a statin alone. Clinical trial safety data are supported by post-marketing surveillance data with more than 9.5 million patient-years of treatment.

Dr. Lonn: Ezetimibe has been extremely well tolerated in clinical trials as well as in routine practice. There is a small number of patients who develop muscle side effects, but this is relatively rare. The ENHANCE trial actually serves to endorse the safety of ezetimibe, because ezetimibe was not associated with any significant side effects in that study.

Dr. Mancini: Given the totality of the evidence, I do not have any significant concerns about the safety of ezetimibe. It has had a very good tolerability profile and I think we have had enough clinical experience with this agent to be comfortable with its safety.

Dr. Grégoire: When he presented the ENHANCE results at the ACC meeting, the lead author did emphasize that there were no safety concerns with the drug. The earlier studies say the same thing. I do not think we need to be concerned about safety. We do need definitive proof that LDL lowering with ezetimibe will reduce clinical events, but we will have that information upon completion of the IMPROVE-IT trial. In the meantime, we need to focus on reaching LDL targets and I think for this we have no choice but to use ezetimibe in patients who do not reach goals on statins alone.