The HDL-C Conundrum: When More of the "Good" Isn't So Great

Internal Medicine World Report, July 2007, Volume 0, Issue 0

But New Agents Show the Jury is Still Out

By Daniel M. Keller, PhD

NEW ORLEANS—When is more of the "good" stuff not so good? In several trials presented at the American College of Cardiology annual meeting, raising the levels of "good" high-density lipoprotein cholesterol (HDL-C) failed to reverse the progression of coronary atherosclerosis.

The ILLUSTRATE (Investigation of Lipid Level Management Using Coronary Ultrasound to Assess Reduction of Atherosclerosis by CETP Inhibition and HDL Elevation) and 2 RADIANCE (Rating Atherosclerotic Disease Change by Imaging with a New CETP Inhibitor) studies compared groups of patients receiving torcetrapib plus atorvastatin (Lipitor) to atorvastatin mono-therapy. Torcetrapib was in development as a means to raise serum HDL-C levels, which it did. However, all development and clinical trials were halted when the drug was found to be associated with cardiovascular (CV) adverse events in the ILLUMINATE (Investigation of Lipid Level Management to Understand Its Impact in Atherosclerotic Events) study, which included 15,000 patients.

CETP Inhibition

Torcetrapib acts by inhibiting cholesteryl ester transfer protein (CETP), which mediates the transfer of cholesteryl ester from HDL to triglyceride-rich lipoproteins, which are cleared by low-density lipoprotein cholesterol (LDL-C) receptors in the liver.

In ILLUSTRATE, adding torcetrapib to atorvastatin did not slow the progression of coronary plaque buildup despite a 61% increase in HDL-C and a 20% decrease in LDL-C levels relative to atorvastatin monotherapy.

Torcetrapib raised systolic blood pressure (BP) by an average of 4.6 mm Hg. But Erik Stroes, MD, of the Department of Vascular Medicine at the Academic Medical Center of the University of Amsterdam, the Netherlands, doubts that this BP elevation alone would explain the negative findings seen in the different trials with this drug. "The blood pressure elevation was relatively modest. If you see that after 1 year there was a more than 60% increase in event is very unlikely that that was related only to the blood pressure increase," he said.

Daniel Rader, MD, professor of medicine and director of preventive cardiology at the University of Pennsylvania in Philadelphia, said, "The increase in blood pressure is a marker, a sign, that the molecule [the drug] does do something to the vasculature that is not healthy."

N Engl J Med

And the composition of HDL may be at least as important as the amount. HDL contains a variety of proteins that may regulate thrombotic, inflammatory, and complement systems, and it will be instructive to see if other inhibitors of CETP do not elevate BP but still exert adverse effects on vascular function (. 2007;356: 1364-1366).

J Clin Invest

"HDL does seem to be a platform on which other proteins assemble," Dr Rader said, adding that a study which used a proteomics approach showed that HDL from people with coronary artery disease had very different protein composition from HDL of healthy people (. 2007;117:746-756). Several of the proteins are involved in regulation of the complement system and in protecting tissue from proteolysis, suggesting antiinflammatory and antiatherogenic roles.

N Engl J Med

The RADIANCE 1 and 2 trials also compared torcetrapib plus atorvastatin with atorvastatin monotherapy to assess effects on carotid intima-media thickness (CIMT) in 12 predefined carotid segments. RADIANCE 1 (. 2007;356:1620-1630) followed 850 patients with heterozygous familial hypercholesterolemia. RADIANCE 2 studied 752 patients with mixed hyperlipidemia. Neither specified HDL-C levels for inclusion.

The torcetrapib/atorvastatin groups had significantly elevated HDL-C (52%-63%) and decreased LDL-C (18%-20%) levels compared with the atorvastatin-alone group, but both trials failed to meet their primary end points, the annualized rate of change in maximum CIMT in the 12 segments. "It remains to be investigated whether this is a consequence of the molecule torcetrapib or whether the concept of CETP inhibition is a flawed hypothesis," said lead investigator John Kastelein, MD, PhD, of the Academic Medical Center of the University of Amsterdam.

Considering the lack of effect of torcetrapib on coronary atherosclerosis in ILLUSTRATE or in RADIANCE, Dr Rader said, "These studies suggested that torcetrapib did not substantially worsen atherosclerosis, as assessed by imaging, but nevertheless in the large clinical outcome trials did increase cardiovascular events and total mortality, and so the mechanism of that remains unclear."

CSL-111 and ApoA-1 Milano


The ERASE (Effect of Reconstituted High-Density Lipoprotein on Atherosclerosis—Safety and Efficacy) trial investigated the role of HDL-C in CV disease with a new product, CSL-111 (. 2007;297:1675-1682). CSL111 is a reconstituted HDL made of apolipoprotein (Apo)A-1 isolated from human plasma and phosphatidylcholine derived from soybean that chemically and biologically resembles native HDL.

In ERASE, patients with an acute coronary syndrome event in the previous 2 weeks were randomized to receive 4 weekly infusions of CSL-111, either 40 mg/kg (n = 111) or 80 mg/kg (n = 12), or to placebo (n = 60). Changes in liver function led to a discontinuation of the 80-mg/kg arm. Patients were assessed using intravascular ultrasound (IVUS) and quantitative coronary angiography (QCA) at baseline and at 2 to 3 weeks after the final infusion.

The treatment failed to meet the primary efficacy end point—percent change in atheroma volume by IVUS of the proximal 4 cm of the target coronary artery versus placebo, but the percent change in atheroma volume within the CSL-111 arm compared with baseline was significant.



Cross-sections of the arteries were characterized according to the plaque present: calcific, fibrotic, fibrohypoechoic, hypoechoic, and normal. There were differences between the CSL-111 and placebo arms in the secondary end points of QCA ( = .03) and plaque characterization index ( = .01). Lead investigator Jean-Claude Tardif, MD, of the Montreal Heart Institute, said, "We think that when we take both the primary and secondary end points together, this really is strongly suggestive of a favorable, rapid treatment effect...with CSL-111."

Dr Tardif noted that studies infusing the apolipoprotein ApoA-1 Milano—an HDL variant with seemingly highly antiatherogenic properties—showed similar results to his trial. He said that when ApoA-1 Milano was administered to 47 patients, a 4.2% decrease in atheroma volume was evident compared with a 3.4% reduction using CSL-111.

While cautioning that ERASE was a small trial and not definitive, Christie Ballantyne, MD, director, Center for Cardiovascular Disease Prevention, Baylor College of Medicine, Houston, noted that the trends were in the right direction. "It was encouraging, if we ...put this in the context with ApoA-1 Milano," he said.

Dr Rader concurred, saying, "The ERASE trial is an example of the type of intervention that I personally am fairly enthusiastic about...some form of so-called synthetic HDL as a way of trying to rather acutely impact on atherosclerotic plaque, perhaps induce regression, induce stabilization in a way that would reduce the fairly high incidence of events that occur after presentation with acute coronary syndrome."

Clinical Implications

Daniel Rader, MD

Dr Rader cautioned that while "HDL is strongly epidemiologically inversely associated with risk," in any individual patient "a high HDL is no guarantee against developing cardiovascular disease." He added, however, that a high HDL-C value should not be assumed to overcome the risk conferred by a high LDL-C value. "We still need to treat LDL aggressively, and we shouldn't let a high HDL deter us from treating someone who we otherwise would have treated," he advised.

The other big issue for physicians is whether they should try to intervene and raise HDL-C in patients with low levels. "After you've aggressively treated the LDL side of things...I think it is appropriate to think about intervening on the HDL side after lifestyle, using niacin for example, potentially a fibrate," he said.

The adverse outcomes using torcetrapib will make development of other, similar molecules more difficult. Experts at the meeting agreed that trials with earlier, surrogate markers, such as vascular imaging, are advisable before moving into large clinical outcomes studies, for both safety and economic reasons.

"When you see blood pressure going up you should have said, 'Whoa, let's slow down here,'" Dr Ballantyne said. "Anyone who has a cardiovascular drug that raises blood pressure this much, we're worried, and with good reason."