- Email: [email protected]
Preclinical safety evaluation of cerivastatin, a novel HMG-CoA reductase inhibitor
Preclinical Safety Evaluation of Cerivastatin, a Novel HMG-CoA Reductase Inhibitor Eckhard von Keutz,
DVM, PhD,
and Gerhard Schlu¨ter,
MD
Cerivastatin is a new but structurally distinct 3-hydroxy3-methylglutaryl coenzyme A (HMG-CoA ) reductase inhibitor (“statin”). It effectively decreases low-density lipoprotein (LDL) cholesterol at 1% of the doses of other currently available statins. The toxicology of cerivastatin was evaluated in a comprehensive program of studies including: (1) single- and multiple-dose toxicity studies in rats, mice, minipigs, dogs, and monkeys; (2) reproductive toxicity studies in rats and rabbits; (3) in vitro and in vivo mutagenicity assays in rats and mice; and (4) carcinogenicity studies in rats and mice. In addition, studies were undertaken to investigate the effects of cerivastatin on lens opacity, testicular tissue, and hemorrhage in dogs. Oral administration of single and multiple doses of cerivastatin over periods ranging from 4 weeks to 24 months was generally well tolerated. Adverse effects were similar to those observed with other statins and primarily involved the liver and muscle tissue. At the high doses used in the toxicologic studies, cerivastatin caused elevations in serum transaminases and creatine phosphokinase levels as well as some degeneration of muscle fibers in rats, mice, dogs, and minipigs. In dogs, the species most sensitive to statins, cerivastatin caused
erosions and hemorrhages in the gastrointestinal tract, bleeding in the brain stem with fibroid degeneration of vessel walls in the choroid plexus, and lens opacity. Apart from minor morphologic changes in the testicular tissue of dogs—the only organ for which a comparably low margin of safety was observed— cerivastatin had no significant effects on the male or female reproductive system. Cerivastatin also caused no primary embryotoxic or teratogenic effects. With the exception of cerivastatin-induced effects on the eyes and testicles, administration of mevalonic acid reversed the toxicologic effects of cerivastatin, indicating that the toxic effects were related to its mode of action and not to any intrinsic toxicity of the molecule itself. There was no evidence that cerivastatin had any mutagenic effects and, in contrast to other statins, high doses of cerivastatin did not induce tumors in rats. The main metabolite of cerivastatin was well tolerated systemically in all animals, including dogs. Overall, cerivastatin has a similar toxicologic profile to other statins and is a well-tolerated HMG-CoA reductase inhibitor. Q1998 by Excerpta Medica, Inc. Am J Cardiol 1998;82:11J–17J
erivastatin is an enantiomerically pure pyridine derivative that is structurally distinct from other C 3-hydroxy-3-methylglutaryl coenzyme A (HMG-
biosynthesis pathway that are essential for fetal development,9 and there is evidence to suggest that some members of the class may have teratogenic potential.10,11 In addition, the inhibition of HMG-CoA reductase affects the formation of ubiquinone and dolichol, which are involved in electron transport and glycoprotein synthesis, respectively. At high doses, most HMG-CoA reductase inhibitors have been shown to promote the growth of tumors in rodents, particularly in the liver, which is the primary site of drug action.12 The efficacy of cerivastatin at ultra-low doses, coupled with its high hepatic selectivity, means there is very low systemic exposure to the drug, thus decreasing the potential for toxicity.13 The preclinical toxicology test program for cerivastatin was designed to identify: (1) target organs for toxicity; (2) the toxicologic parameters that need to be monitored clinically; (3) the populations that may be at increased risk; and (4) the mechanism of the toxic effects and the doses at which they occur. The program investigated whether the drug has carcinogenic, mutagenic, or teratogenic potential, and whether it might impair fertility or reproductive performance. In addition to the routine toxicology program required for drugs intended for chronic use, part of the program was tailored to the specific features of the HMG-CoA reductase inhibitor class of compounds. Tests for evidence of systemic toxicity, reproductive
CoA) reductase inhibitors (“statins”). (It is “enantiomerically pure” because there is a single active compound; it is either 100% levo or 100% dextro, and there is no racemic mix.) As a class, HMG-CoA reductase inhibitors have been shown to have a low risk of severe side effects after chronic exposure.1 Toxicologic studies have shown that the liver, kidney, muscle, the nonglandular stomach, and lymphatic tissue are potential target organs and tissues. The central nervous system, the eyes, the testes, and the thyroid gland may also be affected.2– 8 The similarity of the toxicologic profiles of current HMG-CoA reductase inhibitors suggests that there is a close relation between toxicity and the inhibition of HMG-CoA reductase. Inhibition of sterol synthesis would be expected to have a significant impact on a wide range of tissue and cell systems in the body. For example, inhibition of cholesterol synthesis could have an adverse effect on reproduction by decreasing the supply of circulating cholesterol and other products of the cholesterol From the Institute of Toxicology, PH-Product Development, Bayer AG, Wuppertal, Germany. Address for reprints: Eckhard von Keutz, DVM, PhD, Institute of Toxicology, PH-Product Development, Bayer AG, PO Box 101709, Aprather Road, D-42096, Wuppertal, Germany. ©1998 by Excerpta Medica, Inc. All rights reserved.
0002-9149/98/$19.00 PII S0002-9149(98)00424-X
11J
toxicity, and genotoxicity were extended to include the main metabolite of cerivastatin, M23, which is only found in humans and has an inhibitor activity similar to the parent drug. This article describes the extensive preclinical toxicology program designed to assess the safety of cerivastatin and the M23 metabolite and discusses some of the pathologic mechanisms that underlie side effects, such as elevated hepatic transaminases and effects on muscle tissue, that are occasionally seen in patients treated with statins.14,15
RESULTS
METHODS The species used in the toxicologic studies included rats, mice, rabbits, dogs, minipigs, and monkeys. The animals were given cerivastatin orally by gavage, in capsules, or through diet. In addition, cerivastatin was administered intravenously in rats and dogs to support absolute bioavailability studies in human volunteers. Acute and repeat dose toxicologic studies were undertaken in which cerivastatin, at a wide range of doses from 0.008 –300 mg/kg/day, was administered as either a single dose or as multiple doses over periods ranging from 4 weeks to 24 months. M23 was administered either orally or intravenously to different animal species over a period of 4 weeks in systemic toxicity tests. The effects of cerivastatin on male and female fertility in rats were investigated as were the embryotoxicity and teratogenicity of cerivastatin in rats and rabbits and its potential to cause perinatal/postnatal toxicity in rats. Genotoxicity tests included the Ames test, the hypoxanthine guanidine phosphoribosyl transferase (HGPRT) forward mutation test, the chromosomal aberration test, the unscheduled DNA synthesis test (UDS), and a spindle-inhibiting test in human lymphocytes. The in vivo micronucleus assay and the dominant lethal test in mice were also conducted. Both short- and long-term carcinogenicity studies were conducted; in a short-term carcinogenicity study in mice, cerivastatin was given after diethylnitrosamine had been administered intraperitoneally (once a week for 10 weeks) to initiate liver carcinogenesis. Traditional long-term carcinogenicity studies (up to 2 years) were performed in rats and mice. Lovastatin, an older statin with a known toxicologic profile, served as the reference drug in a number of the studies specifically designed to address the toxicologic features of cerivastatin. In some of the pivotal studies, cerivastatin was supplemented with mevalonic acid (the product of enzyme inhibition) to discriminate between an exaggeration of pharmacodynamic effects and intrinsic toxic properties of the drug. Finally, studies investigated specific toxicologic endpoints known for HMG-CoA reductase inhibitors, such as testicular toxicity, lens opacities, and bleeding in dogs. To assess the biologic significance of testicular lesions in dogs, a 12-month experiment investigated the reproducibility, incidence, and severity of lesions 12J THE AMERICAN JOURNAL OF CARDIOLOGYT
using high doses of cerivastatin (up to the lethal dose of 0.1 mg/kg per day). A study to investigate the reversibility of the lesion involved unilateral castration after 12 months of treatment followed by a recovery period of 3 months. Testicular function was determined by measurement of serum hormone levels (testosterone, luteinizing hormone), ejaculate volume, total sperm count, spermatozoa motility, and morphology during the treatment and recovery periods.
Effects on the liver: The effects of cerivastatin on the liver ranged from no change in primates, to subtle elevations of serum transaminases in dogs and minipigs, characteristic morphologic alterations in the rodent liver, and finally, to single-cell necrosis in dogs, the latter occurring only at lethal dosages (0.1 mg/kg per day). The changes in rodents consisted of cellular atypia characterized by cytoplasmic and nuclear pleomorphism in the periportal region of the liver lobule and foci of cellular alteration, which appeared either as basophilic or eosinophilic areas randomly distributed throughout the liver lobule. No similar changes were seen in dogs, minipigs, or monkeys, and since rodents respond differently to inhibition of the enzyme compared with these animals and humans, these results are not considered to be clinically relevant. Effects on the muscle: Elevations of serum creatine phosphokinase after treatment with cerivastatin were observed in rats, dogs, and minipigs and coincided with degeneration of muscle fibers. It has been suggested that statins cause intracellular ubiquinone deficiency, which interferes with normal cellular respiration in muscle, thus causing myopathy.16 –18 In studies of the effects of cerivastatin in dogs, measurements of ubiquinone in serum and various tissues, including muscles, revealed a marked decrease that was completely inhibited by concurrent administration of mevalonic acid. Effects on the gastrointestinal tract: In rats and mice, acanthosis (hyperplasia of the squamous epithelium), and hyperkeratosis of the nonglandular stomach (forestomach) was observed. There was no evidence of acanthosis in the glandular portion of the rodent stomach, nor in other parts of the rodent gastrointestinal tract containing squamous epithelium (esophagus or anorectal junction). Since such changes were not observed in the gastrointestinal squamous epithelium of any other species tested either, it appears that this lesion is species-specific and is limited to an anatomical structure not found in dogs, monkeys, or humans. The change appears to be directly linked to the biochemical mechanism of action of this class of compounds and requires the direct contact of the forestomach squamous epithelium with high concentrations of inhibitor for prolonged periods of time and is therefore not considered to be indicative of risk to humans. The only other effects to occur in the gastrointes-
VOL. 82 (4B)
AUGUST 27, 1998
FIGURE 1. Hemorrhage in the brain stem of dogs after administration of cerivastatin (0.1 mg/ kg/day p.o.).
FIGURE 2. Testicular lesion in dogs after administration of cerivastatin (0.1 mg/kg/day p.o.).
tinal tract were erosions and hemorrhages, which were confined to dogs receiving high doses of cerivastatin (0.1 mg/kg/day). Effects on the kidney: There was evidence of tubular degeneration to the kidneys of minipigs. However, this effect occurred only at doses that were in the lethal range for these animals (2– 4 mg/kg/day) and which were well in excess of the therapeutic doses. Effects on the central nervous system: Dogs exposed to the lethal dose of 0.1 mg/kg/day of cerivastatin developed hemorrhages in the brain stem with evidence of characteristic statin-induced fibroid degeneration of vessel walls in the choroid plexus (Figure 1).
Effects on the eye: Cerivastatin caused lens opacities in some of the dogs exposed to very high doses of the drug (0.1 mg/kg/day). Effects on the reproductive system: Extensive investigations showed cerivastatin had no significant impact on the male or female reproductive system in the rat. There were also no indications for a primary embryotoxic or teratogenic potential in rats and rabbits (Table I). Focal testicular atrophy was observed in dogs. This consisted of atrophy of the seminiferous epithelium with oligospermia, vacuolation of the seminiferous epithelium, and atypical or multinucleated giant sper-
A SYMPOSIUM: ADVANCES IN HMG-COA REDUCTASE INHIBITION
13J
matids (Figure 2). However, the Sertoli cells and the Leydig cells were viable and the spermatogonia were intact. The lesion was graded slight to very slight, did not progress with time, was not consistently reproducible, occurred at a low and variable incidence, and had a poor dose response. There was no correlation between the presence of lesions and serum testosterone and luteinizing hormone levels and, despite morphologic findings, there was no effect on testicular function (semen quality) up to lethal doses. One testicular specimen from a dog showed histologic evidence of multifocal degeneration of the seminiferous epithelium after chronic treatment with cerivastatin (0.1 mg/kg/day) but the absence of similar findings in the remaining testicle after 13 weeks of recovery suggests that the lesion resolves once treatment is stopped. There was no evidence of similar lesions in any of the other animal species. Mutagenicity potential of cerivastatin: No evidence of mutagenicity was seen in any of the assays performed with cerivastatin including 2 in vivo tests in mice (Table II). Carcinogenic potential of cerivastatin: The liver is the primary target organ for development of statininduced tumors. Liver adenomas and carcinomas were observed in mice treated with high dietary doses (6.75–11.5 mg/kg body weight/day) of cerivastatin but not at lower doses (1.32–2.31 mg/kg body weight/day; Table III). These findings confirmed the results of the short-term carcinogenicity study in mice in which oral cerivastatin (0.5 mg/kg/day) did not enhance the carcinogenic effect of diethylnitrosamine in the liver. In contrast to mice, rats showed no increase in any tumor type up to the maximum tolerated dose of 0.1– 0.2 mg/kg body weight/day—well in excess of the therapeutic doses of cerivastatin. Effects of supplementation with mevalonic acid: Administration of mevalonic acid compensates for the inhibition of HMG-CoA reductase. Toxic effects caused by the biochemical action of cerivastatin can therefore be distinguished from intrinsic toxic properties of the drug itself. Most of the toxic effects observed with cerivastatin were prevented or amelio-
TABLE I Results of Reproductive Toxicity Studies with Cerivastatin NOAEL (mg/kg/day; p.o.) Study Type
Animal
Male and female fertility (Segment I) Developmental toxicity (Segment II) Developmental toxicity (Segment II) Perinatal/postnatal toxicity (Segment III)
Maternal 0.03
0.1
Charles River rat Himalayan rabbit Sprague Dawley rat
0.16
0.72
0.15
0.75
0.1
0.1
NOAEL 5 no adverse effect levels; p.o. 5 by mouth. In the Segment I study, at a dose 0.3 mg/kg/day, a marginal reduction in fetal weight and delay in bone development was observed; the length of gestation was marginally prolonged, stillbirths were increased, and the survival rate up to day 4 postpartum was decreased. In the Segment III study, early postpartum mortality of 2 F1 litters at the highest dose was observed.
TABLE II Results of Mutagenicity Studies with Cerivastatin Test System
Remarks
In vitro Ames test; Salmonella typhimurium, Escherichia coli HGPRT forward mutation test; CHO UDS test; primary rat hepatocytes Chromosomal aberration; CHO Spindle inhibiting test; human lymphocytes In vivo Micronucleus test; mouse Dominant lethal test; mouse
Liver Adenomas/carcinomas Thyroid Adenomas/carcinomas Nonglandular stomach Papillomas Lung Adenomas Lymphatic system Malignant lymphomas Harderian gland Adenomas
Negative Negative
rated by co-administration of mevalonic acid (Table IV). The only exceptions were the induction of lens opacities and focal degenerations in the testes of dogs. Toxicity of M23: The results of toxicity tests on M23 showed no evidence of reproductive toxicity or mutagenicity, and systemic tolerability was significantly better than that of the parent compound. Whereas an oral dose of 0.1 mg/kg per day cerivastatin proved
Cerivastatin
Lovastatin
Pravastatin
Simvastatin
Fluvastatin
M
M/R
R
M/R
—
—
—
—
R
R
—
M
M
M
M/R
—
M
—
M
—
—
—
M
—
—
—
—
—
M
—
M 5 mouse; R 5 rat. *Increased incidence of tumors observed in bioassay with rats and/or mice.
14J THE AMERICAN JOURNAL OF CARDIOLOGYT
Negative Negative Negative Negative Negative
HGPRT 5 hypoxanthine guanidine phosphoribosyl transferase; UDS 5 unscheduled DNA synthesis; CHO 5 Chinese hamster ovary.
TABLE III Results of Carcinogenicity Test with Cerivastatin and Other Marketed Statins* Target Organ Tumor Type
Fetal
Wistar rat
VOL. 82 (4B)
AUGUST 27, 1998
TABLE IV Prevention/Amelioration of Cerivastatin-Induced Toxic Effects by Administration of Mevalonic Acid Lactone*
TABLE VI NOAEL and LEL Values for Toxicologic Targets in Multidose Studies with Cerivastatin
Species
Target Organ Typical Effect
Rat
General Lethality Liver Transaminases increases single cell necrosis Foci of cellular alteration Musculature Necrosis/hyperkeratosis Nonglandular stomach Acanthosis, hyperkeratosis Lymphatic system Edema, necrosis Eyes Cataracts Testicles Degeneration
Dog
√
√
√
√
√
NT
√
√
√
NT
NT
√
NT
X
NT
X
√ 5 reverse/amelioration by mevalonic acid; X 5 no reverse/amelioration by mevalonic acid; NT 5 not a target organ in this species. *Dose of mevalonic acid lactone not optimized.
lethal to dogs, M23 tested under identical conditions was well tolerated (Table V). Safety margins of cerivastatin: Safety factors were calculated on the basis of real exposure data (plasma concentration), expressed as maximum concentrations (Cmax) or area under the curve. Table VI shows the comparisons between Cmax values achieved in humans treated with a 0.3 mg dose of cerivastatin and the Cmax levels at the “no adverse effect” and the “lowest effective dose” levels for various toxicologic target organs. Based on disposition, systemic exposure, and metabolism in dogs (which is the most appropriate and sensitive animal species for the safety evaluation of cerivastatin), sufficient safety factors for most of the target organs (expressed by multiples of human Cmax) can be provided. These can be further increased by a factor of 2.6 if the higher and more relevant free plasma concentrations of cerivastatin in dogs compared with humans are taken into account. In the rat carcinogenicity study, at a dose of 0.1– 0.2 mg/kg body weight/day, an exposure (area under the curve) was achieved that exceeded human exposure by a factor of 2–3. As in dogs, the free concentration of cerivastatin in the rat plasma is higher than in humans, which adds an additional safety factor of 3.6.
Target Organ Species
Dose Multiples of (mg/kg/day; p.o.) Cmax to Humans* NOAEL†
Liver (transaminases) Dog 0.01 Monkey .0.1 Testicles Rat .5 Dog ,0.008 Monkey .0.1 Musculature Dog 0.025 Monkey .0.1 CNS, GI tract (bleeding) Rat .5 Dog 0.07 Monkey .0.1 Lymphatic system (edema, necrosis) Rat .5 Dog 0.025 Monkey .0.1 Eyes (cataracts) Rat .5 Dog 0.07 Monkey .0.1
LEL
NOAEL
LEL
0.025 —
1.7 .5
6.3 —
— 0.008 —
.571 ,1.4 .5
— 1.4 —
0.07 —
6.3 .5
11 —
— 0.1 —
.571 11 .5
— 23 —
— 0.07 —
.571 6.3 .5
— 11 —
— 0.1 —
.571 11 .5
— 23 —
NOAEL 5 no adverse effect level; LEL 5 lowest effect level; p.o. 5 per os (by mouth); Cmax 5 maximum concentration; CNS 5 central nervous system; GI 5 gastrointestinal. *Based on free plasma concentrations of cerivastatin; multiples to humans are higher by a factor of 2.6 for dogs and by a factor of 3.6 for rats (additional safety factor). † . indicates that the effect was not observed at the highest dose tested.
DISCUSSION AND CONCLUSIONS The administration of cerivastatin at critical dosages resulted in a spectrum of treatment-related changes in the liver, the muscle, the gastrointestinal tract, the central nervous system, the eye, the testicle, and the kidney (Table VII); this toxicologic profile is similar to that of other statins.2–7,19 –22 Given the importance of HMG-CoA reductase for normal biologic functions, prolonged administration of high levels of potent inhibitors would be expected to produce a wide spectrum of adverse effects in a variety of tissues. Inhibition of this enzyme results in decreased synthesis of mevalonate, which serves as a precursor for the biosynthesis of sterol. It is also an intermediate for the formation of ubiquinone and dolichols, which are involved in electron transport and glycoprotein synthesis, respectively.
TABLE V Comparison of the Toxicity of Cerivastatin and Metabolite M23 Study Type Acute toxicity; mouse i.v. 4-week toxicity; rat i.v. 4-week toxicity; dog p.o. Genotoxicity
Cerivastatin
Metabolite M23
LD50 approx. 230 mg/kg NOAEL .0.08 mg/kg/day Mortality at 0.1 mg/kg/day Negative
LD50 approx. 500 mg/kg NOAEL 5 1.5 mg/kg/day NOAEL .0.1 mg/kg/day Negative
i.v. 5 intravenous; p.o. 5 per os (by mouth); LD50 5 median lethal dose; approx 5 approximately; NOAEL 5 no adverse effect level.
A SYMPOSIUM: ADVANCES IN HMG-COA REDUCTASE INHIBITION
15J
TABLE VII Target Organs in Multidose Toxicologic Studies with HMG-CoA Reductase Inhibitors Target Organ* Liver Gallbladder Kidney Musculature Nonglandular stomach Lymphatic system Central nervous system Eyes (lens) Testicles Thyroid
Cerivastatin
Lovastatin
Pravastatin
Simvastatin
Fluvastatin
√ — √† √ √ √ √ √ √ —
√ √ √‡ √ √ √ √ √ √ —
√ — √ √ √ √ √ √ — √
√ √‡ √‡ √ √ √ √ — √ √
√ √ — √ √ √ — √ √ √
HMG-CoA 5 3-hydroxy-3-methylglutaryl coenzyme A. *Some additional tissues with higher incidence of tumors and/or with hemorrhages not included. † In minipigs only. ‡ In rabbits only.
There were, however, considerable differences in the reactions of various animal species to the drug. Whereas monkeys tolerated cerivastatin up to the highest doses tested without any effect, dogs were highly sensitive to cerivastatin and were the only animals in which there was evidence of hemorrhages in the brain stem and gastrointestinal tract. As the primary target organ for all statins, the liver is an important site for drug-related adverse effects. There were some similarities between the transient transaminase elevations seen in dogs receiving cerivastatin and the transaminase elevations observed in a small percentage of patients receiving statins.19 –21 These patients are also asymptomatic, and clinical evidence suggests that the elevations resolve spontaneously, despite continued drug administration. Since elevations in liver transaminases occur with all HMGCoA reductase inhibitors, this is obviously a class effect. There is evidence that the initial elevation of transaminases always involves alanine aminotransferase, an enzyme known to occur solely in the cytosol of hepatocytes. Since HMG-CoA reductase inhibitors interfere with synthesis and/or physiologic stability of membranes, the increased levels of alanine transaminase in the plasma may be due to leakage through the liver cell membrane. The fact that at the highest toxicologic doses, single-cell necrosis occurred along with an elevation of plasma glutamate dehydrogenase (an enzyme located solely in the mitochondria of liver cells) may then be explained by the decrease in the ubiquinone content of the liver cell affecting mitochondrial respiration. The morphologic changes observed in the liver of rodents treated with cerivastatin are similar to those seen with other HMG-CoA reductase inhibitors4,5 and are not considered to be important for human safety, since rodents respond differently to inhibition of the enzyme compared with other animal species and humans. Most statins promote the growth of liver tumors in rodents, particularly when administered at high doses. In contrast to other statins, cerivastatin only produced liver tumors in mice, a species known to be particularly sensitive to the formation of liver tumors. 16J THE AMERICAN JOURNAL OF CARDIOLOGYT
In addition, and importantly, the metabolism of cerivastatin in mice is totally different from humans and therefore, in the case of cerivastatin, the mice data have no relevance to the human risk assessment. The completely negative carcinogenicity study in rats, a species in which the liver was highly exposed to the drug, suggests that cerivastatin can be considered free of carcinogenic risk. Cerivastatin was also shown to have no mutagenic potential. The favorable toxicologic profile of cerivastatin is shared by its major M23 metabolite, which was extremely well tolerated in animals, including dogs, which have proved most sensitive to the parent compound. Elevations of serum creatine phosphokinase and degeneration of muscle fibers observed in animals treated with cerivastatin have also been observed in a small number of patients treated with inhibitors of HMG-CoA, especially in those concurrently receiving fibric acid derivatives, cyclosporine, nicotinic acid, erythromycin, or azole antifungals.23 In dogs, toxic doses of cerivastatin decreased the intracellular ubiquinone content of muscle tissue, thereby affecting the mitochondrial respiratory chain. It has been reported that decreases in serum ubiquinone concentrations in humans do not result in decreased levels in muscle tissue during short-term statin treatment.24 This does not mean that the reductions in muscle ubiquinone caused by statin therapy are not the cause of myopathy, even though muscle symptoms did not develop during the study. The observation that elevations in creatine phosphokinase and signs of myopathy occur more frequently in patients receiving concomitant treatment with fibric acid derivatives and cyclosporine may be explained by a pharmacokinetic interaction between the drugs, which are metabolized by the cytochrome P-450 (CYP3A4) pathway, leading to decreased clearance of the older statins and hence in increased plasma concentrations and tissue exposure.25 Apart from evidence of erosions and hemorrhages in the gastrointestinal tract of dogs treated with cerivastatin, the only other drug-related effects were hyperplasia and hyperkeratosis of the nonglandular
VOL. 82 (4B)
AUGUST 27, 1998
stomachs of rodents (an anatomic structure that has no counterpart in humans). The effects of cerivastatin on other important tissues such as the central nervous system, eye, and kidney were broadly similar to those reported for the statins as a whole, but unlike other statins, cerivastatin had no effect on thyroid tissue or the gallbladder (Table VII). Consistent with studies of other HMG-CoA reductase inhibitors, the testicular tissue of dogs was the only organ for which cerivastatin had a comparatively low margin of safety. Importantly, the minor morphologic alterations in dogs were not associated with impaired testicular function even at doses that were otherwise lethal to the animals. Preclinical findings in dogs treated with lovastatin and simvastatin triggered studies of hormones and semen in men.26,27 Sperm motility was the only parameter slightly, but significantly, decreased in men with type II hyperlipoproteinemia receiving lovastatin treatment; other indicators of testicular function were unaffected.26 In a study of the short-term treatment of familial hypercholesterolemic patients with simvastatin, there were no effects on sperm quality and serum levels of sex hormones.27 Testicular function (hormones, sperm quality) was unaffected after 6 months of treatment with cerivastatin, thus confirming the results of the semen analysis in dogs. The administration of mevalonic acid, the precursor of cholesterol, reversed or ameliorated most of the toxicologic effects of cerivastatin, confirming that the toxic effects were indeed related to its inhibitory effects on cholesterol biosynthesis and not to intrinsic toxicity. The fact that mevalonic acid did not reverse pathologic changes in the lens and in the testicle has also been observed with other HMG-CoA reductase inhibitors,4,5 and it may be that concentrations of mevalonic acid that can be achieved in these compartments are insufficient to compensate for inhibition of the enzyme. In conclusion, the extensive range of preclinical toxicology studies with cerivastatin show that the adverse effects observed at high doses are most likely linked to the prolonged and sustained inhibition of HMG-CoA reductase. At therapeutic doses cerivastatin is a well-tolerated HMG-CoA reductase inhibitor with a toxicology profile similar to other statins. 1. Shear CL, Franklin FA, Stinnett S, Hurley DP, Bradford RH, Chremos AN, Nash DT, Langendorfer A. Expanded clinical evaluation of lovastatin (EXCEL) study results. Circulation 1992;85:1293–1303. 2. Gerson RJ, Allen HL, Lankas GR, MacDonald JS, Alberts AW, Bokelman DL. The toxicity of a fluorinated-biphenyl HMG-CoA reductase inhibitor in beagle dogs. Fundam Appl Toxicol 1991;16:320 –329.
3. Gerson RJ, MacDonald JS, Alberts AW, Chen J, YudKovitz JB, Greenspan MD, Rubin LF, Bokelman DL. On the etiology of subcapsular lenticular opacities produced in dogs receiving HMG-CoA reductase inhibitors. Exp Eye Res 1990; 50:65–78. 4. MacDonald JS, Gerson RJ, Kornbrust DJ, Kloss MW, Prahalada S, Berry PH, Alberts AW, Bokelman DL. Preclinical evaluation of lovastatin. Am J Cardiol 1988;62(suppl):16J–27J. 5. Gerson RJ, MacDonald JS, Alberts AW, Kornbrust DJ, Majka JA, Stubbs RJ, Bokelman DL. Animal safety and toxicology of simvastatin and related hydroxymethylglutaryl-coenzyme A reductase inhibitors. Am J Med 1989;87(suppl):28S– 38S. 6. Keim RG. Toxicology and carcinogenicity of pravastatin. FDA Endocrinologic and Metabolic Drugs Advisory Committee on Pravastatin. Miller Reporting Co., Washington DC, 1990. 7. Hartman HA, Myers LA, Evans M, Robinson RL, Engstrom RG, Tse FLS. The safety evaluation of fluvastatin, an HMG-CoA reductase inhibitor, in beagle dogs and rhesus monkeys. Fundam Appl Toxicol 1996;29:48 – 62. 8. Walsh KM, Albassam MA, Clarke DE. Subchronic toxicity of atorvastatin, a hydroxymethylglutaryl-coenzyme a reductase inhibitor, in beagle dogs. Toxicol Pathol 1996;24:468 – 476. 9. Dobs AS, Sarma PS, Schteingart D. Long-term endocrine function in hypercholesterolemic patients treated with pravastatin, a new 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitor. Metabolism 1993;42:1146 –1152. 10. Minsker DH, MacDonald JS, Robertson RT, Bokelman DL. Mevalonate supplementation in pregnant rats suppresses the teratogenicity of mevinolinic acid, an inhibitor of 3-hydroxy-3-methylglutaryl coenzyme A reductase. Teratology 1983;28:449 – 456. 11. FDA. Summary basis for approval of NDA 20-261 (fluvastatin). Pharmacology Review of Amendment Conducted August, 1988. 1993. 12. Newman TB, Hulley SB. Carcinogenicity of lipid-lowering drugs. JAMA 1996;275:55– 60. 13. Angerbauer R, Bischoff H, Steinke W, Ritter W. BAY w 6228. Hypolipidemic HMG-CoA reductase inhibitor. Drugs Fut 1994;19:537–541. 14. Schalke BB, Schmidt B, Toyka K, Hartung HP. Pravastatin-associated inflammatory myopathy. N Engl J Med 1992;327:649 – 650. 15. Tobert JA. Efficacy and long-term adverse effect pattern of lovastatin. Am J Cardiol 1988;62:28 –34. 16. Willis RA, Folkers K, Tucker J, Ye CQ, Xia LJ, Tamagawa H. Lovastatin decreases coenzyme Q levels in rats. Proc Natl Acad Sci USA 1990;87:8928 – 8930. 17. Folkers K, Langsjoen P, Willis R, Richardson P, Xia LJ, Ye CQ, Tamagawa H. Lovastatin decreases coenzyme Q levels in humans. Proc Natl Acad Sci USA 1990;87:8931– 8934. 18. Grinlanda G, Oradei A, Manto A. Evidence of plasma CoQ10-lowering effect by HMG-CoA reductase inhibitors: a double-blind, placebo-controlled study. J Clin Pharmacy 1993;33:226 –229. 19. FDA. Summary basis for approval of NDA 19-643 (lovastatin). Pharmacol Rev 1987. 20. FDA. Summary basis for approval of NDA 19-898 (pravastatin). Pharmacol Rev 1989. 21. FDA. Summary basis for approval of NDA 19-766 (simvastatin). Pharmacol Rev 1990. 22. Physicians’ Desk Reference, Edition 50. Montvale, NJ: Medical Economics Company, 1996. 23. Pierce ILR, Wysowski DK, Gross TP. Myopathy and rhabdomyolysis associated with lovastatin-gemfibrozil combination therapy. JAMA 1990;264:71–75. 24. Laaksonen R, Jokelainen K, Sahi T, Tikkanen MJ, Himberg JJ. Decreases in serum ubiquinone concentrations do not result in reduced levels in muscle tissue during short-term simvastatin treatment in humans. Clin Pharmacol Ther 1995; 57:62– 66. 25. Smith PF, Eydelloth RS, Grossmann SJ, Stubbs RJ, Schwartz MS, Germershausen JI, Vyas KP, Kari PH, MacDonald JS. HMG-CoA reductase inhibitorinduced myopathy in the rat: cyclosporine A interaction and mechanism studies. J Pharmacol Exp Ther 1991;257:1225–1235. 26. Farnsworth WH, Hoeg JM, Maher M, Brittain EH, Sherins RJ, Brewer HB Jr. Testicular function in Type II hyperlipoproteinemic patients treated with lovastatin (mevinolin) or neomycin. J Clin Endocrinol Metab 1987;65:546 –550. 27. Purvis K, Tollefsrud A, Rui H, Haug E, Norseth J, Viksmoen L, Ose L, Lund H. Short-term effects of treatment with simvastatin on testicular function in patients with heterozygous familial hypercholesterolaemia. Eur J Clin Pharmacol 1992;42:61– 64.
A SYMPOSIUM: ADVANCES IN HMG-COA REDUCTASE INHIBITION
17J
DVM, PhD,
and Gerhard Schlu¨ter,
MD
Cerivastatin is a new but structurally distinct 3-hydroxy3-methylglutaryl coenzyme A (HMG-CoA ) reductase inhibitor (“statin”). It effectively decreases low-density lipoprotein (LDL) cholesterol at 1% of the doses of other currently available statins. The toxicology of cerivastatin was evaluated in a comprehensive program of studies including: (1) single- and multiple-dose toxicity studies in rats, mice, minipigs, dogs, and monkeys; (2) reproductive toxicity studies in rats and rabbits; (3) in vitro and in vivo mutagenicity assays in rats and mice; and (4) carcinogenicity studies in rats and mice. In addition, studies were undertaken to investigate the effects of cerivastatin on lens opacity, testicular tissue, and hemorrhage in dogs. Oral administration of single and multiple doses of cerivastatin over periods ranging from 4 weeks to 24 months was generally well tolerated. Adverse effects were similar to those observed with other statins and primarily involved the liver and muscle tissue. At the high doses used in the toxicologic studies, cerivastatin caused elevations in serum transaminases and creatine phosphokinase levels as well as some degeneration of muscle fibers in rats, mice, dogs, and minipigs. In dogs, the species most sensitive to statins, cerivastatin caused
erosions and hemorrhages in the gastrointestinal tract, bleeding in the brain stem with fibroid degeneration of vessel walls in the choroid plexus, and lens opacity. Apart from minor morphologic changes in the testicular tissue of dogs—the only organ for which a comparably low margin of safety was observed— cerivastatin had no significant effects on the male or female reproductive system. Cerivastatin also caused no primary embryotoxic or teratogenic effects. With the exception of cerivastatin-induced effects on the eyes and testicles, administration of mevalonic acid reversed the toxicologic effects of cerivastatin, indicating that the toxic effects were related to its mode of action and not to any intrinsic toxicity of the molecule itself. There was no evidence that cerivastatin had any mutagenic effects and, in contrast to other statins, high doses of cerivastatin did not induce tumors in rats. The main metabolite of cerivastatin was well tolerated systemically in all animals, including dogs. Overall, cerivastatin has a similar toxicologic profile to other statins and is a well-tolerated HMG-CoA reductase inhibitor. Q1998 by Excerpta Medica, Inc. Am J Cardiol 1998;82:11J–17J
erivastatin is an enantiomerically pure pyridine derivative that is structurally distinct from other C 3-hydroxy-3-methylglutaryl coenzyme A (HMG-
biosynthesis pathway that are essential for fetal development,9 and there is evidence to suggest that some members of the class may have teratogenic potential.10,11 In addition, the inhibition of HMG-CoA reductase affects the formation of ubiquinone and dolichol, which are involved in electron transport and glycoprotein synthesis, respectively. At high doses, most HMG-CoA reductase inhibitors have been shown to promote the growth of tumors in rodents, particularly in the liver, which is the primary site of drug action.12 The efficacy of cerivastatin at ultra-low doses, coupled with its high hepatic selectivity, means there is very low systemic exposure to the drug, thus decreasing the potential for toxicity.13 The preclinical toxicology test program for cerivastatin was designed to identify: (1) target organs for toxicity; (2) the toxicologic parameters that need to be monitored clinically; (3) the populations that may be at increased risk; and (4) the mechanism of the toxic effects and the doses at which they occur. The program investigated whether the drug has carcinogenic, mutagenic, or teratogenic potential, and whether it might impair fertility or reproductive performance. In addition to the routine toxicology program required for drugs intended for chronic use, part of the program was tailored to the specific features of the HMG-CoA reductase inhibitor class of compounds. Tests for evidence of systemic toxicity, reproductive
CoA) reductase inhibitors (“statins”). (It is “enantiomerically pure” because there is a single active compound; it is either 100% levo or 100% dextro, and there is no racemic mix.) As a class, HMG-CoA reductase inhibitors have been shown to have a low risk of severe side effects after chronic exposure.1 Toxicologic studies have shown that the liver, kidney, muscle, the nonglandular stomach, and lymphatic tissue are potential target organs and tissues. The central nervous system, the eyes, the testes, and the thyroid gland may also be affected.2– 8 The similarity of the toxicologic profiles of current HMG-CoA reductase inhibitors suggests that there is a close relation between toxicity and the inhibition of HMG-CoA reductase. Inhibition of sterol synthesis would be expected to have a significant impact on a wide range of tissue and cell systems in the body. For example, inhibition of cholesterol synthesis could have an adverse effect on reproduction by decreasing the supply of circulating cholesterol and other products of the cholesterol From the Institute of Toxicology, PH-Product Development, Bayer AG, Wuppertal, Germany. Address for reprints: Eckhard von Keutz, DVM, PhD, Institute of Toxicology, PH-Product Development, Bayer AG, PO Box 101709, Aprather Road, D-42096, Wuppertal, Germany. ©1998 by Excerpta Medica, Inc. All rights reserved.
0002-9149/98/$19.00 PII S0002-9149(98)00424-X
11J
toxicity, and genotoxicity were extended to include the main metabolite of cerivastatin, M23, which is only found in humans and has an inhibitor activity similar to the parent drug. This article describes the extensive preclinical toxicology program designed to assess the safety of cerivastatin and the M23 metabolite and discusses some of the pathologic mechanisms that underlie side effects, such as elevated hepatic transaminases and effects on muscle tissue, that are occasionally seen in patients treated with statins.14,15
RESULTS
METHODS The species used in the toxicologic studies included rats, mice, rabbits, dogs, minipigs, and monkeys. The animals were given cerivastatin orally by gavage, in capsules, or through diet. In addition, cerivastatin was administered intravenously in rats and dogs to support absolute bioavailability studies in human volunteers. Acute and repeat dose toxicologic studies were undertaken in which cerivastatin, at a wide range of doses from 0.008 –300 mg/kg/day, was administered as either a single dose or as multiple doses over periods ranging from 4 weeks to 24 months. M23 was administered either orally or intravenously to different animal species over a period of 4 weeks in systemic toxicity tests. The effects of cerivastatin on male and female fertility in rats were investigated as were the embryotoxicity and teratogenicity of cerivastatin in rats and rabbits and its potential to cause perinatal/postnatal toxicity in rats. Genotoxicity tests included the Ames test, the hypoxanthine guanidine phosphoribosyl transferase (HGPRT) forward mutation test, the chromosomal aberration test, the unscheduled DNA synthesis test (UDS), and a spindle-inhibiting test in human lymphocytes. The in vivo micronucleus assay and the dominant lethal test in mice were also conducted. Both short- and long-term carcinogenicity studies were conducted; in a short-term carcinogenicity study in mice, cerivastatin was given after diethylnitrosamine had been administered intraperitoneally (once a week for 10 weeks) to initiate liver carcinogenesis. Traditional long-term carcinogenicity studies (up to 2 years) were performed in rats and mice. Lovastatin, an older statin with a known toxicologic profile, served as the reference drug in a number of the studies specifically designed to address the toxicologic features of cerivastatin. In some of the pivotal studies, cerivastatin was supplemented with mevalonic acid (the product of enzyme inhibition) to discriminate between an exaggeration of pharmacodynamic effects and intrinsic toxic properties of the drug. Finally, studies investigated specific toxicologic endpoints known for HMG-CoA reductase inhibitors, such as testicular toxicity, lens opacities, and bleeding in dogs. To assess the biologic significance of testicular lesions in dogs, a 12-month experiment investigated the reproducibility, incidence, and severity of lesions 12J THE AMERICAN JOURNAL OF CARDIOLOGYT
using high doses of cerivastatin (up to the lethal dose of 0.1 mg/kg per day). A study to investigate the reversibility of the lesion involved unilateral castration after 12 months of treatment followed by a recovery period of 3 months. Testicular function was determined by measurement of serum hormone levels (testosterone, luteinizing hormone), ejaculate volume, total sperm count, spermatozoa motility, and morphology during the treatment and recovery periods.
Effects on the liver: The effects of cerivastatin on the liver ranged from no change in primates, to subtle elevations of serum transaminases in dogs and minipigs, characteristic morphologic alterations in the rodent liver, and finally, to single-cell necrosis in dogs, the latter occurring only at lethal dosages (0.1 mg/kg per day). The changes in rodents consisted of cellular atypia characterized by cytoplasmic and nuclear pleomorphism in the periportal region of the liver lobule and foci of cellular alteration, which appeared either as basophilic or eosinophilic areas randomly distributed throughout the liver lobule. No similar changes were seen in dogs, minipigs, or monkeys, and since rodents respond differently to inhibition of the enzyme compared with these animals and humans, these results are not considered to be clinically relevant. Effects on the muscle: Elevations of serum creatine phosphokinase after treatment with cerivastatin were observed in rats, dogs, and minipigs and coincided with degeneration of muscle fibers. It has been suggested that statins cause intracellular ubiquinone deficiency, which interferes with normal cellular respiration in muscle, thus causing myopathy.16 –18 In studies of the effects of cerivastatin in dogs, measurements of ubiquinone in serum and various tissues, including muscles, revealed a marked decrease that was completely inhibited by concurrent administration of mevalonic acid. Effects on the gastrointestinal tract: In rats and mice, acanthosis (hyperplasia of the squamous epithelium), and hyperkeratosis of the nonglandular stomach (forestomach) was observed. There was no evidence of acanthosis in the glandular portion of the rodent stomach, nor in other parts of the rodent gastrointestinal tract containing squamous epithelium (esophagus or anorectal junction). Since such changes were not observed in the gastrointestinal squamous epithelium of any other species tested either, it appears that this lesion is species-specific and is limited to an anatomical structure not found in dogs, monkeys, or humans. The change appears to be directly linked to the biochemical mechanism of action of this class of compounds and requires the direct contact of the forestomach squamous epithelium with high concentrations of inhibitor for prolonged periods of time and is therefore not considered to be indicative of risk to humans. The only other effects to occur in the gastrointes-
VOL. 82 (4B)
AUGUST 27, 1998
FIGURE 1. Hemorrhage in the brain stem of dogs after administration of cerivastatin (0.1 mg/ kg/day p.o.).
FIGURE 2. Testicular lesion in dogs after administration of cerivastatin (0.1 mg/kg/day p.o.).
tinal tract were erosions and hemorrhages, which were confined to dogs receiving high doses of cerivastatin (0.1 mg/kg/day). Effects on the kidney: There was evidence of tubular degeneration to the kidneys of minipigs. However, this effect occurred only at doses that were in the lethal range for these animals (2– 4 mg/kg/day) and which were well in excess of the therapeutic doses. Effects on the central nervous system: Dogs exposed to the lethal dose of 0.1 mg/kg/day of cerivastatin developed hemorrhages in the brain stem with evidence of characteristic statin-induced fibroid degeneration of vessel walls in the choroid plexus (Figure 1).
Effects on the eye: Cerivastatin caused lens opacities in some of the dogs exposed to very high doses of the drug (0.1 mg/kg/day). Effects on the reproductive system: Extensive investigations showed cerivastatin had no significant impact on the male or female reproductive system in the rat. There were also no indications for a primary embryotoxic or teratogenic potential in rats and rabbits (Table I). Focal testicular atrophy was observed in dogs. This consisted of atrophy of the seminiferous epithelium with oligospermia, vacuolation of the seminiferous epithelium, and atypical or multinucleated giant sper-
A SYMPOSIUM: ADVANCES IN HMG-COA REDUCTASE INHIBITION
13J
matids (Figure 2). However, the Sertoli cells and the Leydig cells were viable and the spermatogonia were intact. The lesion was graded slight to very slight, did not progress with time, was not consistently reproducible, occurred at a low and variable incidence, and had a poor dose response. There was no correlation between the presence of lesions and serum testosterone and luteinizing hormone levels and, despite morphologic findings, there was no effect on testicular function (semen quality) up to lethal doses. One testicular specimen from a dog showed histologic evidence of multifocal degeneration of the seminiferous epithelium after chronic treatment with cerivastatin (0.1 mg/kg/day) but the absence of similar findings in the remaining testicle after 13 weeks of recovery suggests that the lesion resolves once treatment is stopped. There was no evidence of similar lesions in any of the other animal species. Mutagenicity potential of cerivastatin: No evidence of mutagenicity was seen in any of the assays performed with cerivastatin including 2 in vivo tests in mice (Table II). Carcinogenic potential of cerivastatin: The liver is the primary target organ for development of statininduced tumors. Liver adenomas and carcinomas were observed in mice treated with high dietary doses (6.75–11.5 mg/kg body weight/day) of cerivastatin but not at lower doses (1.32–2.31 mg/kg body weight/day; Table III). These findings confirmed the results of the short-term carcinogenicity study in mice in which oral cerivastatin (0.5 mg/kg/day) did not enhance the carcinogenic effect of diethylnitrosamine in the liver. In contrast to mice, rats showed no increase in any tumor type up to the maximum tolerated dose of 0.1– 0.2 mg/kg body weight/day—well in excess of the therapeutic doses of cerivastatin. Effects of supplementation with mevalonic acid: Administration of mevalonic acid compensates for the inhibition of HMG-CoA reductase. Toxic effects caused by the biochemical action of cerivastatin can therefore be distinguished from intrinsic toxic properties of the drug itself. Most of the toxic effects observed with cerivastatin were prevented or amelio-
TABLE I Results of Reproductive Toxicity Studies with Cerivastatin NOAEL (mg/kg/day; p.o.) Study Type
Animal
Male and female fertility (Segment I) Developmental toxicity (Segment II) Developmental toxicity (Segment II) Perinatal/postnatal toxicity (Segment III)
Maternal 0.03
0.1
Charles River rat Himalayan rabbit Sprague Dawley rat
0.16
0.72
0.15
0.75
0.1
0.1
NOAEL 5 no adverse effect levels; p.o. 5 by mouth. In the Segment I study, at a dose 0.3 mg/kg/day, a marginal reduction in fetal weight and delay in bone development was observed; the length of gestation was marginally prolonged, stillbirths were increased, and the survival rate up to day 4 postpartum was decreased. In the Segment III study, early postpartum mortality of 2 F1 litters at the highest dose was observed.
TABLE II Results of Mutagenicity Studies with Cerivastatin Test System
Remarks
In vitro Ames test; Salmonella typhimurium, Escherichia coli HGPRT forward mutation test; CHO UDS test; primary rat hepatocytes Chromosomal aberration; CHO Spindle inhibiting test; human lymphocytes In vivo Micronucleus test; mouse Dominant lethal test; mouse
Liver Adenomas/carcinomas Thyroid Adenomas/carcinomas Nonglandular stomach Papillomas Lung Adenomas Lymphatic system Malignant lymphomas Harderian gland Adenomas
Negative Negative
rated by co-administration of mevalonic acid (Table IV). The only exceptions were the induction of lens opacities and focal degenerations in the testes of dogs. Toxicity of M23: The results of toxicity tests on M23 showed no evidence of reproductive toxicity or mutagenicity, and systemic tolerability was significantly better than that of the parent compound. Whereas an oral dose of 0.1 mg/kg per day cerivastatin proved
Cerivastatin
Lovastatin
Pravastatin
Simvastatin
Fluvastatin
M
M/R
R
M/R
—
—
—
—
R
R
—
M
M
M
M/R
—
M
—
M
—
—
—
M
—
—
—
—
—
M
—
M 5 mouse; R 5 rat. *Increased incidence of tumors observed in bioassay with rats and/or mice.
14J THE AMERICAN JOURNAL OF CARDIOLOGYT
Negative Negative Negative Negative Negative
HGPRT 5 hypoxanthine guanidine phosphoribosyl transferase; UDS 5 unscheduled DNA synthesis; CHO 5 Chinese hamster ovary.
TABLE III Results of Carcinogenicity Test with Cerivastatin and Other Marketed Statins* Target Organ Tumor Type
Fetal
Wistar rat
VOL. 82 (4B)
AUGUST 27, 1998
TABLE IV Prevention/Amelioration of Cerivastatin-Induced Toxic Effects by Administration of Mevalonic Acid Lactone*
TABLE VI NOAEL and LEL Values for Toxicologic Targets in Multidose Studies with Cerivastatin
Species
Target Organ Typical Effect
Rat
General Lethality Liver Transaminases increases single cell necrosis Foci of cellular alteration Musculature Necrosis/hyperkeratosis Nonglandular stomach Acanthosis, hyperkeratosis Lymphatic system Edema, necrosis Eyes Cataracts Testicles Degeneration
Dog
√
√
√
√
√
NT
√
√
√
NT
NT
√
NT
X
NT
X
√ 5 reverse/amelioration by mevalonic acid; X 5 no reverse/amelioration by mevalonic acid; NT 5 not a target organ in this species. *Dose of mevalonic acid lactone not optimized.
lethal to dogs, M23 tested under identical conditions was well tolerated (Table V). Safety margins of cerivastatin: Safety factors were calculated on the basis of real exposure data (plasma concentration), expressed as maximum concentrations (Cmax) or area under the curve. Table VI shows the comparisons between Cmax values achieved in humans treated with a 0.3 mg dose of cerivastatin and the Cmax levels at the “no adverse effect” and the “lowest effective dose” levels for various toxicologic target organs. Based on disposition, systemic exposure, and metabolism in dogs (which is the most appropriate and sensitive animal species for the safety evaluation of cerivastatin), sufficient safety factors for most of the target organs (expressed by multiples of human Cmax) can be provided. These can be further increased by a factor of 2.6 if the higher and more relevant free plasma concentrations of cerivastatin in dogs compared with humans are taken into account. In the rat carcinogenicity study, at a dose of 0.1– 0.2 mg/kg body weight/day, an exposure (area under the curve) was achieved that exceeded human exposure by a factor of 2–3. As in dogs, the free concentration of cerivastatin in the rat plasma is higher than in humans, which adds an additional safety factor of 3.6.
Target Organ Species
Dose Multiples of (mg/kg/day; p.o.) Cmax to Humans* NOAEL†
Liver (transaminases) Dog 0.01 Monkey .0.1 Testicles Rat .5 Dog ,0.008 Monkey .0.1 Musculature Dog 0.025 Monkey .0.1 CNS, GI tract (bleeding) Rat .5 Dog 0.07 Monkey .0.1 Lymphatic system (edema, necrosis) Rat .5 Dog 0.025 Monkey .0.1 Eyes (cataracts) Rat .5 Dog 0.07 Monkey .0.1
LEL
NOAEL
LEL
0.025 —
1.7 .5
6.3 —
— 0.008 —
.571 ,1.4 .5
— 1.4 —
0.07 —
6.3 .5
11 —
— 0.1 —
.571 11 .5
— 23 —
— 0.07 —
.571 6.3 .5
— 11 —
— 0.1 —
.571 11 .5
— 23 —
NOAEL 5 no adverse effect level; LEL 5 lowest effect level; p.o. 5 per os (by mouth); Cmax 5 maximum concentration; CNS 5 central nervous system; GI 5 gastrointestinal. *Based on free plasma concentrations of cerivastatin; multiples to humans are higher by a factor of 2.6 for dogs and by a factor of 3.6 for rats (additional safety factor). † . indicates that the effect was not observed at the highest dose tested.
DISCUSSION AND CONCLUSIONS The administration of cerivastatin at critical dosages resulted in a spectrum of treatment-related changes in the liver, the muscle, the gastrointestinal tract, the central nervous system, the eye, the testicle, and the kidney (Table VII); this toxicologic profile is similar to that of other statins.2–7,19 –22 Given the importance of HMG-CoA reductase for normal biologic functions, prolonged administration of high levels of potent inhibitors would be expected to produce a wide spectrum of adverse effects in a variety of tissues. Inhibition of this enzyme results in decreased synthesis of mevalonate, which serves as a precursor for the biosynthesis of sterol. It is also an intermediate for the formation of ubiquinone and dolichols, which are involved in electron transport and glycoprotein synthesis, respectively.
TABLE V Comparison of the Toxicity of Cerivastatin and Metabolite M23 Study Type Acute toxicity; mouse i.v. 4-week toxicity; rat i.v. 4-week toxicity; dog p.o. Genotoxicity
Cerivastatin
Metabolite M23
LD50 approx. 230 mg/kg NOAEL .0.08 mg/kg/day Mortality at 0.1 mg/kg/day Negative
LD50 approx. 500 mg/kg NOAEL 5 1.5 mg/kg/day NOAEL .0.1 mg/kg/day Negative
i.v. 5 intravenous; p.o. 5 per os (by mouth); LD50 5 median lethal dose; approx 5 approximately; NOAEL 5 no adverse effect level.
A SYMPOSIUM: ADVANCES IN HMG-COA REDUCTASE INHIBITION
15J
TABLE VII Target Organs in Multidose Toxicologic Studies with HMG-CoA Reductase Inhibitors Target Organ* Liver Gallbladder Kidney Musculature Nonglandular stomach Lymphatic system Central nervous system Eyes (lens) Testicles Thyroid
Cerivastatin
Lovastatin
Pravastatin
Simvastatin
Fluvastatin
√ — √† √ √ √ √ √ √ —
√ √ √‡ √ √ √ √ √ √ —
√ — √ √ √ √ √ √ — √
√ √‡ √‡ √ √ √ √ — √ √
√ √ — √ √ √ — √ √ √
HMG-CoA 5 3-hydroxy-3-methylglutaryl coenzyme A. *Some additional tissues with higher incidence of tumors and/or with hemorrhages not included. † In minipigs only. ‡ In rabbits only.
There were, however, considerable differences in the reactions of various animal species to the drug. Whereas monkeys tolerated cerivastatin up to the highest doses tested without any effect, dogs were highly sensitive to cerivastatin and were the only animals in which there was evidence of hemorrhages in the brain stem and gastrointestinal tract. As the primary target organ for all statins, the liver is an important site for drug-related adverse effects. There were some similarities between the transient transaminase elevations seen in dogs receiving cerivastatin and the transaminase elevations observed in a small percentage of patients receiving statins.19 –21 These patients are also asymptomatic, and clinical evidence suggests that the elevations resolve spontaneously, despite continued drug administration. Since elevations in liver transaminases occur with all HMGCoA reductase inhibitors, this is obviously a class effect. There is evidence that the initial elevation of transaminases always involves alanine aminotransferase, an enzyme known to occur solely in the cytosol of hepatocytes. Since HMG-CoA reductase inhibitors interfere with synthesis and/or physiologic stability of membranes, the increased levels of alanine transaminase in the plasma may be due to leakage through the liver cell membrane. The fact that at the highest toxicologic doses, single-cell necrosis occurred along with an elevation of plasma glutamate dehydrogenase (an enzyme located solely in the mitochondria of liver cells) may then be explained by the decrease in the ubiquinone content of the liver cell affecting mitochondrial respiration. The morphologic changes observed in the liver of rodents treated with cerivastatin are similar to those seen with other HMG-CoA reductase inhibitors4,5 and are not considered to be important for human safety, since rodents respond differently to inhibition of the enzyme compared with other animal species and humans. Most statins promote the growth of liver tumors in rodents, particularly when administered at high doses. In contrast to other statins, cerivastatin only produced liver tumors in mice, a species known to be particularly sensitive to the formation of liver tumors. 16J THE AMERICAN JOURNAL OF CARDIOLOGYT
In addition, and importantly, the metabolism of cerivastatin in mice is totally different from humans and therefore, in the case of cerivastatin, the mice data have no relevance to the human risk assessment. The completely negative carcinogenicity study in rats, a species in which the liver was highly exposed to the drug, suggests that cerivastatin can be considered free of carcinogenic risk. Cerivastatin was also shown to have no mutagenic potential. The favorable toxicologic profile of cerivastatin is shared by its major M23 metabolite, which was extremely well tolerated in animals, including dogs, which have proved most sensitive to the parent compound. Elevations of serum creatine phosphokinase and degeneration of muscle fibers observed in animals treated with cerivastatin have also been observed in a small number of patients treated with inhibitors of HMG-CoA, especially in those concurrently receiving fibric acid derivatives, cyclosporine, nicotinic acid, erythromycin, or azole antifungals.23 In dogs, toxic doses of cerivastatin decreased the intracellular ubiquinone content of muscle tissue, thereby affecting the mitochondrial respiratory chain. It has been reported that decreases in serum ubiquinone concentrations in humans do not result in decreased levels in muscle tissue during short-term statin treatment.24 This does not mean that the reductions in muscle ubiquinone caused by statin therapy are not the cause of myopathy, even though muscle symptoms did not develop during the study. The observation that elevations in creatine phosphokinase and signs of myopathy occur more frequently in patients receiving concomitant treatment with fibric acid derivatives and cyclosporine may be explained by a pharmacokinetic interaction between the drugs, which are metabolized by the cytochrome P-450 (CYP3A4) pathway, leading to decreased clearance of the older statins and hence in increased plasma concentrations and tissue exposure.25 Apart from evidence of erosions and hemorrhages in the gastrointestinal tract of dogs treated with cerivastatin, the only other drug-related effects were hyperplasia and hyperkeratosis of the nonglandular
VOL. 82 (4B)
AUGUST 27, 1998
stomachs of rodents (an anatomic structure that has no counterpart in humans). The effects of cerivastatin on other important tissues such as the central nervous system, eye, and kidney were broadly similar to those reported for the statins as a whole, but unlike other statins, cerivastatin had no effect on thyroid tissue or the gallbladder (Table VII). Consistent with studies of other HMG-CoA reductase inhibitors, the testicular tissue of dogs was the only organ for which cerivastatin had a comparatively low margin of safety. Importantly, the minor morphologic alterations in dogs were not associated with impaired testicular function even at doses that were otherwise lethal to the animals. Preclinical findings in dogs treated with lovastatin and simvastatin triggered studies of hormones and semen in men.26,27 Sperm motility was the only parameter slightly, but significantly, decreased in men with type II hyperlipoproteinemia receiving lovastatin treatment; other indicators of testicular function were unaffected.26 In a study of the short-term treatment of familial hypercholesterolemic patients with simvastatin, there were no effects on sperm quality and serum levels of sex hormones.27 Testicular function (hormones, sperm quality) was unaffected after 6 months of treatment with cerivastatin, thus confirming the results of the semen analysis in dogs. The administration of mevalonic acid, the precursor of cholesterol, reversed or ameliorated most of the toxicologic effects of cerivastatin, confirming that the toxic effects were indeed related to its inhibitory effects on cholesterol biosynthesis and not to intrinsic toxicity. The fact that mevalonic acid did not reverse pathologic changes in the lens and in the testicle has also been observed with other HMG-CoA reductase inhibitors,4,5 and it may be that concentrations of mevalonic acid that can be achieved in these compartments are insufficient to compensate for inhibition of the enzyme. In conclusion, the extensive range of preclinical toxicology studies with cerivastatin show that the adverse effects observed at high doses are most likely linked to the prolonged and sustained inhibition of HMG-CoA reductase. At therapeutic doses cerivastatin is a well-tolerated HMG-CoA reductase inhibitor with a toxicology profile similar to other statins. 1. Shear CL, Franklin FA, Stinnett S, Hurley DP, Bradford RH, Chremos AN, Nash DT, Langendorfer A. Expanded clinical evaluation of lovastatin (EXCEL) study results. Circulation 1992;85:1293–1303. 2. Gerson RJ, Allen HL, Lankas GR, MacDonald JS, Alberts AW, Bokelman DL. The toxicity of a fluorinated-biphenyl HMG-CoA reductase inhibitor in beagle dogs. Fundam Appl Toxicol 1991;16:320 –329.
3. Gerson RJ, MacDonald JS, Alberts AW, Chen J, YudKovitz JB, Greenspan MD, Rubin LF, Bokelman DL. On the etiology of subcapsular lenticular opacities produced in dogs receiving HMG-CoA reductase inhibitors. Exp Eye Res 1990; 50:65–78. 4. MacDonald JS, Gerson RJ, Kornbrust DJ, Kloss MW, Prahalada S, Berry PH, Alberts AW, Bokelman DL. Preclinical evaluation of lovastatin. Am J Cardiol 1988;62(suppl):16J–27J. 5. Gerson RJ, MacDonald JS, Alberts AW, Kornbrust DJ, Majka JA, Stubbs RJ, Bokelman DL. Animal safety and toxicology of simvastatin and related hydroxymethylglutaryl-coenzyme A reductase inhibitors. Am J Med 1989;87(suppl):28S– 38S. 6. Keim RG. Toxicology and carcinogenicity of pravastatin. FDA Endocrinologic and Metabolic Drugs Advisory Committee on Pravastatin. Miller Reporting Co., Washington DC, 1990. 7. Hartman HA, Myers LA, Evans M, Robinson RL, Engstrom RG, Tse FLS. The safety evaluation of fluvastatin, an HMG-CoA reductase inhibitor, in beagle dogs and rhesus monkeys. Fundam Appl Toxicol 1996;29:48 – 62. 8. Walsh KM, Albassam MA, Clarke DE. Subchronic toxicity of atorvastatin, a hydroxymethylglutaryl-coenzyme a reductase inhibitor, in beagle dogs. Toxicol Pathol 1996;24:468 – 476. 9. Dobs AS, Sarma PS, Schteingart D. Long-term endocrine function in hypercholesterolemic patients treated with pravastatin, a new 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitor. Metabolism 1993;42:1146 –1152. 10. Minsker DH, MacDonald JS, Robertson RT, Bokelman DL. Mevalonate supplementation in pregnant rats suppresses the teratogenicity of mevinolinic acid, an inhibitor of 3-hydroxy-3-methylglutaryl coenzyme A reductase. Teratology 1983;28:449 – 456. 11. FDA. Summary basis for approval of NDA 20-261 (fluvastatin). Pharmacology Review of Amendment Conducted August, 1988. 1993. 12. Newman TB, Hulley SB. Carcinogenicity of lipid-lowering drugs. JAMA 1996;275:55– 60. 13. Angerbauer R, Bischoff H, Steinke W, Ritter W. BAY w 6228. Hypolipidemic HMG-CoA reductase inhibitor. Drugs Fut 1994;19:537–541. 14. Schalke BB, Schmidt B, Toyka K, Hartung HP. Pravastatin-associated inflammatory myopathy. N Engl J Med 1992;327:649 – 650. 15. Tobert JA. Efficacy and long-term adverse effect pattern of lovastatin. Am J Cardiol 1988;62:28 –34. 16. Willis RA, Folkers K, Tucker J, Ye CQ, Xia LJ, Tamagawa H. Lovastatin decreases coenzyme Q levels in rats. Proc Natl Acad Sci USA 1990;87:8928 – 8930. 17. Folkers K, Langsjoen P, Willis R, Richardson P, Xia LJ, Ye CQ, Tamagawa H. Lovastatin decreases coenzyme Q levels in humans. Proc Natl Acad Sci USA 1990;87:8931– 8934. 18. Grinlanda G, Oradei A, Manto A. Evidence of plasma CoQ10-lowering effect by HMG-CoA reductase inhibitors: a double-blind, placebo-controlled study. J Clin Pharmacy 1993;33:226 –229. 19. FDA. Summary basis for approval of NDA 19-643 (lovastatin). Pharmacol Rev 1987. 20. FDA. Summary basis for approval of NDA 19-898 (pravastatin). Pharmacol Rev 1989. 21. FDA. Summary basis for approval of NDA 19-766 (simvastatin). Pharmacol Rev 1990. 22. Physicians’ Desk Reference, Edition 50. Montvale, NJ: Medical Economics Company, 1996. 23. Pierce ILR, Wysowski DK, Gross TP. Myopathy and rhabdomyolysis associated with lovastatin-gemfibrozil combination therapy. JAMA 1990;264:71–75. 24. Laaksonen R, Jokelainen K, Sahi T, Tikkanen MJ, Himberg JJ. Decreases in serum ubiquinone concentrations do not result in reduced levels in muscle tissue during short-term simvastatin treatment in humans. Clin Pharmacol Ther 1995; 57:62– 66. 25. Smith PF, Eydelloth RS, Grossmann SJ, Stubbs RJ, Schwartz MS, Germershausen JI, Vyas KP, Kari PH, MacDonald JS. HMG-CoA reductase inhibitorinduced myopathy in the rat: cyclosporine A interaction and mechanism studies. J Pharmacol Exp Ther 1991;257:1225–1235. 26. Farnsworth WH, Hoeg JM, Maher M, Brittain EH, Sherins RJ, Brewer HB Jr. Testicular function in Type II hyperlipoproteinemic patients treated with lovastatin (mevinolin) or neomycin. J Clin Endocrinol Metab 1987;65:546 –550. 27. Purvis K, Tollefsrud A, Rui H, Haug E, Norseth J, Viksmoen L, Ose L, Lund H. Short-term effects of treatment with simvastatin on testicular function in patients with heterozygous familial hypercholesterolaemia. Eur J Clin Pharmacol 1992;42:61– 64.
A SYMPOSIUM: ADVANCES IN HMG-COA REDUCTASE INHIBITION
17J