NIH pulls plug on AIM-HIGH trial with niacin (niacin+statin)
May 27, 2011 | Lisa Nainggolan|
Bethesda, MD (updated with commentary) - A trial of high-dose extended-release niacin (Niaspan, Abbott) given in addition to statin therapy in patients with a history of cardiovascular disease, high triglycerides, and low levels of HDL cholesterol has been halted prematurely, 18 months ahead of schedule, because niacin offered no additional benefits in this patient population .
There was also a small, unexplained increase in ischemic stroke in the niacin group, in the
Atherothrombosis Intervention in Metabolic Syndrome with Low HDL Cholesterol/High Triglyceride and Impact on Global Health Outcomes (AIM-HIGH) study, according to a statement from the National Heart Lung and Blood Institute (NHLBI), which sponsored it.
In a media telebriefing today, Dr Susan B Shurin (acting director of the NHLBI) and two investigators from the trial stressed that the findings apply only to the specific patient population studied in AIM-HIGH and that they could not be extrapolated to others.
While some outside commentators agreed with this advice, others felt this was a nail in the coffin for the whole concept of treating newer lipid markers associated with CV risk, such as triglycerides and HDL.
FDA will not change labeling for niacin at the moment
Despite treatment with statin therapy for elevated LDL-cholesterol levels, those with low levels of HDL cholesterol remain at significant risk for cardiovascular events, and AIM-HIGH was designed to examine whether raising HDL using extended-release niacin would be beneficial in such patients. AIM-HIGH was a five-year study of almost 3500 patients, and results were originally expected in September 2012.
The decision to stop the trial was made at a regularly scheduled meeting of the study's independent data and safety monitoring board (DSMB) on April 25, 2011. The DSMB concluded that "high-dose, extended-release niacin offered no benefits beyond statin therapy alone in reducing cardiovascular-related complications in this trial. The rate of clinical events was the same in both treatment groups, and there was no evidence that this would change by continuing the trial."
During the telebriefing, Shurin noted that the increased stroke risk did play a role in the decision to stop the trial, but she observed, "The overall frequency of stroke was less than 1%, and previous studies do not suggest that stroke is a potential complication of niacin.
"The FDA is aware of the findings and is recommending no change in labeling or practice [regarding the use of extended-release niacin alone or in combination with simvastatin or other statins] pending further analysis of the data ," she said, adding that the NHLBI would soon "begin the important task of analyzing" the findings in detail. Full results of AIM-HIGH will be presented at a major scientific meeting and published in a peer-reviewed journal at a future date, she noted.
Is this the death of the HDL hypothesis? Not quite, say investigators
AIM-HIGH was based on data from observational studies and a few small clinical studies and the fact that use of high-dose extended-release niacin "is now common practice [but] the evidence did not seem to be as solid as we thought it should be," Shurin explained. "Although we did not see the expected clinical benefit, we have answered an important scientific question about treatment for cardiovascular disease," she noted.
And during the call she stressed-together with investigators from the trial-that patients who were not in the AIM-HIGH trial should not stop taking high-dose, extended-release niacin without talking to their doctor first. "We don't see that the study results here would recommend any general change in the approach to use of niacin, and the FDA concurs with that, at least so far," Shurin commented. "We would strongly recommend that any individual who is taking niacin have that conversation with their physician, but there is nothing in here that would be compelling." The NHLBI is now notifying all primary-care physicians of the findings, she added.
When asked by a reporter what patients should know, AIM-HIGH coinvestigator Dr William E Boden (Buffalo General Hospital, NY) said: "We tell them what the [National Cholesterol Education Program] NCEP recommendations have been for the past decade: the most important determinant of dyslipidemic risk is elevated LDL, and the most important therapeutic imperative is to get the LDL down with statins."
He pointed out that in AIM-HIGH, the extended-release niacin "produced the predicted effects on all lipids measured, increasing HDL levels by 20% and reducing triglycerides by around 25%. While high-dose niacin raised HDL, it did not affect cardiovascular events."
The results from AIM-HIGH "clearly underscore the importance of LDL reduction, and the fact that these patients were so incredibly well-treated with respect to LDL [average LDL at baseline was 71 g/dL] indicates that we could not, in that subpopulation, demonstrate incremental benefit of HDL raising. That doesn't mean that there aren't patients out there who might benefit from HDL raising, it's just that we don't have evidence from trials to guide us," he noted.
"The AIM-HIGH study population was not the very highest-risk population, and that clearly is why we can't generalize these findings to patients we didn't study in the AIM-HIGH trial," Boden added.
"Many patients in the real world have far higher LDL levels at baseline than we encountered. There is an enormous unmet need of increased residual risk, even among patients who take statins, and one of the challenges we as physicians and cardiologists face is to try to address that unmet need therapeutically," he said, noting, "It's difficult to speculate about whether HDL therapies are completely ineffective."
Another AIM-HIGH investigator, Dr Jeffrey Probstfield (University of Washington, Seattle), concurred: "We have to focus specifically on the fact that we had a very special population in this study. The only thing that we can say about alteration of HDL is in relation to the population we addressed in this study. There may be other populations who will respond differently."
But others disagree . . .
When asked whether she thought these findings meant the "death" of the HDL hypothesis, Shurin replied: "This sends us a bit back to the drawing board in terms of trying to figure out how to approach this hypothesis. Either the approach to raise HDL was not effective or HDL is just not a good target for therapy in the setting of other lipid-lowering agents."
Dr Darren McGuire (UT Southwestern Medical Center at Dallas) told heartwire: "In the absence of data to the contrary, clinicians accepted possibilities that led to niacin and other drugs being used to treat biomarkers associated with CV risk, such as triglycerides and HDL."
But now, "in the wake of negative trials of fibric acids, we have a futility finding with niacin and, in this case, at least some chance of a possibly serious adverse risk for stroke. In this setting, I think there is no justification to continue niacin therapy. I don't understand why patients should not stop the drug. There is no danger to stopping the drug to my knowledge and no withdrawal syndrome requiring tapering or other management."
"We can await findings from [the Heart Protection Study 2 Treatment of HDL to Reduce the Incidence of Vascular Events] HPS-2-THRIVE to make any further clinical decisions," McGuire adds.
"I think there is no justification to continue niacin therapy. I don't understand why patients should not stop the drug."
HPS-2-THRIVE is another large international trial of high-dose, extended-release niacin, which is still ongoing, with results expected in 2013. Probstfield said that although the dose of niacin used in this trial is "comparable" to that employed in AIM-HIGH, the preparation is different and the patient population "is quite a bit different; it's a whole panorama of HDL levels, all-comers."
Others not involved with AIM-HIGH and polled by heartwire had mixed views. Dr Sanjay Kaul (Cedars Sinai Hospital, Los Angeles, California)-who last week sat on the FDA advisory panel that debated fenofibrate (Trilipix, Abbott), another agent used with a statin to reduce triglycerides and increase HDL cholesterol, about which there is uncertainty-said it was "very reasonable for the DSMB to recommend stopping AIM-HIGH, as it met the criterion of 'futility'."
However, "one could make the argument that the potential benefits might be seen in high-risk patients such as ACS or dyslipidemic patients with elevated TG plus low HDL," he told heartwire. "Perhaps subgroup analysis in such cohorts, especially the dyslipidemic cohort, might be instructive."
Nevertheless, AIM-HIGH now adds to the list of trials that have shown that HDL is "a very complex story," Kaul said, citing the Action to Control Cardiovascular Risk in Diabetes (ACCORD) Lipid trial with fenofibrate and the disappointing results with torcetrapib in the Investigation of Lipid Level Management to Understand its Impact in Atherosclerotic Events (ILLUMINATE) study, in which this investigational agent-which did raise HDL levels as well as reduce triglycerides and LDL-actually increased the rate of cardiovascular events.
"The findings from ILLUMINATE, ACCORD-Lipid, and AIM-HIGH raise questions regarding whether 'residual risk' in patients optimally treated with statins is modifiable. I hope together these findings will strengthen the argument for a change in the regulatory standard for approving lipid-modifying therapies from a lipid biomarker-centric to clinical-outcomes-based standard," Kaul noted.
Also, Kaul added that, by his calculation, the stroke finding in AIM-HIGH represents a 2.3-fold increased risk, which is "concerning."
But Dr Daniel Rader (University of Pennsylvania, Philadelphia) told heartwire he does not believe the difference in ischemic stroke between arms in the trial was real: "there was no compelling evidence of harm."
"It will be interesting to see if there are subgroups that appear to benefit [from niacin]," Radar added, noting that he will await the outcome of "the much larger and more definitive" HPS-2 THRIVE study before making any firm decisions.
"I feel clinically that it is premature to withdraw patients from niacin therapy at this time," he said, but countered, "On the other hand, I may be a bit more selective on who I start on niacin pending the results of HPS-2."
Dr Roger Blumenthal (Johns Hopkins University School of Medicine, Baltimore, MD) commented: "The results are very disappointing [and] call into question whether the statistically significant benefits seen in prior carotid intima-media thickness (IMT) studies with niacin truly have clinical relevance. Most of us expected that the trial would be positive based on the consistency of the subclinical atherosclerosis imaging results. We all await the publication of these important clinical trial results."
AIM-HIGH enrolled 3414 participants in the US and Canada with a history of cardiovascular disease, low HDL cholesterol, and high triglycerides, who were all prescribed simvastatin and who were also randomized to either high-dose, extended-release niacin in gradually increasing doses up to 2000 mg per day (n=1718) or placebo (n=1696). Of the participants, 515 were given a second LDL-cholesterol-lowering drug, ezetimibe (Zetia, Merck/Schering-Plough), in order to maintain LDL-cholesterol levels at the target range between 40 and 80 mg/dL.
However, the combination treatment did not reduce the five-component end point-fatal or nonfatal MI, strokes, hospitalizations for acute coronary syndrome, or revascularization procedures; the primary end point was an annualized rate of 5.8% in the high-dose-niacin group vs 5.6% in placebo group.
Most of the primary-end-point events were hospitalizations or revascularizations (around 140 events), although there were also a "substantial" number of nonfatal MIs and around 40 deaths, the telebriefing was told.
The investigators said they did not have sufficient information yet to determine the role that ezetimibe played in the subgroup that also received that drug.
During the 32-month follow-up period, there were 28 strokes (1.6%) reported among participants taking high-dose, extended-release niacin vs 12 strokes (0.7%) in the control group. Nine of the 28 strokes in the niacin group occurred in participants who had discontinued the drug at least two months and up to four years before their stroke.
All AIM-HIGH study participants have been informed of the results and will be scheduled for clinic visits within the next 2.5 months, the doctors said. Participants will be followed for an additional 12 to 18 months.
Does niacin have effects on different subgroups of HDL?
Asked during the telebriefing about niacin's possible effects on different subgroups of HDL, Boden said: "Niacin is thought to raise the more beneficial subgroups of HDL. We have a very extensive substudy within the trial that remains under analysis, so at some point in the future we can give you much more detail about this."
Radar concurred, stating that it will be informative to see if any subgroups, such as those with LDL above the median, high triglycerides or high lipoprotein(a) levels, for example, appear to benefit from niacin. Blumenthal agrees: "We will need to see if there are some subgroups that had any trend for benefit such as those with a very low HDL cholesterol or a moderately high triglyceride level or a high Lp(a)." However, Rader also cautioned that the small size of the AIM-HIGH trial "may make this difficult."
AIM-HIGH was funded by the NHLBI with additional support from Abbott. McGuire consults for Sanofi-Aventis, Roche and Genentech. Rader has consulted for Abbott in the past, and currently consults for Merck. Kaul is a consultant for Hoffman-La Roche, which is conducting a trial on dalcetrapib, an HDL-modifying drug.
1. National Institutes of Health. NIH stops clinical trial on combination cholesterol treatment [press release]. May 26, 2011. Available here.
2. Food and Drug Administration. FDA statement on the AIM-HIGH trial. May 26, 2011. Available here.
· Low HDL, even in statin-treated patients, associated with increased CVD risk
[Lipid/Metabolic > Lipid/Metabolic; Dec 22, 2010]
· New trial of fenofibrates needed, says FDA panel
[Lipid/Metabolic > Lipid/Metabolic; May 19, 2011]
· ARBITER 6-HALTS: Final results from complete study population published
[Lipid/Metabolic > Lipid/Metabolic; Apr 14, 2010]
· Statins do not eliminate risk of low HDL-cholesterol levels
[Lipid/Metabolic > Lipid/Metabolic; Nov 17, 2009]
· ARBITER 6-HALTS: HDL raising with niacin superior to ezetimibe
[Lipid/Metabolic > Lipid/Metabolic; Nov 15, 2009]