2',3'-Dideoxyinosine and 2',3'-didehydro-3'-deoxythymidine are nucleoside analogue reverse transcriptase inhibitors used in the management of human immunodeficiency virus infection. This study examined the potential developmental toxicity of ddI, d4T and ddI/d4T combinations administered orally to timed-mated mice during embryo/fetal development. Drugs were administered alone or in combination at ~50, 200 or 400 times their human therapeutic doses.
Timed-mated CD-1 mice (24/group) were dosed by gavage with vehicle control (Maalox TC), ddI (360, 1440 or 2880 mg/kg/day), d4T (60, 240, or 480 mg/kg/day), or ddI/d4T (360/60, 1440/240 or 2880/480 mg/kg/day) on gestational days 6 through 16. The experimental design represented a ten-group subset of a 4 X 4 factorial performed in two replicates.
Total daily doses were equally divided (morning and afternoon) and given at least 6 hours apart. Dose volume was 10 ml/kg body weight at each administration for a total daily dose volume of 20 ml/kg/day. The oral route corresponds to a commonly used route in human patients.
Timed-mated mice were monitored at regular intervals for clinical signs of toxicity, feed consumption, and body weight At necropsy (gd 17), clinical signs, maternal body weight, liver weight, gravid uterine weight, pregnancy status and number of corpora lutea were recorded. In each gravid uterus, the numbers of prenatal deaths (resorptions and/or dead fetuses) and live fetuses were recorded. Live fetuses were weighed, sexed, and examined for morphological anomalies (external, visceral, and skeletal).
Data were analyzed for differences among all ten treatment groups and for the following interactions: DDI X D4T (the overall drug interaction), DDI X D4T X REP, and COMBO X REP where REP is the replicate variable and COMBO is the treatment variable for drug combinations. Significant replicate interactions resulted in analysis of data within each replicate. Significant overall and pairwise drug interactions resulted in evaluation of individual ddI/d4T groups to determine the nature of the interaction (synergism, potentiation or antagonism). Experimental groups were also divided into three blocks of four treatment groups (i.e., vehicle control plus low, mid, and high doses of ddI, d4T or ddI/d4T). Each 4-group block was analyzed for dose-response trends and differences among groups. When a group-wise comparison was significant, individual groups within that block were compared to the vehicle control group. In addition, significance of the whole model (drug main effects plus interaction) resulted in comparison of effects for each ddI/d4T drug combination to the corresponding dosages of ddI alone and d4T alone.
There was no maternal morbidity or mortality, and clinical signs were generally unremarkable. At necropsy (gd 17), 22-24 females per group were confirmed pregnant (92-100% per group). There were no significant effects on gravid uterine weight, maternal body weight or weight change with the following exceptions. From gd 9 to 12, maternal body weight change showed an increasing trend across the drug combination groups. Maternal corrected weight change (i.e., gestational weight change minus gravid uterine weight) differed among groups for the analysis of d4T effects. Significant increases occurred at 60 and 480 mg d4T/kg/day, but in the absence of a dose-related trend.
Maternal liver weights (absolute and relative) showed increasing dose-related trends following exposure to d4T alone or ddI/d4T combinations. Absolute liver weight was significantly elevated above controls at 480 mg d4T/kg/day and at the high-dose combination. Relative liver weight was elevated above controls at 240 and 480 mg d4T/kg/day, and at the high-dose combination. Absolute (high-dose combination) and relative maternal liver weights (mid- and high-dose combinations) were significantly increased relative to the constituent doses of ddI. These results are consistent with previously reported reversible hepatic hypertrophy in rats and monkeys following repeated oral exposure to d4T, possibly an adaptive response (Schilling et al., 1995). Maternal liver weights were not affected by ddI alone and there were no drug interactions.
Maternal relative feed consumption (g of feed/kg of body weight/day) did not differ among groups during pre-treatment (gd 0 to 6). For the treatment period (gd 6 to 17), increasing dose-related trends were noted for ddI, d4T and ddI/d4T combinations. Significant differences among groups were noted for ddI and ddI/d4T combinations, but not for d4T. Feed consumption was significantly higher than controls at 1440 and 2880 mg ddI/kg/day, and at the high-dose combination. In relation to maternal body weight change (gd 9 to 12, see above), it is worth noting that maternal relative feed consumption (gd 9 to 12) showed increasing trends that approached significance for ddI, d4T and ddI/d4T combinations. Otherwise, maternal relative feed consumption was unaffected.
There were no significant main effects or interactions for the number of corpora lutea per dam, number of implantation sites per dam or percent pre-implantation loss per dam.
Treatment was not associated with increased incidences of resorptions, late fetal deaths, non-live implantation sites (i.e., resorptions plus late fetal deaths) or adversely affected implantation sites (i.e., nonlive plus malformed). For the percent of litters with 100% nonlive implantation sites (full litter loss), some comparisons were statistically significant, but the pattern of findings lacked biological relevance. Full litter loss was noted only in the control group (2/22 litters) and mid-dose combination group (1/23 litters), so that the incidence within each treated group was always lower than the concurrent control group.
Among live litters, the number of live fetuses per dam and percent male fetuses were unaffected. Fetal body weight was not affected by ddI or d4T, but a significant decreasing trend was noted across the ddI/d4T combinations. Low, mid and high combinations were 98.6%, 93.9% and 93.9% of control, respectively. There were no drug interactions.
Incidences of fetal malformations were not adversely affected by exposure to ddI, d4T, or ddI/d4T, and there was no evidence for drug interactions. For percent litters with visceral malformations, the overall effect (i.e., difference among groups) was significant for d4T. Notably, one fetus from one litter had hydrocephaly in the low d4T group (60 mg d4T/kg/day), but no other visceral malformations were found in this study. Thus, there were no significant adverse effects on the percent malformed fetuses or percent litters with malformed fetuses analyzed by general classes of malformations (external, visceral, or skeletal) or for all types combined.
Similarly, there was no clear evidence for a treatment-related increase in the incidence of fetuses with morphological variations. Percent fetuses with skeletal variations showed a significant decreasing trend for ddI, due to a lower incidence of rib anomalies. Percent fetuses with external variations had a significant interaction (COMBO X REP), apparently due to group-wise differences and an elevated incidence at the mid-dose combination only in REP II.
The percent litters with any type of variation was unaffected. The percent litters with external variations exhibited significant replicate interactions (DDI X D4T X REP and COMBO X REP). In REP II, there was a significant group-wise comparison for d4T and an increase at 240 mg d4T/kg/day. Also in REP II, there was a significant group-wise comparison among the ddI/d4T combinations and an increase for the mid-dose combination. There were no clear dose-response patterns in either replicate or for both combined. Thus, the significant replicate interaction terms were due to spurious distribution of low incidence findings, (open eye and hematoma) across replicates. There was no evidence for overall or pairwise drug interactions in either replicate or for both replicates combined. For percent litters with visceral variations, the overall DDI X D4T drug interaction term was significant, as were pairwise drug interactions at the mid and high-dose combinations. Findings were limited to enlarged lateral ventricles of the brain and bilateral distended ureters. Differences from control at the mid- and high dose combinations were not significant and smaller than predicted by the additive model (antagonism). The biological relevance of this interaction is questionable considering the absence of significant main effects or trends. The percent litters with skeletal variations showed a significant main effect for d4T and a decreased incidence only at 60 mg d4T/kg/day.
In summary, all treatments were well tolerated by the dams. Biological activity was demonstrated by increased maternal liver weight following exposure to d4T or ddI/d4T combinations, consistent with previous studies of d4T. Otherwise, there were no well-defined drug effects or drug interactions for maternal or developmental toxicity endpoints at doses up to ~400 times the therapeutic dose.