Cancer and Metastasis Reviews 17: 317–324, 1999. © 1999 Kluwer Academic Publishers. Printed in the Netherlands. Peptide growth factors and prostate cancer: New models, new opportunities Barbara A. Foster 1 , Paula J. Kaplan 1 and Norman M. Greenberg 1,2 1 Department of Cell Biology, 2 Scott Department of Urology, Baylor College of Medicine, One Baylor Plaza, Houston, TX, USA Key words: fibroblast growth factor, insulin-like growth factor, TRAMP, prostate, probasin Abstract Prostate cancer is the leading form of newly diagnosed cancer cases in men in the United States. However, the molec- ular mechanisms contributing to the initiation, progression and ultimate development of metastatic and androgen independent disease are poorly understood. This is due in part to the difficulty in obtaining clinical samples repre- senting early disease and the lack of animal models that recapitulate the full spectrum of the clinical disease. To this end we have developed and characterized the Transgenic Adenocarcinoma of Mouse Prostate (TRAMP) animal model that expresses the oncogene SV40 T antigen specifically in the epithelium of the prostate. TRAMP devel- ops spontaneous autochthonous prostate cancer compelte with distant site metastasis and can progress to androgen independent disease. Changes in the fibroblast growth factor (FGF) axis and the insulin-like growth factor (IGF) axis have been examined during prostate cancer progression utilizing the TRAMP model and these data generally support observations reproted in the clinical disease. Moreover, we report novel changes in the FGF axis and IGF axis utilizing TRAMP. Thus, TRAMP can be used as a potent tool in understanding the mechanism of prostate cancer initiation and progression. Although the American Cancer Society estimated 334,500 new prostate cancer cases and 41,800 prostate cancer deaths for 1997, progress in understanding the biology of prostate cancer has been slow because of (1) our limited understanding of the non-steroidal fac- tors that control the growth and differentiation of the prostate gland, (2) the inherent difficulties involved in studying how changes in molecular signaling mecha- nisms contribute to such a heterogeneous disease in a heterogeneous population, and (3) the paucity of ani- mal models that temporally and spatially develop the progressive stages associated with clinical prostate can- cer. To this end, we have developed a novel research system based on the ability to perturb gene expression in a prostate-specific manner to facilitate detailed and comprehensive analysis of the molecular mechanisms involved in the genesis, progression, metastasis and androgen independent growth of prostate cancer. The central feature of our system is the use of the rat probasin (rPB) gene promoter to direct hormonally- and developmentally-regulated expression of heterologous genes specifically to the prostatic epithelium in transgenic mice [1]. This rPB system was used to develop the Transgenic Adeno- carcinoma of Mouse Prostate (TRAMP) model in C57BL/6 inbred mice [2]. A rPB-SV40 T antigen fusion transgene was used for the TRAMP model because the T antigen oncoprotein abrogates the func- tional expression of p53 and Rb tumor suppressor genes that have been observed to be inactivated in human prostate cancer. The TRAMP mice develop sponta- neous progressive autochthonous prostate cancer with complete penetrance that frequently metastasizes to the lymph nodes and lung and less frequently, to the kid- ney, adrenal glands, liver and bone [3]. Following cas- tration at 12 weeks of age approximately 80% of mice develop primary tumors within 3 months while 20% of TRAMP mice show a durable response and remain tumor free (4). With the establishment of the TRAMP model of autochthonous and metastatic prostate cancer, we are uniquely positioned to test the hypothesis that specific