by Michael O'Riordan
November 18, 2009 — Adding extended-release niacin (Niaspan, Abbott) to statin therapy results in a significant regression of atherosclerosis as measured by carotid intima-media thickness (IMT), whereas the addition of ezetimibe (Zetia, Merck/Schering-Plough) to statin therapy did not, according to an eagerly anticipated study .
The results, from the Arterial Biology for the Investigation of the Treatment Effects of Reducing Cholesterol 6: HDL and LDL Treatment Strategies in Atherosclerosis (ARBITER 6-HALTS) study, were presented at the American Heart Association 2009 Scientific Sessions and published simultaneously online in the New England Journal of Medicine.
"The proper framework of the ARBITER 6-HALTS study is to compare the effectiveness of two drugs," lead investigator Dr Allen Taylor (Medstar Research Institute, Washington, DC) told heartwire . "In this regard, the data are clear--niacin is clearly superior to ezetimibe. That's good news for patients, and it's also good news for doctors who need to know how different drugs compare against one another so they can practice evidence-based medicine."
Dr James Stein (University of Wisconsin, Madison), who was not part of the ARBITER 6-HALTS trial, told heartwire that while the study tells clinicians little, if anything, about ezetimibe, the results are good news for niacin.
"If you practice evidence-based medicine in 2009, after high-dose statin therapy, niacin is your number-two drug," he said. "The amount of evidence available for niacin, although it's less than for high-dose statins, dwarfs the evidence available for any of the prevention interventions that we currently have available. It certainly dwarfs the evidence available for ezetimibe and fibrates, and we really think that niacin should be the number-two drug for all patients who are taking a statin and that doctors should rarely be using ezetimibe these days."
Dr Steven Nissen (Cleveland Clinic, OH), also commenting on the results for heartwire , called ARBITER 6-HALTS a classic "comparative-effectiveness" study and said there have been calls in the US legislature for such trials for the past few years.
"Now, here it is," he said. "Niacin is a 50-year-old drug, and you can buy it over the counter at your local pharmacy. When you have an inexpensive therapy like this--there are issues about being able to tolerate high-dose niacin, but if you get patients to tolerate it--niacin looks to be a better strategy."
Significant Difference in IMT
The results of the ARBITER 6-HALTS study have been keenly awaited, but not because clinicians questioned the efficacy of niacin. Previous studies have shown that niacin, alone or in combination with other drugs, including simvastatin, was associated with plaque stabilization or a modest regression of atherosclerosis.
Instead, the ARBITER 6-HALTS study was anticipated because investigators compared two distinct lipid-modifying strategies in patients who had or who were at high risk for vascular disease but who had LDL-cholesterol levels <100 mg/dL and moderately low HDL-cholesterol levels (<50 mg/dL). Among the 363 patients enrolled in the study, half were randomized to a treatment that further reduced LDL-cholesterol levels with the addition of ezetimibe (10 mg/daily) to statin therapy, while the other half were randomized to HDL-raising therapy with extended-release niacin (2000 mg/daily).
Coupled with the unique treatment approaches, many wondered how ezetimibe would fare given its rash of negative press from recent clinical trials, among them the controversial ENHANCE study and the SEAS trial.
In ARBITER 6-HALTS, men and women with known vascular disease or coronary-disease risk equivalents treated with statins for at least three months prior to randomization were included in the study. At baseline, the mean LDL-cholesterol level was 84 mg/dL and 81 mg/dL in the niacin- and ezetimibe-treatment arms, respectively. After 14 months of treatment, mean HDL-cholesterol levels among the niacin-treated patients increased 18.4% to 50 mg/dL. In the ezetimibe arm, LDL-cholesterol levels further declined 19.2% to 66 mg/dL.
Compared with ezetimibe, treatment with niacin resulted in significantly larger changes in the mean and maximal carotid IMT over 14 months. From baseline, niacin resulted in a significant regression of mean and maximal carotid IMT when measured at eight months and 14 months, whereas there was no significant change in mean or maximal IMT among those treated with ezetimibe. Although not powered for clinical outcomes, there were more major cardiovascular events in the ezetimibe arm compared with those in the niacin arm (nine events vs two events, respectively; p=0.04).
Addressing the results, Taylor said the findings answer an important clinical question in secondary prevention. "If my patient is on a statin, then where do I go from here?" he said. "These results suggest that niacin is superior in a head-to-head trial with ezetimibe."
Stein, a researcher with experience in IMT studies, said the amount of regression with niacin is significant.
"The difference between niacin and ezetimibe in the IMT measurements is very large," he told heartwire . "The difference of 0.017 mm over 14 months is on the order of statin therapy vs placebo. It's essentially as effective as a statin and would be expected to translate into a large difference in cardiovascular events, which was observed in this study, although they occurred in a study that was short and small."
In a post hoc analysis exploring the relationship between changes in LDL-cholesterol levels and mean carotid IMT, investigators observed a "significant inverse relationship" between changes in LDL-cholesterol levels and mean carotid IMT in the ezetimibe arm, "such that a paradoxical increase in the carotid IMT was seen in patients with greater reductions in LDL cholesterol."
Taylor said his group performed the post hoc analysis based on the null effect with ezetimibe because they wanted to determine whether any patients, particularly those with the lowest on-treatment LDL-cholesterol levels, might have benefited from the LDL-cholesterol–lowering drug. The inverse relationship is counterintuitive, he acknowledged, and while it is only a weak to modest association, he said it raises some concerns about possible unintended biologic effects with the agent.
Relying on Surrogate End Points for Clinical Decisions
Speaking with heartwire , Dr Sanjay Kaul (Cedars Sinai Medical Center, Los Angeles) cautioned against making too much of the overall results, especially since ARBITER 6-HALTS was a small, short-duration, surrogate-end-point study that was prematurely stopped (40% of patients did not undergo the final IMT measurement at 14 months). He said that physicians should take these issues into account when interpreting the findings.
He expressed particular concern about the study's early termination, as did an editorial by Drs Roger Blumenthal and Erin Michos (Johns Hopkins University, Baltimore, MD) , saying there are only three reasons a trial should be stopped, and these are futility, unacceptable safety profile, or a large treatment effect where the ethical imperative is to protect the trial participants.
While the treatment effect between niacin and ezetimibe achieved statistical significance at the time of the interim analysis, he said that differences in carotid IMT do not justify early termination, because it is unclear what these differences actually mean. Although he also criticized the early termination of JUPITER, the rosuvastatin study, in 2008, that study was at least terminated early because of differences in hard clinical outcomes.
"I don't think we should be relying on surrogate-end-point studies to make clinical decisions," he said. Kaul noted that the use of carotid IMT remains controversial and that the United States Preventive Services Task Force recently concluded that the evidence available to assess benefits and harms using nontraditional risk factors, including carotid IMT, as a risk-prediction tool in asymptomatic men and women is insufficient. He said it is still unknown at this point if the difference in IMT outcomes between ezetimibe and niacin will translate into clinically relevant outcomes.
"The results of this study do not provide any new insights and are unlikely to change my practice of prescribing ezetimibe only as a third-choice agent after trying statins as first choice and niacin, fibrates, or resins as second choice," he said.
To heartwire , Taylor said the trial was stopped because of a precisely defined prespecified interim analysis showing niacin to be superior to ezetimibe. Nissen called the discussion about the early termination "insider baseball," saying that while it can be debated whether or not the study should have been stopped, he said, "No reasonable person is going to look at the results and believe that the early termination had any impact on what happened. There are points to be made, but the real question is which strategy is best, and this study provides a hands-down winner."
Strengths and Weaknesses
To heartwire , Stein said the fact that such a large difference was able to be shown in such a short period of time tells doctors that the treatment effect was large. Although they differ on the significance of the IMT findings, Stein and Kaul both said the post hoc findings should be regarded with caution given the potential for confounding in these analyses.
According to Stein, while some might see the study as an indictment of ezetimibe, ARBITER 6-HALTS "tells us nothing about ezetimibe's effectiveness, because there was no placebo arm, but instead tells us that niacin is more effective than ezetimibe for preventing the progression of carotid atherosclerosis." One could argue, he said (although he doesn't believe the argument), that the lack of progression in the ezetimibe arm means the much-maligned drug was "doing something."
"At the end of the day, ARBITER 6 is really just an IMT study that looks at different strategies of lowering LDL vs raising HDL cholesterol," Stein told heartwire . "But we have to remember that niacin does a lot more than raise HDL cholesterol--that's what makes the study a little more complicated than saying it is a direct comparison of LDL lowering vs HDL raising. Niacin raises HDL cholesterol, but it also lowers LDL cholesterol, lowers triglycerides, and makes LDL particles bigger, so it’s really a multiple lipoprotein intervention."
To heartwire , Taylor agreed, but said that clinicians primarily select niacin to raise HDL cholesterol and rarely use the drug to lower LDL cholesterol or triglyceride levels. Data from clinical trials, including HATS, show the benefit of niacin to be primarily from changes in HDL cholesterol, he said.
Dr Richard Karas (Tufts Medical Center, Boston, MA), who was also not part of the study, told heartwire that given the relationship between HDL cholesterol and the risk of cardiovascular events, the results are in line with what would be expected from a drug that increases HDL cholesterol approximately 20%. By increasing HDL cholesterol to 50 mg/dL, clinicians are able to significantly decrease the residual risk of patients treated with statins alone, whereas the extra boost from LDL lowering has little impact on that risk. Like Stein, he said that ARBITER 6-HALTS tells clinicians little about ezetimibe and that they must wait until outcomes studies are completed.
The Outcomes Studies
IMPROVE-IT, the large clinical-outcomes study with ezetimibe, is expected in 2012 but likely to be later, while SHARP, a study of ezetimibe in patients with renal failure, is also under way. The two much-anticipated outcome studies for niacin are AIM-HIGH, due out in 2011, and HPS2-THRIVE, which is expected to be completed in 2013.
As an aside, editorialists Drs John Kastelein (Academic Medical Center, Amsterdam, the Netherlands) and Michel Bots (University Medical Center, Utrecht, the Netherlands) , as well as Nissen, raised concerns about whether IMPROVE-IT would ever reach completion. More than 5000 hard clinical end points are needed for the study to reach statistical significance, an unusually high number given that past studies, including JUPITER, required a few hundred events. Based on these concerns, many clinicians, among them Kaul and Nissen, reiterated their displeasure with an approval process that allowed ezetimibe to be on the market for more than a decade before any hard clinical end-point data are available.
Results Likely to Be Correct
In an editorial accompanying the published study, Kastelein and Bots write that the ARBITER findings are likely to be correct, although they, like others, suggest the magnitude of difference might be exaggerated because the trial was stopped early. Still, they argue in favor of a treatment approach with niacin.
"Together, the results available to date provide support for the concept that the use of statins to reduce LDL to target levels with the subsequent addition of a drug to raise HDL-cholesterol levels (niacin), rather than a drug to lower LDL-cholesterol levels (ezetimibe), is a more effective treatment for patients at high cardiovascular risk," they write.
In the second editorial, Blumenthal and Michos say there is substantial evidence for initial use of a statin in patients with an increased risk of vascular disease, but if the lipid goal can't be met with a statin alone, the choices of adjunctive therapy include a fibrate, niacin, a bile-acid sequestrant, n-3 fatty-acid supplements, or ezetimibe. For their money, though, they "support the use of niacin as the preferred adjunctive agent to be used in combination with the maximal dose of a potent statin in persons who have low levels of HDL cholesterol and established cardiovascular disease."
Abbott sponsored the ARBITER 6-HALTS study. Taylor reports receiving lecture fees from Abbott. Stein reports serving on the data monitoring committee for a nonniacin, Abbott-sponsored study, receiving a research grant from Siemens and SonoSite, and receiving royalties from the University of Wisconsin for carotid ultrasound and risk prediction not related to the ARBITER 6-HALTS study. Kastelein reports receiving consulting fees, lecture fees, and grant support from Merck and lecture fees from Schering-Plough. Bots reports lecture fees and grant support from AstraZeneca.
- Taylor AJ, Villines TC, Stanck EJ, et al. Extended-release niacin or ezetimibe and carotid intima-media thickness. N Engl J Med 2009; DOI:10.1056/NEJMoa907569. Available at: http://www.nejm.org.
- Blumenthal RS, Michos ED. The HALTS trials--halting atherosclerosis or halted too early? N Engl J Med 2009; 361:2178-2180.
- Kastelein JJ, Bots ML. Statin therapy with ezetimibe or niacin in high-risk patients. N Engl J Med 2009; 361: 2180-2183.