A SIMULATION STUDY OF THE INFLUENCE OF STUDY DESIGN ON THE ESTIMATIONOF BENCHMARK DOSES FOR DEVELOPMENTAL TOXICITY

The benchmark dose (BMD)(4) approach is emerging as replacement to det ermination of the No Observed Adverse Effect Level (NOAEL) in noncance r risk assessment. This possibility raises the issue as to whether cur rent study designs for endpoints such as developmental toxicity, optim ized for detecting pair wise comparisons, could be improved for the pu rpose of calculating BMDs. In this paper, we examine various aspects o f study design (number of dose groups, dose spacing, dose placement, a nd sample size per dose group) on BMDs for two endpoints of developmen tal toxicity (the incidence of abnormalities and of reduced fetal weig ht). Design performance was judged by the mean-squared error (reflecti ve of the variance and bias) of the maximum likelihood estimate (MLE) from the log-logistic model of the 5% added risk level (the likely tar get risk for a benchmark calculation), as well as by the length of its 95% confidence interval (the lower value of which is the BMD). We fou nd that of the designs evaluated, the best results were obtained when two dose levels had response rates above the background level, one of which was near the ED(05), were present. This situation is more likely to occur with more, rather than fewer dose levels per experiment. In this instance, there was virtually no advantage in increasing the samp le size from 10 to 20 litters per dose group. If neither of the two do se groups with response rates above the background level was near the ED(05), satisfactory results were also obtained, but the BMDs tended t o be more conservative (i.e., lower). If only one dose level with a re sponse rate above the background level was present, and it was near th e ED(05), reasonable results for the MLE and BMD were obtained, but he re we observed benefits of larger dose group sizes. The poorest result s were obtained when only a single group with an elevated response rat e was present, and the response rate was much greater than the ED(05). The results indicate that while the benchmark dose approach is readil y applicable to the standard study designs and generally observed dose -responses in developmental assays, some minor design modifications wo uld increase the accuracy and precision of the BMD.