January 15th, 2013

Is HDL Fool’s Gold?

Merck recently announced that Tredaptive, its compound of extended-release niacin and laropiprant, failed to meet the primary-endpoint goal in the HPS2-THRIVE trial, a comparison with statin therapy alone. We at CardioExchange asked Dr. Peter P. Toth for his perspective on recent trials of niacin, including HPS2-THRIVE and AIM-HIGH. Here is a sampling of his remarks. (See also previous remarks on this topic from William E. Boden and Harlan M. Krumholz.)

In AIM-HIGH, patients had a baseline LDL-cholesterol level of 71 mg/dL, a non-HDL level of 107 mg/dL, and an HDL-cholesterol of 35 mg/dL, and most were chronically treated with a statin. This is not the clinical scenario where niacin is used. It is typically used in patients with low HDL-c, elevated triglycerides, and often residual LDL-c elevation despite statin therapy. The AIM-HIGH patients essentially had pristine control of their atherogenic lipoprotein burden and isolated, mildly low HDL-c. The Framingham study showed that in the absence of insulin resistance, such patients do not have elevated risk compared to patients with more elevated HDL-c.

In HPS-2 THRIVE, the stated hypothesis was “Does niacin combined with extended- release niacin/laropiprant 2 g daily prevent vascular events in high-risk patients who are receiving intensive LDL-lowering treatment?” The trial was not designed to ascertain whether or not raising HDL with niacin affects risk, as low HDL was not an inclusion criterion. It’s also possible that laropiprant may adversely affect vasculoprotective or atheroprotective pathways that niacin therapy influences, but that is conjectural. Gratefully, the FDA insisted that the niacin/laropiprant combination be evaluated in a large randomized prospective study before it considers drug approval. In the type of patient population studied in HPS-2 THRIVE, clearly the addition of Tredaptive to statin therapy did not improve outcomes. Subsequent investigation will have to determine whether Tredaptive offset potential benefit from the combination of statin/niacin used in the trial.

We simply don’t yet have a trial involving high-risk patients specifically with low HDL and high triglycerides to determine whether or not niacin reduces event rates beyond that observed against a background of statin therapy. I’d suggest an average baseline HDL of 28 or 30 mg/dL and triglycerides >200 mg/dL to really put niacin to the test. Otherwise, it’s like trying to treat a cancer with a drug that is not designed to intercept any growth or proliferative feature of the cell type in question. Who would design such a trial?

In ARBITER 2 and 6, the use of adjuvant niacin clearly affected rates of progression of carotid intima-media thickness. Granted, we do not know whether CIMT regression correlates with reductions in cardiovascular events, but most of us believe regression is a good thing. Niacin does appear to limit atherosclerosis disease progression against a statin background. Is the benefit from raising HDL-c? No one can answer that yet. I believe the antiatherogenic efficacy of niacin stems from its effect on all components of the lipid profile.

My concern is that physicians will write HDL therapeutics off before we even begin to understand it. There is evidence that HDL antagonizes inflammation, oxidation, thrombosis, and endothelial dysfunction; it is a modulator of insulin sensitivity and participates in the activation of islet-cell insulin secretion; it comprises many species with variable functionality. In addition, five decades of epidemiology have shown that low HDL is a categorical risk factor for CHD regardless of race, sex, or ethnicity. The list of suggestive evidence goes on.

I personally do not believe that CETP inhibition will reduce cardiovascular event rates because (1) studies with torcetrapib and dalcetrapib revealed no benefit; (2) I see no positive role for inhibiting indirect cholesterol transport to the liver; and (3) epidemiologically, low CETP activity correlates with increased risk for CHD events in multiple studies around the world. We know that HDL’s capacity to stimulate macrophage efflux capacity in humans correlates with risk for CHD-related events. This is an extremely difficult issue to investigate, characterize, and apply therapeutically. It is simply premature to conclude that HDL is fool’s gold. The issue is far more complex than we have thus far envisioned. For now, I will continue to use niacin in patients with low HDL-c, elevated triglycerides, and residual elevations in LDL-c despite appropriate, or maximally tolerated, doses of statins.

What do you think the future of HDL-modifying therapy will be?

2 Responses to “Is HDL Fool’s Gold?”

  1. Neville Wilson, M.B., Ch.B., D.Fam. Med., M.Sc., D.Obs (COG) SA., B.A. says:

    I agree wholeheartedly with Dr.Peter Toth. My understanding from the Framingham data as far back as 1977, showed that HDL was a more potent predictive factor for risk than LDL.
    Considering the anti-atherogenic properties of HDL in respect of inhibition of monocyte adhesion, inhibition of LDL oxidation, and anti-thrombotic potential, the reverse transport capabilities of HDL, mediated by CETP, should not be perceived to be its only, or major, benefit. Reliance, therefore, on the future success of experimental CETP inhibitors is surely misplaced, and hardly a safe or economically viable pharmacological pursuit !
    As stated previously, I will continue to use Niacin in conjunction with HDL enhancing behavioural changes in at risk patients.

  2. Pablo Corral, MD says:

    I totally agree with Dr. Peter Toth. “We can’t fix what it isn’t broke”. We have to remember CDP, FATS, HATS, CLAS, Arbiter, Oxford Niaspan ….