July 16th, 2014

HPS2-THRIVE: The Final Chapter in the Niacin Story?


Today’s publication of findings from the HPS2-THRIVE trial raises uncomfortable questions for physicians. Even if some subgroups who may benefit from niacin are yet to be studied, did the drug deserve to generate almost a billion dollars in sales? Could we have learned sooner about the significant increases in adverse effects that HPS2-THRIVE found?

Kudos to the marketers. They took a drug that is unpleasant to take and transformed it into one of America’s more lucrative drug franchises, convincing many that the proprietary form of the drug was worth the cost. HPS2-THRIVE revealed that extended-release niacin, even with the laropiprant to blunt the flushing, could not be tolerated by a third of the participants in the run-in period, and adherence across the duration of the trial was less than 80%. (Click here for CardioExchange’s news coverage of the trial.)

From a regulatory standpoint, the FDA hit the nail on the head when it refused to approve Merck’s combination of niacin and laropiprant in 2008, seemingly in an effort to defer a decision until the outcomes data from HPS2-THRIVE became available. In contrast, the European Medicines Agency did approve the drug, but it promptly withdrew its approval in 2013 as the top-line results of HPS2-THRIVE were announced.

Although the rationale behind the FDA’s decision is not publicly available, it appeared to mark a deviation from the agency’s earlier willingness to approve lipid-lowering drugs on the basis of surrogate endpoints. The FDA’s approach to the novel PCSK9 inhibitors, which powerfully lower LDL but have an uncertain effect on real-world outcomes, represents an important opportunity for the agency to reaffirm its commitment to emphasizing meaningful efficacy data.

Lastly, the commercial success of AbbVie’s niacin product, Niaspan, is quite a story. The company booked more than $900 million in sales of the drug in 2012, the first full year after AIM-HIGH was halted, with price increases offsetting declining prescriptions for the drug. It went on to clock $650 million in sales in 2013, after the HPS2-THRIVE results were first announced. Although the generic competition, which finally arrived at the end of 2013, will rapidly deplete AbbVie’s cash cow, physicians who continue to prescribe niacin in great numbers may now be asking themselves why — and at what cost to patients and the health care system?

Share your thoughts with us here.

22 Responses to “HPS2-THRIVE: The Final Chapter in the Niacin Story?”

  1. Gervasio Antonio Lamas, MD says:

    Niaspan, like Lovaza, and like diltiazem (post MI) a generation before, are triumphs of marketing. Slim data, aggressive and attractive reps, money flowing to speakers, nice dinners, fooled me and others for years. I am quite annoyed because we should have known better. We have to remember that clinical trials are key. Surrogate endpoints of “particle size” and “particle number” are nonsense without solid prospective outcomes data.

    We can’t forget that although big pharma contributes to research and discovery, their fiduciary responsibility is to the shareholder, not to the patient.

    My advice to clinical trialists doing commercially sponsored research is from the old sergeant in Hill Street Blues,”Be careful out there”.

    My advice to politicians is to double the NIH and FDA budgets.

    Is anybody listening?

  2. Peter Zuendorf, MD says:

    Sure, the hype about niacin is not warranted.
    Still I hold that important questions remain:
    1. Its possible value for people with very low HDL
    2. and in people with intolerance or non response to statins.

    Similarly the hype about ezetemibe is not justified. LDL still is only a surrogate. After many years, there are no valid prospective data on ezetemibe. (I presume that this is one of the main motives of the new ACC lipid guidelines. )

  3. Peter Zuendorf, MD says:

    PS: as to the question of too low LDL-levels: so called primitive peoples (Papua, Amazone Indians) usually have an LDL below 30 mg/dl, levels hardly ever seen even in treated patients …

  4. The niacin story can teach us a lot–if we let it. Dr. Lamas’ point about marketing hype is one important lesson. Here are five other lessons.


  5. Pablo Corral, MD says:

    The combination of Niacin and Laropiprant,that combination is the one with bad side effects, not Niacin. Niacin is a vitamin, so it has no bad side effects when taken in the doses that doctors recommend. If you combine Niacin with a drug like Laropiprant is when the problem appears.

    • Ronald Hirsch, MD says:

      You are correct Pablo- vitamins have no bad side effects in doses that doctors recommend, because that dose is Zero, an even lower dose than a placebo’s dosage.

  6. Donald M Lloyd-Jones, MD, ScM says:

    Nicholas and Harlan – Nice perspectives. Obviously I agree. DLJ

  7. H Robert Silverstein, MD says:

    What if a study treated patients with ROSUVASTATIN 40 mg and the goal/PERFECT/disease-reversing non-HDL cholesterol reached was 90 (HDL 44 & LDL 63 as in HPS2-THRIVE). And then it was decided to add 80 mg of ATORVISTATIN to look for improved atherosclerotic end points. Since disease stabilizing/reversing values had already been obtained before adding the second lipid lowering medication, no doubt only further toxicity would be apparent, as was the case in HPS2-THRIVE. It is disappointing to see such a large study created by such important lipid specialists. Frankly, I consider HPS2-THRIVE unethical and the authors should be appropriately criticized for this pointless undertaking. I felt the same about AIM-HIGH. HRS, MD, FACC

  8. Jing Li, PhD says:

    Agree with Dr. Downing and Dr. Krumholz.

  9. H. Joel Gorfinkel, MD says:

    Last nail in the niacin coffin.

  10. Leon Hyman, Ms M.D. says:

    Any one who still uses niacin to prevent cardiovascular disease, probably still believes that bleeding, blistering and purging can cure most diseases. This comment is a bit sarcastic.

  11. Are you people serious? You believe that HPS2-THRIVE’s failure is about niacin? What if we did a study of lipitor plus Vioxx to see if Vioxx stopped the muscle aches of lipitor. While Lipitor reduced the MI incidence by 30 to 40%, the Vioxx would increase the MI incidence by 150%. The study would show increased events with this combination. Would that mean that we should stop using Lipitor?

    Laropiprant is an anti-prostaglandin and as with every other member of this class of drugs likely has an adverse effect on cardiovascular events. A study from Yale New Haven demonstrated that Laropiprant blocks the niacin effect on platelet function.

    Any conclusion reached about niacin based on HPS-THRIVE is therefore bogus.

    AIM-HIGH is also massively flawed. The placebo in this study was niacin. The treatment dose was too high at 2,000 mg. As an intent to treat trial, the 1/4th of the subjects who stopped the drug were considered as though they were still taking it. The average baseline LDL was 70. And finally, a large percentage of the subjects were in China, ignoring the knowledge that Chinese subjects have been shown to be more sensitive to the toxicities of niacin.

    A statistical fact that is ignored by both of these studies (and all of the editorials)is that a study’s failure to disprove the null hypothesis does not ipso facto prove the null hypothesis. Although both AIM-HIGH and THRIVE failed to disprove the null hypothesis that Niacin does not work, they did not come close to proving the null hypothesis.

    A meta-analysis was published in JACC in 2012: “The Current State of Niacin in Cardiovascular Disease Prevention A Systematic Review and Meta-Regression” Paul M. Lavigne, MD, Richard H. Karas, MD, PHD
    Boston, Massachusetts. This study included the flawed AIM-HIGH trial plus all prior niacin trials with an outcome of hard cardiovascular events. The result was a dramatic reduction in heart attacks among those randomized to niacin.

    I plan on continuing to do what I can to prevent heart attacks. I will continue to use niacin in appropriately selected subjects.

  12. H Robert Silverstein, MD says:

    ND, HMK, & DL-J: I am sufficiently insecure or irritatingly ignored for being correct, that I request you 3 comment on W L Blanchet & my points above. HRS, MD,FACC

  13. While I agree that you can always find flaws in research, I find the flaws in AIM-HIGH and HPS2-THRIVE to be exceptionally glaring. I have come up with two metaphorical studies that I think capture the essence of my concern.

    My biggest concern about AIM-HIGH is that they used niacin as the placebo. 2,000 mg of niacin as the study drug and 200 mg of niacin as the placebo. Lets do a metaphorical study with aspirin: 800 mg of aspirin as the study drug and 80 mg of aspirin as the placebo. We would find that the 800 mg aspirin dose studied had no benefit with increased toxicity compared to the “placebo”. The conclusion would be that aspirin does not work and has increased side effects compared to placebo. The reality as we know it now is that aspirin reduces heart attacks by about 20%, colon cancer by 35% and all cancers by about 20% however 80 mg works as well as any dose. If you recall, we used to dose aspirin at 1,300 mg a day while today we know that 80 mg does just as well. I have found that niacin in a dose as low as 250 mg has resulted in stabilizing calcified plaque. This dose is only 50 mg above the placebo dose in AIM-HIGH.

    I have a metaphorical study for HPS2-THRIVE. HPS2-THRIVE was a study of a stain vs statin plus niacin plus Laropiprant (Laropiprant is an antiprostaglandin designed to prevent the flushing from niacin). Lets do a metaphorical study using niacin vs niacin plus lipitor (which sometimes causes muscle soreness) plus Vioxx (an antiprostaglandin designed to treat muscle soreness). As we know from several studies, lipitor has been shown to reduce heart attacks by 20 to 40% (depending on the study you chose) while vioxx was taken off the market a decade ago due to a 150% increase in heart attack risk. This metaphorical study would show that lipitor plus vioxx increases heart attacks compared to niacin alone. Should we then take lipitor off the market? I think not. We should instead look at lipitor independently, not combined with the potentially dangerous drug Vioxx. However when we use niacin in combination with a potentially dangerous drug Laropiprant, everyone seems to want to throw niacin out as the problem.

    Niacin suffers from a lack of good studies. There are almost no legitimate niacin studies large enough to reach a meaningful outcome relating to coronary death. We then get two fatally flawed studies that are big enough to distract any meta-analysis of niacin use.

    I find it interesting that, based on a recent meta-analysis by Keene et al niacin without statin reduces MI by 31% while niacin plus statin had a small but statistically inconclusive benefit compared to statin alone. The observation could easily be made that statin adds no additional benefit to niacin alone.

  14. Pablo Corral, MD says:

    My point is that we can’t compare the AE of Niacin in AIM-HIGH from HPS2-THRIVE.

    First because the trials weren’t similar and you have a lot of different things (the Chinese population, the use of laropiprant in HPS2-THRIVE).

    Second you can find that only slight significant raise of infections you find in AIM HIGH. All the rest serious AE reported in HPS2-THRIVE didn’t find statistical significance in AIM HIGH.

    So you can’t extrapolate and fusion all the AE of this two trials; it is not a small detail the use of laropiprant.

    There are different extended-release niacin preparations and the HPS-2 study saw many more serious adverse events than did in AIM-HIGH. I believe that laropiprant might be partially responsible for this. I believe these two studies are fundamentally different in terms of the study populations and the agents employed.

  15. Without repeating what he wrote, I agree with William Blanchett. If we stop using niacin on the basis of the two trials under discussion, we could well be throwing out the proverbial baby with the bath water. There have been nearly 900 studies of niacin, and a vast majority of them have supported its effectiveness in raising HDL. Moreover, nearly all work on HDL in the past decade continues to show that low HDL-C is an independent risk factor for CVD. In addition, low HDL-C is associated with cognitive impairment, as well as a number of neurodegenerative diseases.

    Many cardiovascular researchers and clinicians appear to be jumping on the anti-niacin bandwagon, but they shouldn’t be. I think that before we drop niacin and stop worrying about low HDL, we need to complete a series of studies that are much more definitive than those that have become the current (and far too limited) focus of the discussion. The HPS2-THRIVE and AIM-HIGH studies are simply not an adequate basis for concluding that niacin should not be used.

  16. Enrique Guadiana, Cardiology says:

    I always wonder why many believe risk is a two way avenue. If you have low cholesterol you are at risk for CV disease, then if you increase HDL you resolve the problem. If you have green eyes probably you are caucasian so if you use green contact glasses, does this make you become a caucasian? The relationship of high HDL and good outcomes is not always causal – effect. It’s more important how you increase HDL (excercice, glucose control, insuline resistance, etc).

  17. I agree Enrique, however the fact that increasing HDL sometimes does not work e.g. CTIP inhibitors, does not mean that it never works. The fact that there is a nice meta analysis of niacin studies by Paul M. Lavigne, MS and Richard H. Karas, MD, PHD published in JACC in 2013 demonstrating a significant reduction in coronary events among those on niacin, comparable or greater than the benefit we see with statin therapy, should influence our discussion and conclusions.

  18. Enrique Guadiana, Cardiology says:

    True Dr. Blanched. My point is I agree with the statement from Dr John M Mandrola “The niacin failure, taken together with failures of many other potent HDL-raising drugs, strongly suggest that the relationship of high HDL and good outcomes is not causal. HDL may be a risk marker but it is not a risk factor.”

    Usually the most significative raise in HDL cholesterol is when we address the prinary causes ej. Diabetes, metabolic syndrome, etc. not with HDL raising drugs. Also I wonder how much of the CV benefit is from the lipid lowering drug and the change in the lipid profile and how much is from unintended effects from the drugs.

  19. I agree Enrique. Indeed the 2013 meta-analysis in JACC showing a significant benefit from niacin did not correlate the level of benefit with with level of improvement in HDL cholesterol.

    When I use niacin, I begin it independent of HDL if the patient is demonstrating progression of calcified plaque by EBT imaging. I have been able to demonstrate, in two separate retrospective analyses, stabilization of calcified plaque, (and the associated reduction in coronary events), by adding niacin to the preventive protocol when plaque is otherwise progressing.