Molecular determinants of the cytotoxicity of platinum compounds: the contribution of in silico research.

Gene expression profiling of tumors allows the establishment of relationships between gene expression profiles and sensitivity to anticancer drugs. In an attempt to study the molecular determinants of the activity of platinum compounds, we explored the publicly available databases of the National Cancer Institute (NCI; http://dtp.nci.nih.gov), which allow access to the gene expression profiles of the 60 cell lines for which drug cytotoxicity patterns already existed. Using this database, we have conducted an in silico research to identify the genes the expression of which was positively or negatively correlated to the sensitivity to four platinum compounds (cisplatin, carboplatin, oxaliplatin and tetraplatin). Important similarities were noticed between cisplatin and carboplatin on one hand, and tetraplatin and oxaliplatin on the other hand. In the restricted panel of 1416 genes and molecular markers, we identified 204 markers, among which 120 corresponded to identified genes, that significantly correlated (P < 0.001) with the cytotoxicity of at least one platinum compound. For example, the functionality of the p53-activated pathway appeared positively correlated with the cytotoxicity of all platinum compounds. More specific are the positive correlations between RAS gene mutations and MYC expression and the cellular sensitivity to oxaliplatin. Among the parameters already known as related to the sensitivity to platinum compounds, we identified, in the complete set of 9400 genes, numerous significant relationships, such as the negative correlations between ERB-B2 and BCL-X(L) expressions and the cytotoxicity of the platinum compounds. Public databases mining, therefore, appears to be a valuable tool for the identification of determinants of anticancer drug activity in tumors.