Replicability in EFDT Studies: Implications of Context Dependence for Virtual Control Groups (VCGs).

Area of Science:

• Toxicology
• Developmental Biology
• Reproductive Toxicology

Background:

• Reliable interpretation of toxicology studies requires reproducibility.
• Virtual Control Groups (VCGs) are discussed for nonclinical toxicology, emphasizing variable matching.
• Biological implications of inter-study variability are underexplored, especially in embryo-fetal developmental toxicity (EFDT) studies.

Purpose of the Study:

• To identify and organize sources of non-treatment variability in EFDT studies.
• To explore the biological implications of this variability on study outcomes.
• To assess the feasibility of constructing reliable VCGs and using historical control data (HCD).

Main Methods:

• Literature review and analysis of previous EFDT variability studies.
• Identification and categorization of 69 potential sources of non-treatment variability.
• Organization of variables into five domains: animals, pre-study practices, microenvironment, macroenvironment, and study design/procedures.

Main Results:

• Numerous intrinsic and extrinsic variables significantly influence EFDT pregnancy outcomes and fetal development.
• Variability sources differ in controllability and standardization potential across laboratories.
• EFDT outcomes are context-dependent, influenced by both treatment and study environment.
• Strict replicability of control-group outcomes across studies is biologically improbable due to inherent variability.

Conclusions:

• Inherent variability in EFDT studies makes strict control-group outcome replicability biologically improbable.
• This variability impacts expectations for study concordance and the construction of VCGs and HCD.
• Developing reliable VCGs is most feasible within individual laboratories with aligned variability domains.