Niacin for Reduction of Cardiovascular Risk

Transcription

1 correspondence Niacin for Reduction of Cardiovascular Risk To the Editor: The lack of benefit for treatment with niacin and laropiprant that is described in the Heart Protection Study 2 Treatment of HDL to Reduce the Incidence of Vascular Events (HSP2- THRIVE) (July 17 issue) 1 was expected, given the low level of low-density lipoprotein (LDL) cholesterol (63 mg per deciliter) and the normal level of high-density lipoprotein (HDL) cholesterol (44 mg per deciliter) prior to randomization. With statin therapy, a high pretreatment LDL cholesterol level (190 mg per deciliter) is associated with an absolute reduction in the risk of myocardial infarction and death from cardiovascular causes of approximately 1.2 percentage points yearly, an intermediate level (120 to 150 mg per deciliter) with a reduction of approximately 0.6 percentage points yearly, and a low level (approximately 100 mg per deciliter) with a reduction of approximately 0.3 percentage points yearly (Fig. 1). 2,3 With nonstatin therapy (niacin, 4 gemfibrozil, fish oil, and partial ileal bypass), the relationship between the LDL cholesterol level and the absolute risk reduction is similar to that observed with statins. The absolute risk reduction with niacin laropiprant in HSP2-THRIVE falls near the regression line for statins, suggesting that the results are those expected on the basis of the low LDL cholesterol level prior to randomization. I contend that an LDL cholesterol lowering lipid-based treatment would probably not reduce the risk of myocardial infarction or death from cardiovascular causes in the HPS2-THRIVE population. In addition, the normal HDL cholesterol this week s letters 1940 Niacin for Reduction of Cardiovascular Risk 1944 Spread of Artemisinin Resistance in Malaria 1945 Epidemiology of Blunt Head Trauma in Children level prior to randomization allowed less scope for a treatment that increases this level. This Yearly Absolute Risk Reduction (percentage points) TNT HPS2- THRIVE HIT POSCH LDL Cholesterol Level in Control Group (mg/dl) 4S CDP Niacin GISSI-P LIPID HPS CARE CARDS IDEAL Figure 1. Absolute Reduction in the Risk of Myocardial Infarction or Death from Cardiovascular Causes in Trials of Secondary Prevention with Statin Therapy versus Nonstatin Therapy. The graph shows the yearly absolute reduction in the risk of myocardial infarction or death from cardiovascular causes in statin-based secondary prevention (open symbols), as compared with nonstatin-based secondary prevention (solid symbols), as a function of the low-density lipoprotein (LDL) cholesterol level before randomization (specifically the LDL cholesterol level in the control group). Symbols are labeled with the respective trial acronyms. The LDL cholesterol level in the Coronary Drug Project (CDP) trial was calculated with the assumption that the high-density lipoprotein (HDL) cholesterol level was 40 mg per deciliter, since the HDL cholesterol and LDL cholesterol levels were not measured. The HPS2-THRIVE trial was added to the previously published figure as a filled circle. To convert the values for cholesterol to millimoles per liter, multiply by The abbreviation 4S denotes Scandinavian Simvastatin Survival Study, CARDS Collaborative Atorvastatin Diabetes Study, CARE Cholesterol and Recurrent Events, GISSI-P Gruppo Italiano per lo Studio della Sopravvivenza nell Infarto Miocardico Prevenzione, HIT HDL Intervention Trial, HPS Medical Research Council British Heart Foundation Heart Protection Study, IDEAL Incremental Decrease in Endpoints through Aggressive Lipid Lowering, LIPID Long-Term Intervention with Pravastatin in Ischemic Disease, POSCH Program on the Surgical Control of the Hyperlipidemias, and TNT Treating to New Targets. 1940

2 correspondence trial should have been performed in persons with a lower HDL cholesterol level, in those with a higher LDL cholesterol level, in those with small, dense LDL cholesterol, or in some combination of these groups. Christopher M. Rembold, M.D. University of Virginia Charlottesville, VA crembold@virginia.edu 1. Landray MJ, Haynes R, Hopewell JC, et al. Effects of extendedrelease niacin with laropiprant in high-risk patients. N Engl J Med 2014;371: Rembold CM. To statin or to non-statin in coronary disease considering absolute risk is the answer. Atherosclerosis 2007; 195: Rembold CM. To statin or to non-statin in coronary disease considering absolute risk is the answer. Atherosclerosis 2008; 200: Coronary Drug Project Research Group. Clofibrate and niacin in coronary heart disease. JAMA 1975;231: conclude that HDL cholesterol is not an independent cardiovascular risk factor. Fernando Meneses, M.D. Jerusalen Clinic Quito, Ecuador femete@yahoo.es 1. Lloyd-Jones DM. Niacin and HDL cholesterol time to face facts. N Engl J Med 2014;371: Khera AV, Cuchel M, de la Llera-Moya M, et al. Cholesterol efflux capacity, high-density lipoprotein function, and atherosclerosis. N Engl J Med 2011;364: Camont L, Chapman MJ, Kontush A. Biological activities of HDL subpopulations and their relevance to cardiovascular disease. Trends Mol Med 2011;17: Camont L, Lhomme M, Rached F, et al. Small, dense highdensity lipoprotein-3 particles are enriched in negatively charged phospholipids: relevance to cellular cholesterol efflux, antioxidative, antithrombotic, anti-inflammatory, and antiapoptotic functionalities. Arterioscler Thromb Vasc Biol 2013;33: Khera AV, Patel PJ, Reilly MP, Rader DJ. The addition of niacin to statin therapy improves high-density lipoprotein cholesterol levels but not metrics of functionality. J Am Coll Cardiol 2013;62: To the Editor: In his editorial on HPS2-THRIVE, Lloyd-Jones questions the hypothesis that the HDL cholesterol level is a causal cardiovascular risk factor, noting that an increase in the plasma levels of HDL cholesterol with various therapies has not resulted in cardiovascular benefit. 1 However, the studies on which he bases this conclusion did not quantify subpopulations of patients who have an HDL cholesterol level with demonstrated antiatherogenic properties and did not measure cholesterol efflux capacity. It has been shown that cholesterol efflux capacity from macrophages, a measure of HDL cholesterol function, is independent of the plasma level of HDL cholesterol, 2 because only small, dense, protein-rich HDL cholesterol particles appear to show potent atheroprotective properties across the spectrum of subpopulations of patients with various HDL cholesterol levels. 3 The atheroprotective activity of HDL cholesterol was strongly interrelated with the subpopulation with HDL3 cholesterol but not with the subpopulation with HDL2 cholesterol. 4 Also, in a small, prospective, randomized study that measured plasma levels of HDL cholesterol, cholesterol efflux capacity, and the HDL inflammatory index, the addition of niacin to statin therapy led to favorable changes in the patients lipid profiles without a demonstrable effect on HDL cholesterol functionality. 5 It is, therefore, premature to To the Editor: In HPS2-THRIVE, there was a trend toward better outcomes in subgroups with a high baseline LDL cholesterol level ( 58 mg per deciliter) or a high apolipoprotein B level. With niacin, the percentage reduction in the LDL cholesterol level becomes less with low pretreatment levels, whereas statins are associated with a fairly constant percentage reduction over a wide range of baseline levels. 1 Data from Hong Kong show no average reduction with niacin in the LDL cholesterol level among patients with a baseline level below approximately 65 mg per deciliter, and some patients had an increase (Fig. 1). Similar relationships were seen for non-hdl cholesterol and apolipoprotein B. Niacin increases the HDL cholesterol level substantially but has modest effects on the apolipoprotein A-I level (mean [±SD] increase of 4.4±9.6% in the Hong Kong study), which may be a better marker of HDL particle functionality than the HDL cholesterol level. 2 We might predict that when the benefits of reducing the LDL cholesterol level are lost at low baseline levels, the potential benefit from the small increase in the apolipoprotein A-I level could be offset by adverse effects, such as increased glucose levels. The role of laropiprant in the HPS2-THRIVE outcome is uncertain, but a troublesome exanthematous rash developed in 14% of the Hong Kong Chinese patients who re- 1941

3 Percentage Change in LDL Cholesterol Level ceived niacin laropiprant, which was not seen with niacin alone. 3 Miao Hu, Ph.D. Brian Tomlinson, M.D. Chinese University of Hong Kong Shatin, Hong Kong btomlinson@cuhk.edu.hk Dr. Tomlinson reports receiving research funding and lecture fees from Merck Sharp & Dohme and Abbott Laboratories. No other potential conflict of interest relevant to this letter was reported. 1. Nicholls SJ, Brandrup-Wognsen G, Palmer M, Barter PJ. Meta-analysis of comparative efficacy of increasing dose of atorvastatin versus rosuvastatin versus simvastatin on lowering levels of atherogenic lipids (from VOYAGER). Am J Cardiol 2010; 105: Al-Hijji M, Martin SS, Joshi PH, Jones SR. Effect of equivalent on-treatment apolipoprotein levels on outcomes (from the AIM-HIGH and HPS2-THRIVE). Am J Cardiol 2013;112: Yang YL, Hu M, Chang M, Tomlinson B. A high incidence of exanthematous eruption associated with niacin/laropiprant combination in Hong Kong Chinese patients. J Clin Pharm Ther 2013;38: R 2 =0.26 P< Pretreatment LDL Cholesterol Level (mg/dl) Figure 1. Niacin Laropiprant and LDL Cholesterol Levels in the Hong Kong Study. The graph shows the relationship between the baseline LDL cholesterol level and percentage change in the LDL cholesterol level after treatment with 2 g of extended-release niacin and 40 mg of laropiprant for 8 weeks in 121 patients from Hong Kong. The solid line shows the best fit to a cubic equation. To convert the values for cholesterol to millimoles per liter, multiply by To the Editor: In his editorial on HPS2-THRIVE, Lloyd-Jones states that the HDL cholesterol level should be considered only as a risk marker for coronary heart disease and that increasing the HDL cholesterol level does not affect cardiovascular risk. This opinion is largely based on the results of HPS2-THRIVE and the Atherothrombosis Intervention in Metabolic Syndrome with Low HDL/High Triglycerides: Impact on Global Health Outcomes (AIM-HIGH) 1 study. In our opinion, this conclusion is premature. In both randomized, controlled trials, niacin (which increases the HDL cholesterol level) was compared with placebo but with the concomitant use of statins and with low mean LDL cholesterol levels at baseline (1.63 mmol per liter in HPS2-THRIVE and 1.91 mmol per liter in AIM-HIGH). Moreover, half the patients in HPS2-THRIVE had a normal HDL cholesterol level (mean, 1.14 mmol per liter). This probably explains the modest increase in the HDL cholesterol level (0.16 mmol per liter) and decrease in the LDL cholesterol level ( 0.25 mmol per liter) and, in turn, the lack of a significant effect on major vascular events. Moreover, this design can lead to more drug interactions and, therefore, more side effects in the group receiving multiple drugs. Finally, the atheroprotective properties of HDL cholesterol are based not only on concentration but also (and probably more) on the protein composition of the HDL cholesterol particle. 2 Therefore, drugs that substantially increase HDL cholesterol levels or alter HDL-protein composition could have atheroprotective effects. 3 Inge A.M. van den Oever, M.D. Michael T. Nurmohamed, M.D., Ph.D. Jan van Breemen Research Institute Reade Amsterdam, the Netherlands i.vd.oever@reade.nl Willem F. Lems, M.D., Ph.D. VU University Medical Center Amsterdam, the Netherlands 1. The HPS2-THRIVE Collaborative Group.. Niacin in patients with low HDL cholesterol levels receiving intensive statin therapy. N Engl J Med 2011;365: [Erratum, N Engl J Med 2012;367:189.] 2. Navab M, Reddy ST, Van Lenten BJ, Fogelman AM. HDL and cardiovascular disease: atherogenic and atheroprotective mechanisms. Nat Rev Cardiol 2011;8: van Eijk IC, de Vries MK, Levels JH, et al. Improvement of lipid profile is accompanied by atheroprotective alterations in high-density lipoprotein composition upon tumor necrosis factor blockade: a prospective cohort study in ankylosing spondylitis. Arthritis Rheum 2009;60:

4 correspondence To the Editor: The results of HPS2-THRIVE show that the addition of extended-release niacin and laropiprant to statins does not reduce cardiovascular risk, leading the editorialist to conclude that HDL cholesterol is futile as a therapeutic target. Because the patients in HPS2-THRIVE were assigned to the combination of niacin and laropiprant, it is not possible to ascertain which agent is responsible for the unexpected adverse results. Knockout mice lacking the prostaglandin D2 receptor DP1 (inhibited by laropiprant) have an increase in atherogenesis, 1,2 thrombogenesis, 1 aneurysm formation, 1 and hypertensive response to angiotensin II. 1 The niacin-induced reduction of 10 mg per deciliter in the LDL cholesterol level that was observed in HPS2-THRIVE should translate into a reduction in cardiovascular risk of 6%. 3 The absence of a cardiovascular benefit despite the reduction in the LDL cholesterol level might be explained by the aforementioned laropiprant-induced deleterious effects. The question of whether the lack of therapeutic efficacy in HPS2-THRIVE is due to laropiprant or niacin could have been addressed only with a 2-by-2 factorial design. We concur with the lack of clinical efficacy of niacin laropiprant, but we cannot conclude that HDL cholesterol is futile as a therapeutic target. Carlos G. Santos-Gallego, M.D. Juan Badimon, Ph.D. Icahn School of Medicine at Mount Sinai New York, NY carlos.santos-gallego@mssm.edu 1. Song WL, Stubbe J, Ricciotti E, et al. Niacin and biosynthesis of PGD₂ by platelet COX-1 in mice and humans. J Clin Invest 2012;122: Strack AM, Carballo-Jane E, Wang SP, et al. Nicotinic acid and DP1 blockade: studies in mouse models of atherosclerosis. J Lipid Res 2013;54: Cholesterol Treatment Trialists (CTT) Collaboration. Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170,000 participants in 26 randomised trials. Lancet 2010;376: To the Editor: Two recent trials (HPS2-THRIVE and AIM-HIGH 1 ) show a higher incidence of infection among patients taking extended-release niacin than among patients taking placebo. Neither the trials nor the accompanying editorial attempt to explain this finding. Immune cells express niacin receptors (Niacr1 and Niacr2). In rodents, 0.4 g per day, g per day, 3 or 2.9 g per day 4 (human-equivalent dose) of niacin induces immunosuppressive regulatory T cells 3 ; increases the circulating levels of immunosuppressive interleukin-10 3 ; inhibits macrophage differentiation and migration, 3 neutrophil migration, 4 and leukocyte adhesion 4 ; reduces circulating levels of proinflammatory interleukin-17, 3 high-mobility group B box 1, 2 and matrix metalloproteinase 9 2 ; and suppresses clinical inflammation. 2-4 The immunosuppressive effects 3 are Niacr1-mediated and disappear in Niacr1 / mutant mice. In addition to the immunosuppressive effects of niacin on the host in vivo, 2-4 niacin also induces virulence-associated traits in certain bacterial pathogens in vitro. 5 The immunosuppressive effects of niacin 2-4 and possible enhancement of bacterial pathogenicity 5 might help explain the higher incidence of infection seen with niacin in the above trials. Lawrence Mayer, M.D. 16 Hudson Rd. Lexington, MA lim3@cornell.edu 1. Anderson TJ, Boden WE, Desvigne-Nickens P, et al. Safety profile of extended-release niacin in the AIM-HIGH trial. N Engl J Med 2014;371: Ye X, Chopp M, Liu X, et al. Niaspan reduces high-mobility group box 1/receptor for advanced glycation endproducts after stroke in type-1 diabetic rats. Neuroscience 2011;190: Singh N, Gurav A, Sivaprakasam S, et al. Activation of Gpr109a, receptor for niacin and the commensal metabolite butyrate, suppresses colonic inflammation and carcinogenesis. Immunity 2014;40: Ferreira RG, Matsui TC, Gomides LF, et al. Niacin inhibits carrageenan-induced neutrophil migration in mice. Naunyn Schmiedebergs Arch Pharmacol 2013;386: Edwards RL, Bryan A, Jules M, Harada K, Buchrieser C, Swanson MS. Nicotinic acid modulates Legionella pneumophila gene expression and induces virulence traits. Infect Immun 2013;81: Dr. Landray and Colleagues Reply: Despite effective statin-based LDL cholesterol lowering therapy, patients with atherosclerotic disease remain at high risk for vascular events. HPS2- THRIVE showed that the addition of extendedrelease niacin plus laropiprant to statin-based therapy did not significantly reduce the risk of vascular events among such patients. However, 1943

5 this niacin-based regimen was associated with significant and clinically important harms, including excesses of bleeding and infection that had not been associated with niacin previously. Rembold and Hu and Tomlinson suggest that no further benefit of lowering the LDL cholesterol level should be expected from the addition of niacin, since the mean LDL cholesterol level was 63 mg per deciliter in patients receiving statin-based therapy at baseline. However, in the Cholesterol Treatment Trialists Collaboration meta-analysis of statin trials, 1 the proportional reductions in the risk of vascular events observed for a given absolute reduction in the LDL cholesterol level were similar among persons who had an initial LDL cholesterol level below 70 mg per deciliter and those who had a higher level. Moreover, even among the 4900 participants in HPS2-THRIVE who had a baseline HDL cholesterol level of less than 35 mg per deciliter, assignment to niacin-based therapy was not associated with significant benefit. Overall in HPS2-THRIVE, assignment to the group that received niacin plus laropiprant in addition to statin-based therapy (simvastatin at a dose of 40 mg daily, with or without the addition of 10 mg of ezetimibe daily) produced only modest differences in lipid levels: an LDL cholesterol level that was 10 mg per deciliter lower and an HDL cholesterol level that was 6 mg per deciliter higher, as compared with statin-based therapy alone. On the basis of randomized trials of statins, we would expect that a reduction of 10 mg per deciliter in the LDL cholesterol level would produce a proportional reduction of 5 to 6% in the rates of vascular events. No similar evidence from a randomized trial exists for the effects of raising the HDL cholesterol level, but if the association with vascular disease risk in observational studies is causal and half of the risk is reversible within a few years (as with LDL cholesterol lowering therapy), then an increase of 6 mg per deciliter might plausibly produce a reduction in risk of 4 to 5%. In that case, the combined lipid changes might have been expected to reduce the risk of vascular events by approximately 10%, which is entirely compatible with the observed result. Consequently, we agree that it is premature to conclude that raising the HDL cholesterol level does not reduce risk. Martin J. Landray, Ph.D., F.R.C.P. Richard Haynes, D.M., M.R.C.P. Jane Armitage, F.R.C.P., F.F.P.H. University of Oxford Oxford, United Kingdom jane.armitage@ctsu.ox.ac.uk for the HPS2-THRIVE Collaborative Group Since publication of their article, the authors report no further potential conflict of interest. 1. Cholesterol Treatment Trialists (CTT) Collaboration. Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170,000 participants in 26 randomised trials. Lancet 2010;376: Spread of Artemisinin Resistance in Malaria To the Editor: Ashley et al. (July 31 issue) 1 recently described the spread of Plasmodium falciparum resistant to artemisinin across Southeast Asia but not Africa. This and other studies 2,3 included patients with uncomplicated malaria, whereas young children with severe malaria in Africa represent a key population of interest with the highest mortality burden. From the placebo group of our recent (2011 to 2013) randomized, controlled trial in Uganda, 4 we quantified parasite-clearance kinetics using identical methods (standardized parasite-clearance estimator) 5 among 91 children with severe falciparum malaria treated with intravenous artesunate. The median clearance half-life was 4.6 hours (interquartile range, 3.6 to 5.9) (Fig. 1). Slowly clearing infections, defined by Ashley et al. as those with a clearance half-life of more than 5 hours, occurred in 46% of patients in our study, as compared with 0 to 3% of Africans with uncomplicated malaria and 5 to 73% of Cambodians at the epicenter of resistance in their study. 1 We hypothesize that host factors, disease severity, choice of drug, and route of administration affect clearance kinetics in vivo, in addition to parasite resistance. These factors are relevant for monitoring arte- 1944