RESUMO
Patient tumor explants established subcutaneously in serial passage in nude mice were characterized for their sensitivity towards 11 standard cytotoxic anti-cancer agents. The latter include the alkylating agents cyclophosphamide, ifosfamide, mitomycin C, cisplatin and CCNU, the antimetabolites 5-FU and methotrexate; the topoisomerase II inhibitors adriamycin and etoposide, as well as the tubulin binders paclitaxel and vindesine. The mean number of tumors treated with any of the various drugs was 54 (range 31-78). The tumor xenografts' gene expression profiles were determined using the Affymetrix HG-U133 plus 2.0 mRNA expression array representing approximately 38.500 human genes. The hypothesis was that the correlation of drug response to gene expression would identify gene signatures that can predict the drug response of individual tumors to these agents. Predictive gene signatures were found and subsequently verified using the leave-one-out cross-validation (LOOCV) technique. Tumors were considered as responsive if the drugs effected a tumor volume inhibition to less than 11-41% of the volume of vehicle control tumors (T/C%). The median cut-off over all drugs was a T/C of 25%. Using these criteria, on average one third of the test tumors were sensitive (responders) and two thirds were resistant (non-responders). The bio-informatic analysis yielded predictive gene signatures consisting of 42-129 genes (mean for the 11 drugs: 87 genes). On average, the response rate for predicted responders (83%) was 2.45 fold higher than that for all test tumors (random testing, 34%). This increase of response rates, following signature-guided testing, was consistent for all 11 agents. Conversely, 94% of the predicted non-responders (range: 84-100%) proved to be non-responders in nude mouse studies while the proportion of non-responders among all test tumors was approximately 66%. The majority of genes (59%) making up the predictive gene signatures had an unknown function. Known genes were implicated in cell proliferation, apoptosis, DNA repair, cell cycle, metabolism and transcription. The predictive gene signatures presented here for 11 cytotoxic agents have the potential, if employed in the clinic, to substantially increase tumor response rates compared to empirical drug treatment. However they need to be further validated, both preclinically and clinically.
