Abstract
Recent technological advances in the field of big data have increased our capabilities to query large databases and combine information from different domains and disciplines. In the area of preclinical studies, initiatives like SEND (Standard for Exchange of Nonclinical Data) will also contribute to collect and present nonclinical data in a consistent manner and increase analytical possibilities. With facilitated access to preclinical data and improvements in analytical algorithms there will surely be an expectation for organisations to ensure all the historical data available to them is leveraged to build new hypotheses. These kinds of analyses may soon become as important as the animal studies themselves, in addition to being critical components to achieve objectives aligned with 3Rs. This article proposes the application of meta-analyses at large scale in corporate databases as a tool to exploit data from both preclinical studies and in vitro pharmacological activity assays to identify associations between targets and tissues that can be used as seeds for the development of causal hypotheses to characterise of targets. A total of 833 in-house preclinical toxicity studies relating to 416 compounds reported to be active (pXC50 ≥ 5.5) against a panel of 96 selected targets of interest for potential off-target non desired effects were meta-analysed, aggregating the data in tissue-target pairs. The primary outcome was the odds ratio (OR) of the number of animals with observed events (any morphology, any severity) in treated and control groups in the tissue analysed. This led to a total of 2139 meta-analyses producing a total of 364 statistically significant associations (random effects model), 121 after adjusting by multiple comparison bias. The results show the utility of the proposed approach to leverage historical corporate data and may offer a vehicle for researchers to share, aggregate and analyse their preclinical toxicological data in precompetitive environments.
Highlights
A critical part of the drug development process is the assessment of toxicity of a candidate at preclinical level in order to predict its safety profile
Histopathology animal data was extracted from historical preclinical toxicity studies available in GSK in-house database with records organised at observation level
A systematic review of literature available should be carried out to explore the biological plausibility of the association and gather the information needed to establish a causal explanation which could be represented in the form of an Adverse Outcome Pathway (AOP)
Summary
A critical part of the drug development process is the assessment of toxicity of a candidate at preclinical level in order to predict its safety profile. This is an area with high dependency on animal testing and still, despite the experimental effort, safety is a primary cause of attrition in the pharmaceutical industry. Androgen Receptor agonists) is made available as supporting information. This dataset includes the results of 38 preclinical studies, annotated by species, administration route and duration, with the counts of animals with events observed in control and treated groups by tissue analysed. The meta-analyses can be reproduced with the R code provided
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
