The paradigm of individualized drug therapy based on an individual's unique genetic make-up is especially desirable in the field of hematology/oncology, where the therapeutic index is often narrow, and the consequences of chemotherapy drug toxicity can be life-threatening. If clinicians could better predict which individuals are at the greatest risk of suffering chemotherapy-related toxicities, this could greatly impact the overall care of cancer patients. For cancer drugs, development of toxicities is not universally predictable among all individuals being treated. Evidence suggests that genetic differences between individuals, and between populations, might explain a significant portion of this heterogeneity. For example, Asians have significantly higher rates of platinum chemotherapy-related toxicities compared to other populations. This observation suggests that genetic polymorphisms unique to, or present at higher rates in, Asians might explain this difference. If one could identify "platinum toxicity susceptibility polymorphisms" in populations and individuals, prediction of platinum toxicity risk based on genotype might be possible. Hypothesis: Germline genetic polymorphisms are critical determinants of platinum chemotherapy-related toxicity. We have developed an experimental model employing EBV-transformed B-lymphoblastoid cell lines (LCLs) from healthy individuals of different ethnic backgrounds that permits elucidation of genetic determinants of chemotherapy-related toxicity susceptibility via an unbiased genome-wide approach. In this proposal, we apply this model toward discovery of germline single nucleotide polymorphisms (SNPs) governing toxicity susceptibility to two platinum chemotherapies, cisplatin and carboplatin. We will first identify SNPs which associate with in vitro cytotoxicity in LCLs from Asian individuals, since Asians represent an "enriched toxicity phenotype" population. Then, using SNPs identified in Asians, in conjunction with genotype information from Caucasians and Africans, we will identify SNPs existing "across-populations" which may predict platinum toxicity susceptibility in any individual, regardless of ethnicity. "Cross population" SNPs will then be subjected to validation in our second aim, which establishes a clinical trial enrolling cisplatin-treated patients to determine whether the pre-clinically derived "cisplatin toxicity" SNPs predictably associate with development of cisplatin-related nephrotoxicity. The trial will also permit identification of additional, novel, "cisplatin nephrotoxicity" SNPs through a secondary genome-wide association analysis of patient samples. This information could lead to more judicious use of chemotherapy in cancer patients, potentially allowing chemotherapies to be chosen and/or dosed based on an individual's personal risk of developing side- effects, balanced against the potential benefits. On a population scale, results of this work have the potential to reduce the significant burden of illness that can be conferred by chemotherapy. For the trainee, the proposal represents a comprehensive learning experience in both basic pharmacogenetic discovery and clinical trial design.

F32CA136123

2010-01-01

2010-06-30