e13549 Background: Adverse drug reactions (ADRs) are a significant obstacle in cancer chemotherapy. In spite of the advances in the development of drugs for cancer therapy, ADRs compromise their potential beneficial effects in many patients by imposing reduction of dose or cessation of treatment. Prediction of susceptibility to ADRs in individual patients could significantly improve treatment outcomes by more effectively choosing which drugs to treat with and which dose is appropriate. The pharmacokinetics (PK) of many drugs has been extensively studied and numerous enzymes and transporters that are involved in their activation and processing have been identified. However, when their role in individual susceptibility to ADRs has been tested in numerous studies, the variants of these proteins have not been able to consistently predict specific ADRs. Therefore, additional studies are required for a reliable prediction of toxicity. Methods: We used mouse strains with precisely defined limited genomic differences to evaluate genetic control of ADRs to three drugs commonly used in cancer chemotherapy - irinotecan, gemcitabine, and doxorubicin. We compared strain distribution of susceptibility to ADRs caused by each of these drugs and tested whether there is a correlation with genotypes at the major processing/transport (PK) genes that were reported in humans to affect toxicity of these drugs. Results: The major genetic differences in toxic reactions caused by these drugs do not correlate with the major PK loci and that the strain distribution pattern of susceptibility vs. resistance for each drug is different. Conclusions: This data suggests that a significant part of genetic susceptibility to ADRs is controlled by genes other than the presently known PK-related genes, and that many responsible genes are drug specific, with some possible overlaps. We hypothesize that these genes are likely involved in downstream pharmacodynamic (PD) processes and have remained largely unknown, because most studies in humans have been limited to known PK-related genes. We are proceeding towards identification of these novel genes, as they could help to optimize the selection of therapy for individual patients.