Objectives: To better understand the gene expression patterns in tumor-associated stroma, laser-capture-microdissections from clinical specimens were analyzed by genome-wide-expression microarray technology. The epithelial-stromal interaction plays a critical role in prostate development, reactive changes, and tumorigenesis. Diverse microarray technologies have been used to characterize the molecular changes in prostate cancer. Even though these gene expression studies are compromised by the heterogeneity of the tumor, as well as by the difficulty associated with collecting appropriate counterparts to represent normal prostate cells, the gene array data from tumors have shown promising results. Currently, little is known about the tumor-associated stromal gene expression profile in prostate cancer. Methods: Matching benign and malignant epithelial cell-related stroma cells were subjected to microarray platforms. Results: The prostatatic stroma expressed several osteogenic molecules. In particular, one of the genes, OSF2, was upregulated in tumor-associated stroma compared with benign epithelial cell associated stroma, which was further validated by immunohistochemical examination. Conclusions: These data show that the combination of laser capture dissection with computational enhancement of epithelial and stromal microarray data is a useful tool to assess gene expression changes in prostate cancer stroma.