In silico prediction of genotoxicity and carcinogenicity
Client
A global manufacturer of consumer products.
Background
As part of their overall regulatory requirements, the client was asked to provide an assessment of the known or predicted genotoxicity and carcinogenicity of a number of ingredients in their products. Bibra were provided with a list of several substances and asked to carry out this work following the genetic toxicology in silico (GIST) protocol. This formed part of a broader programme of work that we conducted for this client, in which we went on to investigate the potential metabolic transformation products of these ingredients, and the potential toxicity of those metabolites.
Project goals
To evaluate a large group of substances, following the GIST protocol established by Hasselgren et al. (2019). This protocol was developed through a consortium of toxicologists (involving Pete Watts, one of our Toxicology Directors), computational scientists, and regulatory scientists across several industries. Following this protocol allowed us to combine the genotoxicity and carcinogenicity data from expert group reviews, the peer-reviewed literature, and (Q)SAR predictions from two reliable software packages to reach an overall weight-of-evidence conclusion.
Approach and outcome
Following the GIST protocol, we conducted searches in a wide range of sources, including our in-house database, TRACE, to identify expert group reviews or peer-reviewed publications containing relevant data on the genotoxicity and carcinogenicity of the ingredients of interest. These data were summarised and scored for their reliability and our confidence in the result.
(Q)SAR predictions for genotoxicity and carcinogenicity were then generated using Leadscope and Derek Nexus. Models (statistical and rule-based) were selected to predict bacterial and mammalian mutagenicity, in vitro and in vivo chromosome damage, and carcinogenicity in mice and rats. The results were added to the same GIST framework, and again scored for reliability and confidence.
The final part of the process saw the overall scoring for each genotoxicity and carcinogenicity endpoint, considering both literature and in silico findings. This then allowed for a final conclusion to be made on the genotoxic and carcinogenic potential of these substances.
