Registration Dossier

Administrative data

Endpoint:
toxicity to reproduction: other studies
Type of information:
experimental study
Adequacy of study:
supporting study
Reliability:
4 (not assignable)
Rationale for reliability incl. deficiencies:
secondary literature

Data source

Reference
Reference Type:
publication
Title:
Predictive models of prenatal developmental toxicity from ToxCast high-throughput screening data
Author:
Sipes N, Martin M, Reif D, Kleinstreuer N, Judson R, Singh A, Chandler K, Dix D, Kavlock R, and Knudsen T
Year:
2011
Bibliographic source:
TOXICOLOGICAL SCIENCES 124(1), 109–127 (2011)

Materials and methods

Test guideline
Qualifier:
no guideline followed
Principles of method if other than guideline:
To test the hypothesis that developmental toxicity in guideline animal studies captured in the ToxRefDB database would correlate with cell-based and cell-free in vitro high-throughput screening (HTS) data to reveal meaningful mechanistic relationships and provide models identifying chemicals with the potential to cause developmental toxicity. The work is intended to indicate the utility of HTS assays for developing pathway-level models predictive of developmental toxicity.

It evaluates a statistical model with inputs from regulatory data packages and in vitro bioactivity assays. Due to the nature of this study no new data specific to the toxicology of the test substance are presented
GLP compliance:
no
Type of method:
other: In silico

Test material

Reference
Name:
Unnamed
Type:
Constituent

Results and discussion

Any other information on results incl. tables

This study is the first attempt to construct predictive models of developmental toxicity based on broad spectrum profiling of biological activity in HTS assays. Results of this study demonstrate the following findings: (1) individual species-specific models are necessary for predicting developmental toxicity in pregnant rats and rabbits, (2) plausible cellular targets and pathways can be linked to specific endpoint toxicity, (3) toxicity endpoints cluster together based on similar biological process associations indicating potential similarities in developmental stage or processes, (4) xenobiotic metabolism plays a role in developmental toxicity, (5) there is no clear trend between in vivo chemical dose and assay characteristics, and (6) this analysis demonstrates the capability of using HTS assays to predict developmental toxicity. Taken together, these data indicate for the first time that ToxCast HTS of a large number of compounds can produce in vitro bioactivity profiles that can predict, with a BA of over 70%, in vivo developmental toxicity potential. 

The Phase I chemical library data set was run through each species model for ranking, and these rankings were then visualized as ToxPi. The model score for each chemical was based on the in vitro activity profile from the HTS assays of the selected composite assay set for each model. Higher scoring chemicals are predicted to be more likely developmentally toxic than chemicals with lower ranking scores.

Applicant's summary and conclusion

Conclusions:
This study is the first attempt to construct predictive models of developmental toxicity based on broad spectrum profiling of biological activity in high-throughput screening (HTS) assays. The generated data indicate for the first time that ToxCast HTS of a large number of compounds can produce in vitro bioactivity profiles that can predict, with a BA of over 70%, in vivo developmental toxicity potential.
Executive summary:

EPA’s ToxCast project is profiling the in vitro bioactivity of chemicals to assess pathway-level and cell-based signatures that correlate with observed in vivo toxicity. The developmental toxicity in guideline animal studies captured in the ToxRefDB database was hypothesised to correlate with cell-based and cell-free in vitro high-throughput screening (HTS) data to reveal meaningful mechanistic relationships and provide models identifying chemicals with the potential to cause developmental toxicity. To test this hypothesis, a statistical association was built based on HTS and in vivo developmental toxicity data from ToxRefDB. Univariate associations were used to filter HTS assays based on statistical correlation with distinct in vivo endpoint. This revealed 423 total associations with distinctly different patterns for rat and rabbit across multiple HTS assay platforms. From these associations, linear discriminant analysis with cross-validation was used to build the models. Species-specific models of predicted developmental toxicity revealed strong balanced accuracy (> 70%) and unique correlations between assay targets such as transforming growth factor beta, retinoic acid receptor, and G-protein– coupled receptor signalling in the rat and inflammatory signals, such as interleukins (IL) (IL1a and IL8) and chemokines (CCL2), in the rabbit. Species-specific toxicity endpoints were associated with one another through common Gene Ontology biological processes, such as cleft palate to urogenital defects through placenta and embryonic development. This work indicates the utility of HTS assays for developing pathway-level models predictive of developmental toxicity.

This study evaluates a statistical model with inputs from regulatory data packages and in vitro bioactivity assays.  Due to the nature of this study no new data specific to the toxicology of the test substance are presented.

The generated data indicate that ToxCast HTS of a large number of compounds can produce in vitro bioactivity profiles that can predict, with a BA of over 70%, in vivo developmental toxicity potential.