CARCINOscreen®: New short-term prediction method for hepatocarcinogenicity of chemicals based on hepatic transcript profiling in rats

Abstract

-Carcinogenicity is one of the most serious toxic effects of chemicals, and highly accurate methods for predicting carcinogens are strongly desired for human health. Here, we developed a new prediction system named "CARCINOscreen®" for evaluating the carcinogenic potentials of chemicals using the gene expression profiles of liver tissues from rats after a 28-day repeated dose toxicity study. The prediction formula was generated using a support vector machine with predictive genes selected from 68 training chemical datasets; a predictive score was then calculated to predict the carcinogenic potentials of the tested chemicals. To ensure the accuracy of the prediction system, the chemicals were divided into three groups (Groups 1 to 3) according to the resulting hepatic gene expression profiles, and a prediction formula was generated for each group. The prediction system was capable of predicting the carcinogenicity of training carcinogens and non-carcinogens with an accuracy of 92.9% to 100%. The final prediction result was determined based on the maximum prediction value obtained with three independent prediction formulas to build up the CARCINOscreen®. The system was able to predict carcinogenicity accurately in 94.1% of the 68 training chemicals. An external validation trial was performed with 16 chemicals, consisting of various carcinogens targeting rat liver or other organs and non-carcinogens. The system identified 68.8% of all the chemicals and 100% of the rat liver carcinogens as carcinogens. Thus, the CARCINOscreen®, a novel system for predicting hepatocarcinogenicity, is a promising tool for the prediction of rat liver carcinogens.

Simpler alternative to CARCINOscreen^® based on quantitative PCR (qPCR)

“…A total of 51 chemicals, including 29 for trainingphase trials and 22 for validation-phase trials, were selected from among the chemicals used in previous studies (Matsumoto et al, 2009(Matsumoto et al, , 2014. The names of the chemicals and their toxicological profiles are shown in Table 1 and Table 2.…”

“…Among them, animal experiments and CARCINOscreen ® prediction for all chemicals in training-phase and two chemicals in validation-phase, 4-aminoazobenzene and o-nitrotoluene, were not conducted for this study but these were described in a previous report (Matsumoto et al, 2014).…”

“…Microarray analysis and prediction using the standard CARCINOscreen ® Microarray analysis was performed as described previously (Matsumoto et al, 2014). Briefly, three types of custom arrays, Toxarray III ver.2 and Agilent Whole Rat Genome Microarrays 8x60K Toxplus ver.1 and ver.2, and the gene-expression-based carcinogenicity prediction system, CARCINOscreen ® , were applied to the data obtained in this study (Matsumoto et al, 2014).…”

“…This system consists of three prediction formulae generated from 15 characteristically predictive genes. The accuracy of this microarray-based CARCINOscreen ® for carcinogenicity prediction is reported as 94.1% for training chemical sets, therefore, this prediction system may be a promising tool for predicting hepatocarcinogenicity of chemicals in rats (Matsumoto et al, 2009(Matsumoto et al, , 2011(Matsumoto et al, and 2014. However, microarray analysis requires specialized equipment and skilled bioinformatics approaches for data analysis, limiting its versatility for the toxicological applications.…”

“…For this purpose, we tried to select the minimal gene set by using microarray data obtained with F344 rat previously reported (Matsumoto et al, 2014) and verify the performance of the selected gene sets by qPCR based trial as the training phase. Then the qPCR-based system was evaluated for its performance and an applicability to widely-used rat strain (SD rats) by the chemicals with no prediction results of the microarray based CARCINOscreen ® as the validation-phase.…”