HPS2-THRIVE: As Niacin Fails, HDL's Role Is Debated Anew

July 16, 2014

OXFORD, UK ( Updated with commentary ) — More than a year after its public presentation, the Heart Protection Study 2—Treatment of HDL to Reduce the Incidence of Vascular Events (HPS2-THRIVE) is now in print[1], accompanied by an editorial that doesn't mince words: it's time the cardiovascular community face the facts about niacin therapy[2]. The truth, based on randomized, controlled clinical trials, is that niacin, which is given to raise HDL-cholesterol levels, carries an "unacceptable toxicity profile for the majority of patients and it should not be used routinely," editorialist Dr Donald Lloyd-Jones (Northwestern University Feinberg School of Medicine, Chicago, IL) argues.

Lloyd-Jones goes further, throwing cold water on the once-promising field of drug therapies to raise HDL-cholesterol levels.

"The consistent findings of a lack of benefit of raising HDL cholesterol with the use of niacin when added to effective LDL-cholesterol–lowering therapy with statins seriously undermine the hypothesis that HDL cholesterol is a causal risk factor," he states. "The failure (to date) of cholesteryl-ester-transfer-protein [CETP] inhibitors, such as torcetrapib and dalcetrapib , to show any reduction in cardiovascular risk, despite the marked increases in HDL cholesterol associated with these drugs, lends further credence to the notion that HDL cholesterol is unlikely to be causal."

HPS2-THRIVE, along with other trials and studies, including AIM-HIGH , has cast a pall over the HDL hypothesis, that being the previously widespread belief that agents designed to raise HDL-cholesterol levels could reduce the residual risk of cardiovascular disease among patients treated with optimal medical therapy, including statins.

The results of HPS2-THRIVE aren't new, although they were surprising when they were first announced in December 2012. Merck, the study's sponsor, revealed that the combination of extended-release niacin and the antiflushing agent laropiprant failed to reduce the risk of major vascular events in patients with well controlled LDL-cholesterol levels. The study was later presented at the American College of Cardiology 2013 Scientific Sessions in San Francisco, CA and is now published in the July 17, 2014 issue of the New England Journal of Medicine.


The HPS2-THRIVE investigators, led by Dr Jane Armitage (Oxford University, UK), randomized 25 673 patients with vascular disease to 2 g of extended-release niacin and 40 mg of laropiprant or matching placebo for a median of 3.9 years. At baseline, the mean HDL cholesterol was 44 mg/dL and the mean LDL-cholesterol level was 63 mg/dL. As reported by heartwire , there was no significant impact on the primary outcome of major vascular events, defined as nonfatal MI, death from coronary causes, stroke, or arterial revascularization (14.5% for niacin/laropiprant vs 15.0% for placebo; p=0.29).

Treatment with niacin, however, did adversely affect patients who received the therapy. Overall, there was a 1.0% higher absolute risk of gastrointestinal events, a 0.7% higher absolute risk of musculoskeletal events, and a 1.4% higher absolute risk of infection. Skin-related and bleeding adverse events were also significantly higher. For those without diabetes at baseline, there was a significant 32% higher relative risk of developing the disease (1.2% higher absolute risk). There was also a significant 55% higher relative risk of impaired glucose control among those with diabetes at baseline (3.7% higher absolute risk).

"If we had seen the [HPS2-THRIVE] results several years ago, the results would have been very surprising," Dr Jeffrey Berger (New York University Langone Medical Center) told heartwire . Although he was not affiliated with HPS2-THRIVE, Berger pointed out that many researchers in the field had been skeptical that niacin would have an impact on clinical outcomes. He said that other trials, including those with CETP inhibitors and AIM-HIGH, did not even hint at a clinical benefit for raising HDL cholesterol.

"AIM-HIGH really put a large dampening on niacin," said Berger. Still, he doesn't view these negative studies as a death rattle for niacin and other HDL-cholesterol–raising therapies. "I think what it tells us is that raising HDL when LDL cholesterol is so low is not going to be beneficial. Whether or not this type of approach would be helpful in subjects whose LDL cholesterol is not very low remains in doubt."

Dr John Kastelein (Academic Medical Center, Amsterdam, the Netherlands) also sees a possible role for niacin in medical therapy. He does not believe AIM-HIGH or HPS2-THRIVE have "completely excluded" the drug when the agent is used correctly. "Nobody does trials as well as the Oxford group, let me say that, but when you look at the baseline LDL- and HDL-cholesterol levels in this trial, who on this planet would ever throw 2 g of nicotinic acid at a patient with that lipoprotein profile? No one. No one would do that."

If you're a patient and you have those [baseline] lipids on statins, I would say, 'Wow! I'm done.'

He said the changes induced with niacin in HPS2-THRIVE—on average, HDL cholesterol was increased 6 mg/dL and LDL cholesterol decreased 10 mg/dL—were so small that in his clinical practice he would not even have bothered to continue with treatment.

"To be quite frank," Kastelein told heartwire , "I think that when trials are so removed from what we do in daily clinical practice, it becomes silly. If you're a patient and you have those [baseline] lipids on statins, I would say, 'Wow! I'm done.' You can see how these trials are becoming too far removed from reality."

HPS2-THRIVE included 11 000 Chinese patients, and these particular patients had a higher risk of side effects than other patient subgroups. "Compare that with a lipid clinic in Chicago, where you have a patient who is 110 kg and despite statin therapy still has elevated triglycerides and elevated LDL cholesterol," said Kastelein. "My problem with both trials [AIM-HIGH and HPS2-THRIVE] is that they have so vigorously pretreated with statin therapy that baseline lipids were essentially normal. This means that any change you induce with an agent is, by definition, going to be minimal."

Dr William Boden (Samuel Stratton VA Medical Center, Albany, NY) made a similar statement, telling heartwire he does not believe this study or AIM-HIGH, which he led, closes the door on the HDL hypothesis. "This was an all-comers study—it was a very important trial—that didn't really target a population of patients in whom one would expect to see a benefit from niacin," he said. Looking at the normal baseline lipids, Boden, like Kastelein, said no clinician would ever consider using niacin or fibrates or any other intervention.

"It's another negative study, and I guess if you're a niacin naysayer, you could say, we have two negative trials, end of story," said Boden. "Or you could say both of these studies have some design issues or concerns. Given that they were broadly inclusive with no predefined lipid criteria, particularly for low HDL cholesterol, I don't know if you can say this violates the HDL hypothesis."

The flaws inherent in AIM-HIGH and HPS2-THRIVE, with lower but relatively normal baseline HDL-cholesterol levels, are also inherent in the studies testing the CETP inhibitors. "If you're on the relatively flat part of the event curve, why would you expect to see a benefit?" asked Boden.

One Doctor Sees the End of Niacin

For Berger, he pointed to data from Mendelian randomization studies that suggest genetic polymorphisms in HDL cholesterol were not associated with an increased risk of cardiovascular events. "What that tells me is that HDL cholesterol as a number might not be as important as we once thought," he said. "People shouldn't take away from this that HDL is not important. What it should do, in my mind, is to get people to think about functionality. It's not just about how much HDL cholesterol you have, but how well HDL is working."

He added that several studies have shown that functionality, or "HDL flux," is more important than the HDL number for determining the risk of cardiovascular events. Berger also said the clinical trials could still be designed if researchers are able to identify a subgroup of patients where niacin might be beneficial, such as the patient with low HDL cholesterol and high triglycerides or high LDL-cholesterol levels. "I think we have to think about high-risk groups, not entire populations," said Berger. "We need to move away from that."

A Very Small Role Might Still Be Possible

I think we have to think about high-risk groups, not entire populations.

In the editorial, Lloyd-Jones says niacin might have a role in the patient with contraindications for statin therapy who has high LDL cholesterol but emphasized that statins are the cornerstone of treatment in the new clinical guidelines. Niacin might have a part as a fourth-line agent—after lifestyle changes, fibrates, and pharmaceutical-grade fish oil—in patients with high triglycerides who are at risk of pancreatitis. Armitage and the HPS2-THRIVE investigators address this point in the paper and conclude that even when using niacin in these possibly relevant subgroups, physicians need to balance the potential for benefit against the observed hazards.

Boden pointed out that AIM-HIGH and HPS2-THRIVE used different proprietary formulations of extended-release niacin, with the HPS2-THRIVE investigators using laropiprant to reduce flushing, whereas AIM-HIGH simply used aspirin. While many of the off-target effects can be attributed to niacin, such as cellulitis/skin infections, gastrointestinal side effects, and increasing blood glucose levels/dysglycemia, it's impossible to be certain if all the off-target effects are the domain of niacin and not laropiprant. "There seems to be a larger safety signal in HPS2-THRIVE that we didn't see in AIM-HIGH, although there were some similarities," he said.

To heartwire , Dr Prediman K Shah (Cedars Sinai Medical Center, Los Angeles, CA), who was not involved in the study, the purists might argue that laropiprant caused some of the side effects or even interfered with the beneficial aspects of niacin, but this is unlikely to resurrect the drug. Instead, he believes HPS2-THRIVE is the last nail in the coffin for niacin. So far, the only compound to work thus far in the field of HDL is the apolipoprotein (apo) A-1 Milano variant of the HDL complex. In epidemiological studies, the carriers of this particular variant have a low risk of cardiovascular disease. In one landmark study, researchers led by Dr Steven Nissen (Cleveland Clinic, OH) showed that five weeks of treatment with intravenous recombinant apoA-1 Milano reduced coronary atheroma as measured by intravascular ultrasound.

"Nothing else so far has panned out," said Shah. "That's a real big disappointment. The only thing left are the trials with the new CETP inhibitors, and quite frankly, I'm a little skeptical about what those results are likely to be. The rest of the field is replete with failures."

A lot of our preconceived notions turned out to be totally disproven.

Torcetrapib was abandoned when it was shown to increase the risk of death and cardiovascular events, while dalcetrapib had no impact on clinical events. Two other agents are also being tested—evacetrapib (Lilly, Indianapolis, IN) and anacetrapib (Merck, Whitehouse Station, NJ)—but Shah said that since these two trials are still ongoing, their effect is likely to be either small or neutral. In Japan, a third CETP inhibitor is still early in development.

"It's clear that cholesterol levels associated with HDL are not going to be a good biomarker of clinical benefit," said Shah. "For example, infusions of apoA-1 Milano actually resulted in a reduction in HDL, just like the natural carriers, and yet we see a profound biological effect. I think we have to get away from the idea that we can measure HDL cholesterol and define beneficial impact. We need to go beyond that. There's a lot of work that still needs to be done in the HDL field. A lot of our preconceived notions turned out to be totally disproven."

What Lies Ahead?

For Kastelein, he is more excited about the effects of the new inhibitors of proprotein convertase subtilisin-kexin type 9 (PCSK9). As reported previously, these new agents significantly reduce LDL-cholesterol levels. For the average HPS2-THRIVE patient, where the baseline LDL cholesterol was 63 mg/dL, what type of risk reduction would be observed if LDL cholesterol were reduced another 60%? "That's a question I'm interested in," said Kastelein.

In terms of bang for the research buck, Kastelein said there has been a shift in attention away from raising HDL cholesterol toward studies evaluating the role of apoA, remnant cholesterol, and triglyceride-rich lipoproteins in cardiovascular disease. LDL-cholesterol–lowering therapies, among them PCSK9 inhibitors and the use of small interfering RNA to turn off genes involved in cholesterol synthesis, are also garnering a lot of excitement, as researchers aim to lower LDL cholesterol to previously unheard of levels.

There is also a resurgence in interest in apoC3, a protein encoded by APOC3. As reported by heartwire in June, two independent papers published in the New England Journal of Medicine showed that individuals with a genetic mutation in APOC3 not only have lower plasma triglyceride levels but they also have a decreased risk of developing coronary heart disease. "This has resulted in a swing back toward triglycerides, but not with the old drugs," said Kastelein. "I think fibrates and nicotinic acid, while there still might be a place for them, they've had their time in the sun. But everybody is now looking for drugs that can lower apoC3."

HPS2-THRIVE was sponsored by grants from Merck, the UK Medical Research Council, the British Heart Foundation, Cancer Research UK (to the University of Oxford) and the British Heart Foundation Centre of Research Excellence. Armitage has no conflicts of interest; disclosures for the coauthors are listed in the paper. Lloyd-Jones has no conflicts of interest.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.