A ~hero.sc~/cru,s;.s,9 I ( I YY1) S29-S34 p 1991 Elsevier Scientific Publishers
ATHEKC)
S29 Ireland,
Ltd. All rights reserved
0021.YlSO/91/$03.50
0467-I
Lipoprotein particle analysis comparing simvastatin and fenofibrate J.M. Bard ‘, H.J. Parra I, G. Luc ‘, R. Camare 2, 0. Ziegler 3, C. Dachet 4, E. Bruckert ‘, P. Douste-Blazy 2, P. Drouin ‘, B. Jacotot 4, J.L. De Gennes s, U. Keller ’ and J.C. Fruchart ’ ’Serhu. Institut Pasteur, Lille (France), .zINSERM UlOl, H&ital Purpan, Toulouse (France), -’Sen,ice de Mgdecine G&&ale G, H&ital Jeanne d’Arc, Toul (France), ’ Unit6 de Recherches sur I’Ath&oscl&ose, INSERM lJ32, H&ital Henri Mondor, Cr&eil iFrunccl, ’Senx? d’Endocri~lolo~ie, Hspital Pit%Salp&tri&e, Paris (France) and ’Abteilung fiir Endokrinologie und Stoffwechsel, Kantonsspital, Base1 (Switzerland)
Summary
This study compares the effects of fenofibrate and simvastatin in primary hypercholesterolemia, with particular regard to lipoprotein particles, as defined by their apolipoprotein composition: LpAI, LpAII : AI, LpE: B, LpCIII : B. This was a double-blind study in which patients were randomized to 2 groups, one receiving simvastatin 20 mg once daily and the other receiving fenofibrate 200 mg b.i.d., if their total cholesterol and their LDL cholesterol remained above 7.60 mmol/l (300 mg/dl) and 4.95 mmol/l (195 mg/dl) after a 4-week placebo period. Simvastatin dosage was doubled at the end of 6 weeks of therapy if the LDL-cholesterol level remained above 3.55 mmol/l (140 mg/dl). Analyses were done after 6 and 10 weeks of therapy. Apolipoprotein AI was increased significantly only at week 10 with fenofibrate (+7.4%X Simvastatin had a more pronounced effect than fenofibrate on apolipoprotein B. There was a significant difference between drugs at weeks 6 and 10. No change was observed in the LpAII : AI level with simvastatin, whereas fenofibrate increased these particles quite significantly ( + 13.9 and + 22.3%). The drugs had opposite effects on LpAI (+ 2.5 and +5.6% with simvastatin; - 12.8 and - 15.1% with fenofibrate). LP E: B (- 33.0 and -40.8% with simvastatin; - 53.8 and - 52.2% with fenofibrate) and LpCIII : B (-23.8 and -31.8% with simvastatin; -35.1 and -43.5% with fenofibrate) were decreased by both drugs, but fenofibrate was significantly more effective in reducing these particles
Correspondence to: Dr. J.M. Bard, Serlia,
1 rue du Pr. Calmette,
F-59019
Lille CCdex, France,
Tel. 20.87.77.55
s30 than simvastatin the lipoproteins, probably related
Key words:
at week 6. This study suggests that both drugs led to different structural modifications which would not be revealed by total apolipoprotein analysis. These differences to the mechanisms of action of these drugs.
HMG
CoA reductase
inhibitors;
Fenofibrate;
Introduction Fenofibrate and lovastatin (formerly known as mevinolin) are potent medications for the treatment of primary hypercholesterolemia [ 1,2]. Simvastatin is an analog of lovastatin and shares with this compound the capability to inhibit 3-hydroxy3-methylglutaryl coenzyme A (HMG-CoA) reductase, the rate-limiting enzyme in cholesterol synthesis. In human subjects the lack of cholesterol available to liver cells apparently stimulates the production of hepatic LDL receptors, thus increasing the activity of the receptor-specific pathway for LDL catabolism [3]. The explanation of plasma lowering by fenofibrate is less obvious. Strong evidence now exists that fenofibrate acts on lipoprotein lipase activity and enhances peripheral catabolism of triglyceride-rich lipoproteins [4-61. Lipoproteins are classically defined by their physical properties such as density or electrophoretic mobility. However, the development of immunological methods has clearly established that density classes represent in fact a mixture of particles with the same density but different apolipoprotein (ape) compositions. Thus, lipoproteins may be distinguished on the basis of their apolipoprotein composition. In this new concept, lipoproteins are separated into simple lipoprotein particles, containing 1 apolipoproteins (LpB, lipoprotein particles, LpAl . . . ) and complex containing 2 or more apolipoproteins (LpB: E, LpB : CIII, LpB : CIII : E, LpAI : AI1 . . . > [7]. We have shown that all apolipoprotein AI-containing particles may not have the same role in reverse cholesterol transport [8,9]. Lipoproteins containing apolipoprotein B do not all behave in the same way and depend on the apolipoprotein associated with apo B [lO,ll]. Thus, a study of lipoproteins as defined by their density may not
Lipoprotein
particles;
of are
Apolipoproteins
be sufficient, because changes in particles with different protein composition and the same density will not necessarily be revealed [12,13]. This study examines the effects of simvastatin and fenofibrate in primary hypercholesterolemia, with particular regard to those lipoprotein particles defined by their apolipoprotein composition.
Patients
and methods
Study design This study included a total of 189 patients, aged 18-72 years, with primary hypercholesterolemia. Patients entered the placebo baseline period if their plasma and LDL cholesterol values were above 300 mg/dl and 195 mg/dl, respectively, after 4-6 weeks off all lipid-lowering drugs and following a standard lipid-lowering diet. Patients with plasma triglyceride values more than 350 mg/dl or suffering from type I, III, IV or V hyperlipidemia were not accepted in this study. Premenopausal women, unless surgically sterilized, were excluded, as were patients consuming more than 10 alcoholic drinks per week or suffering from mental problems. Patients with the following illnesses were also excluded from the study: impaired hepatic function, unstable angina and vasospastic (Prinzmetal) angina, obesity (defined as body weight more than 30% of that listed in standard height/ weight tables), myocardial infarction or coronary bypass surgery within the previous 4 months, diabetes mellitus or a fasting blood glucose of more than 140 mg/dl. Secondary hypercholesterolemia was excluded by determination of free thyroxine, TSH, urine protein and hepatic tests. Current therapy with any other investigational drug was avoided during the study. Concurrent use of barbiturates, anticonvulsants, anticoagulants (except antiplatelet drugs), theo-
s31 phylline, quinidine, cimetidine, corticosteroids or regularly used antiacids was not allowed during the study. This was a double-blind randomized study. After the 4-week placebo baseline period (week O), the patients were randomized into 2 groups, one receiving simvastatin 20 mg once a day before the evening meal and the other receiving 200 mg fenofibrate twice a day before the morning meal and the evening meal, if their total cholesterol and LDL-cholesterol values remained above 300 mg/dl and 1% mg/dl, respectively. Simvastatin dosage was doubled at the end of 6 weeks of active treatment if the LDL-cholesterol level remained above 140 mg/dl. Active treatment was always administered concomitantly with the placebo of the other treatment. EDTA plasma was drawn at week 0, after 6 weeks (week 6) and 10 weeks (week 10) of active treatment for apolipoprotein and lipoprotein particle analysis, which was performed in the central laboratory at Institut Pasteur, Lille, France.
Consistency of treatment response among centers was assessed by examining treatment-by-study center interaction for each of the efficacy variables at each visit. An analysis of variance was performed on the ranks of the percentage changes from baseline for a model that included study center, treatment and their interaction as model effects. The interaction was tested at alpha = 0.10. An ANOVA on the ranked values of each of the variables at baseline was performed. The factors in the model were treatment group and study center. The purpose of this analysis was to assess the comparability of the treatment groups at baseline with respect to the variables. All statistical tests were 2-tailed. Results Table 1 describes the results obtained on apolipoproteins. Both drugs tended to increase apolipoprotein A-I but statistical analysis revealed significance only for fenofibrate at week
Apolipoprotein quantification Apolipoproteins AI and B were analysed by immunonephelometric assays (Behring, F.R.G.). Lipoprotein particle quantification Lipoproteins containing apolipoproteins A-II and A-I (Lp A-II : A-I), apolipoproteins C-III and B (Lp C-III: B) or apolipoproteins E and B (Lp E: B), were quantified using a 2-site immunoenzymometric assay [14,15]. Lipoproteins containing apolipoprotein A-I, but free of apolipoprotein A-II (Lp A-II, were quantified using a differential elcctroimmunoassay marketed by Sebia, France. Statist&l analysi.v The analysis used the ‘all-patients-treated’ approach including protocol violators, dropouts and all other patients with any efficacy data both at baseline (week 0) and on treatment. Comparisons between treatment groups were made with respect to the percentage change from baseline. The comparisons were made for each visit using an analysis of variance on the ranks with treatment and study center as model effects. Within-group comparisons were made for each visit using the Wilcoxon signed-rank test.
iniro in,er
cr3_p
camparlson
qrcup
comparison
‘*D
ATHEKC)
S29 Ireland,
Ltd. All rights reserved
0021.YlSO/91/$03.50
0467-I
Lipoprotein particle analysis comparing simvastatin and fenofibrate J.M. Bard ‘, H.J. Parra I, G. Luc ‘, R. Camare 2, 0. Ziegler 3, C. Dachet 4, E. Bruckert ‘, P. Douste-Blazy 2, P. Drouin ‘, B. Jacotot 4, J.L. De Gennes s, U. Keller ’ and J.C. Fruchart ’ ’Serhu. Institut Pasteur, Lille (France), .zINSERM UlOl, H&ital Purpan, Toulouse (France), -’Sen,ice de Mgdecine G&&ale G, H&ital Jeanne d’Arc, Toul (France), ’ Unit6 de Recherches sur I’Ath&oscl&ose, INSERM lJ32, H&ital Henri Mondor, Cr&eil iFrunccl, ’Senx? d’Endocri~lolo~ie, Hspital Pit%Salp&tri&e, Paris (France) and ’Abteilung fiir Endokrinologie und Stoffwechsel, Kantonsspital, Base1 (Switzerland)
Summary
This study compares the effects of fenofibrate and simvastatin in primary hypercholesterolemia, with particular regard to lipoprotein particles, as defined by their apolipoprotein composition: LpAI, LpAII : AI, LpE: B, LpCIII : B. This was a double-blind study in which patients were randomized to 2 groups, one receiving simvastatin 20 mg once daily and the other receiving fenofibrate 200 mg b.i.d., if their total cholesterol and their LDL cholesterol remained above 7.60 mmol/l (300 mg/dl) and 4.95 mmol/l (195 mg/dl) after a 4-week placebo period. Simvastatin dosage was doubled at the end of 6 weeks of therapy if the LDL-cholesterol level remained above 3.55 mmol/l (140 mg/dl). Analyses were done after 6 and 10 weeks of therapy. Apolipoprotein AI was increased significantly only at week 10 with fenofibrate (+7.4%X Simvastatin had a more pronounced effect than fenofibrate on apolipoprotein B. There was a significant difference between drugs at weeks 6 and 10. No change was observed in the LpAII : AI level with simvastatin, whereas fenofibrate increased these particles quite significantly ( + 13.9 and + 22.3%). The drugs had opposite effects on LpAI (+ 2.5 and +5.6% with simvastatin; - 12.8 and - 15.1% with fenofibrate). LP E: B (- 33.0 and -40.8% with simvastatin; - 53.8 and - 52.2% with fenofibrate) and LpCIII : B (-23.8 and -31.8% with simvastatin; -35.1 and -43.5% with fenofibrate) were decreased by both drugs, but fenofibrate was significantly more effective in reducing these particles
Correspondence to: Dr. J.M. Bard, Serlia,
1 rue du Pr. Calmette,
F-59019
Lille CCdex, France,
Tel. 20.87.77.55
s30 than simvastatin the lipoproteins, probably related
Key words:
at week 6. This study suggests that both drugs led to different structural modifications which would not be revealed by total apolipoprotein analysis. These differences to the mechanisms of action of these drugs.
HMG
CoA reductase
inhibitors;
Fenofibrate;
Introduction Fenofibrate and lovastatin (formerly known as mevinolin) are potent medications for the treatment of primary hypercholesterolemia [ 1,2]. Simvastatin is an analog of lovastatin and shares with this compound the capability to inhibit 3-hydroxy3-methylglutaryl coenzyme A (HMG-CoA) reductase, the rate-limiting enzyme in cholesterol synthesis. In human subjects the lack of cholesterol available to liver cells apparently stimulates the production of hepatic LDL receptors, thus increasing the activity of the receptor-specific pathway for LDL catabolism [3]. The explanation of plasma lowering by fenofibrate is less obvious. Strong evidence now exists that fenofibrate acts on lipoprotein lipase activity and enhances peripheral catabolism of triglyceride-rich lipoproteins [4-61. Lipoproteins are classically defined by their physical properties such as density or electrophoretic mobility. However, the development of immunological methods has clearly established that density classes represent in fact a mixture of particles with the same density but different apolipoprotein (ape) compositions. Thus, lipoproteins may be distinguished on the basis of their apolipoprotein composition. In this new concept, lipoproteins are separated into simple lipoprotein particles, containing 1 apolipoproteins (LpB, lipoprotein particles, LpAl . . . ) and complex containing 2 or more apolipoproteins (LpB: E, LpB : CIII, LpB : CIII : E, LpAI : AI1 . . . > [7]. We have shown that all apolipoprotein AI-containing particles may not have the same role in reverse cholesterol transport [8,9]. Lipoproteins containing apolipoprotein B do not all behave in the same way and depend on the apolipoprotein associated with apo B [lO,ll]. Thus, a study of lipoproteins as defined by their density may not
Lipoprotein
particles;
of are
Apolipoproteins
be sufficient, because changes in particles with different protein composition and the same density will not necessarily be revealed [12,13]. This study examines the effects of simvastatin and fenofibrate in primary hypercholesterolemia, with particular regard to those lipoprotein particles defined by their apolipoprotein composition.
Patients
and methods
Study design This study included a total of 189 patients, aged 18-72 years, with primary hypercholesterolemia. Patients entered the placebo baseline period if their plasma and LDL cholesterol values were above 300 mg/dl and 195 mg/dl, respectively, after 4-6 weeks off all lipid-lowering drugs and following a standard lipid-lowering diet. Patients with plasma triglyceride values more than 350 mg/dl or suffering from type I, III, IV or V hyperlipidemia were not accepted in this study. Premenopausal women, unless surgically sterilized, were excluded, as were patients consuming more than 10 alcoholic drinks per week or suffering from mental problems. Patients with the following illnesses were also excluded from the study: impaired hepatic function, unstable angina and vasospastic (Prinzmetal) angina, obesity (defined as body weight more than 30% of that listed in standard height/ weight tables), myocardial infarction or coronary bypass surgery within the previous 4 months, diabetes mellitus or a fasting blood glucose of more than 140 mg/dl. Secondary hypercholesterolemia was excluded by determination of free thyroxine, TSH, urine protein and hepatic tests. Current therapy with any other investigational drug was avoided during the study. Concurrent use of barbiturates, anticonvulsants, anticoagulants (except antiplatelet drugs), theo-
s31 phylline, quinidine, cimetidine, corticosteroids or regularly used antiacids was not allowed during the study. This was a double-blind randomized study. After the 4-week placebo baseline period (week O), the patients were randomized into 2 groups, one receiving simvastatin 20 mg once a day before the evening meal and the other receiving 200 mg fenofibrate twice a day before the morning meal and the evening meal, if their total cholesterol and LDL-cholesterol values remained above 300 mg/dl and 1% mg/dl, respectively. Simvastatin dosage was doubled at the end of 6 weeks of active treatment if the LDL-cholesterol level remained above 140 mg/dl. Active treatment was always administered concomitantly with the placebo of the other treatment. EDTA plasma was drawn at week 0, after 6 weeks (week 6) and 10 weeks (week 10) of active treatment for apolipoprotein and lipoprotein particle analysis, which was performed in the central laboratory at Institut Pasteur, Lille, France.
Consistency of treatment response among centers was assessed by examining treatment-by-study center interaction for each of the efficacy variables at each visit. An analysis of variance was performed on the ranks of the percentage changes from baseline for a model that included study center, treatment and their interaction as model effects. The interaction was tested at alpha = 0.10. An ANOVA on the ranked values of each of the variables at baseline was performed. The factors in the model were treatment group and study center. The purpose of this analysis was to assess the comparability of the treatment groups at baseline with respect to the variables. All statistical tests were 2-tailed. Results Table 1 describes the results obtained on apolipoproteins. Both drugs tended to increase apolipoprotein A-I but statistical analysis revealed significance only for fenofibrate at week
Apolipoprotein quantification Apolipoproteins AI and B were analysed by immunonephelometric assays (Behring, F.R.G.). Lipoprotein particle quantification Lipoproteins containing apolipoproteins A-II and A-I (Lp A-II : A-I), apolipoproteins C-III and B (Lp C-III: B) or apolipoproteins E and B (Lp E: B), were quantified using a 2-site immunoenzymometric assay [14,15]. Lipoproteins containing apolipoprotein A-I, but free of apolipoprotein A-II (Lp A-II, were quantified using a differential elcctroimmunoassay marketed by Sebia, France. Statist&l analysi.v The analysis used the ‘all-patients-treated’ approach including protocol violators, dropouts and all other patients with any efficacy data both at baseline (week 0) and on treatment. Comparisons between treatment groups were made with respect to the percentage change from baseline. The comparisons were made for each visit using an analysis of variance on the ranks with treatment and study center as model effects. Within-group comparisons were made for each visit using the Wilcoxon signed-rank test.
iniro in,er
cr3_p
camparlson
qrcup
comparison
‘*D
