Evaluation of non-Animal methods for assessing skin Sensitisation hazard: A Bayesian value-of-information analysis

Maria Leontaridou*, Silke Gabbert, Ekko C. van Ierland, Andrew P. Worth, Robert Landsiedel

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

3 Citations (Scopus)

Abstract

This paper offers a Bayesian Value-of-Information (VOI) analysis for guiding the development of non-Animal testing strategies, balancing information gains from testing with the expected social gains and costs from the adoption of regulatory decisions. Testing is assumed to have value, if, and only if, the information revealed from testing triggers a welfare-improving decision on the use (or non-use) of a substance. As an illustration, our VOI model is applied to a set of five individual non-Animal prediction methods used for skin sensitisation hazard assessment, seven battery combinations of these methods, and 236 sequential 2-Test and 3-Test strategies. Their expected values are quantified and compared to the expected value of the local lymph node assay (LLNA) as the animal method. We find that battery and sequential combinations of non-Animal prediction methods reveal a significantly higher expected value than the LLNA. This holds for the entire range of prior beliefs. Furthermore, our results illustrate that the testing strategy with the highest expected value does not necessarily have to follow the order of key events in the sensitisation adverse outcome pathway (AOP).

Original languageEnglish
Pages (from-to)255-269
JournalATLA-Alternatives To Laboratory Animals
Volume44
Issue number3
Publication statusPublished - 2016

Keywords

  • Animal Testing
  • Bayesian Value-Of-Information Analysis
  • Non-Animal Methods
  • Replacement
  • Sequential Testing Strategies
  • Skin Sensitisation

Fingerprint Dive into the research topics of 'Evaluation of non-Animal methods for assessing skin Sensitisation hazard: A Bayesian value-of-information analysis'. Together they form a unique fingerprint.

Cite this