Change in α1 HDL Concentration Predicts Progression in Coronary Artery Stenosis
Objective— We examined the effects of simvastatin-niacin and antioxidant vitamins on changes in high-density lipoprotein (HDL) subpopulations and alterations in coronary artery stenosis, as assessed by angiography.
Methods and Results— Lipids, lipoproteins, and HDL particles were measured on and off treatment in 123 subjects of the HDL-Atherosclerosis Treatment Study. Patients were assigned to 4 treatment groups, simvastatin-niacin, simvastatin-niacin-antioxidant vitamins, antioxidant vitamins, and placebo. Subjects were followed for 3 years on treatment and then for 2 months off treatment. Simvastatin-niacin significantly increased the 2 large apoA-I–containing HDL subpopulations, α1 and preα1, and significantly decreased the 2 smallest particles, preβ1 and α3, compared with values obtained from the same patients off treatment. Adding antioxidant vitamins to the lipid-modifying agents blunted these effects (not significant). A significant negative correlation (r=−0.235; P<0.01) between the changes in α1 HDL particle concentration and coronary artery stenosis was noted. Subjects in the third tertile (157% increase in α1) had no progression of stenosis in the 3-year follow-up period, whereas subjects in the first tertile (15% decrease in α1) had an average of 2.1% increase in stenosis.
Conclusions— Simvastatin-niacin therapy significantly increased the large apoA-I–containing α1 HDL particles. This increase was significantly associated with less progression of coronary stenosis even after adjusting for traditional risk factors.
Epidemiological studies have shown an association between high levels of low-density lipoprotein cholesterol (LDL-C) and triglyceride (TG) as well as low levels of high-density lipoprotein cholesterol (HDL-C) and premature coronary heart disease (CHD).1–5⇓⇓⇓⇓ Moreover, each 1% reduction in LDL-C level or 1% increase in HDL-C level results in an approximately 1% to 2% reduction in CHD risk.2,6⇓ Accumulating evidence suggests that HDL is a heterogeneous group of particles differing not only in size and composition but also in physiological functions.7–10⇓⇓⇓ Many laboratories have published data indicating that the large HDL subpopulation containing apolipoprotein (apo) A-I only (LpA-I) carries more antiatherogenic properties than the smaller LpA-I:A-II particles.11–14⇓⇓⇓ Recently we have documented that monotherapy with different statins not only normalizes lipids but also shifts the HDL subpopulation profile toward normal.12,15⇓ Although statins increase HDL-C by 4% to 6% and apoA-I by <1%, they increase the large, LpA-I α-1 by 14% to 52% and preα-1 by 19% to 89%, depending on type and dose of statin. Niacin has been reported to raise HDL-C up to 30% and to selectively raise LpA-I levels while slightly decreasing LpA-I:A-II levels.16,17⇓ The precise protective role of HDL-C is not completely understood. There may be value in determining which HDL subpopulation is responsible for the protection against CHD and to define which HDL subpopulations are affected by drugs used for treating lipid disorders.
We have developed a quantitative two-dimensional gel electrophoresis, immunoblot, image analysis method for separating HDL subpopulations in plasma.11,18⇓ Most apoA-I–containing particles have α mobility and have been classified as α1, α2, and α3 with sizes of 11.2, 9.51, and 7.12 nm, respectively. We have demonstrated that α1, along with the preβ and preα mobility particles, contains apoA-I without apoA-II; therefore, these particles are termed LpA-I HDL. ApoA-II is present only in the α2 and α3 HDL subpopulations; consequently, these 2 subpopulations are LpA-I:A-II HDL particles.19 We have reported a strong positive correlation between HDL-C and the α1 HDL subpopulation19 and reported that the concentrations of the large, cholesterol-rich, LpA-I α1 and preα1 HDL subpopulations are lower whereas the concentrations of the TG-rich LpA-I:A-II α3 are higher in patients with CHD compared with controls.11
In the present study, we investigated the effects of simvastatin plus niacin, antioxidant vitamins, and simvastatin-niacin plus antioxidants combination therapy on the apoA-I–containing HDL subpopulation profiles of patients with CHD. We hypothesized that treatment with simvastatin plus niacin would normalize not only the apoB-containing lipoproteins but also the HDL subpopulation profile of patients with CHD.
Subjects and Study Design
We received plasma samples from 123 patients in the HDL-Atherosclerosis Treatment Study (HATS),20 which enrolled 160 patients with clinical coronary disease (defined as previous myocardial infarction, coronary interventions, or confirmed angina) and with 3 or more stenoses of at least 30% of the luminal diameter or 1 stenosis of at least 50%. All patients had low levels of HDL-C (<35 mg/dL in men and ≤40 mg/dL in women), LDL-C levels of ≤145 mg/dL, and TG levels <400 mg/dL. Male patients (n=108) were younger than 63 years and female patients (n=15) were younger than 70 years of age. The trial was carried out in a double-blinded, placebo-controlled fashion with a two-by-two factorial design. The four regimens were as follows: (1) simvastatin plus niacin with antioxidant vitamin placebo (S+N); (2) simvastatin-niacin plus antioxidant vitamins (S+N+antiox); (3) antioxidant vitamins with placebos for simvastatin and niacin (antiox); and (4) all placebos (placebo). Patients randomly received one of these treatments for 36 months and were also seen at 38 months off study medication. An LDL-C target level of ≤140 mg/dL was set for all of the subjects regardless of allocation into the treatment groups. Patients receiving placebo were given 10 to 20 mg of simvastatin if their LDL-C level was 140 mg/dL or higher. The primary end points were change in coronary stenosis (defined by arteriography) and occurrence of a first cardiovascular event (death, myocardial infarction, stroke, or revascularization). We received plasma samples from patients taken at 2 time points, 24 months (on treatment) and 38 months (off treatment).
Treatments have previously been described in detail by Brown et al.20,21⇓ Patients received a mean dose of 13.5±6.3 mg/d simvastatin and 2287±866 mg/d niacin in the S+N group and 12.0±4.3 mg/d simvastatin and 2353±1042 mg/d niacin and antioxidant vitamins in doses of 760±40 IU vitamin E, 950±50 mg vitamin C, and 95±5 μg selenium per day in the S+N+antiox group. In the antioxidant group, subjects received 740±60 IU vitamin E, 925±75 mg vitamin C, and 93±7 μg selenium. To reach the LDL-C goal (<140 mg/dL), 3 subjects in the placebo group and 4 subjects in the antioxidant group had been receiving simvastatin treatment at the 24-month blood-sampling period in 7.9±0.9-mg/d and 11.8±4.9-mg/d mean doses, respectively. All subjects received counseling for a healthy lifestyle and raising HDL levels. Emphasis was also placed on weight loss, smoking cessation, and monounsaturated fat consumption.
Fasting plasma concentrations of TG and total, HDL, and LDL cholesterol were determined by Northwest Lipid Research Laboratories using standard techniques.21,22⇓ ApoA-I and apoA-II values were determined by immuno-turbidometric assays.23 Remnant-like particle cholesterol (RLP-C) measurements were carried out using immuno-separation technique (Japan ImmunoResearch Laboratories).24 Quantitative coronary angiography was performed at the Cardiology Department of the University of Washington, School of Medicine, Seattle, at baseline and at 36 months.20,21⇓
Two-dimensional nondenaturing agarose-polyacrylamide gel-electrophoresis and image analysis for determining the apoA-I–containing and apoA-II–containing HDL subpopulations were carried out on plasma previously stored at −80°C, as described.18,19⇓ For quality-control purposes, a plasma sample stored frozen at −80°C in single-use aliquots was included with each run. The CVs were <10% for the α mobility subpopulations and <15% for the rest of the particles.
Values of the measured parameters of each patient obtained on treatment (24 months) and off treatment (38 months) were compared by the Wilcoxon matched-pair signed-rank test. Pearson correlation analysis was used to estimate the correlation between changes in coronary stenosis and the changes in HDL subpopulation profile. Data were log-transformed to bring the individual distributions closer to normal and make the associations more linear. A multiple regression analysis was also performed when data were adjusted for other independent risk factors. Statmost and SYSTAT software packages were used for the statistical analyses.
In Tables 1 through 4, we present the biochemical parameters of patients off and on treatments and the changes of these parameters after receiving simvastatin plus niacin (S+N), simvastatin-niacin plus antioxidant (S+N+Antiox), antioxidant (Antiox), and placebo (placebo) treatments for 2 years. On S+N treatment (Table 1), there were significant decreases in TC, LDL-C, VLDL-C, RLP-C, plasma TG, and apoA-II concentrations (27%, 36%, 35%, 15%, 27%, and 7%, respectively). HDL-C and apoA-I levels were increased significantly by 20% and 7%, respectively. S+N treatment also caused significant changes in all but 2 apoA-I–containing HDL subspecies. The small preβ1 and α3 particles decreased significantly by 39% and 17%, respectively. The α1, α2, preα1, and preα2 particles significantly increased by 115%, 27%, 311%, and 77%, respectively. There was a significant decrease in the mean value of the preβ1/α1 ratio (58%).
Treatment with simvastatin-niacin plus antioxidants (Table 2) also altered significantly the concentrations of all the measured plasma parameters except apoA-II. The mean decreases in TC, LDL-C, VLDL-C, RLP-C, and TG were 23%, 29%, 38%, 17%, and 27%, respectively. In contrast, HDL-C and apoA-I increased significantly by 21% and 10%, respectively, on this treatment. Preβ1 and α3 decreased significantly (37% and 7%, respectively). The α1, α2, and all of the preα-mobility particles increased significantly by 90%, 23%, 173%, 83%, and 35%, respectively. The preβ1/α1 ratio decreased significantly by 62%.
The third group of patients was treated with antioxidant vitamins (Table 3). During this treatment, none of the measured plasma parameters changed significantly; however, there was a trend for antioxidants to increase concentrations of VLDL-C, RLP-C, and plasma TG. Preβ1 and preβ2 HDL decreased significantly by 22% and 18%, whereas concentrations of the other particles increased, but only the changes in preα2 (16%) reached significance. The ratio of preβ1/α1 tended to decrease.
In the placebo-treated group (Table 4), the mean value of TC and LDL-C decreased significantly by 7% and 9%, and HDL-C increased significantly by 8%. Preβ1 decreased significantly by 32%. The larger spherical particles, α1, α2, preα1, and preα2, increased significantly by 42%, 13%, 97%, and 46%, respectively. The increases in preβ2 and preα3 were not significant. The ratio of preβ1/α1 decreased significantly (41%).
A significant negative correlation was found between the mean change of the large apo-A-I–containing α1 HDL subpopulations and the mean change in coronary stenosis (r=−0.235; P<0.01), and a positive correlation was found between the mean change in preβ1/α1 ratio and the mean change of coronary stenosis (r=0.202; P=0.02) using data from all subjects in the regression analysis. Changes in the concentrations of α1 HDL particles and coronary stenosis were also compared after subjects were divided into tertiles based on the changes in the concentrations of α1 HDL (Table 5). Subjects in the first tertile, whose mean values of α1 HDL decreased by 15%, increased their mean coronary stenosis by 2.1%. Subjects in the second tertile with a 48% increase in α1 had a mean increase in coronary stenosis of 1.4%. Subjects in the third tertile had a mean increase in α1 HDL of 157% with no change in coronary stenosis.
In the present study, we wished to address 3 questions. How does simvastatin and niacin treatment modify the HDL subpopulation profile of patients with hypoalphalipoproteinemia? How does treatment with antioxidants alone or in combination with S+N modify the HDL subpopulation profile in the same subjects? Is there any correlation between changes in HDL subpopulation profiles and coronary artery stenosis?
We received samples collected after 2 months off treatment (38 months) and samples collected at 24 months on treatment of 123 participants of the HATS. We assumed that the plasma biochemical parameters collected 2 months off treatment were similar to those parameters at baseline (0 month). To test this assumption, we compared the lipid and apoA-I values obtained from samples collected at baseline and at 38 months (off treatment). HDL-C was 3.4% higher (P<0.05) at baseline than at 38 months, but no other parameters were significantly different at these 2 time points. We also compared lipid data obtained at 24 and 36 months (time point of arteriography) and found no significant differences in the values obtained at these 2 time points.
We determined the concentrations of apoA-I and apoA-II in all HDL subpopulations by two-dimensional nondenaturing gel electrophoresis, immunoblotting, and image analysis. On both lipid-modifying treatments (S+N and S+N+antiox), the apoA-I contents of the small particles, preβ1 and α3, decreased; in contrast, those of the larger particles increased. We attribute these changes at least partially to CETP activity, which is decreased on statin therapy.25 However, the increases in the mean values of α1 and preα1 (115% and 311%) are much larger than can be achieved by statin monotherapy.15 This finding is in agreement with recently published data on the HATS study indicating a 47% increase in the concentration of apoA-I in the large LpA-I (9.2 to 11.2 nm) HDL.26
Plasma apoA-II concentrations decreased significantly (P<0.01) only in the S+N-treated group. This was the only group with significant increase in the apoA-I:A-II ratio in plasma. This ratio was more complex in the 2 LpA-I:A-II HDL subpopulations, α2 and α3. In the S+N, S+N+antiox, and placebo-treated groups, the apoA-I:A-II ratio increased significantly (P<0.05) in the larger α-2 HDL subpopulation because of a significant increase in apoA-I and no change in apoA-II concentration. In the same 3 groups, both apoA-I and apoA-II decreased in the smaller α3 particles, resulting in no change in the apoA-I/A-II ratio.
There were some differences in the average changes of the measured parameters between the groups treated with S+N versus the group treated with S+N+antiox. The addition of antioxidants to S+N treatment blunted the response for treatment by 22% and 44% in case of α1 and preα1 particles, respectively, despite a similar increase in HDL-C concentration on S+N and S+N+antiox. This finding illustrates how lipid-modifying medications are able to alter the HDL subpopulation profile independently of changing HDL-C level.
We postulate that the size and chemical composition of HDL is more important than HDL-C level in determining risk for CHD.11,27⇓ We hypothesize that the large, cholesterol-rich LpA-I α1 and preα1 particles have the most antiatherogenic properties and are sensitive markers of an antiatherogenic lipoprotein profile. Recently we demonstrated that 4 statins (atorvastatin, simvastatin, pravastatin, and lovastatin), at common therapeutic doses, increased the levels of these HDL particles in patients with CHD.12,15⇓ We have also documented that there is a wide range of individual responses for statin treatments in terms of changing the concentrations of plasma lipids and HDL subpopulations. In this study, we also found large differences in individual responses to the treatments. The individual responses and the mean changes in the concentrations of α1 are presented in the Figure. The mean changes in α1 HDL were positive for all 4 groups. However, some patients responded to the lipid-lowering treatments with decreased concentrations of HDL-C and α1. This broad range of variation in response to lipid-lowering therapy in the measured lipid and HDL parameters is not fully understood. Earlier we found that subjects with low HDL-C or high LDL-C levels responded better to treatment with different statins than subjects with the opposite lipid profile.15 Probably different genetic, lifestyle, or other pathological factors influence the response to lipid-lowering treatments.
After correlating all of the apoA-I–containing HDL subpopulations with changes in coronary stenosis, we found a significant negative correlation (r=−0.235; P<0.01) between the changes in the concentrations of the large α1 HDL subpopulation and changes of coronary stenosis when data from all subjects, regardless of treatment, were compared. This correlation was still significant when the data were adjusted for other risk factors (TC, LDL-C, HDL-C, TG, apoB, age, sex, diabetes, and smoking status). It is worth noting that in the first tertile, with a 15% decrease in α1 and a 2.1% increase in stenosis, 30% of the subjects received all placebos and 50% received antioxidant treatment, whereas 20% received S+N alone or in combination with antioxidants. On the other hand, in the third tertile, with a 157% increase in α1 and no change in the mean level of stenosis, 15% of the patients received all placebos and 8% received antioxidant treatments, whereas 85% received S+N alone or in combination with antioxidants. No significant correlation between changes in other HDL subpopulations and changes in stenosis was noticed. The correlation between the on-treatment concentrations of α1 HDL and changes in coronary stenosis was also negative (r=−0.189) but not significant (P=0.06). We still do not have enough information to speculate whether the high concentration of α1 HDL particles alone or as part of a specific HDL subpopulation profile, marked with high level of α1, is responsible for the protective effect against arteriosclerosis. We have to mention that not only α1 but also all of the preα mobility particles’ concentrations are lower in patients with CHD and increase during statin monotherapy or combination therapy.12,15⇓ In particular, preα1 is sensitive to the above-mentioned therapies, because its concentration increased 3-fold more than α1 on statin and niacin therapy. Our finding that S+N treatment increased the largest HDL particles (α1 and preα1) the most is in line with the observation that HDL2 increased the most on this same treatment.21
We agree with the concept of Fielding that the discoidal preβ1 HDL particles pick up cholesterol at the periphery and mature into large spherical α-mobility particles as cholesterol is esterified by the action of lecithin cholesterol acyltransferase.8 We have preliminary data indicating a significant negative correlation between the preβ1 and α1 HDL subpopulations (unpublished data, 2002). Considering these data, we believe that the preβ1/α1 ratio is a measure of the efficiency of reverse cholesterol transport (RCT), and a significant decrease in this rate indicates enhanced RCT. It is assumed that increased RCT, accompanied by decreased cholesterol concentrations in LDL, VLDL, and remnant-like particles plus decreased plasma TG, represents an improved lipoprotein profile and a decreased risk for atherosclerosis.
Our data indicate that treatment with simvastatin plus niacin normalized not only the lipid but also the HDL subpopulation profile of patients with CHD. The most important changes are probably the significant increases in α1 and preα1 and the significant decrease in preβ1 HDL subpopulations. Moreover, the mean changes in the concentration of α1 and the preβ1/α1 ratio are significantly correlated with the mean change in coronary artery stenosis. These data are consistent with the concept that increasing α1 HDL decreases progression of coronary artery stenosis.
This study was supported by KOS Pharmaceuticals, Inc (Miami, Fla) and by the National Institutes of Health (grant HL-64738).
- Received January 21, 2003.
- Accepted February 19, 2003.
- ↵Gordon DJ, Probstfield JL, Garrison RJ, Neaton JD, Castelli WP, Knoke JD, Jacobs DR, Bangdiwala S, Tyroler HA. High density lipoprotein cholesterol and cardiovascular disease: four prospective American studies. Circulation. 1989; 79: 8–15.
- ↵Goldbourt U, Yaari S, Medalie JH. Isolated low HDL cholesterol as a risk factor for coronary heart disease mortality: a 21-year follow-up of 8000 men. Arterioscler Thromb Vasc Biol. 1997; 17: 107–113.
- ↵Mowri HO, Patsch W, Smith LC, Gotto AM, Patsch JR. Different reactivities of high-density lipoprotein2 subfractions with hepatic lipase. J Lipid Res. 1992; 33: 1269–1279.
- ↵Asztalos BF, Roheim PS, Milani RL, Lefevre M, McNamara JR, Horvath KV, Schaefer EJ. Distribution of ApoA-I-containing HDL subpopulations in patients with coronary heart disease. Arterioscler Thromb Vasc Biol. 2000; 20: 2670–2676.
- ↵Asztalos BF, Horvath KV, McNamara JR, Roheim PS, Rubinstein JJ, Schaefer EJ. Effects of atorvastatin on the HDL subpopulation profile of coronary heart disease patients. J Lipid Res. 2002; 43: 1701–1707.
- ↵Cheung MC, Brown BG, Wolf AC, Albers JJ. Altered particle size distribution of apolipoprotein A-I containing lipoproteins in subjects with coronary heart disease. J Lipid Res. 1991; 32: 383–394.
- ↵Fruchart JC, Ailhaud G. Apolipoprotein A-containing lipoprotein particles: physiological role, quantification, and clinical significance. Clin Chem. 1992; 38: 793–797.
- ↵Asztalos BF, Lefevre M, Foster TA, Tulley R, Widhauser M, Wong L, Roheim PS. Normolipidemic subjects with low HDL cholesterol levels have altered HDL subpopulations. Arterioscler Thromb Vasc Biol. 1997; 17: 1885–1893.
- ↵Brown GB, Zhao XQ, Chait A, Frohlich J, Cheung M, Heise N, Dowdy A, DeAngelis D, Fisher LD, Albers J. Lipid altering or antioxidant vitamins for patients with coronary disease and very low HDL cholesterol? The HDL-Atherosclerosis Treatment Study design. Can J Cardiol. 1998; 14 (suppl A): 6A–13A.
- ↵Cheung MC, Zhao XQ, Chait A, Albers JJ, Brown GB. Antioxidant supplements block the response of HDL to simvastatin-niacin therapy in patients with coronary artery disease and low HDL. Arterioscler Thromb Vasc Biol. 2001; 21: 1320–1326.
- ↵Contois JH, McNamara JR, Lammi-Keefe CJ, Wilson PWF, Massov T, Schaefer EJ. Reference intervals for plasma apolipoprotein A-I determined with a standardized commercial immunoturbidometric assay: results from the Framingham Offspring Study. Clin Chem. 1996; 42: 515–523.
- ↵Guerin M, Lassel TS, Le Goff W, Farnier M, Chapman MJ. Action of atorvastatin in combined hyperlipidemia: preferential reduction of cholesterol ester transfer from HDL to VLDL1 particles. Arterioscler Thromb Vasc Biol. 2000; 20: 189–197.
- ↵Brown GB, Cheung MC, Lee AC, Zhao XQ, Chait A. Antioxidant vitamins and lipid therapy. Arterioscler Thromb Vasc Biol. 2002; 22: 1535–1546.
- ↵Sweetnam PM, Bolton CH, Yarnel IW, Bainton D, Baker IA, Elwood PC, Miller NE. Associations of HDL2 and HDL3 cholesterol subfractions with the development of ischemic heart disease in British men. Circulation. 1994; 90: 769–774.