Prostate cancer is one of the most commonly diagnosed cancers and is a major cause of cancer death in men. Although the majority of the diagnosed prostate cancers will remain localized and never produce clinical symptoms during the lifetime of the host, a subset of these cancers will progress to a more malignant state requiring therapeutic intervention. Acquisition of metastatic ability by prostatic cancer cells is the most lethal aspect of prostatic cancer progression. Once this has occurred, definitive therapy is required before the initially localized metastatic cells escape from the prostate. At present, metastatic prostate cancer is incurable. Therefore, there is an urgent need to develop molecular markers that can be used to predict the metastatic potential of prostate cancers. Using somatic cell hybridization, we have demonstrated that acquisition of metastatic ability requires both the loss of metastasis-suppressor function(s) and the activation of oncogenes. In further studies using micro-cell-mediated chromosomal transfer, we located genes on human chromosome, 8, 10cen-q23, 11p11.2-13, and 17pter-q23, which, when introduced into rat prostatic cancer cells, are capable of suppressing their metastatic ability without affecting their tumorigenicity or growth rate in vivo. Initially we focused upon the human chromosome 11p11.2-13 region to clone metastasis-suppressor gene(s) positionally. One such gene, termed KAI-1, encodes a membrane glycoprotein. KAI-1 has been mapped to the p11.2 region of human chromosome 11 by fluorescence in-situ hybridization analysis. Expression of KAI-1 has been detected in all normal human tissues thus far tested, including prostate tissue. When introduced into rat metastatic prostatic cancer cells, KAI-1 significantly suppressed the metastasis without affecting the tumor growth rate. KAI-1 expression is high in human normal prostate and benign prostatic hyperplasia but is dramatically lower in cancer cell lines derived from metastatic prostate tumors.