At current rates of diagnosis, a man in the United States has a one in five chance that invasive prostate cancer will develop in his lifetime (1). This rate is nearly twice that of lung cancer and three times that of colorectal cancer. Death from prostate cancer is the second leading cause of death from cancer in men in the United States. Almost every man with advanced prostate cancer will undergo androgen ablation therapy and, in time, most will progress. The central characteristic of lethal prostate cancer is androgen independence. These two facts were true in 1941, when therapeutic castration was first described (2, 3), and, unfortunately, are true in 2000. Historically, there has been an inverse relationship between efforts to maximize the efficacy of hormonal therapy for prostate cancer and the outcomes of those efforts: Thousands of patients studied and billions of dollars spent repeatedly show hormonal therapy to have dramatic yet ultimately ineffective therapeutic effects (4). Although a number of growth and survival factors have been implicated in the androgenindependent phenotype of prostate cancer, there has been no translation of these findings to effective therapy (5).

In some androgen-independent prostate cancers, neuroendocrine gene products are expressed, usually as scattered cells or nest of cells throughout the tumor (6). Neuroendocrine cells are characterized by secretory granules rich in neuropeptides, such as chromogranin A; these cells are normally found dispersed in the prostate and may function as paracrine mediators of secretory epithelial function, much like a “soluble nervous system.” It is the continued or reexpression of this phenotype in prostate cancer that has led to investigation of these neuroendocrine gene products as mediators of androgenindependent disease (7-13). Rather than confine this discussion, however, to the classic neuroendocrine phenotype (which, in its small cell or carcinoid manifestations, represents a fraction of prostate cancers), we examine a recent series of related observations about the role of the small bioactive peptides bombesin, endothelin-1 (ET-I), and neurotensin in prostate cancer. These peptides-with compelling biologic effects in prostate cancer-act through specific high-affinity heptahelical, G-proteincoupled receptors (14-19). The enzyme responsible for the degradation of these peptides, endopeptidase 24.1 I, is decreased in androgen-independent prostate cancer (20);