Many genes and signalling pathways controlling cell proliferation, death and differentiation, as well as genomic integrity, are involved in cancer development. New techniques, such as serial analysis of gene expression and cDNA microarrays, have enabled measurement of the expression of thousands of genes in a single experiment, revealing many new, potentially important cancer genes. These genome screening tools can comprehensively survey one tumor at a time; however, analysis of hundreds of specimens from patients in different stages of disease is needed to establish the diagnostic, prognostic and therapeutic importance of each of the emerging cancer gene candidates. Here we have developed an array-based high-throughput technique that facilitates gene expression and copy number surveys of very large numbers of tumors. As many as 1000 cylindrical tissue biopsies from individual tumors can be distributed in a single tumor tissue microarray. Sections of the microarray provide targets for parallel in situ detection of DNA, RNA and protein targets in each specimen on the array, and consecutive sections allow the rapid analysis of hundreds of molecular markers in the same set of specimens. Our detection of six gene amplifications as well as p53 and estrogen receptor expression in breast cancer demonstrates the power of this technique for defining new subgroups of tumors.

Topoisomerase IIalpha (topoIIalpha) is a key enzyme in DNA replication and a molecular target for many anti-cancer drugs called topoII inhibitors. The topoIIalpha gene is located at chromosome band 17q12-q21, close to the ErbB-2 oncogene (HER-2/neu), which is the most commonly amplified oncogene in breast cancer. Because of the physical proximity to ErbB-2, copy number aberrations may also occur in the topoIIalpha gene. These topoIIalpha gene copy number aberrations may be related to the altered chemosensitivity to topoII inhibitors that breast cancers with ErbB-2 amplification are known to have. We used fluorescence in situ hybridization to study copy number aberrations of both topoIIalpha and ErbB-2 in nine breast cancer cell lines and in 97 clinical breast tumors, which were selected for the study according to their ErbB-2 status by Southern blotting. TopoIIalpha-protein expression was studied with Western blot and sensitivity to doxorubicin (a topoII inhibitor) with a 96-well clonogenic in vitro assay. Two of the five cell lines with ErbB-2 gene amplification (SK-BR-3 and UACC-812) showed amplification of topoIIalpha. In MDA-361 cells, ErbB-2 amplification (14 copies/cell) was associated with a physical deletion of topoIIalpha (four copies of chromosome 17 centromere and two copies of topoIIalpha). The topoIIalpha amplification in UACC-812 cells was associated with 5.9-fold-increased topoIIalpha protein expression and 2.5-fold-increased sensitivity to the topoII inhibitor, doxorubicin, whereas the deletion in MDA-361 leads to decreased protein expression (45% of control) and a 2.4-fold-increased chemoresistance in vitro. Of 57 ErbB-2-amplified primary breast carcinomas, 25 (44%) showed ErbB-2-topoIIalpha coamplification and 24 (42%) showed a physical deletion of the topoIIalpha gene. No topoIIalpha copy number aberrations were found in 40 primary tumors without ErbB-2 amplification. TopoIIalpha gene amplification and deletion are common in ErbB-2-amplified breast cancer and are associated with increased or decreased sensitivity to topoII inhibitors in vitro, respectively. These findings may explain the altered chemosensitivity to topoII inhibitors reported in ErbB-2-amplified breast cancers.

Background: Studies by comparative genomic hybridization (CGH) have shown that chromosomal region 17q23 is amplified in up to 20% of primary breast cancers. We used microarray analyses to measure the expression levels of genes in this region and to explore their prognostic importance. Methods: A microarray that contained 4209 complementary DNA (cDNA) clones was used to identify genes that are overexpressed in the MCF-7 breast cancer cell line as compared with normal mammary tissue. Fluorescence in situ hybridization was used to analyze the copy number of one overexpressed gene, ribosomal protein S6 kinase (S6K), and to localize it to the 17q23 region. Northern and western blot analyses were used to measure S6K gene and protein expression, and an enzymatic assay was used to measure S6K activity. Tumor tissue microarray analysis was used to study amplification of S6K and the HER-2 oncogene, another 17q-linked gene, and the relationship between amplification and prognosis was analyzed. The KaplanMeier method was used for data analysis, and the log-rank test was used for statistical analysis. All P values are two-sided. Results: S6K was amplified and highly overexpressed in MCF-7 cells relative to normal mammary epithelium, and protein expression and enzyme activity were increased. S6K was amplified in 59 (8.8%) of 668 primary breast tumors, and a statistically significant association between amplification and poor prognosis (P = .0021) was observed. Amplification of both S6K and HER-2 implied particularly poor