Enhancing reliability of embryo-fetal developmental toxicity studies: A proposed design of replicate studies.

Abstract

Background: This report addresses the reliability of results from rat Embryo-Fetal Developmental Toxicity (EFDT) studies. Recent literature discusses the roles of reproducibility, replicability, and other influences on scientific reliability. Reproducibility is a re-analysis of the original data, while replicability addresses the same question with a separate study of some type. Concordance of rat and rabbit studies has been addressed previously, but replication of single-species EFDT studies was not found in the literature. A modest modification of the rat study is therefore proposed to assess replicability and possibly enhance reliability.

Methods: Regulatory guidelines were consulted and relevant literature was identified through online searches.

Results: Each replicate EFDT (r-EFDT) study in rats would consist of half the mated females of the definitive study. Studies would start at the same or different times in one testing facility. Separate shipments of animals (non-littermates) are required. All other procedures would be protocol-driven. The micro- and macro-environments of the animals would be held as constant as possible. Justification, design options, and interpretation methods are discussed.

Conclusion: Besides adding reliability, other benefits include reduced animal usage, and potentially reduced cost and time to final reports. By reducing the need for repeated studies due to questionable results, this modified study is viewed as a more efficient use of costly resources. The r-EFDT study design could easily be adapted to assess replicability of rabbit EFDT and some general toxicity studies. Future replicate studies are needed to critically evaluate replicability and the overall impact on study reliability.