A Probability Analysis of Historical Pregnancy and Fetal Data from Dutch Belted and New Zealand White Rabbit Strains from Embryo-Fetal Development Studies.

Abstract

Embryo-fetal development (EFD) studies, typically in pregnant rats and rabbits, are conducted prior to enrolling females of reproductive age in clinical trials. Common rabbit strains used are the New Zealand White (NZW) and Dutch Belted (DB). As fetal abnormalities can occur in all groups, including controls, Historical Control Data (HCD) is compiled using data from control groups of EFD studies, and is used along with each study's concurrent control group to help determine whether fetal abnormalities are caused by the test article or are part of background incidences. A probability analysis was conducted on 2014 HCD collected at Charles River Inc., Horsham PA on Covance NZW, Covance DB, and Charles River (CR) NZW rabbits. The analysis was designed to determine the probability of 2 or 3 out of a group of 22 does aborting their litter or of having a fetal abnormality by chance. Results demonstrate that pregnancy parameters and fetal observations differ not only between strains, but between sources of rabbits of the same strain. As a result the probability of these observations occurring by chance in two or three litters was drastically different. Although no one single strain is perfect, this analysis highlights the need to appreciate the inherent differences in pregnancy and fetal abnormalities between strains, and points out that an apparent isolated increased incidence of an observation in one strain will not necessarily be test-article related in another strain. A robust HCD is critical for interpretation of EFD rabbit studies, regardless of the rabbit strain used.