Dr. Jones: Welcome to an expert roundtable discussion on the clinical benefits of niacin for the treatment of dyslipidemia and on the strategies to improve adherence to long-term use. I’m Dr. Peter H. Jones, Associate Professor in the Section of Atherosclerosis and Lipid Research at Baylor College of Medicine and joining me are Dr. Bradley Bale, Medical Director of a heart health program, Grace Clinic, and Clinical Assistant Professor at Texas Tech Health Science Center in Lubbock, Texas; and Ms. Amy Doneen, a Nurse Practitioner and the Medical Director of the Heart Attack and Stroke Prevention Center in Spokane, Washington, and an Adjunct Professor at the Texas Tech Health Science School of Nursing.
The overwhelming and indisputable clinical outcomes benefit of reducing low-density lipoprotein cholesterol (LDL) in high-risk patients with statins has made this drug class first step treatment in all cardiovascular (CV) guidelines around the world. While relative CV risk reductions with statins have been between 25% and 40% in randomized clinical trials over five years, the total CV disease event rate still remain well over 20% for high-risk subjects during that same time.1
This so-called residual risk most likely has many causes. One plausible contributor is persistently low high-density lipoprotein (HDL) cholesterol and/or high levels of triglyceride-rich lipoproteins. Niacin, which is vitamin B3, has been an effective lipid-lowering drug, and, in adequate doses, can increase HDL cholesterol, reduce triglyceride-rich lipoproteins, as well as reduce LDL cholesterol, non-HDL cholesterol, lipoprotein(a) and apolipoprotein (Apo)B.2
The important clinical question is when and how to use niacin in clinical practice, and how to manage the expected flushing reaction that may interfere with compliance and adherence.
Let me start with a question for Dr. Bale. There have been surrogate outcomes trials such as quantitative coronary angiography and carotid intima-media thickness (IMT), suggesting that combining niacin with statins reduces atherosclerosis progression and may actually induce regression of these measures. However, the Atherothrombosis Intervention in Metabolic Syndrome With Low HDL Cholesterol/High Triglyceride and Impact on Global Health Outcomes (AIM-HIGH) trial3 was recently published showing that statin plus niacin versus statin monotherapy didn’t provide incremental outcomes benefit. Brad, can you give us your analysis of this trial?
Dr. Bale: Yes. There are many issues that can be discussed here, but one that I really haven’t heard discussed at this point that I think may be pertinent is that in the Study to Evaluate the Effect of Rosuvastatin on Intravascular Ultrasound-Derived Coronary Atheroma Burden (ASTEROID) trial,4 where they showed regression of coronary disease, and the Arterial Biology for the Investigation of the Treatment Effects of Reducing Cholesterol 6-HDL and LDL Treatment Strategies in Atherosclerosis (ARBITER 6) trial,5 where they showed regression of carotid-intimal thickness with a statin plus niacin, at the end of study total cholesterol (TC):HDL ratio was 2.8 in ASTERIOD and 2.7 in the ARBITER 6. There are a great deal of published data showing that the ApoB/ApoA1 ratio, or the "poor man’s version," which is the TC:HDL ratio, is the most predictive lipid measurement over time for CV events.
In the recent AIM-HIGH trial neither arm got down into the twos for a final TC:HDL ratio. The mean ratio in the niacin arm in AIM-HIGH ended up at 3.1 and in the statin arm it was 3.6. It may be that to have a bigger impact on event risk, the ratio needs to be driven into the twos. Both arms of AIM-HIGH still carried very significant event risk of 16%, which would be considered extremely high-risk. Perhaps the AIM-HIGH study did not aim low enough.
So I think there is an issue trying to compare the ASTROID trial, the ARBITER 6 trial with the AIM-HIGH in terms of the endpoint lipids that were achieved and in particular the TC:HDL ratio. And I’m curious, Dr. Jones and Ms. Doneen, whether that thought went through your heads when you looked at the results of AIM-HIGH.
Dr. Jones: Yes, Amy, what do you think about the results of AIM-HIGH, and how does that impact your feelings about the combination of niacin and statin compared with statin monotherapy?
Ms. Doneen: The most predictive ratio that we know of, as far as predicting events in men and women globally was published in the INTERHEART trial in 2004.6 Vascular disease is a multifactorial disease state and that’s probably the most important point. Additionally, in 2007,7 the Framingham Health Study showed that ApoB/ApoA1 and TC/HDL was the most predictive of CV events over six years. Most recently, in 2010, the Women’s Health Study was analyzed and TC/HDL was the most predictive lipid parameter for women. When we manipulate one variable such as lipids we want to be as inclusive in that application as possible. TC/HDL and ApoB/ApoA1 clearly provide the most value in regard to predictability of CV events in both men and women. In AIM-HIGH, the ratio did not get to a level that has historically proven CV disease stability, TC/HDL <3.0. The ending ratio in the niacin arm of AIM-HIGH was 3.1 and in the statin-only arm was 3.6.
Also in AIM-HIGH, reflecting back to the multifactorial element of this trial, we can learn about stroke predictability from the INTERSTROKE trial. The top cause of ischemic stroke was elevated blood pressure. Additionally, we have to question whether psychosocial factors were addressed adequately. When we deal with patients clinically on a one-on-one basis, we have the opportunity to assess the disease more carefully. And I believe that’s where the clinical application of this knowledge from AIM-HIGH coupled with data that’s been published historically can be taken into account.8 When we analyze the results from AIM-HIGH, which clinically weren’t what we anticipated—we must examine it for its value, it’s that it re-emphasizes the fact that to manipulate one variable only, such as lipids, along with the hypothesis that in either arm those lipids where not manipulated optimally, may not be enough to change events.
Dr. Jones: Yes, the intensive lowering of LDL-C and Apo B in both of the statin arms seems to be the basis of most guidelines. The answer to whether adding niacin on top of that intensive, targeted treatment could add any outcomes benefit was disappointing. It’s just unfortunate we were not able to see the benefits of what niacin does through increasing HDL-C when both randomized arms were targeted to such intensive reductions in LDL-C and ApoB.
In AIM-HIGH, they were using 2g/d of extended-release niacin in the niacin arm, compared with placebo, and that therapy did increase HDL 25%. Niacin also provided additional LDL-C lowering and Apo B reduction when combined with a statin, so, effects on lipids were seen. In light of the AIM-HIGH results, how do both of you now use, or how will you use niacin in CV disease prevention? Let’s start with Dr. Bale.
Dr. Bale: I think before I get to that question it is important to point out that the actual difference in HDL between the statin arm and the niacin arm at the end of the trial turned out to be only 4 mg/dL. The actual difference in the LDL arm turned out to be only 6 mg/dL. This may have been influenced by confounding therapy with ezetimibe, which was used in 22% of the individuals in the statin arm and only 10% in the niacin arm. The difference in HDL and LDL would predict about a 10% reduction in events in favor of niacin. The trial, of course, was not powered to pick up that type of event benefit. They anticipated 25%, but that degree of benefit is unlikely with such a slight difference between the HDL and LDL values.
So I think it’s unfortunate that they allowed therapy with ezetimibe. That medication may have changed some of the study end points. The bottom line is that it is important to point out the trial neither confirmed nor negated the hypothesis that niacin is a good agent. There are lots of reasons to use niacin. It may be excellent therapy for a high-risk population, such as individuals who just suffered an acute coronary event. The AIM-HIGH trial did not address that hypothesis. The AIM-HIGH patients were stable; they weren’t acute or high-risk individuals.
There are practical matters to consider. Statins are marvelous agents and we believe they are the cornerstone of therapy in the reduction of CV risk There are some patients who have trouble tolerating the statins. All of us run into the myalgia problems with statins. They are real and must be taken seriously. Biopsy studies have been perfomed demonstrating actual damage to the muscle fibers. When a patient is taking a statin and complains about muscle pain, it doesn’t matter what the creatin kinase level is; those complaints need to be taken very seriously.
Some of these individuals cannot tolerate even moderate doses of statins. Many can tolerate lower doses. You can get very creative, of course, with rosuvastatin. With a 19-hour half-life it’s not unusual for us to place a patient on half of 5 mg and have them take it every other day. It doesn’t have to be dosed every day and many patients with statin myalgia can get by with that amount of rosuvastatin. Low-dose statins can provide significant benefit but will leave a lot of patients exposed to CV risk from lipids that are not ideally lowered. If the patient has the insulin-resistant condition, which is present in about 70% of people with atherosclerosis, they very likely will have complex lipid issues not only with the HDL but problems with LDL and HDL along with elevated triglycerides.9 They may be unlucky enough to also have inherited a lipoprotein(a) problem. Low dose statin treatment is not going to take care of all those issues. These patients would be excellent candidates for niacin therapy.
We certainly still plan to use niacin in patients. The AIM-HIGH trial did not conclude that niacin should not be used. It didn’t say it was a bad agent or caused harm. The study was stopped for futility, not harm. I think there are many patients who will be candidates for niacin therapy, especially those with the statin myalgia issues.
Dr. Jones: Amy, what are you going to do in your practice?
Ms. Doneen: Yes, I agree. One thing I’d like to point out is that Dr. Philip Barter from the Heart Research Institute in Sydney, Australia really made a great statement in the AIM-HIGH evaluation. Brad mentioned it and I just want to highlight it, that this trial in no means challenges the hypothesis that HDL is an important element in the equation. It doesn’t really challenge the hypothesis because the study lacked the power to tell us that either it did or it didn’t. Nor does the AIM-HIGH trial mitigate what we’ve learned historically with Dr. Greg Brown’s work, which showed beneficial effect of niacin/statin combination on disease stabilization. Additionally, Dr. Allen Taylor's work with cIMT and niacin/statin combination also demonstrates disease benefit.
The concept that we need regression or stability of lesions to halt vascular events is best documented clinically with inflammatory biomarkers coupled with study of structure. So what I’d like to see in trials moving forward and even with the Heart Protection Study 2 Treatment of HDL to Reduce the Incidence of Vascular Events (HPS2-THRIVE),10 is clinical structure analysis whether it is carotid IMT, intravascular ultrasound, something else to show us that the disease is halting and stopping its growth pattern. Intravascular ultrasound–derived radio frequency imaging can visualize necrotic core volume stabilize, coupled with biomarkers that are clinically useful. Clinical biomarkers that provide demonstration of arterial wall stability include high-sensitivity C-reactive protein, microalbumen, albumen/creatinine urine ratio, and lipoprotein-associated phospholipase A2.