Pre- and Postnatal Toxicity of the HMG-CoA Reductase Inhibitor Atorvastatin in Rats

TOXICOLOGICAL SCIENCES ARTICLE NO.

41, 88–99 (1998)

TX972400

Pre- and Postnatal Toxicity of the HMG-CoA Reductase Inhibitor Atorvastatin in Rats J. W. Henck,*,1 W. R. Craft,* A. Black,† J. Colgin,‡ and J. A. Anderson* Departments of *Pathology and Experimental Toxicology, †Pharmacokinetics and Drug Metabolism, and ‡Biometrics, Parke-Davis Pharmaceutical Research Division, Warner-Lambert Company, Ann Arbor, Michigan 48105 Received December 2, 1996; accepted October 22, 1997

droxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase, which catalyzes the conversion of HMG-CoA to mevalonate and constitutes the rate-limiting step in the biosynthesis of cholesterol (Fig. 1). By inhibiting HMG-CoA reductase activity in the liver, atorvastatin increases the cellular demand for cholesterol, resulting in an increase in hepatic low-density lipoprotein (LDL) receptors and an increase in clearance of plasma LDL by the liver (Kovanen et al., 1981). Since most cholesterol is transported in LDL in humans, decreased plasma concentrations of LDL cholesterol would be anticipated in patients with hyperlipidemia. Several inhibitors of HMG-CoA reductase (vastatins) are currently being used clinically in the treatment of hyperlipidemia; atorvastatin has been shown to produce even greater reductions in LDL cholesterol in hyperlipidemic patients than those produced by vastatins currently in use (Nawrocki et al., 1995). Inhibition of HMG-CoA reductase results in reduced levels of cholesterol, a precursor in the biosynthesis of hormones important in the maintenance of pregnancy, parturition, lactation, and maternal behavior (Numan, 1988; Freeman, 1988; Hodgen and Itskovitz, 1988). In addition, cholesterol is required for normal development in mammals both in utero and during the postnatal period (Belknap and Dietschy, 1988; Yount and McNamara, 1991). HMG-CoA reductase inhibition also results in decreased concentrations of mevalonate and its many isoprenoid derivatives, including dolichol, isopentenyl adenine, ubiquinone, haem A, and many other growth-regulating proteins bound to farnesyl and geranylgeranyl residues (Soma et al., 1992; Brewer et al., 1993; Corsini et al., 1995) (Fig. 1). These isoprenoids are involved in membrane synthesis, DNA replication, cellular growth and metabolism, and protein glycosylation, processes crucial to normal development (Quesney-Huneeus et al., 1983; Surani et al., 1983; Farnsworth et al., 1987; Brewer et al., 1993). The majority of vastatins are developmentally toxic in rats. Lovastatin was developmentally toxic in rats when administered during organogenesis, producing a treatment-related increase in the incidence of the malformation gastroschisis (Minsker et al., 1983). Pre- and postnatal administration of

Pre- and Postnatal Toxicity of the HMG-CoA Reductase Inhibitor Atorvastatin in Rats. Henck, J. W., Craft, W. R., Black, A., Colgin, J., and Anderson, J. A. (1998). Toxicol. Sci. 41, 88–99. Atorvastatin is a potent inhibitor of the enzyme 3-hydroxy-3methylglutaryl-coenzyme A (HMG-CoA) reductase, which catalyzes the conversion of HMG-CoA to mevalonate and constitutes the rate-limiting step in the biosynthesis of cholesterol. Steroid hormones derived from cholesterol, as well as mevalonate and its isoprenoid derivatives, provide important contributions to the maternal animal during pregnancy and lactation, as well as to the growth and development of the offspring; these contributions may potentially be influenced by inhibition of HMG-CoA reductase. To investigate the effects of atorvastatin on various aspects of reproduction and development, female Sprague–Dawley rats received 0, 20, 100, or 225 mg/kg daily by gavage from gestation day 7 through lactation day 20. Maternal toxicity, characterized by morbidity/mortality (13%), reduced body weight gain and food consumption, and pathologic lesions in the nonglandular mucosa of the stomach, occurred at 225 mg/kg. Offspring survival at birth and during the neonatal period at 225 mg/kg was reduced relative to control by up to 45%, and 28% of litters had no viable offspring by 10 days postpartum. Additional effects on offspring included reduced body weight during the neonatal and maturation periods (100, 225 mg/kg), delayed appearance of pinnae detachment and incisor eruption (225 mg/kg), impaired rotorod performance (females only; 100, 225 mg/kg), reduced acoustic startle responding (males only; 20, 100, 225 mg/kg), and transient effects on shuttle avoidance (females only; 225 mg/kg). No treatment-related effects were observed on offspring reproduction. In a separate experiment, a single dose of 10 mg/kg atorvastatin administered to female Wistar rats on gestation day 19 or lactation day 13 provided evidence of placental transfer and excretion into the milk. Results of this study indicate that pre- and postnatal administration of atorvastatin to female rats produces developmental toxicity in their offspring via in utero and/or lactational exposure, and in the presence or absence of maternal toxicity. q 1998 Society of Toxicology.

Atorvastatin is a potent inhibitor of the enzyme 3-hy1

To whom correspondence should be addressed at Parke-Davis Pharmaceutical Research, 2800 Plymouth Road, Ann Arbor, MI 48105. Fax: (313) 998-5718. 1096-6080/98 $25.00 Copyright q 1998 by the Society of Toxicology. All rights of reproduction in any form reserved.

AID

FAAT 2400

/

6k24h$$301

88

02-02-98 11:08:16

ftoxa

PRE- AND POSTNATAL TOXICITY OF ATORVASTATIN

89

FIG. 1. The biosynthetic pathway which yields mevalonate, isoprenoids, and cholesterol. The enzyme HMG-CoA reductase catalyzes the rate-limiting step in this pathway; inhibition of this enzyme by agents such as vastatins results in reduced levels of mevalonate and subsequent products. Adapted from Goldstein and Brown (1990), Soma et al. (1992), and Corsini et al. (1995).

lovastatin to rats resulted in decreased postweaning survival and delayed development of reflexes and behavior (FDA, 1987). Simvastatin was not teratogenic, but resulted in reduced fetal body weight when given to female rats during organogenesis (Wise et al., 1990a), reduced fetal and offspring body weights when given prior to mating and throughout gestation (Wise et al., 1990b), and reduced offspring body weight and increased swim maze errors when given perinatally (Minsker et al., 1990). Fluvastatin resulted in reduced offspring body weight and decreased neonatal survival when administered to rats from late gestation throughout lactation (Hrab et al., 1994), and was developmentally toxic, as evidenced by retarded skeletal development and reduced fetal body weights, when administered prior to mating and throughout gestation and lactation (FDA, 1993). No adverse effects on reproduction, fertility, or survival of the conceptus occurred following oral atorvastatin administration to male or female rats prior to and during mating (Dostal et al., 1996). However, oral administration of ator-

AID

FAAT 2400

/

6k24h$$302

02-02-98 11:08:16

vastatin to pregnant rats and rabbits during the period of organogenesis resulted in developmental toxicity, characterized by increased postimplantation loss and decreased fetal body weight, but not teratogenicity, at maternally toxic doses (Dostal et al., 1994). The present study was conducted to further characterize findings of developmental toxicity by extending the treatment period from organogenesis through delivery and weaning of the offspring. This exposure period was designed to evaluate maternal effects during pregnancy and lactation (including effects on nursing and maternal care), as well as effects on delivered offspring, including survival, growth, development, behavior, and reproductive performance. MATERIALS AND METHODS Pre- and Postnatal Toxicity Animals. Sexually mature male and female Sprague–Dawley rats (Crl:CD BR VAF/Plus; Charles River Breeding Laboratories, Inc., Portage,

ftoxa

90

HENCK ET AL.

FIG. 2. Chemical structure of atorvastatin.

MI) were used to evaluate the pre- and postnatal toxicity of atorvastatin. On gestation day (GD) 0, females were 12 to 13 weeks of age and weighed 226 to 314 g; at the time mating was initiated, untreated males were 12 to 13 weeks of age and weighed 340 to 401 g. Following an acclimation period of at least 1 week, animals with no adverse clinical signs were assigned to the study. Each animal received unique identification and was housed individually (except during mating and lactation) in a stainless steel, hanging wire mesh cage. Near the time of parturition, a solid stainless steel plate and bedding were added to the home cage of each female. Food (Purina Certified Lab Rodent Chow No. 5002, Ralston Purina Co., St. Louis, MO) and water were available ad lib throughout the study. Environmental conditions were in accordance with the Guide for the Care and Use of Laboratory Animals (NIH Publication 86-23, 1985). Untreated males and females were paired in a 1:1 ratio until mating was confirmed by the presence of sperm in the vaginal smear (GD 0). The females were then randomly assigned to treatment groups of 30 animals each. Of these, 25 contributed litters which were evaluated through adulthood, while 5 served as satellite animals from which plasma was collected for pharmacokinetic determinations. Test material. Atorvastatin is described chemically as [R-(R*,R*)]-2(4-fluorophenyl)-b,d-dihydroxy-5-(1-methylethyl)-3-phenyl-4-[(phenylamino)carbonyl]-1H-pyrrole-1-heptanoic acid calcium salt (2:1) (Fig. 2). It was supplied as a white amorphous powder by the Clinical Pharmaceutical Operations Department of the Parke-Davis Pharmaceutical Research Division, Warner-Lambert Company (Ann Arbor, MI), and had a stated active moiety of 91.2%. Dosing aliquots were prepared by suspending bulk atorvastatin in aqueous 0.5% methylcellulose (MC) at specific concentrations based on the active moiety content. Samples from each concentration of dosing suspension were analyzed periodically for drug concentration during the treatment period; results were within 10% of the intended value. Treatment. F0 generation female rats received 0 (0.5% MC vehicle), 20, 100, or 225 mg/kg atorvastatin daily by oral gavage from GD 7 through Lactation Day (LD) 20 at a dose factor of 10 mL/kg body wt. Dose volumes were based on the most recent individual body weights, with the exception that dose volumes continued to be based on the GD 15 body weight until the LD 0 body weight was obtained. The treatment period was selected to cover maternal and developmental events occurring during embryonic and fetal development, parturition, and lactation. F1 generation offspring were exposed to atorvastatin in utero and postnatally during lactation. The oral route of administration was selected because it is the intended clinical route for atorvastatin. Doses for this study, with an anticipated dosing period of approximately 5 weeks, were selected based on results of a 2-week oral dose range-finding study and a 13-week oral toxicity study in nonpregnant female rats (unpublished data), as well as the teratology study in rats previously cited (Dostal et al., 1994). No mortality occurred at 250 mg/kg in the 2-week study, but five females were euthanized moribund at 225 mg/kg in the 13-week study, and one female died at 300 mg/kg in the

AID

FAAT 2400

/

6k24h$$302

02-02-98 11:08:16

teratology study. No-adverse-effect doses in these studies were 70, 20, and 100 mg/kg, respectively. The greatest emphasis was placed on the teratology study; increased postimplantation loss at 300 mg/kg indicated that this dose was too high to achieve a sufficient number of live litters in the present study. Therefore, the high dose of 225 mg/kg in the 13-week study was adopted. The high dose of 225 mg/kg was expected to cause some maternal toxicity without excessive mortality, while the low dose of 20 mg/kg was expected to represent a no-adverse-effect dose, as recommended in ICH guidelines (1994). F0 generation. F0 generation animals were monitored for clinical observations, body weight, and food consumption. Each dam was handled daily during gestation in an attempt to minimize maternal stress and any ensuing effects on offspring during the postnatal period. All F0 females were allowed to deliver and rear their offspring. Each F0 female was euthanized following weaning or death of her litter, and a gross necropsy examination was conducted. F1 generation. F1 generation offspring were evaluated for survival, clinical observations, growth, development, behavior, and reproduction. On postnatal day (PND) 4, each litter was culled to eight offspring (four/sex, whenever possible), and on PND 21, two/sex/litter were euthanized for gross necropsy examination. The PND of acquisition of the following developmental landmarks was noted in all surviving offspring: pinnae detachment, lower incisor eruption, eye opening, testes descent, and vaginal opening. Behavior tests conducted in postweaning and/or mature offspring (one/ sex/litter) included rotorod, acoustic startle, motor activity, and two-way active avoidance. Rotorod performance (Omni-Rotor Treadmill, Omnitech Electronics, Inc., Columbus, OH) was evaluated on PND 28. Each animal received three training trials and three test trials at a constant rotor speed of 10 rpm, with a maximum duration of 60 s/trial. Mean time on the rotorod was utilized for group mean comparison. Locomotor activity was evaluated in an automated activity monitor (Digiscan Animal Activity Monitor, Omnitech Electronics, Inc., Columbus, OH) on PND 42 during one 4-min test session, designed to evaluate initial response to the novel environment of the activity monitor, as well as to observe the rate of habituation to that environment over the four consecutive 1-min test periods. The 4-min length of the test session has been demonstrated in our laboratory to identify changes in the rate of response to a novel open field environment (Henck et al., 1995, 1996). The acoustic startle response, as well as prepulse inhibition and habituation to an acoustic startle stimulus, was evaluated on PND 43. Each animal was given 70 trials using the SR-LAB Startle Response System (San Diego Instruments, San Diego, CA); the trials alternated between a background noise of 70 dB (total of 30 trials), a noise level tone of 120 dB (total of 20 trials), and a prepulse tone of 90 dB followed in 100 ms by a noise level tone of 120 dB (total of 20 trials), separated by intertrial intervals ranging from 5 to 30 s. Peak response for the first 120dB trial, the rate of habituation over 20 120-dB trials, and percentage

ftoxa

91

PRE- AND POSTNATAL TOXICITY OF ATORVASTATIN response inhibition resulting from the prepulse tones were evaluated. On 4 consecutive days during PN week 9, animals were tested for acquisition of a learned behavior (Learning Phase) and, after a 3-day rest period, for memory retention (Recall Phase) using an active avoidance paradigm in a two-compartment shuttle box (Shuttle Scan, Omnitech Electronics, Columbus, OH). Animals were tested for 25 trials on each testing day. The conditioned stimulus was light and the unconditioned stimulus was footshock with a duration of 10 s. Footshock levels were 0.33 mA for males and 0.22 mA for females; different levels were required due to the sexually dimorphic response to footshock in rats (Beatty and Beatty, 1970; Denti and Epstein, 1972). Parameters which were evaluated included percentage avoidance response, percentage escape response, and mean shock duration (the amount of time, up to 10 s/trial, for which footshock was endured prior to an avoidance or an escape). Sexually mature offspring (one/sex/litter, excluding animals tested for behavior) were evaluated for reproductive potential by 1:1 pairings of nonlittermates from the same treatment group. Mated F1 females underwent cesarean section on GD 21; maternal and fetal parameters (including examination for external abnormalities) were evaluated. Following behavior testing or evaluation of reproductive potential, surviving F1 offspring were euthanized and subjected to gross necropsy evaluation. Pharmacokinetics On LD 8 of the pre- and postnatal study, venous blood was obtained under anesthesia from five F0 control females 4 h postdose and from three to five F0 females/atorvastatin treatment group at 0 (predosing), 2, 4, 8, 12, and 24 h postdose, for plasma drug concentration analysis. LD 13 was selected as a time during which the volume of milk produced by the rat is adequate for analytical procedures (Fiorotto et al., 1991). In a separate set of experiments, sexually mature female Wistar rats [Crl:(WI)BR VAF/Plus; Charles River Breeding Laboratories, Inc., Kingston, NY] were used to evaluate the placental transfer and milk excretion of radiolabeled atorvastatin. Although the strain of rat was different from that used to evaluate pre- and postnatal toxicity, the answers obtained in Wistar rats regarding the existence of placental transfer and milk excretion in the rat as a species were considered applicable to Sprague–Dawley rats. To evaluate placental transfer, six rats were given a single dose of 10 mg/kg of a [14C]atorvastatin suspension in 0.5% methylcellulose (average radioactive dose 62 mCi) by oral gavage on GD 19. Animals were anesthetized and heparinized blood samples were collected by cardiac puncture 6 h postdose; whole blood was centrifuged, and the resultant plasma was analyzed by liquid scintillation spectrometry for radioactivity content. The 6-h timepoint was selected based on time to maximum atorvastatin plasma concentrations in a bioavailability study in rats (unpublished data). Following blood collection, animals were euthanized, and the following tissues were obtained for analysis of radioactivity content by liquid scintillation spectrometry following combustion: maternal liver, placenta, whole fetuses (pooled samples from 1/2 of the fetuses/each litter), and fetal liver (pooled samples from the remaining 1/2 of the fetuses/each litter). To determine whether atorvastatin was excreted into rat milk, a second set of six female Wistar rats was given a single dose of 10 mg/kg of a [14C]atorvastatin suspension in 0.5% methylcellulose (average radioactive dose 48 mCi) by oral gavage on LD 13. Each female’s litter was removed approximately 3 h postdose to facilitate milk letdown. Approximately 5 min prior to milk collection at 6 h postdosing, each female was anesthetized and primed with approximately 1.5 IU of oxytocin. Approximately 1.5 mL of milk was collected over a maximum 30-min interval. Immediately following milking, heparinized blood samples were collected by cardiac puncture; whole blood was centrifuged, and the resultant plasma was analyzed for radioactivity content. Following blood collection, animals were euthanized, and their livers were harvested for analysis of radioactivity content, as indicated previously. An additional three female rats and their nursing offspring were anesthetized 6 h postdose on LD 13, and heparinized

AID

FAAT 2400

/

6k24h$$302

02-02-98 11:08:16

blood samples were collected by cardiac puncture for subsequent analysis of maternal and offspring plasma radioactivity content. Following blood collection, animals were euthanized, and maternal and offspring livers were harvested for analysis of radioactivity content. Statistical Analysis To control for the multiplicity of statistical comparisons (i.e., reduce the number of false positive conclusions), all sets of maternal, reproductive, developmental, and behavioral response measures (parameters) were divided into distinct classes based on the relationship of the parameters. For example, all maternal body weight measurements comprised one class, while all developmental landmark data comprised another. The level of significance for each comparison within the class was 0.05 divided by the square root of the number of parameters in the class (Tukey et al., 1985), to provide an approximate classwise significance level of 5%. The basic trend test employed was the sequential trend test, using the rank dose scale and rank-transformed data (Park, 1985; Tukey et al., 1985). This test is equivalent to sequential application of a Kruskal–Wallis oneway analysis of variance by ranks (Kruskal and Wallis, 1952), with the treatment effects being evaluated by dose-trend tests which have contrast coefficients for equally spaced (ranked) treatment groups. The sequential linear trend test is designed to detect monotone changes in dose response: it does not detect trend reversal or curvature. If the true dose response was not monotonic, the trend test was considered not sensitive enough to detect the treatment effect. A trend reversal test was then conducted, consisting of a quadratic trend test performed at the two-tailed 1% classwise significance level. If the trend reversal test was significant and the high-dose trend test was not significant, the treatment groups were compared to the control by Dunnett’s test (Dunnett, 1955, 1964), using rank-transformed data, and performed at the 5% classwise significance level. The monotonic dose–response relationship tested by the trend test was not considered realistic for the activity monitor parameters because they have been demonstrated with several drugs to exhibit a U-shaped dose– response curve (Iversen and Iversen, 1981). Hence, Dunnett’s test on ranktransformed data was performed in place of the trend test as the main analytical method. In addition to trend analysis, acoustic startle data were subjected to profile analysis (Johnson and Wichern, 1982) to evaluate the acoustic startle response by treatment group interaction (parallelism test) and, secondarily, to address the question of equal group effects; the raw data were used for this analysis. The methods of Greenhouse and Geisser (1959) were incorporated in the profile analysis to produce conservative tests for the parallelism hypothesis.

RESULTS

F0 Generation Maternal toxicity, characterized by mortality and reduced body weight and food consumption, occurred at 225 mg/kg (Table 1). Two animals in this group died, one animal was euthanized in moribund condition, and one animal was euthanized due to inability to deliver. Ten dams at 225 mg/kg were euthanized by LD 10 because they had no viable offspring remaining. The most common treatment-related gross pathologic finding in dams at 225 mg/kg found dead or euthanized prior to scheduled termination was abnormal surface of the nonglandular mucosa of the stomach. Body weight gain at 225 mg/kg was significantly (p õ 0.0189) less than control by 14% during the gestation treatment period, but was significantly (p õ 0.0189) greater than control

ftoxa

92

HENCK ET AL.

TABLE 1 Maternal Parameters in Dams Treated with Atorvastatin from GD 7 through LD 20 Dose (mg/kg) Parameter

0

20

100

225

No. sperm-positive females No. died/euthanized prior to scheduled termination Body weight change (g)c GD 7–21 LD 0–21 Food consumption (g/animal)c GD 7–21 LD 0–21 Gestation duration (days)c Litter size (liveborn)c Postimplantation loss (%)c No. euthanized because no viable offspring remained

30

30

30

30

2a

0

1b

4

(23) 128 { 4.4 (24) 17.3 { 3.09

(22) 130 { 4.7 (22) 18.5 { 4.05

(23) 135 { 3.5 (23) 30.3 { 3.84

(20) 100 { 4.9* (8) 36.4 { 9.70*

(23) 352 (22) 1125 (24) 21.9 (24) 15.0 (24) 9.0

{ { { { {

6.3 14.9 0.08 0.44 1.39

(22) 360 (22) 1088 (22) 22.2 (22) 14.7 (22) 10.6

0

{ { { { {

8.3 34.3 0.13 0.81 1.86

0

(23) 362 (23) 1067 (24) 22.0 (24) 15.0 (24) 10.2 0

{ { { { {

5.8 18.7 0.07 0.62 2.33

(20) (7) (18) (18) (18)

325 866 21.9 14.1 11.3

{ { { { {

6.6* 79.5* 0.08 0.67 2.41

10

Note. GD, gestation day; LD, lactation day. a Deaths attributed to gavage error. b Euthanized due to prolapsed uterus. c (N) Mean { standard error; treatment groups compared statistically to vehicle control. * p õ 0.0189 (0.05 divided by the square root of seven individual parameters in the maternal body weight change class), different from vehicle control for trend test.

by twofold during the lactation treatment period. Food consumption at 225 mg/kg was significantly (p õ 0.0189) less than control by 8% during the gestation treatment period and by 23% during the lactation treatment period. No significant differences from control occurred for duration of gestation, litter size, or postimplantation loss at 225 mg/kg. No maternal parameters were affected by atorvastatin treatment at 20 or 100 mg/kg. F1 Generation Survival at birth and on PND 4 and 21 was significantly (p õ 0.0289) less than control at 225 mg/kg by 5, 36, and 45%, respectively (Table 2). The majority of deaths in this group occurred between PND 0 (including stillbirths) and 10, and were not preceded by clinical signs of morbidity. Survival during the maturation period (PND 21–91) was also significantly (p õ 0.05) less than control for females only, although only two females in this group died, as compared to no deaths in the control group. Group mean body weights of male and female offspring (Table 2) were significantly (p õ 0.0289) less than control at 100 mg/kg by 10–15% on PND 4 and 21, and at 225 mg/ kg they were significantly (p õ 0.0289) less by 9–37% on PND 0, 4, and 21. Group mean body weight gain during PND 4–21 was also significantly (p õ 0.0289) less than control for males and females at 100 and 225 mg/kg, by 15–16 and 32–40%, respectively. While body weights at

AID

FAAT 2400

/

6k24h$$303

02-02-98 11:08:16

100 mg/kg were essentially comparable to control by PN days 42 and 70 for males and females, respectively, body weights at 225 mg/kg continued to be less than control, and were significantly (p õ 0.0289) less for PN day 91 by 6 and 8% for males and females, respectively. Pinnae detachment and eye opening were significantly (p õ 0.0224) delayed at 225 mg/kg relative to control by approximately 1 and 1.5 days, respectively (Table 2). The incidence of neonatal offspring (°21 days of age) with the anatomical variation dilated renal pelvis was increased at 225 mg/kg relative to control and other atorvastatin groups. The incidence of dilated renal pelvis was comparable between control and atorvastatin groups after PND 21. Rotorod performance of females at 100 and 225 mg/kg was significantly (p õ 0.05) less than control by 37 and 53%, respectively (Fig. 3). No treatment-related effects were apparent on rotorod performance for males or on motor activity for males and females. For acoustic startle testing, although not statistically significant by trend analysis, group mean maximum input voltage for Noise Level (120 dB) Trial 1 for males at 20, 100, and 225 mg/kg was suppressed relative to control in a nondose-related manner by 48, 32, and 59%, respectively (Table 3). In addition, maximum input voltage for Noise Level Trial 2 for males at 225 mg/kg was suppressed by 50% (data not shown). This reduced response to an acoustic stimulus appeared to be transient, however, as responding in subse-

ftoxa

93

PRE- AND POSTNATAL TOXICITY OF ATORVASTATIN

TABLE 2 Physical Parameters in Offspring from Dams Treated with Atorvastatin from GD 7 through LD 20 Dose (mg/kg) Parameter Survival (%)a,b PN day 0 PN days 0–4 (preculling) PN days 4–21 PN days 21–91 Males Females Body weight (g)a PN day 0b Males Females PN day 4b Males Females PN day 21b Males Females PN day 42 Males Females PN day 70 Males Females PN day 91b Males Females Developmental landmarksa,b Pinnae detachment Lower incisor eruption Eye opening Testes descent Vaginal opening Incidence of dilated renal pelvis in neonates (° PN day 21)

0

20

(24) 98.7 { 0.73 (24) 98.7 { 0.54 (24) 100

(22) 97.0 { 2.36 (22) 98.4 { 0.58 (22) 98.3 { 1.25

(24) 95.8 { 1.80 (23) 96.4 { 1.28 (23) 100 { 0

(24) 95.8 { 2.88 (24) 100

(21) 100 (22) 100

(23) 100 (23) 100

225

(18) 94.0 { 1.77* (18) 63.3 { 10.3* (14) 55.4 { 13.3* (7) 100 (8) 87.5 { 12.5*

6.6 { 0.12 6.3 { 0.12

(24) (24)

6.4 { 0.12 6.0 { 0.08

(18) (18)

5.8 { 0.19* 5.4 { 0.16*

(24) 10.2 { 0.21 (24) 9.6 { 0.20

(21) 10.4 { 0.29 (22) 9.6 { 0.27

(23) (23)

9.2 { 0.32* 8.6 { 0.30*

(12) (14)

7.2 { 0.66* 6.4 { 0.52*

(24) 57.4 { 0.98 (24) 54.5 { 0.91

(21) 58.0 { 1.70 (22) 53.5 { 1.71

(23) 49.4 { 1.70* (23) 46.4 { 1.58*

(24) (24)

6.4 { 0.09 6.1 { 0.08

100

(22) (22)

(7) 40.5 { 4.57* (8) 34.2 { 4.31*

(24) 229 (24) 176

{ 4.2 { 2.2

(21) 235 (22) 177

{ 3.9 { 3.6

(23) 219 (23) 165

{ 5.7 { 2.8

(7) 198 (7) 146

{ 12.3 { 7.5

(24) 438 (24) 267

{ 6.7 { 3.4

(21) 447 (22) 269

{ 6.0 { 5.0

(23) 424 (23) 259

{ 7.7 { 4.1

(7) 402 (7) 246

{ 16.8 { 9.2

(24) 525 (24) 306

{ 8.4 { 4.6

(21) 538 (22) 305

{ 7.1 { 5.8

(23) 511 (23) 294

{ 8.5 { 4.3

(7) 493 (7) 280

{ 16.9* { 13.6*

(24) (24) (24) (24) (24)

{ { { { {

(22) (22) (22) (21) (22)

{ { { { {

(23) (23) (23) (23) (23)

{ { { { {

2.9 11.5 14.5 21.3 32.7

0.10 0.15 0.16 0.38 0.49

0.9%

2.5 10.7 14.5 20.9 31.6

0.15 0.21 0.19 0.36 0.26

3.3%

2.8 11.9 15.2 20.8 33.4

0.8%

0.10 0.14 0.19 0.31 0.89

(14) (8) (8) (7) (7)

3.7 11.6 16.1 22.2 34.1

{ { { { {

0.29** 0.45 0.50** 0.92 1.11

6.1%

Note. GD, gestation day; LD, lactation day; PN, postnatal. a (Litter N) Mean { standard error; sexes combined unless indicated otherwise. b Treatment groups compared statistically to vehicle control. * p õ 0.0289 (0.05 divided by the square root of seven individual parameters in the maturation body weight class), different from vehicle control for trend test. ** p õ 0.0224 (0.05 divided by the square root of five individual parameters in the developmental landmarks class), different from vehicle control for trend test.

quent trials was in general comparable to that of controls (data not shown). No treatment-related effects were apparent for Noise Level trials for females, and statistical comparison across 20 Noise Level trials revealed no treatment-related difference in the rate of habituation to repeated acoustic stimuli for males or females. The degree of response inhibition following a prepulse stimulus for Prepulse Trial 1 (defined as the percentage response inhibition relative to the preceding Noise Level trial) is also given in Table 4. No treatment-related effect on response inhibition was evident

AID

FAAT 2400

/

6k24h$$303

02-02-98 11:08:16

for males. However, response inhibition of all atorvastatin treatment groups of females was greater than that of controls for Trial 1, and, when all 20 trials were considered, group mean response inhibition across all trials was 68, 86, 80, and 84% for the controls and at 20, 100, and 225 mg/kg, respectively (data not shown). All shuttle avoidance parameters for males were comparable between atorvastatin groups and controls for both the Learning and Recall phases of testing. Although not statistically significant, group mean shock duration for females at

ftoxa

94

HENCK ET AL.

FIG. 3. Group mean and standard error rotorod performance in offspring from dams administered atorvastatin daily by gavage from gestation day 7 through lactation day 20. Each animal received three trials, with a maximum duration of 60 s/trial. The mean of these three trials was used to calculate the group mean. *Significantly different from control mean by trend test, p õ 0.0500.

225 mg/kg was 68% and twofold greater than control on test days 3 and 4 of the Learning Phase, respectively (Fig. 4). In addition, group mean percentage avoidance on test day 4 for females at 225 mg/kg was 65% less than control. All other shuttle avoidance parameters for females of all atorvastatin groups were comparable to control for both the Learning and Recall Phases of testing. No treatment-related effects were apparent on F1 generation reproductive parameters, including copulation and fertility indices, number of corpora lutea, litter size, and pre- and postimplantation losses, or on F2 fetal parameters, including body and placental weights, sex ratio, and external appearance (Table 4).

Pharmacokinetics Maternal plasma pharmacokinetic parameters obtained in the pre- and postnatal study on LD 8 are given in Table 5. Mean plasma Cmax and AUC(0–24) values increased with increasing atorvastatin dose; this increase was more than proportional to dose. Pharmacokinetic data from the placental transfer and milk excretion experiments are given in Table 6. 14C Radioactivity was detected in fetal rats and fetal livers 6 h following oral administration of [14C]atorvastatin to female rats on GD 19, indicating that atorvastatin and/or its metabolites undergo placental transfer. Mean fetal liver concentration was 0.176 mg eq/g, while mean concentration of the entire fetus was

TABLE 3 Group Mean Acoustic Startle Parameters for the First of 20 Noise Level and 20 Prepulse Trials, Including Percentage Response Inhibition, in Offspring from Dams Treated with Atorvastatin from GD 7 through LD 20 Atorvastatin (mg/kg)

Sex

0 0 20 20 100 100 225 225

Males Females Males Females Males Females Males Females

Maximum input voltage, Noise Level Trial 1a (24) (23) (20) (20) (22) (23) (6) (6)

973 1106 502 664 665 1163 396 716

{ { { { { { { {

Maximum input voltage, Prepulse Trial 1a

148.9 191.0 120.7 121.8 88.2 154.3 114.0 148.3

(24) (24) (20) (21) (22) (23) (6) (6)

a

315 406 232 190 257 318 146 77

{ { { { { { { {

78.2 66.4 69.9 55.5 70.8 68.1 64.1 12.6*

(N) Mean { standard error, first Noise Level (120 dB) and Prepulse (90 dB followed in 100 ms by 120-dB tone) trials. Percentage inhibition in response relative to Noise Level Trial 1. * p õ 0.0500, different from vehicle control for trend test.

b

AID

FAAT 2400

/

6k24h$$303

02-02-98 11:08:16

ftoxa

% Response inhibitionb 68 63 54 71 61 73 63 89

95

PRE- AND POSTNATAL TOXICITY OF ATORVASTATIN

TABLE 4 Reproductive Parameters in Offspring from Dams Treated with Atorvastatin from GD 7 through LD 20 Dose (mg/kg) Parameter

0

No. pairs cohabited Copulation index (%)a,b Fertility index (%)b,c Corpora lutead Litter sized Preimplantation lossd,e Postimplantation loss (%)d,f F2 Fetal parameters Body weight (g), live fetusesd Males Females Sex ratio (%)d Males Females Placenta weight (g)d

(18) (20) (18) (24)

23 91.3 95.2 17.4 { 13.2 { 13.0 { 9.0 {

20

0.62 1.06 3.69 1.39

(17) (17) (17) (22)

21 95.2 85.0 17.9 { 15.2 { 6.2 { 10.6 {

100

0.63 0.53 2.16 1.86

(20) (20) (20) (24)

23 91.3 95.2 18.1 { 15.3 { 9.6 { 10.2 {

225

0.64 0.64 2.67 2.33

(6) (6) (6) (18)

7 100 85.7 16.0 { 12.0 { 10.0 { 11.3 {

0.86 1.65 6.82 2.41

(19) 5.4 { 0.08 (19) 5.1 { 0.08

(17) 5.4 { 0.09 (17) 5.0 { 0.07

(20) 5.4 { 0.06 (20) 5.1 { 0.07

(6) 5.4 { 0.12 (6) 5.1 { 0.04

(19) 47.4 { 2.84 (19) 52.6 { 2.84 (19) 0.47 { 0.015

(17) 46.9 { 2.17 (17) 53.1 { 2.17 (17) 0.48 { 0.015

(20) 48.8 { 2.92 (20) 51.2 { 2.92 (20) 0.48 { 0.008

(6) 43.4 { 3.85 (6) 56.6 { 3.85 (6) 0.54 { 0.038

Note. All treatment group means are statistically comparable to control. a (Number females with positive mating divided by number cohabited) 1 100. b Group mean comparisons conducted. c (Number pregnant females divided by number with positive mating) 1 100. d (N) Mean { standard error; treatment groups compared statistically to vehicle control. e [(Number of corpora lutea 0 implant sites)/corpora lutea] 1 100. f [(Number of implant sites 0 viable fetuses)/implant sites] 1 100.

0.035 mg eq/g, suggesting hepatic extraction of radioactivity. Pregnant rats also had high liver radioactivity concentrations, with a mean of 26.7 mg eq/g, relative to mean placental and plasma values of 0.111 mg eq/g and 0.178 mg eq/mL, respectively. 14 C Radioactivity was detected in nursing pups 6 h following oral administration of [ 14C]atorvastatin to lactating rats on LD 13, indicating that atorvastatin equivalents undergo milk excretion in lactating rats. Lactating rats whose pups were removed 3 h prior to milk collection had mean plasma and milk values of 0.099 and 0.102 mg eq/mL, respectively, for a milk:plasma ratio of 1.56. The mean liver concentration for these females was 8.22 mg eq/g. Nursing pups that remained with the dam through 6 h postdose had mean plasma and liver concentrations of 0.050 mg eq/mL and 0.037 mg eq/g, respectively. Dams of these pups had mean plasma and liver concentrations of 0.077 mg eq/mL and 10.7 mg eq/g, respectively. DISCUSSION

Maternal toxicity, characterized by mortality, suppression of body weight gain and food intake, and pathologic lesions, occurred at 225 mg/kg. The plasma AUC(0–24) achieved at this dose was 22 times the human AUC at the highest recommended therapeutic dose of 80 mg/kg (Parke-Davis,

AID

FAAT 2400

/

6k24h$$303

02-02-98 11:08:16

1996). Lesions observed in the nonglandular mucosa of the stomach of females at 225 mg/kg were similar to those described in lactating rats treated with lovastatin (MacDonald et al., 1988; Minsker et al., 1990) or fluvastatin (FDA, 1993; Hrab et al., 1994). Stomach lesions observed with fluvastatin were shown to be specific to rodents and were considered due to contact irritation (Robison et al., 1994). Maternal body weight gain was less than that of controls at 225 mg/ kg during the gestation treatment period, but was actually greater during lactation, a pattern similar to that observed with lovastatin (FDA, 1987). It is unclear in the present study why weight gain increased during lactation when food consumption was suppressed to an even greater extent than it was during gestation. Maternal body weight gain was also greater than control at 100 mg/kg during lactation, but was comparable to control during gestation; no treatment-related effects on food consumption occurred at this dose. The biological significance of the dose-related increase in maternal body weight gain during lactation cannot be ascertained, but is not considered to represent maternal toxicity. Hrab and co-workers (1994) determined that mevalonic acid supplementation of perinatally administered fluvastatin prevented maternal toxicity, mortality, and cardiac myopathy; in addition, MacDonald and co-workers (1988) found that coadministration of mevalonic acid prevented lovastatin-induced lesions in the gastric mucosa. Hrab and co-workers (1994)

ftoxa

96

HENCK ET AL.

FIG. 4. Group mean and standard error shuttle avoidance parameters during the Learning Phase in female offspring from dams administered atorvastatin daily by gavage from gestation day 7 through lactation day 20. Data are mean shock duration for 25 trials/animal, with a maximum duration of 10 s/ trial, and percent of 25 trials which elicited an avoidance response. Although not statistically significant when tested by trend analysis for each day of the Learning Phase, group mean shock duration at 225 mg/kg was 68% and twofold greater than control on days 3 and 4 of the Learning Phase, respectively, and percent avoidance responding was 65% less than control on day 4.

concluded that the adverse maternal effects observed with fluvastatin were due to exaggerated pharmacologic activity resulting from inhibition of the conversion of HMG-CoA to mevalonate. Because atorvastatin employs the same mechanism of action as fluvastatin, it is likely that inhibition of HMG-CoA reductase was also responsible for atorvastatininduced maternal toxicity. No treatment-related malformations were observed in pups stillborn or found dead on PND 0, consistent with the lack of malformations in the teratology study in rats with atorvastatin (Dostal et al., 1994); in contrast, although postimplantation loss was increased in the teratology study at the highest dose of 300 mg/kg, no treatment-related effects occurred on fetal survival in the present study. However, survival at birth and throughout the preweaning period was significantly reduced at the maternally toxic dose of 225 mg/ kg, and all offspring died in 28% of the litters. Treatment of female rats during late gestation and throughout lactation with lovastatin (FDA, 1987) or fluvastatin (FDA, 1993; Hrab et al., 1994) also resulted in reduced offspring survival during the preweaning period at maternally toxic doses. This increase in mortality was reversed by supplementation of fluvastatin with mevalonic acid (Hrab et al., 1994), implicating inhibition of HMG-CoA reductase as a possible mechanism of action. The window of susceptibility to mortality

AID

FAAT 2400

/

6k24h$$304

02-02-98 11:08:16

resulting from atorvastatin exposure was immediately prior to birth (resulting in stillbirths) through the early postnatal period. When the events occurred which resulted in stillbirths and early postnatal death is unknown. Pharmacokinetic data indicate that atorvastatin can be transferred from the dam to the fetus during late gestation, and can also be excreted into the milk during lactation, providing two routes by which atorvastatin could have contributed to neonatal mortality. However, the contribution of the dam and the

TABLE 5 Mean Plasma Pharmacokinetic Parameters Obtained on LD 8 from Dams Treated with Atorvastatin Beginning on GD 7a Dose (mg/kg)

Cmax (ng eq/mL)

tmax (h)

AUC(0–24) (ng eqrh/mL)

20 100 225

104 (57.8) 1050 (85.4) 5780 (98.4)

2.4 (37.3) 3.2 (83.8) 2.0 (0.0)

803 (38.6) 6130 (83.7) 21600 (77.3)

Note. Cmax , maximum observed plasma atorvastatin concentration; tmax , time to reach Cmax ; AUC(0–24), area under the plasma concentration–time curve from time 0 to 24 h postdose. a Data presented as mean (% relative standard deviation, RSD); N Å 3 (225 mg/kg) or 5 (20 and 100 mg/kg).

ftoxa

PRE- AND POSTNATAL TOXICITY OF ATORVASTATIN

TABLE 6 Placental Transfer and Milk Secretion of Atorvastatin Equivalent Concentration 6 H after a Single 10 mg/kg Oral Dose of [14C]Atorvastatina Pregnant and fetal rats (N Å 6 litters) Maternal

Fetal

Liver

Plasma

Placentab

Fetusb

Liverb

26.7 (15)

0.178 (18)

0.111 (12)

0.035 (26)

0.176 (16)

Lactating rats (N Å 6 litters) Maternal Liver

Plasma

Milk

8.22 (47)

0.099 (62)

0.102 (25) [1.56]c

Lactating rats and pups (N Å 3 litters) Maternal Liver

Plasma

10.7 (7)

0.077 (9)

Neonatal Plasmab 0.050 (41)

Liverb 0.037 (92)

a

Data presented as mean (% RSD) mg eq/g or mg eq/mL. Pooled samples. c [Mean plasma:milk ratio]. b

impact of maternal toxicity at this dose must also be considered. It is of interest that in a study which included perinatal administration of fluvastatin alone or fluvastatin supplemented with mevalonic acid, there was a higher incidence of dead pups without milk in the stomach among groups administered fluvastatin alone (Hrab et al., 1994), indicating that pups did not suckle and/or dams did not nurse the pups. Offspring body weights were reduced relative to those of controls from birth through some portion of the maturation period at 100 and 225 mg/kg, indicating that growth suppression occurred at these doses. Pinnae detachment and eye opening were delayed at 225 mg/kg; these indications of developmental delay are not unexpected, since achievement of preweaning landmarks is highly correlated with body weight (ICH, 1994). The incidence of dilated renal pelvis was increased in offspring at 225 mg/kg; while this is not an uncommon finding in rodent developmental toxicity studies, it is unclear if it represents a developmental delay or a biologically significant anomaly (Kavlock et al., 1988). Developmental delay appears to be a likely scenario in this study, however, because dilated renal pelvis was no longer observed after PND 21. Various effects on behavior were detected in offspring from atorvastatin-treated groups. Rotorod performance of females at 100 and 225 mg/kg was significantly impaired.

AID

FAAT 2400

/

6k24h$$304

02-02-98 11:08:16

97

Although impairment may have been due to the reduced body weights of these animals, rotorod performance in males was unaffected in the presence of similar reductions in body weight gain. The acoustic startle response in males of all atorvastatin groups was suppressed relative to control during the first one or two trials only of a multiple trial test session. Although the initial response to a noise level tone was unaffected in females of all atorvastatin groups, the maximum response generated during the first prepulse trial at 225 mg/ kg was less than that of controls; in contrast, response inhibition resulting from a prepulse stimulus was elevated for the first trial and most remaining trials. It is unclear why response inhibition would be increased at 225 mg/kg; a deficit in hearing would be expected to decrease response inhibition because the ability to detect a prepulse tone would be impaired (Wecker et al., 1985), while a deficit in motor function would be expected to reduce the response to the initial noise level tone (Fechter and Young, 1983). Prepulse inhibition was unaffected in males and females, indicating that atorvastatin was not ototoxic under the testing conditions, but rather may have affected reactivity to initial acoustic stimuli. Transient effects on shuttle avoidance (increased mean shock duration and decreased percent avoidance) occurred during the learning phase in females only at 225 mg/ kg; no effects on retention of the learned response (memory) were observed in males or females. Transient effects on behavior were also observed with simvastatin; in a swim maze test of learning and memory, female offspring from dams treated with simvastatin during organogenesis had more errors than controls at 7 to 8 weeks of age, but not at 10 to 11 weeks of age (Wise et al., 1990b). Behavioral effects in offspring from dams given lovastatin prior to mating and during gestation included delays in development of the righting reflex, negative geotaxis, auditory startle, and swimming ontogeny, as well as a reduction in the mean latency period for open field (females of all dose groups) (FDA, 1987); offspring from dams given lovastatin during late gestation and lactation displayed a reduction in the acoustic startle response, decreased responsiveness to the free-fall righting reflex, and a reduced number of fecal pellets in the open field (FDA, 1987). Behavioral changes which occurred at 100 and 225 mg/kg in the present study may have been related to growth suppression and developmental delay, although a direct effect on the central nervous system cannot be entirely ruled out. Developmental toxicity induced by HMG-CoA reductase inhibitors is prevented in vitro or in vivo by mevalonic acid, but not by cholesterol or a specific inhibitor of cholesterol synthesis (Minsker et al., 1983; Surani et al., 1983; Carson and Lennarz, 1979; Hrab et al., 1994). Results of these studies indicate that products of the mevalonate pathway other than cholesterol could be involved in the mechanism of action governing developmental toxicity. HMG-CoA reductase

ftoxa

98

HENCK ET AL.

activity demonstrates a specific developmental pattern: a sharp rise in activity occurs immediately prior to birth (which may coincide with the increased sterol synthesis required for rapid tissue hypertrophy), followed by an equally sharp postnatal decline and low levels of activity throughout most of the suckling period; activity levels rise again following weaning (McNamara et al., 1972; Yount and McNamara, 1991). Pharmacokinetic data in the present study indicated not only that atorvastatin was transferred across the placenta on gestation day 19, but also that concentrations were higher in the fetal liver than in the whole fetus. Thus, the potential existed for atorvastatin to directly influence the late gestational increase in fetal hepatic cholesterol synthesis required for cell growth and proliferation, synthesis and maintenance of membranes, and synthesis of hormones and bile acids (Quesney-Huneeus et al., 1983; Belknap and Dietschy, 1988). The influence of HMG-CoA reductase inhibitors on isoprenoid derivatives of mevalonate has not been characterized during late gestation in the rat; however, given the important role these compounds play in membrane biogenesis, DNA replication, cellular growth and metabolism, and protein glycosylation (Quesney-Huneeus et al., 1983; Surani et al., 1983; Farnsworth et al., 1987; Brewer et al., 1993), the potential exists for reductase inhibitors to alter development by reducing levels of isoprenoids. Alteration of maternal levels of steroids synthesized from cholesterol could potentially play a role in developmental toxicity resulting from pre- and postnatal exposure to HMGCoA reductase inhibitors, although that role has not yet been elucidated. Lactation is the most important means by which neonates obtain cholesterol; therefore, hormonal disruption of lactation could adversely affect neonatal survival and development. Because the activity of HMG-CoA reductase and hepatic sterol synthesis is already low in suckling rat fetuses, it is unlikely that reductase inhibition by atorvastatin during this period would produce adverse developmental effects directly, despite the fact that direct exposure occurs via the milk. It is more likely that if developmental toxicity results from events which occur after parturition, those events are related to the maternal animal and may involve such factors as the quantity and quality of milk and maternal behavior, including nursing. Results from this study indicate that maternal toxicity occurred at 225 mg/kg, and that developmental toxicity, characterized by increased preweaning mortality, reduced body weight, developmental delay, and behavioral changes, was of the greatest magnitude at this dose. Reduced offspring body weight and behavioral changes also occurred at 100 mg/kg, a dose which did not result in adverse effects on the maternal animal. Although diminished acoustic startle responding occurred in male offspring at 20 mg/kg, the effect was transient, and no other apparent treatment-related effects occurred in dams or offspring at this dose. Therefore, 20

AID

FAAT 2400

/

6k24h$$304

02-02-98 11:08:16

mg/kg was considered a no-adverse-effect dose for developmental toxicity under the conditions of this study, while 100 mg/kg was considered a no-adverse-effect dose for maternal toxicity. ACKNOWLEDGMENTS The authors thank Debra Sherman and Roxann Alonzo for technical assistance, and Y. Y. Shum for plasma drug concentration analyses.

REFERENCES Beatty, W. W., and Beatty, P. A. (1970). Hormonal determinants of sex differences in avoidance behavior and reactivity to electric shock in the rat. J. Comp. Physiol. Psychol. 3, 446–455. Belknap, W. M., and Dietschy, J. M. (1988). Sterol synthesis and low density lipoprotein clearance in vivo in the pregnant rat, placenta, and fetus. J. Clin. Invest. 82, 2077–2085. Brewer, L. M., Sheardown, S. A., and Brown, N. A. (1993). HMG-CoA reductase mRNA in the post-implantation rat embryo studied by in situ hybridization. Teratology 47, 137–146. Carson, D. D., and Lennarz, W. J. (1979). Inhibition of polyisoprenoid and glycoprotein biosynthesis causes abnormal embryonic development. Proc. Natl. Acad. Sci. USA 76, 5709–5713. Corsini, A., Maggi, F. M., and Catapano, A. L. (1995). Pharmacology of competitive inhibitors of HMG-CoA reductase. Pharmacol. Res. 31, 9– 27. Denti, A., and Epstein, A. (1972). Sex differences in the acquisition of two kinds of avoidance behavior in rats. Physiol. Behav. 8, 611–615. Dostal, L. A., Schardein, J. L., and Anderson, J. A. (1994). Developmental toxicity of the HMG-CoA reductase inhibitor, atorvastatin, in rats and rabbits. Teratology 50, 387–394. Dostal, L. A., Whitfield, L. R., and Anderson, J. A. (1996). Fertility and general reproduction studies in rats with the HMG-CoA reductase inhibitor, atorvastatin. Fundam. Appl. Toxicol. 32, 285–292. Dunnett, C. W. (1955). A multiple comparison procedure for comparing several treatments with a control. J. Am. Stat. Assoc. 50, 1096–1121. Dunnett, C. W. (1964). New tables for multiple comparisons with a control. Biometrics 20, 482–491. Farnsworth, W. H., Hoeg, J. M., Maher, M., Brittain, E. H., Sherins, R. J., and Brewer, H. B., Jr. (1987). Testicular function in Type II hyperlipoproteinemic patients treated with lovastatin (mevinolin) or neomycin. J. Clin. Endocrinol. Metab. 65, 546–550. FDA (1987). Summary basis of approval of NDA 19-643 (lovastatin), 1– 28. FDA (1993). Summary basis of approval of NDA 20-261 (fluvastatin), 51– 54, and Amendment, pp. 1 and 2. Fechter, L. D., and Young, J. S. (1983). Discrimination of auditory from nonauditory toxicity by reflex modulation audiometry: Effects of triethyltin. Toxicol. Appl. Pharmacol. 70, 216–227. Fiorotto, M. L., Burrin, D. G., Perez, M., and Reeds, P. J. (1991). Intake and use of milk nutrients by rat pups suckled in small, medium, or large litters. Am. J. Physiol. 260, R1104–R1113. Freeman, M. E. (1988). The ovarian cycle of the rat. In The Physiology of Reproduction, Vol. 2 (E. Knobil and J. D. Neill, Eds.), pp. 1893–1928. Raven, New York. Goldstein, J. L., and Brown, M. S. (1990). Regulation of the mevalonate pathway. Nature 343, 425–430.

ftoxa

PRE- AND POSTNATAL TOXICITY OF ATORVASTATIN Greenhouse, S. W., and Geisser, S. (1959). On methods in the analysis of profile data. Psychometrika 32, 95–112. Henck, J. W., Petrere, J. A., and Anderson, J. A. (1995). Developmental neurotoxicity of CI-943, a novel antipsychotic. Neurotoxicol. Teratol. 17, 13–24. Henck, J. W., Frahm, D. T., and Anderson, J. A. (1996). Validation of automated behavioral test systems. Neurotoxicol. Teratol. 18, 189–197. Hodgen, G. D., and Itskovitz, J. (1988). Recognition and maintenance of pregnancy. In The Physiology of Reproduction, Vol. 2 (E. Knobil and J. D. Neill, Eds.), pp. 1995–2021. Raven, New York. Hrab, R. V., Hartmen, H. A., and Cox, R. H., Jr. (1994). Prevention of fluvastatin-induced toxicity, mortality, and cardiac myopathy in pregnant rats by mevalonic acid supplementation. Teratology 50, 19–26. ICH (1994). International Conference on Harmonisation; Guideline on detection of toxicity to reproduction for medicinal products. Fed. Reg. 59, 48746–48752. Iversen, S. D., and Iversen, L. L. (1981). Behavioral Pharmacology. Oxford University Press, New York. Johnson, R. A., and Wichern, D. W. (1982). Applied Multivariate Statistical Analysis. Prentice–Hall, New York. Kavlock, R. J., Hoyle, B. R., Rehnberg, B. F., and Rogers, E. H. (1988). The significance of the dilated renal pelvis in the nitrogen-exposed rat fetus: Effects on morphology and function. Toxicol. Appl. Pharmacol. 94, 287–296. Kovanen, P. T., Bilheimer, D. W., Goldstein, J. L., Jaramillo, J. J., and Brown, M. S. (1981). Regulatory role for hepatic low density lipoprotein receptors in vivo in the dog. Proc. Natl. Acad. Sci. USA 78, 1194–1198. Kruskal, W. H., and Wallis, W. A. (1952). Use of ranks in one-criterion variance analysis. J. Am. Stat. Assoc. 47, 583–621. MacDonald, J. S., Gerson, R. J., Kornbrust, D. J., Kloss, M. W., Prahalda, S., Berry, P. H., Alberts, A. W., and Bokelman, D. L. (1988). Preclinical evaluation of lovastatin. Am. J. Cardiol. 62, 16J–27J. McNamara, D. J., Quackenbush, F. W., and Rodwell, V. W. (1972). Regulation of hepatic 3-hydroxy-3-methyl-glutaryl coenzyme A reductase. J. Biol. Chem. 247, 5805–5810. Minsker, D. H., MacDonald, J. S., Robertson, R. T., and Bokelman, D. L. (1983). Mevalonate supplementation in pregnant rats suppresses the teratogenicity of mevinolinic acid, and inhibitor of 3-hydroxy-3-methylglutaryl-coenzyme A reductase. Teratology 28, 449–456.

AID

FAAT 2400

/

6k24h$$305

02-02-98 11:08:16

99

Minsker, D. H., Robertson, R. T., Bokelman, D. L., Akutsu, S., and Fujii, T. (1990). Simvastatin (MK-0733): Oral late gestation and early lactation study in rats. Oyo Yakuri 39, 169–179. Nawrocki, J. W., Weiss, S. R., Davidson, M. H., Sprecher, D. L., Schwartz, S. L., Lupien, P.-J., Jones, P. H., Haber, H. E., and Black, D. M. (1995). Reductions of LDL cholesterol by 25% to 60% in patients with primary hypercholesterolemia by atorvastatin, a new HMG-CoA reductase inhibitor. Arterioscler. Thromb. Vasc. Biol. 15, 678–682. Numan, M. (1988). Maternal behavior. In The Physiology of Reproduction, Vol. 2 (E. Knobil and J. D. Neill, Eds.), pp. 1569–1645. Raven, New York. Park, Y. C. (1985). Nonparametric sequential trend test. Proceedings of the Tenth Annual SAS Users Group International Conference, pp. 809–813. Parke-Davis, Division of Warner-Lambert Company (1996). Lipitor (atorvastatin calcium) tablets. [package insert] Quesney-Huneeus, V., Galick, H. A., and Siperstein, M. D. (1983). The dual role of mevalonate in the cell cycle. J. Biol. Chem. 258, 378–385. Robison, R. L., Suter, W., and Cox, R. H. (1994). Carcinogenicity and mutagenicity studies with fluvastatin, a new, entirely synthetic HMGCoA reductase inhibitor. Fundam. Appl. Toxicol. 23, 9–20. Soma, M. R., Corsini, A., and Paoletti, R. (1992). Cholesterol and mevalonic acid modulation in cell metabolism and multiplication. Toxicol. Lett. 64/ 65, 1–15. Surani, M. A. H., Kimber, S. J., and Osborn, J. C. (1983). Mevalonate reverses the developmental arrest of preimplantation mouse embryos by Compactin, an inhibitor of HMG CoA reductase. J. Embryol. Exp. Morph. 75, 205–223. Tukey, J. W., Ciminera, J. L., and Heyse, J. F. (1985). Testing the statistical certainty of a response to increasing doses of a drug. Biometrics 41, 295– 301. Wecker, J. R., Isor, J. R., and Foss, J. A. (1985). Reflex modification as a test for sensory function. Neurobehav. Toxicol. Teratol. 7, 733–738. Wise, L. D., Majka, J. A., Robertson, R. T., et al. (1990a). Simvastatin (MK-0733): Oral teratogenicity study in rats pre- and postnatal observation. Oyo Yakuri 39, 143–158. Wise, L. D., Minsker, D. H., Robertson, R. T., Bokelman, D. L., Akutsu, S., and Fujii, T. (1990b). Simvastatin (MK-0733): Oral fertility study in rats. Oyo Yakuri 39, 127–141. Yount, N. Y., and McNamara, D. J. (1991). Dietary regulation of maternal and fetal cholesterol metabolism in the guinea pig. Biochim. Biophys. Acta 1085, 82–90.

ftoxa