[Microarray analysis of tumor xenograft model].

F. Fujita,M. Shirane,K. Mori,M. Koike,M. Fujita
2010-08-01
Abstract:We carried out gene expression profiling of forty human tumor cells for research choice method of the most fitting anticancer drug, using unsupervised hierarchal clustering analysis. This clustering analysis is based on a tumor growth inhibition panel of nine antitumor drugs (MMC, CDDP, ACNU, CPT-11, CPA, FT-207, UFT, 5'-DFUR and ADM) for forty human cancers. These cancers(eleven stomach, seven colon, six breast, three pancreas, five lung, two esophageal carcinomas, one liver, one renal cell carcinoma, one uterus, two ovarian, and one melanoma) have been maintained by serial s. c. passages in nude mice of the same sex of donor patients. Nine antitumor drugs were divided into two groups, a 5-FU-related drug group (5'-DFUR, FT-207 and UFT) and another group. On the other hands, forty cells were clustered into four groups. By using GeneChip (Hu95Av2, Affymetrix), we investigated gene expression profiling of the matched tumor cells and selected specific genes in each group. Interestingly, a pathway analysis revealed that expressions of p53-related genes were up-regulated in the 5-FU-sensitive groups. This result suggested that chemosensitivity was predicted by gene expression profiling of tumor cells. We considered that microarray analysis would be a good tool for further tailor-made medications.
What problem does this paper attempt to address?