The pattern of chromosomal aberrations and their significance in prostate cancer are poorly understood. We studied 23 prostate cancer and 10 benign prostatic hyperplasia (BPH) specimens by fluorescence in situ hybridization (FISH) using pericentromeric repeat-specific probes for 10 different chromosomes. The aims of the study were: 1) to compare the sensitivity of FISH and DNA flow cytometry in aneuploidy detection, 2) to determine which chromosome copy number changes are most common, and 3) which probe combinations would be most effective in aneuploidy diagnosis. Disaggregated tumor cells from formalin-fixed, paraffin-embedded tissues were pretreated with our newly developed method based on hot glycerol solution to improve probe penetration. All BPH specimens were diploid by DNA flow cytometry and showed no numerical chromosome aberrations by FISH. In prostate cancer, flow cytometry showed abnormal DNA content in 35% of cases, whereas 74% were abnormal by FISH. Aberrant copy number of chromosomes 8 (48% of cases), X (43% of cases), and 7 (39% of cases) were most common. Ninety-four percent of all aneuploid cases would have been detected with these three probes alone. Simple chromosome losses were uncommon but in DNA tetraploid tumors relative losses (trisomy or disomy) of several chromosomes were often found, suggesting progression of prostate cancer through tetraploidization followed by losses of selected chromosomes. In conclusion, our results indicate that FISH using three selected chromosome-specific probes is two to three times more sensitive than flow cytometric DNA content analysis in aneuploidy detection.