Chromosomal aberrations, including deletions, gains and amplifications are thought to be a mechanism for the development and progression of cancer. Many frequent alterations have been described in prostate cancer, but only a few definitive target genes have been identified. The identification of target genes could lead to the development of diagnostic and/or prognostic markers as well as new targets for therapy. According to comparative genomic hybridisation (CGH), the HIF1A locus (14q23.2) is amplified in the prostate cancer cell line PC-3, and the EZH2 locus (7q36.1) in the prostate cancer xenograft LuCaP41. The locus of EIF3S3 (8q23.3­24.11) is gained in up to 80% of advanced prostate cancers. In order to determine whether these genes could be the target genes of the amplifications, fluorescence in situ hybridisation (FISH) was used to assess their amplification frequencies. In addition, the expression of EZH2 was studied by quantitative real-time reverse transcription polymerase chain reaction (Q-RT-PCR) and immunohistochemistry (IHC). No amplifications of HIF1A were found in clinical prostate cancer samples. Only PC-3 had an amplification of the gene, indicating that the generally observed overexpression of the protein in prostate cancer is not due to gene amplification. The frequencies of EIF3S3 and EZH2 gains/amplifications were significantly higher in advanced disease, reaching over 50% in hormone-refractory prostate cancer. The alterations were weakly associated with poor progression-free survival in prostatectomy-treated patients. The expression of EZH2 was higher in advanced disease and especially in samples with high-level amplification of the gene (p<0.05). Eight out of ten xenografts contained a gain of EZH2, including one with a high-level amplification of 7­10 copies. This sample also showed markedly higher mRNA expression by Q-RT-PCR. A trend towards increased expression was seen in the xenografts with gain of EZH2, although it was not statistically significant. Both EIF3S3 and EZH2 may be considered putative target genes for the gains of their respective loci. Finally, array comparative genomic hybridisation (aCGH) and cDNA microarrays were used to screen prostate cancer cell lines and xenografts for genome wide copy number and expression alterations. The copy number alterations (CNAs) and their frequencies were generally consistent with earlier data obtained by CGH of the same samples, although due to the better resolution some aberrations were narrowed down or shown to consist of several smaller aberrations interrupted by regions of normal copy number. Previously unreported frequent copy number alterations were also found, for example, gains in 1q21.2­23.1, 9p13­q21 and 16p. The amplicon in 9p13 was verified by FISH. cDNA microarrays showed that even a modest increase in copy number significantly affects the expression of the altered genes, indicating that simple gains may have a larger impact in prostate cancer than previously thought. Several putative target genes for copy number aberrations were also identified by cDNA microarray analysis, including POGZ (gain 1q21.3), ITGA4 (loss 2q31.3), FZD6 (gain 8q22.3), UBE2R2 (gain 9p13.3), and RBBP6 (gain 16p13.3). In conclusion, EIF3S3 and EZH2 may be considered putative target genes of the amplifications at 8q23­24 and 7q36.1, respectively. In addition, EIF3S3 could be used as prognostic marker. Array-CGH (aCGH) detects smaller regions of copy number alterations than CGH and may be used to narrow down known alterations and detect novel CNAs. When it is used together with expression arrays, putative target genes of the CNAs may be directly identified. Even a low-level copy number alteration appears to affect the expression of the altered genes.