DR. TARDIF: Well, yes. Again there have been trials that excited us. I’m referring to the Helsinki Hearts Study8 and the Veterans Affairs High-Density Lipoprotein Intervention Trial (VA-HIT),9 where gemfibrozil provided good results both in primary and secondary prevention. But then there has been a series of fairly major disappointments, and I’m referring to the Fenofibrate Intervention and Event Lowering in Diabetes (FIELD) study10 with fenofibrate, which was a fairly large study, and more recently the Action to Control Cardiovascular Risk in Diabetes (ACCORD) that also failed to demonstrate benefits of adding fenofibrate to statin.11
Actually in the FIELD study the result was a bit strange in a sense that there was actually a trend in the right direction for reduced rates of myocardial infarction, but that was counterbalanced by a trend in the wrong direction for mortality, the overall the result was disappointing. And then in the ACCORD Study,11 the overall the result was neutral. Some people have made a case from the subgroup analysis where the subgroup of patients with high triglycerides and low HDL had a trend in the right direction for clinical events with a P value of 0.06. I think we need to be careful with that. When you’ve spent all your alpha on the primary point that failed, and then you start looking at sub-groups with a nominal P value of 0.06, this is strictly hypothesis generating.
So I think that some people have made too much of a case of P value. So to answer your question, I think fibrates can still have a role for patients who have very high triglyceride values, for example to prevent pancreatitis. As far as the addition of a fibrate to a statin for secondary prevention of cardiovascular events, especially following the results of FIELD and ACCORD, the place of fibrates for the time being is fairly limited. I would change my mind if there was a prospective study testing fibrate in patients with high triglycerides, low HDL that would prospectively demonstrate on a pre-specified endpoint that we improve outcomes. For the time being I don’t think we have that.
DR. RADER: It would be nice if someone would fund that study.
DR. TARDIF: Absolutely.
DR. RADER: So before we move on to new therapies in development, I wonder if we could just briefly address the question, what’s the general approach in clinical practice at this time in the patient with coronary disease and low HDL? Is it primarily or purely aggressive reduction of LDL and apoB-containing lipoproteins, or is there a role for adding something like niacin or fibrate? Could I just quickly ask each of you to give your take on where we are currently before we talk about new therapies? Peter?
DR. TOTH: Yes, I think the approach for patients with coronary disease and low HDL is to treat with a statin plus niacin. But we have to remember that aggressive lifestyle modification in patients with low HDL can also be very helpful. We know that cigarette smoking is associated with lower HDL-C,12 while cessation can raise HDL 15% to 20%.13 There is evidence to suggest that in cigarette smokers LCAT is inhibited to some degree. When LCAT is inhibited, cholesterol transported out of cells is not esterified, leading to impaired maturation of HDL and increased clearance from the circulation. We also know that cigarette smoking potentiates insulin resistance by augmenting tumor necrosis factor-alpha production by adipocytes. This also results in less HDL biogenesis by the liver and adipose tissue and increased rates of HDL clearance secondary to hepatic lipase dependent lipolysis of triglyceride enriched HDL particles.
In addition to smoking cessation, exercise can increase serum levels of HDL-C in a dose-response manner.14 The Health, Risk Factors, Exercise Training and Genetics (HERITAGE) Family Study15 showed that for patients with baseline insulin resistance, exercise reduced insulin resistance, promoted weight loss, and induced HDL elevation. Dietary alterations can promote HDL elevation, so I think lifestyle modification is important. Does it always work? No.
Beyond that, of course statin, niacin, and then plus or minus fibrate if the patient still has residual hypertriglyceridemia. There are no clinical trial data that demonstrates that triple therapy provides incremental risk reduction over and above one or two drug therapy. However, if we pay attention to guideline recommendations where we have to meet risk stratified LDL and non-HDL targets, most certainly there are more complex CAD patients who warrant therapy with two, three, or even more drugs if they have any hope of attaining their risk stratified goals for all of their lipoprotein fractions.
DR. RADER: Bryan, your general take on that?
DR. BREWER: I would agree completely. I think that we have to maximize all the ways to change the lifestyle of the patient and try to increase the HDL. I think that in many patients, particularly with diabetes, or the metabolic syndrome, the practicing cardiologist is only looking at the LDL-C level. I think in the management of these patients we have to achieve the LDL goal but it is also important to reach the non–HDL-C, or apoB goal, or LDL particle number goal. Reducing all of the atherogenic apoB lipoproteins is of great importance to try to reduce the significant residual risk in the statin treated diabetic and metabolic syndrome patient.
It is important to remember that in those patients in whom you have reached the LDL-C target goal, such as the patients in the Treating to New Targets (TNT) trial16 that Phil Barter reviewed, the HDL remains an independent risk factor. So even if we maximize the LDL-C and triglyeride reduction, I think that HDL remains an important risk factor. In those patients with a strong family history of CAD, I think that they are still a candidate for niacin. There are less data to support the use of a fibrate in these patients but I think that you still have to consider HDL as an important independent risk factor. Thus, the high risk cardiovascular patient may require the addition of a second and third drug to reduce the risk of cardiovascular events.
DR. RADER: Jean-Claude, anything to add?
DR. TARDIF: Well, very little. I think it's aggressive use of statins, and as Peter said, lifestyle changes, so that includes smoking cessation, weight reduction, exercise, but as we all know it’s very difficult to induce ongoing changes in lifestyle for many of our patients.
DR. RADER: None of you mentioned my favorite way to raise HDL and that’s alcohol. So my questions are, does the HDL raising of alcohol contribute to the apparent benefit of alcohol in terms of reducing cardiovascular risk? In other words does it raise a good form of HDL? And do you ever consider suggesting alcohol to patients with coronary disease and low HDL?
DR. TOTH: That’s an interesting question, Dan, because Framingham17 has also shown that among the patients who drink moderately, there is a dose response relationship between amount of alcohol consumed daily and level of serum HDL-C.18 Some people believe this underlies the so-called French Paradox, but there is skepticism about that. Alcohol appears to be a weak inhibitor of CETP and has also been shown to potentiate hepatic apoA-1 production and HDL biogenesis.
I actually wrote a paper in Circulation about five years ago on approaches to raising HDL.19 One of the things I mentioned toward the end was consideration of four to six ounces of wine with the evening meal because the average adult is responsible enough to handle this. I was assailed through emails with people accusing me of being irresponsible and promoting alcoholism. What was I thinking?
It was such a disproportionate response because if you think about it, can the average person responsibly drink one or two glasses of wine? The answer is in fact, yes, they can. So in some cases I don’t think it’s such an unreasonable recommendation.
DR. BREWER: It’s a tricky business.
DR. TARDIF: It is a tricky business.
DR. RADER: So I’d like to turn now to new HDL therapies and spend a fair amount of time on the gorilla in the room, which is CETP inhibition. So I wonder if we could just start, Bryan, maybe you can take this and just remind us of the initial discovery of CETP deficiency and essentially what led to the concept that CETP inhibition would raise HDL, and if you could also comment on CETP deficient patients and their cardiovascular risk.
DR. BREWER: There was a great deal of interest generated in terms of the HDL hypothesis with the discovery of patients who had complete CETP deficiency. The lipoprotein profile in patients with complete CETP deficiency was very exciting in the sense that these individuals had very high levels of HDL, greater than 100 mg/dL,20 and low levels of LDL. So it looked like the ideal lipoprotein profile. Unfortunately it was not easy to conclude whether the presence of the high HDL lipoprotein profile was associated with reduced atherosclerosis and clinical events since there were only a small number of patients, primarily in Japan, who were identified with complete CETP deficiency. Secondly, a number of the CETP deficient patients also had other comorbidities including defects in hepatic lipase. So it became difficult to use the CETP deficient patients to effectively assess whether the inhibition of CETP in fact would lead to a significant reduction in cardiovascular risk, even though the lipoprotein profile looked very encouraging,
Nevertheless, the lipoprotein profile in the CETP deficient patients led to a great deal of interest in trying to develop a small molecule to inhibit CETP to change the lipoprotein profile to reflect what is present in the patients with complete CETP deficiency. There was also a great deal of discussion of whether the HDL generated with CETP inhibition was in fact good HDL or was it a bad HDL. It ultimately became clear that the only way we would definitively answer the question of whether CETP inhibition was a good target for reducing cardiovascular events disease would be clinical morbidity and mortality trials.
DR. RADER: So with that discovery many companies developed programs for CETP inhibition, and the first CETP inhibitor that basically got into full scale clinical development was torcetrapib. Jean-Claude, I wonder if you could just briefly trace for us the history of torcetrapib.
DR. TARDIF: Torcetrapib was a powerful CETP inhibitor21 that raised HDL-C values by about 100% on higher dosages. Early on in the program there was a signal in terms of blood pressure raising, but the consensus was that if we lowered the dosage we could find sort of a sweet spot between HDL-C elevation and minimal impact on blood pressure. So the gamble or the hypothesis was that raising of HDL-C with this drug would much more than offset the downside of a slight increase in blood pressure.
So torcetrapib entered a large drug development program. There were three imaging studies; two in the carotid circulation using carotid intima-media thickness, called RADIANCE 122 and RADIANCE 2,23 and there was one coronary intravascular ultrasound (IVUS) study called the ILLUSTRATE study.24 There was also one large clinical outcome study called the ILLUMINATE study.21,25 All three imaging studies failed to demonstrate a benefit on the primary endpoint. There was no evidence of benefit of adding torcetrapib to statin therapy to reach a reasonable LDL-C goal in terms of slowing atherosclerosis progression in the coronary or the carotid circulation. Even in the smaller imaging studies, compared to the large outcome study, there was also a trend in the wrong direction for clinical events.
Then on December 2, 2006 the study was stopped. There was a 25% increase of the primary endpoint of hard outcomes with torcetrapib added to atorvastatin compared to atorvastatin alone. There was a 58% increase in the risk of dying with the combination of torcetrapib plus atorvastatin compared to atorvastatin alone. Actually when you look at event curves, the curves diverge, not in favor of torcetrapib, early, in a matter of months. When this study was stopped the median exposure to torcetrapib was about 18 months.
So the question was, was it the molecule, was it the class, was it the HDL hypothesis all together? I think what we’ve learned is that certainly torcetrapib was not a clean drug; it was turning on a number of bad genes and turning off a number of good genes. In a nutshell, it was, for example, turning on the CYP11B2 aldosterone synthase gene that was resulting in an increase of aldosterone secretion by adrenal glands, leading to electrolyte changes, a blood pressure increase, and potential deleterious vascular and ventricular negative effects.
What part that off-target toxicity played in the disappointing results of torcetrapib is not certain, but certainly these effects that I just described, probably at least in part, explain the negative findings with torcetrapib. Now, what we have learned is that there are a number of newer CETP inhibitors or modulators that are not associated with this target toxicity. And now I think we have the right tools, drugs to test the hypothesis that inhibiting CETP is going to have favorable effects on atherosclerosis in clinical outcomes.
DR. RADER: Could I just ask you, as an IVUS expert—there was a post-hoc analysis of the torcetrapib IVUS trial24 with torcetrapib suggesting that the greater the increase in HDL, the less progression of disease. So it’s been interpreted that this is at least evidence for the benefit of the HDL raising associated with CETP inhibition. How do you interpret that?
DR. TARDIF: Well I have to be careful because I was a co-author on that paper. But that being said, simply put, I think one needs to be very careful when you’re trying to find a positive finding, or put a positive spin on a neutral or negative study, because at the end of the day you spent your alpha on your primary endpoint and you fail. So the rest, in my opinion, needs to be taken with a grain of salt.
DR. TOTH: At least you’re consistent.
DR. TARDIF: Yes, absolutely. I think we need to be careful. Some people say the hypothesis makes sense, but that there was something else preventing the full blown beneficial effects on torcetrapib, which is a nice hypothesis. I think the proof of the pudding will be in these cleaner CETP drugs that we have that will be tested. I think ultimately that will be the answer. I mean there are other hypotheses generating results that have been observed. For example, even the ILLUMINATE study went in the right direction.
DR. RADER: Remarkably, despite this, there have been additional drugs that have continued in development. The one that is the furthest along in its outcomes trial is dalcetrapib. Peter, I wonder if you could just briefly review what we know about dalcetrapib,26-28 its lipid effects, mechanism, and where we are with the clinical trials.
DR. TOTH: Dalcetrapib is a second generation CETP inhibitor and, unlike torcetrapib, does not form a stable covalent complex with CETP and HDL. It appears to be a modulator of CETP activity. It has shown very nice safety—it does not raise blood pressure, doesn’t activate aldosterone synthase, does not disturb electrolyte balance—and it does provide a maximum 31.42% increase in HDL when used at 600 mg and a maximum 36.45% increase in HDL when used at 900 mg. 27