MARCAR aims to establish reliable early markers and molecular classification of tumors in non genotoxic carcinogenesis, applying a mechanism-based approach. Benefits are improved drug safety, more efficient drug development, and progress with 3 R issues. MARCAR focuses on rodent liver, the major target organ of non-genotoxic carcinogens (NGC). However, this approach will facilitate predictions for other organs. Innovative industry-relevant experimental models will be developed: (i) Transgenic mice nulled orhumanized for NGC-responsive nuclear receptors, (ii) transgenic mice excreting marker proteins in urine, including oxidative stress reporter mice to monitor ROS production, (iii) transgenic mice allowing in vivo bioirnaging of preneoplastic lesions (PNL) and tumors, and (iv) primary cultures of human/rodent hepatocytes and their co-cultivation with mesenchymal cells which modulate carcinogenesis by NGC. In these systems, novel molecular technologies will be used for profiling thegenome, epigenome, transcriptome and proteome/phosphoproteome to provide global molecularsignatures of NGC-induced signalling. These complementary technologies in conjunction withstandardized data management and advanced marker identification algorithms will lead to molecular classification of tumors and PNL, and identify predictive molecular signatures of spontaneous vs drug induced tumors. Biomarker panels indicating the mechanisms of GC action will be derived and translated into assay systems. The validity, predictivity and robustness of biornarkers will be tested in collaboration with EFPIA, employing well—characterized pre—climcal models. For clinical translation humanized/human systems and materials from clinical trials will be tested for NGC effects on homologues of rodent biomarkers. The provision of high-quality deliverables is facilitated by complementary expertise in carcinogenesis/bioanalysis, in conjunction with know-how and in kindcontributions of EFPIA partners.