- Email: [email protected]
A Transplacental Carcinogenicity Bioassay in CD-1 Mice with Zidovudine
FUNDAMENTAL AND APPLIED TOXICOLOGY ARTICLE NO.
38, 195–198 (1997)
FA972342
A Transplacental Carcinogenicity Bioassay in CD-1 Mice with Zidovudine Kenneth M. Ayers, Carla E. Torrey, and David J. Reynolds Glaxo Wellcome Inc., Five Moore Drive, Research Triangle Park, North Carolina 27709 Received April 11, 1997; accepted June 17, 1997
A Transplacental Carcinogenicity Bioassay in CD-1 Mice with Zidovudine. Ayers, K. M., Torrey, C. E., and Reynolds, D. J. (1997). Fundam. Appl. Toxicol. 38, 195–198. In oral carcinogenicity bioassays, zidovudine (ZDV) induced vaginal epithelial cell tumors in mice given 30 or 40 mg/kg/day and rats given 300 mg/kg/day. To determine if lifetime exposure to ZDV, beginning perinatally, would alter this pattern of carcinogenicity, two groups of 60 pregnant CD-1 mice were given 20 or 40 mg/kg/day of ZDV in 0.5% methyl cellulose from Gestation Day 10 through Lactation Day 21. At weaning, 2 pups per sex from each of 35 litters in each group were assigned to the study and given 20 or 40 mg/kg/day of ZDV in the drinking water until 17 –35 days of age, followed by daily gavage for 24 months. Two additional groups of 60 pregnant CD-1 mice each were given 40 mg/kg/day of ZDV daily from Gestation Day 10 through Lactation Day 21; in one, ZDV treatment was halted at weaning and in the other, treatment was stopped 90 days after weaning. Two other groups of 60 pregnant CD-1 mice were left untreated (environmental control) or were given 0.5% methyl cellulose beginning on Gestation Day 10 (vehicle control). Vehicle control progeny received plain drinking water for 17 –35 days postweaning and then 0.5% methyl cellulose daily by gavage for 24 months. ZDV treatment did not affect survival or body weight in either sex. In females given 20 or 40 mg/kg/day of ZDV for 24 months there was mild macrocytic anemia. Similar, non-dose-related changes were seen in males in these groups. ZDV-related tumor findings were limited to the vagina, where there were 2 and 11 vaginal squamous cell carcinomas in mice given 20 or 40 mg/kg/day of ZDV daily, respectively. This incidence was not remarkably different from that seen in previously reported bioassays. It was concluded that lifetime oral treatment of mice with ZDV, beginning perinatally, did not alter the previously reported pattern of carcinogenicity and that under the conditions tested ZDV was not a transplacental carcinogen. q 1997 Society of Toxicology.
Zidovudine (ZDV)1 was the first antiviral agent licensed for the treatment of acquired immunodeficiency syndrome and has been used in adults and children since 1987. A placebo-controlled trial showed that treatment with ZDV 1 Zidovudine (3*-azido-3*-deoxythymidine) has previously been called AZT and azidothymidine. Retrovir is the tradename for ZDV.
dramatically reduced the rate of transmission of HIV from mother to infant (Connor et al., 1995) and ZDV is now also approved for use in the prevention of maternal:fetal HIV transmission. The treatment regimen (Physicians’ Desk Reference, 1996) includes oral ZDV beginning between 14 and 34 weeks of gestation (100 mg po five times daily until the start of labor), iv ZDV during labor (2 mg/kg iv for 1 hr, then 1 mg/kg/hr iv until clamping of umbilical cord), and administration of ZDV to the newborn after birth (2 mg/kg po every 6 hr, continuing through 6 weeks of age). In oral carcinogenicity bioassays (Ayers et al., 1996), ZDV treatment, begun when animals were 42 days of age, was shown to induce vaginal epithelial cell tumors in mice after 22 months of dosing at 30 or 40 mg/kg/day and after 24 months of dosing in rats given 300 mg/kg/day. The transplacental carcinogenicity study described here was conducted to determine if lifetime exposure to ZDV, beginning perinatally, would alter this pattern of carcinogenicity. MATERIALS AND METHODS Six groups of 60 male and 60 female CD-1 mice2 (79 days of age) were pair housed (1 male and 1 female per cage) for a period of up to 14 days. Paired females were examined for the presence of a vaginal plug daily. If a plug was observed that day was designated as Day 0 of gestation. If a plug was not seen, the female was paired again with the appropriate male. This procedure was repeated until each female was inseminated or the 14-day cohabitation period was over. Bred females were randomly assigned to the six treatment groups using a table of random numbers generated in SAS. Prior to Gestation Day 16, dams were housed in shoebox cages and provided with Alpha-Dri bedding.3 At weaning, all pups from one litter were group housed in shoebox cages (males in one cage, females in another) for a 17- to 35-day period and provided food4 and tap water ad libitum. There were six treatment groups and the dosing regimen for each is summarized in Table 1. Once daily gavage dosing of the dams with vehicle or ZDV suspended in 0.5% methyl cellulose (mg/ml) began on Gestation Day 10 and continued throughout gestation, parturition, and lactation. [Note: ZDV had previously been shown to be present in milk (Ruprecht et al., 1990).] Two males and 2 females from each of 35 litters in each group
2 CD-1 mice were purchased from Charles River Breeding Laboratories, Raleigh, NC. 3 Alpha-Dri bedding, Shepherd Specialty Papers, Kalamazoo, MI. 4 Wayne Lab Blox, Allied Mills Inc., Chicago, IL 60606.
195
AID
FAAT 2342
/
6k1f$$$$21
08-20-97 08:34:59
0272-0590/97 $25.00 Copyright q 1997 by the Society of Toxicology. All rights of reproduction in any form reserved.
ftoxa
196
AYERS, TORREY, AND REYNOLDS
TABLE 1 Dosing Regimen for Mice Given Zidovudine in a Transplacental Carcinogenicity Study Group
Substance
Dose (mg/kg/day)
Dose regimen
1
Environmental control
0
2
0
3
Vehicle control, 0.5% methyl cellulose Zidovudine
20
4
Zidovudine
40
5
Zidovudine
40
6
Zidovudine
40
Left untreated throughout gestation, lactation, and 24-month postweaning period. Gestation Day 10 through Lactation Day 21. Postweaning via drinking water, then by daily gavage for 24 months. Gestation Day 10 through Lactation Day 21. Postweaning via drinking water, then by daily gavage for 24 months. Gestation Day 10 through Lactation Day 21. Postweaning via drinking water, then by daily gavage for 24 months. Gestation Day 10 through Lactation Day 21. Postweaning via drinking water, then by daily gavage until Postnatal Day 90. No further treatment for remainder of 24-month period. Gestation Day 10 through Lactation Day 21. No further treatment for remainder of 24-month period.
were selected for inclusion in the study to give a total of 70 male and 70 female progeny in each dose group. At weaning, ZDV was administered in the drinking water (concentration varied based on water consumption of first-born pups) for 17–35 days, following which gavage dosing began. Dose Day 1 of the bioassay was considered to be the day gavage dosing was initiated in the progeny. Mice were monitored daily for signs of toxicity and weekly for the presence of externally visible masses. Body weight was measured in dams on Gestation Days 0, 6, 10, 12, 15, and 17 and Lactation Days 1, 4, 7, 14, and 21. Body weight of progeny was determined when the pups were 1, 4, 7, 14, and 21 days of age and at weekly intervals thereafter until body weight plateaued, after which it was measured once monthly. Measurements for evidence of absorption of ZDV were carried out by radioimmunoassay (Tadapalli et al., 1990) on: (a) plasma samples obtained 15 min after dosing from the first three dams in groups 3–6 at weaning, (b) 30 min after the lights were turned on in the morning from as many pups as necessary to provide adequate samples for analysis, and (c) from three separate pooled samples taken 15 min postdose from 3 males and 3 females each in groups 3 and 4 at the scheduled termination of the study. At the euthanization, blood samples were obtained from 10 males and 10 females per group to measure the following: hematocrit (PCV), total erythrocyte count (RBC), hemoglobin (Hb), total leukocyte count (WBC), mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), and red cell distribution width (RDW). The study was terminated at 24 months. All mice found dead, euthanized for humane reasons, or euthanized at the scheduled study termination point were examined grossly and representative portions of approximately 50 separate tissues and organs were taken, fixed in 10% neutral-buffered Formalin, and processed for light microscopic examination. These included all major organs and tissues, including the full male and female reproductive tract. All protocol-specified tissues from all mice in groups 1, 2, and 4 were examined microscopically. The vagina only was examined from all animals in groups 3, 5, and 6. Survival analyses were limited to groups 3 and 4 since only these groups received ZDV for the duration of their lifetimes. Survival was analyzed by life-table techniques using Kaplan–Meier product limit estimates, Cox– Tarone binary regression of life tables, and Gehan–Breslow nonparametric methods (Thomas et al., 1977). The incidences of all tumors (increased or decreased in a treated group by §2 when compared to the vehicle control group) were analyzed by the Cochran–Armitrage method for trend (where applicable) and Fisher–Irwin exact test for control versus treatment comparisons (Thakur et al., 1985). In the event of any possible intercurrent mortality differences due to toxicity among the treated groups, survival-adjusted
AID
FAAT 2342
/
6k1f$$$$22
08-20-97 08:34:59
tumor analyses were performed on increased or decreased incidences by logistic regression of tumor prevalences (Dinse and Lagakos, 1983) or the exact prevalence test (Ali, 1990). When certain tumor types were assigned as cause of death or morbidity and others were considered incidental, the incidental tumors were censored and those considered as contributing to death or morbidity were analyzed by life-table and prevalence techniques (Ali, 1990). The score statistics and their variances from the above were used to compute the combined evidence in each case as described by Gart et al. (Gart et al., 1986).
RESULTS
There were no statistically significant (p õ 0.05) effects of treatment on survival in either sex in groups 3 and 4. Survival was decreased (p Å 0.0073) in group 5 females compared to the vehicle control. This was not considered to be drug-related, since this group was given ZDV only until Postnatal Day 90. There were no ZDV-related signs of toxicity and treatment had no effect on body weight. As shown in Table 2, compared to the vehicle control group, there was a mild doserelated macrocytic anemia at euthanization in group 3 and 4 females, as evidenced by a decrease (6–15%) in PCV and RBC and an accompanying increase (4–10%) in MCV and MCH. Similar changes were seen in males in these groups, but the effects were not dose-related. RDW was also increased in groups 3 and 4, but the alterations also did not show a dose relationship. No treatment-related changes were noted in WBC. No gross pathologic findings were attributed to drug treatment. Histopathologically, there were no treatment-related neoplastic or nonneoplastic findings in any tissue or organ in males. ZDV-related findings in females were limited to the vagina, where there were 11 (p Å 0.0002) vaginal squamous cell carcinomas in group 4. None of the tumors metastasized. The earliest vaginal tumor was seen after 694 days (23 months) of dosing. Two vaginal squamous cell carcino-
ftoxa
197
TRANSPLACENTAL CARCINOGENICITY BIOASSAY WITH ZIDOVUDINE
TABLE 2 Terminal (24 Months) Group Means ({SD) Hematology Data for Mice Given Zidovudine in a Transplacental Carcinogenicity Study Group (dose) (mg/kg/day) Group 1 (EC)a Male Female Group 2 (VC)b Male Female Group 3 (20) Male Female Group 4 (40) Male Female Group 5 (40) Male Female Group 6 (40) Male Female a b
WBC (103/ml)
RBC (106/ml)
PCV (%)
Hb (g/dl)
MCV (m)
MCH (pg)
MCHC (g/dl)
RDW (%)
2.5 (1.1) 3.4 (1.5)
8.6 (1.1) 8.7 (0.8)
41.2 (4.2) 42.5 (3.0)
13.6 (1.4) 14.3 (1.0)
48.2 (3.2) 49.0 (2.8)
15.8 (0.8) 16.5 (0.9)
32.9 (0.8) 33.8 (1.0)
14.3 (1.5) 13.5 (1.0)
2.7 (1.4) 2.2 (0.8)
9.4 (1.6) 8.0 (2.0)
44.4 (5.5) 40.4 (7.6)
14.6 (2.0) 13.1 (2.7)
47.9 (4.3) 51.7 (6.5)
15.7 (1.0) 16.6 (1.6)
32.8 (1.1) 32.2 (1.4)
14.2 (1.9) 15.4 (3.1)
3.0 (4.1) 3.1 (2.5)
7.4 (1.9) 7.5 (0.7)
37.0 (8.3) 40.1 (2.6)
12.4 (2.8) 13.0 (1.0)
50.3 (3.5) 53.6 (2.7)
16.8 (1.3) 17.3 (0.6)
33.5 (1.1) 32.3 (0.9)
16.3 (3.5) 16.2 (2.0)
1.6 (0.6) 2.2 (2.0)
7.8 (0.9) 6.8 (0.6)
40.4 (3.0) 37.9 (2.4)
13.5 (1.0) 12.5 (0.8)
52.0 (3.1) 55.5 (2.9)
17.3 (0.9) 18.3 (0.8)
33.3 (0.8) 32.9 (0.8)
15.0 (1.9) 15.0 (0.8)
2.2 (0.8) 2.3 (1.2)
8.3 (1.4) 7.9 (1.5)
38.7 (5.6) 38.5 (6.2)
12.9 (2.0) 12.5 (2.3)
47.1 (2.8) 49.2 (3.8)
15.6 (0.6) 16.0 (0.7)
33.2 (0.9) 32.5 (1.3)
13.6 (1.1) 15.2 (4.7)
3.1 (4.5) 2.8 (2.1)
9.2 (1.1) 8.2 (1.0)
42.3 (4.8) 40.1 (3.9)
14.1 (1.7) 13.2 (1.3)
46.2 (2.0) 48.8 (3.5)
15.4 (0.7) 16.1 (1.0)
33.4 (0.9) 33.0 (1.0)
14.9 (3.3) 13.8 (1.1)
EC, environmental control. VC, vehicle control.
mas were observed in group 3. Focal squamous epithelial dysplasia was noted in the vaginas of two group 3 and six group 4 animals and there was an increase in the incidence of moderate vaginal squamous epithelial hyperplasia in group 4. A vaginal squamous cell carcinoma was also seen in one group 5 animal. Such tumors, while rare, do occur spontaneously. Since this animal was given ZDV for only 90 days postweaning, and we have shown that ZDV-related vaginal tumors occur only after at least 19 months of daily oral dosing (Ayers et al., 1996), it is likely that this single occurrence was of spontaneous origin. As shown in Table 3, mean peak ZDV plasma concentrations at term in animals given 40 mg/kg were approximately 37 mg/ml. Mean ZDV levels in dams given 40 mg/kg on Lactation Day 21 were approximately 14 to 15 mg/ml. In pups given 40 mg/kg in the drinking water during the postweaning period, mean ZDV plasma concentrations ranged from approximately 0.04 to 0.14 mg/ml. For comparison, the average steady-state concentration in humans at the recommended daily dose (100 mg q 4 hr) is 0.62 mg/ml. DISCUSSION
Ayers et al. (1996) reported that ZDV, given orally or intravaginally, induced vaginal epithelial neoplasms in rodents in lifetime carcinogenicity bioassays. In a subsequent study, Olivero et al. (1994) showed a positive correlation between the dose of ZDV, proliferation of cells in the basal
AID
FAAT 2342
/
6k1f$$$$22
08-20-97 08:34:59
vaginal epithelium, and incorporation of ZDV into vaginal DNA. They speculated that these findings may play a role in the ability of ZDV to induce vaginal tumors in mice. At a meeting of a panel of experts at the National Institutes of Health on January 14, 1996, to discuss the transplacental carcinogenicity of ZDV (Rowe, 1997), scientists from the National Cancer Institute reported the preliminary results of
TABLE 3 Mean ({SD) Zidovudine (ZDV) Plasma Concentrations Group 3 4 5 6 3 3 4 4 5 5 3 4
damsa damsa damsa damsa male pupsb female pupsb male pupsb female pupsb male pupsb female pupsb malesc malesc
Dose (mg/kg/day) 20 40 40 40 20 20 40 40 40 40 20 40
ZDV (mg/ml) 12.6 15.3 14.1 15.0 0.09 0.04 0.10 0.14 0.10 0.07 14.5 36.6
{ { { { { { { { { { { {
Note. No data available for females. a 15 min postdose (Lactation Day 21). b 30 min after lights turned on (Postweaning Day 14). c 15 min postdose at scheduled termination of study (24 months).
ftoxa
0.8 2.9 1.3 1.2 0.04 0.01 0.10 0.13 0.06 0.04 2.0 1.8
198
AYERS, TORREY, AND REYNOLDS
a study in which daily po doses of 12.5 and 25 mg of ZDV (Ç1000 mg/kg nonpregnant body weight or Ç450 mg/kg of terminal body weight) were given to pregnant mice from Gestation Days 12 to 18 only (Olivero, O. A., Diwan, B. A., Anderson, L. M., Rice, J. M., Beland, F. A., Yuspa, S. H., and Poirier, M. C., personal communication). They demonstrated that ZDV was incorporated into DNA in various tissues and found high incidences of liver and lung tumors in male offspring and lung and reproductive organ tumors in female offspring 1 year after birth. Based on this evidence, they classified ZDV as a transplacental carcinogen. Also discussed at this meeting were the results of the transplacental carcinogenicity study described here in which lower doses of ZDV were used in a dosing regimen designed to mimic the clinical setting and to exaggerate human exposure levels based on peak plasma concentrations and areaunder-the-concentration time curve (AUC). The high dose used in this study (40 mg/kg/day) had previously been shown to be carcinogenic to the vaginal epithelium (Ayers et al., 1996) and to produce peak plasma concentrations 40-fold higher and an AUC 3-fold higher than the values for these parameters in humans (0.62 and 5.24 mg/ml 1 hr, respectively) at the recommended daily dose (100 mg q 4 hr). The mean peak plasma concentration seen at euthanization in this transplacental study (36.6 mg/ml) was approximately 59fold higher than the average steady-state concentration in humans at the recommended daily dose. The differences noted in peak plasma concentrations at 40 mg/kg/day in the dams and males at euthanization (Table 3) were probably related to the very brief peak (Tmax) that occurs after oral administration of ZDV in mice. Samples taken just before or after this peak would result in significant differences in concentration. This also explains the differences in peak concentration seen at 40 mg/kg/day in the previously reported study (Ayers et al., 1996) and this transplacental carcinogenicity study. In the current study, ZDV again induced tumors in the vagina, but not in other tissues or organs. These findings were virtually identical to those observed in the standard oral carcinogenicity study in mice (Ayers et al., 1996), in which the incidence rate of vaginal epithelial tumors in highdose females was 12% (7 tumors in 60 mice). In the current study, the incidence rate for these tumors at 40 mg/kg/day was 16% (11 tumors in 70 mice). Time of onset for the vaginal tumors was also not affected by lifetime treatment
AID
FAAT 2342
/
6k1f$$$$22
08-20-97 08:34:59
in this study. The earliest that a vaginal tumor was seen in the previously reported study was after 579 days (19 months) of daily dosing. In this transplacental carcinogenicity study the earliest vaginal tumor was seen after 694 days (23 months) of dosing. It was concluded, therefore, that lifetime oral treatment of mice with ZDV, beginning perinatally, did not alter the previously reported pattern of carcinogenicity in mice and that under the conditions tested, ZDV was not a transplacental carcinogen. REFERENCES Ali, M. W. (1990). Exact versus asymptotic tests of trend of tumor prevalence in tumorigenicity experiments. A comparison of p-values for small frequency of tumors. Drug Inform. J. 24, 727–737. Ayers, K. M., Clive, D., Tucker, W. E., Jr., Hajian, G., and de Miranda, P. (1996). Nonclinical toxicology studies with zidovudine: Genetic toxicity tests and carcinogenicity bioassays in mice and rats. Fundam. Appl. Toxicol. 32, 148–158. Connor, E. M., Sperling, R. S., Gelber, R., Kisalev, P., Scott, G., O’Sullivan, M. J., Van Dyke, R., Bey, M., Shearer, W., Jacobson, R. L., et al. (1995). Reduction of maternal–infant transmission of human immunodeficiency virus type I with zidovudine treatment. N. Engl. J. Med. 331, 1173– 1180. Dinse, G. E., and Lagakos, S. W. (1983). Regression analysis of tumor prevalence data. J. R. Stat. Soc. Ser. C (Appl. Stat.). 32, 236–241. Gart, J. J., Krewski, D., Lee, P. N., Tarone, R. E., and Wahrendorf, J. (1986). Statistical methods in cancer research. In The Design and Analysis of Long Term Animal Experiments. IARC Sci. Publ. No. 79, Vol. III, p. 219. Oxford Univ. Press, London. Olivero, O. A., Beland, F. A., Fullerton, N. F., and Poirier, M. C. (1994). Vaginal epithelial damage and expression of preneoplastic markers in mice during chronic dosing with tumorigenic levels of AZT. Cancer Res. 54, 6235–6242. Physicians’ Desk Reference (1996). 50th ed., pp. 1162–1163. Montrale, NJ. Rowe, P. M. (1996). US expert panel reaffirms benefit of perinatal zidovudine. Lancet 349, 258. Ruprecht, R. M., Sharpe, A. H., Jaenisch, R., and Trites, D. (1990). Analysis of 3*-azido-3*-deoxythymidine levels in tissues and milk by isocratic high-performance liquid chromatography. J. Chromatogr. 528, 371–383. Tadapalli, S., Puckett, L., Jeal, S., Kanics, L., and Quinn, R. (1990). Differential assay of zidovudine and its glucuronide metabolite in serum and urine with a radioimmunoassay kit. Clin. Chem. 36, 897–900. Thakur, A. K., Berry, K. J., and Mielke, P. W., Jr. (1985). A FORTRAN program for testing trend and homogeneity in proportions. Comput. Biomed. Res. 19, 229–233. Thomas, D. G., Breslow, N., and Gart, J. J. (1977). Trend and homogeneity analyses of proportions and life table data. Comput. Biomed. Res. 10, 373–381.
ftoxa
38, 195–198 (1997)
FA972342
A Transplacental Carcinogenicity Bioassay in CD-1 Mice with Zidovudine Kenneth M. Ayers, Carla E. Torrey, and David J. Reynolds Glaxo Wellcome Inc., Five Moore Drive, Research Triangle Park, North Carolina 27709 Received April 11, 1997; accepted June 17, 1997
A Transplacental Carcinogenicity Bioassay in CD-1 Mice with Zidovudine. Ayers, K. M., Torrey, C. E., and Reynolds, D. J. (1997). Fundam. Appl. Toxicol. 38, 195–198. In oral carcinogenicity bioassays, zidovudine (ZDV) induced vaginal epithelial cell tumors in mice given 30 or 40 mg/kg/day and rats given 300 mg/kg/day. To determine if lifetime exposure to ZDV, beginning perinatally, would alter this pattern of carcinogenicity, two groups of 60 pregnant CD-1 mice were given 20 or 40 mg/kg/day of ZDV in 0.5% methyl cellulose from Gestation Day 10 through Lactation Day 21. At weaning, 2 pups per sex from each of 35 litters in each group were assigned to the study and given 20 or 40 mg/kg/day of ZDV in the drinking water until 17 –35 days of age, followed by daily gavage for 24 months. Two additional groups of 60 pregnant CD-1 mice each were given 40 mg/kg/day of ZDV daily from Gestation Day 10 through Lactation Day 21; in one, ZDV treatment was halted at weaning and in the other, treatment was stopped 90 days after weaning. Two other groups of 60 pregnant CD-1 mice were left untreated (environmental control) or were given 0.5% methyl cellulose beginning on Gestation Day 10 (vehicle control). Vehicle control progeny received plain drinking water for 17 –35 days postweaning and then 0.5% methyl cellulose daily by gavage for 24 months. ZDV treatment did not affect survival or body weight in either sex. In females given 20 or 40 mg/kg/day of ZDV for 24 months there was mild macrocytic anemia. Similar, non-dose-related changes were seen in males in these groups. ZDV-related tumor findings were limited to the vagina, where there were 2 and 11 vaginal squamous cell carcinomas in mice given 20 or 40 mg/kg/day of ZDV daily, respectively. This incidence was not remarkably different from that seen in previously reported bioassays. It was concluded that lifetime oral treatment of mice with ZDV, beginning perinatally, did not alter the previously reported pattern of carcinogenicity and that under the conditions tested ZDV was not a transplacental carcinogen. q 1997 Society of Toxicology.
Zidovudine (ZDV)1 was the first antiviral agent licensed for the treatment of acquired immunodeficiency syndrome and has been used in adults and children since 1987. A placebo-controlled trial showed that treatment with ZDV 1 Zidovudine (3*-azido-3*-deoxythymidine) has previously been called AZT and azidothymidine. Retrovir is the tradename for ZDV.
dramatically reduced the rate of transmission of HIV from mother to infant (Connor et al., 1995) and ZDV is now also approved for use in the prevention of maternal:fetal HIV transmission. The treatment regimen (Physicians’ Desk Reference, 1996) includes oral ZDV beginning between 14 and 34 weeks of gestation (100 mg po five times daily until the start of labor), iv ZDV during labor (2 mg/kg iv for 1 hr, then 1 mg/kg/hr iv until clamping of umbilical cord), and administration of ZDV to the newborn after birth (2 mg/kg po every 6 hr, continuing through 6 weeks of age). In oral carcinogenicity bioassays (Ayers et al., 1996), ZDV treatment, begun when animals were 42 days of age, was shown to induce vaginal epithelial cell tumors in mice after 22 months of dosing at 30 or 40 mg/kg/day and after 24 months of dosing in rats given 300 mg/kg/day. The transplacental carcinogenicity study described here was conducted to determine if lifetime exposure to ZDV, beginning perinatally, would alter this pattern of carcinogenicity. MATERIALS AND METHODS Six groups of 60 male and 60 female CD-1 mice2 (79 days of age) were pair housed (1 male and 1 female per cage) for a period of up to 14 days. Paired females were examined for the presence of a vaginal plug daily. If a plug was observed that day was designated as Day 0 of gestation. If a plug was not seen, the female was paired again with the appropriate male. This procedure was repeated until each female was inseminated or the 14-day cohabitation period was over. Bred females were randomly assigned to the six treatment groups using a table of random numbers generated in SAS. Prior to Gestation Day 16, dams were housed in shoebox cages and provided with Alpha-Dri bedding.3 At weaning, all pups from one litter were group housed in shoebox cages (males in one cage, females in another) for a 17- to 35-day period and provided food4 and tap water ad libitum. There were six treatment groups and the dosing regimen for each is summarized in Table 1. Once daily gavage dosing of the dams with vehicle or ZDV suspended in 0.5% methyl cellulose (mg/ml) began on Gestation Day 10 and continued throughout gestation, parturition, and lactation. [Note: ZDV had previously been shown to be present in milk (Ruprecht et al., 1990).] Two males and 2 females from each of 35 litters in each group
2 CD-1 mice were purchased from Charles River Breeding Laboratories, Raleigh, NC. 3 Alpha-Dri bedding, Shepherd Specialty Papers, Kalamazoo, MI. 4 Wayne Lab Blox, Allied Mills Inc., Chicago, IL 60606.
195
AID
FAAT 2342
/
6k1f$$$$21
08-20-97 08:34:59
0272-0590/97 $25.00 Copyright q 1997 by the Society of Toxicology. All rights of reproduction in any form reserved.
ftoxa
196
AYERS, TORREY, AND REYNOLDS
TABLE 1 Dosing Regimen for Mice Given Zidovudine in a Transplacental Carcinogenicity Study Group
Substance
Dose (mg/kg/day)
Dose regimen
1
Environmental control
0
2
0
3
Vehicle control, 0.5% methyl cellulose Zidovudine
20
4
Zidovudine
40
5
Zidovudine
40
6
Zidovudine
40
Left untreated throughout gestation, lactation, and 24-month postweaning period. Gestation Day 10 through Lactation Day 21. Postweaning via drinking water, then by daily gavage for 24 months. Gestation Day 10 through Lactation Day 21. Postweaning via drinking water, then by daily gavage for 24 months. Gestation Day 10 through Lactation Day 21. Postweaning via drinking water, then by daily gavage for 24 months. Gestation Day 10 through Lactation Day 21. Postweaning via drinking water, then by daily gavage until Postnatal Day 90. No further treatment for remainder of 24-month period. Gestation Day 10 through Lactation Day 21. No further treatment for remainder of 24-month period.
were selected for inclusion in the study to give a total of 70 male and 70 female progeny in each dose group. At weaning, ZDV was administered in the drinking water (concentration varied based on water consumption of first-born pups) for 17–35 days, following which gavage dosing began. Dose Day 1 of the bioassay was considered to be the day gavage dosing was initiated in the progeny. Mice were monitored daily for signs of toxicity and weekly for the presence of externally visible masses. Body weight was measured in dams on Gestation Days 0, 6, 10, 12, 15, and 17 and Lactation Days 1, 4, 7, 14, and 21. Body weight of progeny was determined when the pups were 1, 4, 7, 14, and 21 days of age and at weekly intervals thereafter until body weight plateaued, after which it was measured once monthly. Measurements for evidence of absorption of ZDV were carried out by radioimmunoassay (Tadapalli et al., 1990) on: (a) plasma samples obtained 15 min after dosing from the first three dams in groups 3–6 at weaning, (b) 30 min after the lights were turned on in the morning from as many pups as necessary to provide adequate samples for analysis, and (c) from three separate pooled samples taken 15 min postdose from 3 males and 3 females each in groups 3 and 4 at the scheduled termination of the study. At the euthanization, blood samples were obtained from 10 males and 10 females per group to measure the following: hematocrit (PCV), total erythrocyte count (RBC), hemoglobin (Hb), total leukocyte count (WBC), mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), and red cell distribution width (RDW). The study was terminated at 24 months. All mice found dead, euthanized for humane reasons, or euthanized at the scheduled study termination point were examined grossly and representative portions of approximately 50 separate tissues and organs were taken, fixed in 10% neutral-buffered Formalin, and processed for light microscopic examination. These included all major organs and tissues, including the full male and female reproductive tract. All protocol-specified tissues from all mice in groups 1, 2, and 4 were examined microscopically. The vagina only was examined from all animals in groups 3, 5, and 6. Survival analyses were limited to groups 3 and 4 since only these groups received ZDV for the duration of their lifetimes. Survival was analyzed by life-table techniques using Kaplan–Meier product limit estimates, Cox– Tarone binary regression of life tables, and Gehan–Breslow nonparametric methods (Thomas et al., 1977). The incidences of all tumors (increased or decreased in a treated group by §2 when compared to the vehicle control group) were analyzed by the Cochran–Armitrage method for trend (where applicable) and Fisher–Irwin exact test for control versus treatment comparisons (Thakur et al., 1985). In the event of any possible intercurrent mortality differences due to toxicity among the treated groups, survival-adjusted
AID
FAAT 2342
/
6k1f$$$$22
08-20-97 08:34:59
tumor analyses were performed on increased or decreased incidences by logistic regression of tumor prevalences (Dinse and Lagakos, 1983) or the exact prevalence test (Ali, 1990). When certain tumor types were assigned as cause of death or morbidity and others were considered incidental, the incidental tumors were censored and those considered as contributing to death or morbidity were analyzed by life-table and prevalence techniques (Ali, 1990). The score statistics and their variances from the above were used to compute the combined evidence in each case as described by Gart et al. (Gart et al., 1986).
RESULTS
There were no statistically significant (p õ 0.05) effects of treatment on survival in either sex in groups 3 and 4. Survival was decreased (p Å 0.0073) in group 5 females compared to the vehicle control. This was not considered to be drug-related, since this group was given ZDV only until Postnatal Day 90. There were no ZDV-related signs of toxicity and treatment had no effect on body weight. As shown in Table 2, compared to the vehicle control group, there was a mild doserelated macrocytic anemia at euthanization in group 3 and 4 females, as evidenced by a decrease (6–15%) in PCV and RBC and an accompanying increase (4–10%) in MCV and MCH. Similar changes were seen in males in these groups, but the effects were not dose-related. RDW was also increased in groups 3 and 4, but the alterations also did not show a dose relationship. No treatment-related changes were noted in WBC. No gross pathologic findings were attributed to drug treatment. Histopathologically, there were no treatment-related neoplastic or nonneoplastic findings in any tissue or organ in males. ZDV-related findings in females were limited to the vagina, where there were 11 (p Å 0.0002) vaginal squamous cell carcinomas in group 4. None of the tumors metastasized. The earliest vaginal tumor was seen after 694 days (23 months) of dosing. Two vaginal squamous cell carcino-
ftoxa
197
TRANSPLACENTAL CARCINOGENICITY BIOASSAY WITH ZIDOVUDINE
TABLE 2 Terminal (24 Months) Group Means ({SD) Hematology Data for Mice Given Zidovudine in a Transplacental Carcinogenicity Study Group (dose) (mg/kg/day) Group 1 (EC)a Male Female Group 2 (VC)b Male Female Group 3 (20) Male Female Group 4 (40) Male Female Group 5 (40) Male Female Group 6 (40) Male Female a b
WBC (103/ml)
RBC (106/ml)
PCV (%)
Hb (g/dl)
MCV (m)
MCH (pg)
MCHC (g/dl)
RDW (%)
2.5 (1.1) 3.4 (1.5)
8.6 (1.1) 8.7 (0.8)
41.2 (4.2) 42.5 (3.0)
13.6 (1.4) 14.3 (1.0)
48.2 (3.2) 49.0 (2.8)
15.8 (0.8) 16.5 (0.9)
32.9 (0.8) 33.8 (1.0)
14.3 (1.5) 13.5 (1.0)
2.7 (1.4) 2.2 (0.8)
9.4 (1.6) 8.0 (2.0)
44.4 (5.5) 40.4 (7.6)
14.6 (2.0) 13.1 (2.7)
47.9 (4.3) 51.7 (6.5)
15.7 (1.0) 16.6 (1.6)
32.8 (1.1) 32.2 (1.4)
14.2 (1.9) 15.4 (3.1)
3.0 (4.1) 3.1 (2.5)
7.4 (1.9) 7.5 (0.7)
37.0 (8.3) 40.1 (2.6)
12.4 (2.8) 13.0 (1.0)
50.3 (3.5) 53.6 (2.7)
16.8 (1.3) 17.3 (0.6)
33.5 (1.1) 32.3 (0.9)
16.3 (3.5) 16.2 (2.0)
1.6 (0.6) 2.2 (2.0)
7.8 (0.9) 6.8 (0.6)
40.4 (3.0) 37.9 (2.4)
13.5 (1.0) 12.5 (0.8)
52.0 (3.1) 55.5 (2.9)
17.3 (0.9) 18.3 (0.8)
33.3 (0.8) 32.9 (0.8)
15.0 (1.9) 15.0 (0.8)
2.2 (0.8) 2.3 (1.2)
8.3 (1.4) 7.9 (1.5)
38.7 (5.6) 38.5 (6.2)
12.9 (2.0) 12.5 (2.3)
47.1 (2.8) 49.2 (3.8)
15.6 (0.6) 16.0 (0.7)
33.2 (0.9) 32.5 (1.3)
13.6 (1.1) 15.2 (4.7)
3.1 (4.5) 2.8 (2.1)
9.2 (1.1) 8.2 (1.0)
42.3 (4.8) 40.1 (3.9)
14.1 (1.7) 13.2 (1.3)
46.2 (2.0) 48.8 (3.5)
15.4 (0.7) 16.1 (1.0)
33.4 (0.9) 33.0 (1.0)
14.9 (3.3) 13.8 (1.1)
EC, environmental control. VC, vehicle control.
mas were observed in group 3. Focal squamous epithelial dysplasia was noted in the vaginas of two group 3 and six group 4 animals and there was an increase in the incidence of moderate vaginal squamous epithelial hyperplasia in group 4. A vaginal squamous cell carcinoma was also seen in one group 5 animal. Such tumors, while rare, do occur spontaneously. Since this animal was given ZDV for only 90 days postweaning, and we have shown that ZDV-related vaginal tumors occur only after at least 19 months of daily oral dosing (Ayers et al., 1996), it is likely that this single occurrence was of spontaneous origin. As shown in Table 3, mean peak ZDV plasma concentrations at term in animals given 40 mg/kg were approximately 37 mg/ml. Mean ZDV levels in dams given 40 mg/kg on Lactation Day 21 were approximately 14 to 15 mg/ml. In pups given 40 mg/kg in the drinking water during the postweaning period, mean ZDV plasma concentrations ranged from approximately 0.04 to 0.14 mg/ml. For comparison, the average steady-state concentration in humans at the recommended daily dose (100 mg q 4 hr) is 0.62 mg/ml. DISCUSSION
Ayers et al. (1996) reported that ZDV, given orally or intravaginally, induced vaginal epithelial neoplasms in rodents in lifetime carcinogenicity bioassays. In a subsequent study, Olivero et al. (1994) showed a positive correlation between the dose of ZDV, proliferation of cells in the basal
AID
FAAT 2342
/
6k1f$$$$22
08-20-97 08:34:59
vaginal epithelium, and incorporation of ZDV into vaginal DNA. They speculated that these findings may play a role in the ability of ZDV to induce vaginal tumors in mice. At a meeting of a panel of experts at the National Institutes of Health on January 14, 1996, to discuss the transplacental carcinogenicity of ZDV (Rowe, 1997), scientists from the National Cancer Institute reported the preliminary results of
TABLE 3 Mean ({SD) Zidovudine (ZDV) Plasma Concentrations Group 3 4 5 6 3 3 4 4 5 5 3 4
damsa damsa damsa damsa male pupsb female pupsb male pupsb female pupsb male pupsb female pupsb malesc malesc
Dose (mg/kg/day) 20 40 40 40 20 20 40 40 40 40 20 40
ZDV (mg/ml) 12.6 15.3 14.1 15.0 0.09 0.04 0.10 0.14 0.10 0.07 14.5 36.6
{ { { { { { { { { { { {
Note. No data available for females. a 15 min postdose (Lactation Day 21). b 30 min after lights turned on (Postweaning Day 14). c 15 min postdose at scheduled termination of study (24 months).
ftoxa
0.8 2.9 1.3 1.2 0.04 0.01 0.10 0.13 0.06 0.04 2.0 1.8
198
AYERS, TORREY, AND REYNOLDS
a study in which daily po doses of 12.5 and 25 mg of ZDV (Ç1000 mg/kg nonpregnant body weight or Ç450 mg/kg of terminal body weight) were given to pregnant mice from Gestation Days 12 to 18 only (Olivero, O. A., Diwan, B. A., Anderson, L. M., Rice, J. M., Beland, F. A., Yuspa, S. H., and Poirier, M. C., personal communication). They demonstrated that ZDV was incorporated into DNA in various tissues and found high incidences of liver and lung tumors in male offspring and lung and reproductive organ tumors in female offspring 1 year after birth. Based on this evidence, they classified ZDV as a transplacental carcinogen. Also discussed at this meeting were the results of the transplacental carcinogenicity study described here in which lower doses of ZDV were used in a dosing regimen designed to mimic the clinical setting and to exaggerate human exposure levels based on peak plasma concentrations and areaunder-the-concentration time curve (AUC). The high dose used in this study (40 mg/kg/day) had previously been shown to be carcinogenic to the vaginal epithelium (Ayers et al., 1996) and to produce peak plasma concentrations 40-fold higher and an AUC 3-fold higher than the values for these parameters in humans (0.62 and 5.24 mg/ml 1 hr, respectively) at the recommended daily dose (100 mg q 4 hr). The mean peak plasma concentration seen at euthanization in this transplacental study (36.6 mg/ml) was approximately 59fold higher than the average steady-state concentration in humans at the recommended daily dose. The differences noted in peak plasma concentrations at 40 mg/kg/day in the dams and males at euthanization (Table 3) were probably related to the very brief peak (Tmax) that occurs after oral administration of ZDV in mice. Samples taken just before or after this peak would result in significant differences in concentration. This also explains the differences in peak concentration seen at 40 mg/kg/day in the previously reported study (Ayers et al., 1996) and this transplacental carcinogenicity study. In the current study, ZDV again induced tumors in the vagina, but not in other tissues or organs. These findings were virtually identical to those observed in the standard oral carcinogenicity study in mice (Ayers et al., 1996), in which the incidence rate of vaginal epithelial tumors in highdose females was 12% (7 tumors in 60 mice). In the current study, the incidence rate for these tumors at 40 mg/kg/day was 16% (11 tumors in 70 mice). Time of onset for the vaginal tumors was also not affected by lifetime treatment
AID
FAAT 2342
/
6k1f$$$$22
08-20-97 08:34:59
in this study. The earliest that a vaginal tumor was seen in the previously reported study was after 579 days (19 months) of daily dosing. In this transplacental carcinogenicity study the earliest vaginal tumor was seen after 694 days (23 months) of dosing. It was concluded, therefore, that lifetime oral treatment of mice with ZDV, beginning perinatally, did not alter the previously reported pattern of carcinogenicity in mice and that under the conditions tested, ZDV was not a transplacental carcinogen. REFERENCES Ali, M. W. (1990). Exact versus asymptotic tests of trend of tumor prevalence in tumorigenicity experiments. A comparison of p-values for small frequency of tumors. Drug Inform. J. 24, 727–737. Ayers, K. M., Clive, D., Tucker, W. E., Jr., Hajian, G., and de Miranda, P. (1996). Nonclinical toxicology studies with zidovudine: Genetic toxicity tests and carcinogenicity bioassays in mice and rats. Fundam. Appl. Toxicol. 32, 148–158. Connor, E. M., Sperling, R. S., Gelber, R., Kisalev, P., Scott, G., O’Sullivan, M. J., Van Dyke, R., Bey, M., Shearer, W., Jacobson, R. L., et al. (1995). Reduction of maternal–infant transmission of human immunodeficiency virus type I with zidovudine treatment. N. Engl. J. Med. 331, 1173– 1180. Dinse, G. E., and Lagakos, S. W. (1983). Regression analysis of tumor prevalence data. J. R. Stat. Soc. Ser. C (Appl. Stat.). 32, 236–241. Gart, J. J., Krewski, D., Lee, P. N., Tarone, R. E., and Wahrendorf, J. (1986). Statistical methods in cancer research. In The Design and Analysis of Long Term Animal Experiments. IARC Sci. Publ. No. 79, Vol. III, p. 219. Oxford Univ. Press, London. Olivero, O. A., Beland, F. A., Fullerton, N. F., and Poirier, M. C. (1994). Vaginal epithelial damage and expression of preneoplastic markers in mice during chronic dosing with tumorigenic levels of AZT. Cancer Res. 54, 6235–6242. Physicians’ Desk Reference (1996). 50th ed., pp. 1162–1163. Montrale, NJ. Rowe, P. M. (1996). US expert panel reaffirms benefit of perinatal zidovudine. Lancet 349, 258. Ruprecht, R. M., Sharpe, A. H., Jaenisch, R., and Trites, D. (1990). Analysis of 3*-azido-3*-deoxythymidine levels in tissues and milk by isocratic high-performance liquid chromatography. J. Chromatogr. 528, 371–383. Tadapalli, S., Puckett, L., Jeal, S., Kanics, L., and Quinn, R. (1990). Differential assay of zidovudine and its glucuronide metabolite in serum and urine with a radioimmunoassay kit. Clin. Chem. 36, 897–900. Thakur, A. K., Berry, K. J., and Mielke, P. W., Jr. (1985). A FORTRAN program for testing trend and homogeneity in proportions. Comput. Biomed. Res. 19, 229–233. Thomas, D. G., Breslow, N., and Gart, J. J. (1977). Trend and homogeneity analyses of proportions and life table data. Comput. Biomed. Res. 10, 373–381.
ftoxa