Niacin‐ER/Statin Combination for the Treatment of Dyslipidemia: Focus on Low High‐Density Lipoprotein Cholesterol

Abstract

Statins are effective drugs for lowering low‐density lipoprotein cholesterol, and their use has been associated with a significant decrease in cardiovascular morbidity and mortality. However, statins are ineffective in lowering plasma triglycerides and lipoprotein(a), or increasing low high‐density lipoprotein cholesterol (HDL‐C) plasma levels, which are independent risk factors for coronary heart disease. Niacin, on the other hand, is the most potent drug available for lowering plasma levels of triglycerides and lipoprotein(a) and raising HDL‐C levels. It follows, then, that a combination of niacin with a statin might be an effective combination in improving all components of the lipid profile. Previous studies have shown that the use of long‐acting niacin with a statin, in dose combinations of niacin‐ER/lovastatin 1000/20 mg or 2000/40 mg once daily, has been effective in favorably modifying low‐density lipoprotein cholesterol, triglycerides, lipoprotein(a), and HDL‐C plasma levels. Dyslipidemias often predate the onset of hypertension, and HDL‐C has been found to be inversely related to the incidence of hypertension. Normalization of lipid components, including the total cholesterol/HDL‐C ratio, is important in the management of hypertensive individuals and patients with the metabolic syndrome or diabetes. Thus, the long‐term treatment of dyslipidemias with these two agents may help to modify risk and reduce cardiovascular morbidity and mortality in these patients over and above benefits achieved by lowering blood pressure

The importance of statins in lowering high levels of low‐density lipoprotein cholesterol (LDL‐C) is well established; several large outcomes trials have clearly demonstrated that long‐term treatment with statins is associated with a significant reduction in cardiovascular (CV) morbidity and mortality. ¹ , ² , ³ , ⁴ , ⁵ , ⁶ , ⁷ , ⁸ , ⁹ However, little attention has been paid to low levels of high‐density lipoprotein cholesterol (HDL‐C) as a significant risk factor for coronary heart disease (CHD) and hypertension.

The National Cholesterol Education Program Adult Treatment Panel III report (ATP‐III) ¹⁰ has designated HDL‐C of <40 mg/dL for men and <50 mg/dL for women as a significant CV risk factor even in the presence of normal LDL‐C. ¹⁰ Low HDL‐C levels are quite prevalent in the general population, found in 39% of men and 15% of women, but in subjects with established CHD, the prevalence can be as high as 64%. ¹⁰ , ¹¹ Increasing low levels of HDL‐C is associated with a significant reduction of CV morbidity and mortality, even in the presence of near‐normal LDL‐C. In a combined analysis of four prospective studies of more than 15,000 patients, a strong association was noted between HDL‐C levels and CHD, accounting for a 2% decrease in the rate of CHD events for each mg/dL increase in HDL‐C level. ¹² Two recent, large epidemiologic studies of 14.1 and 10.8 years of observation in 3110 men ¹³ and 16,130 women, ¹⁴ respectively, who were not hypertensive at study onset, reported that in men in the highest quintile of total cholesterol (TC), non‐HDL‐C, and TC/HDL‐C ratio, the risk of developing hypertension was 23%, 39%, and 54% higher, respectively, compared with men in the lowest quintile (p<0.001). In women, ¹⁴ the relative risks of developing hypertension from the lowest to the highest quintile of baseline lipid levels were 1.12, 1.11, and 1.34 for TC, LDL‐C, and TC/ HDL‐C ratio, respectively (p<0.001 for trend). On the other hand, in men and women in the highest quintile of HDL‐C, the incidence of hypertension was 32% and 19%, respectively, lower (p<0.001). Thus, there appears to be a strong inverse relationship between the development of hypertension and HDL levels.

THE CLINICAL RELEVANCE OF LOW HDL‐C

In a lipid screening study of 8500 middle‐aged men with CHD, 38% were found to have HDL‐C levels <35 mg/dL, and 64% <40 mg/dL, the new cutoff level in the ATP‐III guidelines. ¹⁰ Forty‐one percent of these patients had LDL‐C and HDL‐C levels <130 mg/dL and <35 mg/dL, respectively. ¹⁵ Similar findings were reported in the Scandinavian Simvastatin Survival Study (4S), ¹ where 72% of patients had TC levels <200 mg/dL and HDL‐C <40 mg/dL. The concept that low levels of HDL‐C have a strong and independent association with CHD risk has been demonstrated by several studies. In the Framingham Heart Study, ¹⁶ which followed 2815 persons free of CHD at the beginning of the study for 4 years, the HDL‐C level was inversely related with the incidence of CHD (p<0.001) for men and women, even when other standard risk factors for CHD were considered. The high TC/HDL‐C ratio reported by the Framingham investigators as a risk factor for CHD is also a risk factor for hypertension. Several other epidemiologic studies have demonstrated that HDL‐C is inversely related to the incidence of CHD events and that low HDL‐C is independently predictive of CHD and hypertension. ¹³ , ¹⁴ , ¹⁷ , ¹⁸ , ¹⁹ In addition to low HDL‐C levels, high triglyceride (TG) levels have also proven predictive of CHD in several prospective studies. ²⁰ , ²¹ , ²² In a meta‐analysis of 17 prospective studies including 46,413 men and 10,864 women, high TG levels were associated with CHD, even after correction for other risk factors. ²²

EFFECTS OF TREATMENT ON LOW HDL‐C LEVELS

The therapeutic efficacy of the existing hypolipidemic drugs in raising HDL‐C levels is variable. The use of statins and bile acid sequestrants results in approximately a 5%–15% increase in HDL‐C; fibrates may raise HDL‐C by 15%–25%; and niacin preparations may raise HDL‐C by 15%–35%. ¹¹ , ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ The more recently available statins may be more effective. The importance of raising low HDL‐C levels on the incidence of CHD events has been demonstrated by several prospective, randomized clinical outcomes trials, which demonstrate a significant decrease in CV morbidity and mortality by using fibrates or niacin preparations either alone or in combination with other hypolipidemic drugs. ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ In the Bezafibrate Infarction Prevention (BIP) study, ²⁵ 3090 patients with a previous myocardial infarction (MI) with HDL‐C levels of ≤45 mg/dL, LDL‐C ≤180 mg/dL, and TGs ≤300 mg/dL were treated with bezafibrate 400 mg/d or placebo, and followed for 6.2 years. Bezafibrate increased HDL‐C by 18% and decreased LDL‐C and TGs by 6.5% and 21%, respectively. This resulted in a 9.4% decrease in the incidence of CV morbidity and mortality compared with placebo (nonsignificant). However, in the group with higher baseline TGs (≥200 mg/dL), the CV morbidity and mortality was decreased by 39.5% (p<0.02). In the Veterans Affairs High‐Density Lipoprotein Cholesterol Intervention Trial (VA‐HIT), ²⁶ 2531 men with prior CHD, and mean HDL‐C and LDL‐C levels of 32 mg/dL and 111 mg/dL, respectively, were treated with gemfibrozil 1200 mg/d or matching placebo and were followed for 5.1 years. Gemfibrozil increased HDL‐C by 6% (p<0.001) and decreased TC by 4.0% (p<0.001) and TGs by 31% (p<0.001). These lipid changes resulted in a 22% decrease in CHD mortality or nonfatal MI (p=0.006), attributed mostly to the increased levels of HDL‐C. In addition, CHD events and strokes were decreased by 24% (p<0.001). Similar results were reported by the Helsinki Heart Study Investigators, ²⁷ who reported a 34% reduction in CHD events in patients treated with gemfibrozil for 5 years. In contrast, the results of the Fenofibrate Intervention and Event Lowering in Diabetes (FIELD) study ²⁸ were not as robust. In this trial, 9795 dyslipidemic type 2 diabetic patients with a mean baseline TC of 197 mg/dL (5.04 mmol/L), LDL‐C of 120 mg/dL (3.07 mmol/L), HDL‐C of 43 mg/dL (1.1 mmol/L), TG of 174 mg/dL (1.95 mmol/L), and TC/HDL‐C ratio of 4.0 were randomized to fenofibrate 200 mg/d or placebo and were followed for 5 years. The lipid changes with active treatment were modest and did not result in a significant decrease in CHD mortality. However, the rate of nonfatal MIs and revascularizations was significantly decreased with treatment (p=0.03).

The effects of treatment of dyslipidemias on BP reduction or prevention of hypertension have not been systematically studied. Several small, short‐term studies have demonstrated small decreases in BP with the treatment of dyslipidemias, and treatment of the metabolic syndrome has also led to a decrease in BP. ¹³ , ²⁵ Due to the significant inverse relationship of HDL‐C levels to the development of hypertension, a large prospective study addressing the treatment of low HDL‐C for the prevention of hypertension in populations vulnerable to hypertension development is indicated to answer the question whether by raising HDL‐C levels we may prevent the development of hypertension.

Low HDL‐C and high TC, LDL‐C, and TG levels coexist in many dyslipidemias. The four classes of lipid‐modifying drugs (statins, fibrates, bile acid sequestrants, and niacin) exert their major effect on specific lipid or lipoprotein components. Therefore, the ideal treatment of mixed dyslipidemias should be a combination of lipid‐modifying drugs with complimentary action.

THE EMERGING ROLE OF DRUG COMBINATION TREATMENT OF DYSLIPIDEMIA

Niacin Combinations With Fibrates and Bile Acid Sequestrants

The clinical importance of multidimensional therapy of dyslipidemia is underscored by the high prevalence of low HDL‐C with or without elevated LDL‐C or TG in patients with CHD or at high risk for CHD, or those with hypertension, type 2 diabetes, and the metabolic syndrome. ¹³ , ¹⁴ , ²⁶ In the Cholesterol‐Lowering Atherosclerosis Study (CLAS), ²⁹ 162 patients treated with niacin for 2 years demonstrated a significant decrease in the progression of atherosclerosis and new plaque formation in the coronary arteries (p<0.03) and an increase in HDL‐C by 37%. ²⁹ In the HDL‐Atherosclerosis Treatment Study (HATS), ³⁰ 160 patients with CHD and low HDL‐C levels (31 mg/dL), LDL‐C of 125 mg/dL, and TG of 213 mg/dL were randomized into four treatment groups: 1) simvastatin plus niacin; 2) antioxidants; 3) simvastatin plus niacin plus antioxidants; or 4) placebo, and were followed for 2 years. The niacin + simvastatin combination increased HDL‐C by 26% and decreased LDL‐C and TG by 42% and 36%, respectively (p<0.01). No lipid changes were noted with the antioxidants or placebo. The average coronary artery stenosis progressed by 3.9%, 1.8%, and 0.7% with placebo, antioxidants, and simvastatin + niacin + antioxidants, respectively (p<0.004 vs placebo) and regressed by 0.4% with simvastatin + niacin (p<0.001 vs placebo). In the Stockholm Ischemic Heart Disease Secondary Prevention Study, ³¹ 555 post‐MI patients with dyslipidemia (LDL‐C, 161 mg/dL; HDL‐C, 48 mg/dL) were randomized to a control and a treatment group and were followed for 5 years. The treatment group received a combination of clofibrate with niacin. In the active treatment group, TC and TG levels decreased by 13% and 19%, respectively. Compared with the control group, active treatment resulted in 26% reduction in total mortality and 36% reduction in ischemic heart disease mortality (p<0.01). Ischemic heart disease mortality was decreased by 60% in the patients with the higher TG levels.

Niacin/Statin Combinations

A niacin/statin combination is most effective in decreasing serum LDL‐C, TG, and lipoprotein(a) (Lp[a]) and increasing the HDL‐C levels. ³⁰ , ³² , ³³ , ³⁴ , ³⁵ , ³⁶ , ³⁷ , ³⁸ , ³⁹ , ⁴⁰ Niacin in its standard crystalline form has a low level of patient and physician acceptance due to unpleasant side effects, including cutaneous flushing, pruritus, and gastrointestinal intolerance. ⁴¹ , ⁴² The cutaneous flushing and itching is mediated through the release of prostaglandins and can be ameliorated with the concomitant administration of nonsteroidal anti‐inflammatory drugs (NSAIDs). However, newer, long‐acting, slow‐release preparations have been developed which are better tolerated and frequently do not require the coadministration of NSAIDs. A multicenter trial examined the effects of a long‐acting niacin preparation (niacin‐ER) in combination with lovastatin in patients with mixed dyslipidemia. The main findings of this study have been previously published. ³² In this study, 236 patients with type IIA or type IIB dyslipidemia were randomized to one of four dose‐escalating groups: 1) niacin‐ER 500–2000 mg/d; 2) lovastatin 20–40 mg/d; 3) niacin‐ER/lovastatin 500/20–1000/20 mg/d; and 4) niacin‐ER/lovastatin 500/20–2000/40 mg/d. Patients were followed for 28 weeks. The main baseline demographic characteristics of these patients are listed in Table I. Study patients were 55% male, 87% Caucasian, and 61% had two to three risk factors for coronary artery disease. The effects of treatment on lipid changes at the end of study compared with baseline are listed in Table II. The combination of niacin/lovastatin proved to be more effective than monotherapy with either component in favorably modifying the serum lipids. Niacin was significantly more effective, however, than lovastatin monotherapy in raising HDL‐C levels and decreasing Lp(a) levels. The addition of niacin to lovastatin had an additive effect in favorably modifying the serum lipids.

Table I.

Main Baseline Demographic Characteristics of Study Patients

Characteristic	Niacin‐ER 500–2000 mg (n=61)	Lovastatin 20–40 mg (n=61)	Niacin‐ER/Loastatin 500/20–1000/20 mg (n=57)	Niacin‐ER/Lovastatin 500/20–2000/40 mg (n=57)
Age (yr)	60±1	61±1	59±2	60±2
BMI (kg/m2)	29.2±0.7	29.0±0.8	29.9±0.5	28.6±0.7
Men	28 (46)	39 (64)	31 (54)	32 (56)
Women	33 (54)	22 (36)	26 (46)	25 (44)
LDL‐C (mg/dL)	189.7±4.1	185.6±4.7	192.1±6.0	190.9±4.5
HDL‐C (mg/dL)	46.8±1.3	43.5±1.4	44.8±1.5	45.4±1.6
TG (mg/dL)	195.6±12.1	181.5±9.4	224.4±17.1	212.6±14.9
Lp(a) (mg/dL)	41.0±5.1	42.3±5.4	32.4±5.2	34.3±4.8
Data are presented as mean ± SE or n (%). ER=extended release; BMI=body mass index; LDL‐C=low‐density lipoprotein cholesterol; HDL‐C=high‐density lipoprotein cholesterol; TG=triglycerides; Lp(a)=lipoprotein(a). Adapted with permission from Clin Cardiol. 2003;26:112–118. 32

Table II.

Mean Changes (%) From Baseline by Treatment at Week 28

	Daily Dose (mg)	LDL‐C	HDL‐C	TG	Lp(a)
Niacin‐ER	2000	−13.5	+23.5	−22.9	−24.5
Lovastatin	40	−32.2	+6.4	−20.0	−1.8
Niacin‐ER/Lovastatin	1000/20	−27.6	+21.4	−25.9	−15.7
Niacin‐ER/Lovastatin	2000/40	−41.9	+30.4	−42.9	−19.3
LDL‐C=low‐density lipoprotein cholesterol; HDL‐C=high‐density lipoprotein cholesterol; TG=triglycerides; Lp(a)=lipoprotein(a); ER=extended release. Adapted with permission from Clin Cardiol. 2003;26:112–118. 32

In this study, niacin‐ER either alone or in combination with lovastatin was fairly well tolerated by most patients. Nineteen percent of patients assigned to combination therapy withdrew because of adverse effects, compared with 20% and 10% of those assigned to niacin‐ER and lovastatin monotherapy, respectively. The most common adverse effect was cutaneous flushing. Similar results have been published by other investigators using the same drug combinations in patients with dyslip‐idemia. ³² , ³³ , ³⁴ The results of these studies are summarized in Table III. Similar results with niacin‐ER, either alone or in combination with a statin, have been reported by other investigators in 269, 66, and 814 patients with dyslipidemia, respectively. ³⁵ , ³⁶ , ³⁷

Table III.

Lipid‐Modifying Effects of Niacin‐ER/Statin Combinations Compared With Monotherapy

	Drug Regimen			Mean Changes (%)
Study/Group	n	Dose (mg)	LDL‐C	HDL‐C	TG	Lp(a)
Hunninghake et al. 32
Niacin‐ER	61	2000	−13.5	+23.5	−22.9	−24.5
Lovastatin	61	40	−32.2	+6.4	−20.0	−1.8
Niacin‐ER/lovastatin	57	1000/20	−27.6	+21.4	−25.9	−15.7
Niacin‐ER/lovastatin	57	2000/40	−413	+30.4	−42.9	−19.3
Insull et al.33
Niacin‐ER	31	2000	−33.7	+30.0	−28.6	−13.0
Lovastatin	33	40	−24.4	+9.5	−10.5	+0.7
Niacin‐ER/lovastatin	34	1000/20	−34.7	+20.7	−23.4	−4.6
Niacin‐ER/lovastatin	32	2000/40	−45.6	+29.1	−33.5	−14.3
Bays et al. 34
Niacin‐ER/lovastatin	79	1000/40	−39.0	+17.0	−29.0	−19.0
Niacin‐ER/lovastatin	78	2000/40	−42.0	+32.0	−49.0	−21.0
Atorvastatin	82	40	−49.0	+6.0	−31.0	0
Simvastin	76	40	−39.0	+7.0	−19.0	−2.0
LDL‐C=low‐density lipoprotein cholesterol; HDL‐C=high‐density lipoprotein cholesterol; TG=triglycerides; Lp(a)=lipoprotein(a); ER=extended release. Adapted with permission from Clin Cardiol. 2003;26:112–118, 32 Arch Intern Med. 2004; 164:1121–1127, 33 and Am J Cardiol. 2003;91:667–672. 34

PHARMACOKINETICS AND PHARMACODYNAMICS OF NIACIN‐ER/LOVASTATIN FORMULATION

The absorption and bioavailability of a niacin‐ER/lovastatin combination are bioequivalent to the individual drugs. ³⁸ , ³⁹ , ⁴⁰ Niacin‐ER has a dissolution time of 8–12 hours, which is between the immediate‐ and slow‐release formulations. ⁴³ It undergoes an extensive and saturable first pass metabolism by the liver. Lovastatin also undergoes extensive liver extraction, and <5% of the administered dose reaches the circulation; a peak plasma concentration is achieved 2–4 hours after its administration. ⁴⁴ Food intake increases the absorption by 22%–30% and by 21% for niacin‐ER and lovastatin, respectively. ⁴⁰ , ⁴³ Niacin raises HDL‐C and apolipoprotein A‐I levels by decreasing the hepatic catabolism of apolipoprotein A‐I, and reduces plasma TG and very‐low‐density lipoprotein (VLDL) by inhibiting VLDL‐TG synthesis through the inhibition of both fatty acid synthesis and fatty acid esterification. Decreased VLDL concentration results in decreased levels of LDL‐C, since VLDL is converted to intermediate‐density lipoprotein, which in turn is converted to LDL‐C. ⁴⁵ The hydrogel technology used as the formulation of niacin‐ER allows for slow release of the drug and reduces the incidence and severity of side effects. Lovastatin, on the other hand, is more effective in lowering the plasma levels of LDL‐C through the inhibition of 3‐hydroxy‐3‐methylglutaryl coenzyme A (HMG‐CoA) reductase, which is the rate‐limiting enzyme in converting HMG‐CoA to mevalonate and then to cholesterol.

Safety and Tolerability

Statins are safe and generally well tolerated drugs. In one study, myalgia without creatine kinase elevation was seen in 12.3%, 12.2%, 9.6%, and 0.24% in patients treated with placebo and lovastatin 20, 40, and 80 mg/d, respectively, in 8245 patients treated with lovastatin or placebo for 1–2 years. ⁴⁶ In another study, ³² the most commonly observed adverse events are listed in Table IV. Myalgia was noted in 4% of patients taking niacin/lovastatin 1000/20 mg/d and in 7% of those taking lovastatin 40 mg/d.

Table IV.

Frequency (%) of Adverse Events in Patients Treated With Niacin‐ER/Lovastatin Combination or Monotherapy

	Niacin‐ER/L ovastatin 1000/20 mg	Niacin‐ER/Lovastatin 2000/40 mg	Niacin 2000 mg	Lovastatin 40 mg
Withdrew from study	0	19	20	0
Cutaneous flushing	5.5	5.5	5	0
Myalgia	4	0	2	7
Skin rash, itching	1	1	7	2
Headache	4.5	4.5	9	3
Aspartate aminotransferase ≥3 × ULN	0	1	0	1
Hyperglycemia (>1.3 × ULN [5–10 mg/dL])	2	2	5	7
Deaths*	1 patient	0	0	1 patient
ER=extended release; ULN=upper limit of normal; *not drug related. Adapted with permission from Clin Cardiol 2003;26:112–118. 32

The cutaneous flushing and itching from niacin is prostaglandin mediated and can be successfully treated with the administration of NSAIDs that interfere with the action of cyclooxygenase, the rate‐limiting enzyme in prostaglandin synthesis. Many patients develop tachyphylaxis to the cutaneous flushing within a few days or weeks of drug administration and treatment is usually not necessary. ⁴²

DISCUSSION

The data from this study indicate that a niacin‐ER/lovastatin combination is effective in modifying plasma lipids and is safe and well tolerated. Each component contributed to lipid modification in a different manner. Lovastatin was more effective than niacin‐ER in lowering the LDL‐C levels, whereas niacin‐ER was more effective than lovastatin in lowering plasma TG and Lp(a) levels and raising HDL‐C levels. Among the lipid‐lowering drugs, the statins are the most effective in lowering LDL‐C, and their use has been associated with significant reduction in CV morbidity and mortality. ¹ , ² , ³ , ⁴ , ⁵ , ⁶ , ⁷ , ⁸ , ⁹ However, the effect of several of the statins in raising HDL‐C and decreasing TG and Lp(a) levels is modest. ³² , ³³ , ³⁴ , ³⁵ , ³⁶ , ³⁷ Since dyslipidemias often consist of more than one abnormality, effective treatment may require the use of several medications. Both high TG and low HDL‐C levels are frequently noted in diabetes, the metabolic syndrome, and hypertensive individuals, and are associated with increased CV morbidity and mortality. The BIP ²⁵ and the VA‐HIT ²⁶ studies, using different fibric acids, reported significant reductions in CV morbidity and mortality by raising HDL‐C and decreasing TG levels. Niacin use results in a lowering of TG and increasing HDL‐C levels ²⁹ , ³⁰ , ³² , ³³ , ³⁴ , ³⁵ , ³⁶ , ³⁷ , ³⁸ , ³⁹ , ⁴⁰ and in combination with either colestipol or lovastatin has been associated with regression of CHD. ²⁹ , ³⁰ It is possible that the additive effects of a niacin‐ER/lovastatin combination on plasma lipids will translate into additive benefits on clinical outcomes, as reported in the HATS study. ³⁰ The concept of therapy of dyslipidemias and especially low HDL‐C continues to evolve. ⁴⁷ One approach is the infusion of variants of apolipoprotein A‐I, such as A‐I Milano. In preliminary studies this has reduced plaque volume in hyperlipidemic patients with low HDL‐C. ⁴⁸ Recently, a high‐affinity nicotinic acid receptor in humans (hHM74a) coupled to G protein in adipocytes has been discovered. ⁴⁹ This receptor increased the sensitivity to nicotinic acid. It is hoped that future preparations of nicotinic acid that stimulate this receptor will be effective at lower doses to treat dyslipidemias with increased patient compliance. More effective management of low HDL‐C dyslipidemia in hypertensive and metabolic syndrome patients may improve outcomes over and above the benefits noted with BP lowering. Future studies of the treatment of low HDL‐C in prehypertensive patients could be of significance in determining whether this might actually prevent the development of hypertension.