Data quality in predictive toxicology: reproducibility of rodent carcinogenicity experiments.

Abstract

We compared 121 replicate rodent carcinogenicity assays from the two parts (National Cancer Institute/National Toxicology Program and literature) of the Carcinogenic Potency Database (CPDB) to estimate the reliability of these experiments. We estimated a concordance of 57% between the overall rodent carcinogenicity classifications from both sources. This value did not improve substantially when additional biologic information (species, sex, strain, target organs) was considered. These results indicate that rodent carcinogenicity assays are much less reproducible than previously expected, an effect that should be considered in the development of structure-activity relationship models and the risk assessment process.

Consensus report on the future of animal-free systemic toxicity testing

“…At present, the carcinogenicity hazard of chemicals is determined by a costly and lengthy animal test, the "cancer bioassay", although its relevance for human health is seriously doubted (Alden et al, 1996;Knight, 2007;Gottmann et al, 2001). Results of more than 3.500 cancer bioassays, which cost about € 800,000 per substance and species, are publically available: 53% of all substances tested were positive, suggesting an enormous falsepositive rate, but still some accepted human carcinogens were Figures 4 and 6, to include several specific measures and research lines to be followed.…”

Uncertainties of testing methods: What do we (want to) know about carcinogenicity?

“…In addition to these practical needs, a critical mass of concern regarding the scientific uncertainties of animal test results and their regulatory utility also has been steadily accumulating in recent years (e.g., Basketter et al, 2012;Paparella et al, 2013;Hartung, 2013;Leist et al, 2014;NAS, 2015). Amongst these publications, a poor reproducibility of 57% for the 2-year RCB (Gottmann et al, 2001), and for pharmaceuticals in humans, risk assessment context, the focus was on the distinction of pure uncertainties, which may be reduced with further knowledge versus variability of results that cannot be reduced with further knowledge (WHO, 2014;ECHA, 2012). In the context of risk governance, the term "complexity" is used to describe effects triggered by many causes, inter-related like a network -and "ambiguity" is used for uncertainty stemming from the plurality of scientifically legitimate viewpoints (Renn et al, 2011;IRGC, 2005).…”

“…A value of ~ 57% was published by Gottmann et al (2001) based on the carcinogenic potency database, which contains two components: the National Cancer Institute / National Toxicology Program (NCI/NTP) database and the literature database. For 121 chemicals carcinogenicity studies were available in both components; for each of these chemicals, one assessment was carried out based on the studies in the NCI/NTP part and another assessment based on the studies in the literature database; a substance was considered positive if a positive result was obtained in at least one experiment.…”

Current standing and future prospects for the technologies proposed to transform toxicity testing in the 21st century

“…Due to species differences, humans and animals respond differently to chemical exposures, pointing to the need for uncertainty factors ("safety factors") during risk assessment, factors that can vary from 10-fold up to 10,000-fold (NRC, 2000). Between animal species (rat, mouse, rabbit), toxicity studies show an agreement of about 53-60% (Gottmann et al, 2001;Schardein et al, 1985). So far only a few studies have systematically reviewed the accuracy of animal models for the prediction of human health effects, perhaps because animal studies show a rather limited predictive capacity for human risk (Hartung and Daston, 2009).…”