Not since the discovery of p53 has another molecule received as much attention as PTEN. In the 5 years since the discovery of PTEN, encoding a dual specificity phosphatase tumor suppressor on 10q23, it has been shown to be a susceptibility gene for an inherited cancer syndrome, Cowden syndrome, and for several developmental disorders; it has been shown to play a prominent role in normal murine and human development; and it has been shown to be instrumental in cell cycle arrest, apoptosis, and/or possibly cell migration and cytoskeletal affairs. Initial work on cancer cell lines had suggested that PTEN caused every type of cancer because it was reported that a relatively high frequency of a variety of cancer cell lines, whether derived from solid tumors or hematological malignancies, had homozygous or compound heterozygous genetic alterations involving PTEN. Such data, together with the germ-line human and murine model data, suggested that PTEN mutations occurred "early" in sporadic tumorigenesis. However, subsequent painstaking work in noncultured primary tumors and in careful in vitro overexpression studies over the last 4 years demonstrated that the mechanism of PTEN inactivation can be varied and might be cell type dependent. Furthermore, apart from sporadic endometrial carcinoma, PTEN alteration in noncultured sporadic neoplasias likely occurs "late," promoting progression and metastasis. The article by Davies et al. (Clin Cancer Res., 8: 1904-1914, 2002) sheds light on all of these issues when they report on data that derive from a "triple threat" strategy, i.e., in vitro, in vivo, and ex vivo, to demonstrate that adenoviral infection of PTEN into PTEN-null PC3 prostate cancer cell lines results in decreased metastatic potential without significantly altering tumor size via the predominant mechanism of G(1) cell cycle arrest but not apoptosis.