Feature

Going after the Real Killer:

Metastatic Cancer

Until recently, metastatic disease was considered part of the continuum of cancer progression resulting from accumulated mutations—a late stage of a unified disease process in which primary tumor cells acquire the ability to migrate away from their initiation site to invade and proliferate in different organs. Although it is true that metastases exert their life-threatening effects well after the primary tumor has become a cause for serious concern, recent research indicates that the seeds of metastatic destruction are sown relatively early on. Furthermore, several lines of evidence suggest that metastatic disease operates through molecular mechanisms distinct from those involved in the development of primary tumors

Within CCR, several principal investigators are converging on the importance of research specifically aimed at stopping cancer metastases. "The emphasis to date in cancer research and in pharmaceutical development has been on trying to treat and eradicate the primary cancer," noted Jeffrey Green, M.D., Head of the Transgenic Oncogenesis and Genomics Section in CCR’s Laboratory of Cancer Biology and Genetics. "And the therapeutic strategies for treating primary tumors may not be the same as those needed to treat metastases."

Kent Hunter, Ph.D., Head of the Metastasis Susceptibility Section, which is also in CCR’s Laboratory of Cancer Biology and Genetics, agrees. "For breast cancer and many other cancers, we all focus on the primary tumor. That’s the wrong thing to focus on because, more often than not, you solve the primary tumor with surgical resection. What kills people is metastasis."

Metastasis Suppressor Genes

Patricia Steeg, Ph.D. (Photo: R. Baer)

More than 20 years ago, as a Postdoctoral Fellow new to NCI, Patricia Steeg, Ph.D. (now Head of the Women’s Cancers Section of CCR’s Laboratory for Molecular Pharmacology), launched her quest to study the difference between tumor cells that metastasize and those that do not. She decided to study the differences in gene expression between metastasizing and non-metastasizing cell lines derived from the same tumor, hoping to find genes highly expressed in metastatic lines. It was not until she heard a seminar describing the first tumor suppressor gene, Retinoblastoma (Rb), that she realized the significance of a gene she called Nm23 (non-metastatic gene 23), whose expression was instead reduced in metastatic cell lines. Steeg and her colleagues reintroduced Nm23 into a highly metastatic melanoma cell line and found that although the cells still made primary tumors when injected into mice, there was a 90 percent reduction in metastases. Nm23 would be the first identified metastasis suppressor gene.

"Initially, that was an extraordinarily controversial observation," remembered Steeg ruefully. "People looked at metastasis back then and said it was too heterogeneous and unstable to have consistent molecular pathways underlying it." There are now, however, more than 20 known metastasis suppressor genes. These genes are not effective in stopping the growth of primary tumors, but they do stop spreading and/or growth at a distant site. "You have to come to the conclusion that growth of a primary tumor is fundamentally different than the growth of a metastasis."

And where it has been studied, a number of preclinical drug studies have found differential sensitivity of primary and metastatic growth. "We are trying to treat metastatic disease, but we are not developing drugs for it," cautioned Steeg even as she attempts to redress this therapeutic imbalance.

Jeffrey Green, M.D. (Photo: R. Baer)

A proportion of breast cancers lose expression of the Nm23 gene. Steeg and her colleagues showed that high-dose medroxyprogesterone acetate (MPA)—a synthetic progestin hormone used historically in the treatment of endometrial cancers as well as a component of hormone replacement therapy—works atypically through a class of steroid receptors (glucocorticoid receptors) not normally associated with progestin to turn expression of the Nm23 gene back on. The researchers went on to demonstrate in a mouse model of breast cancer metastasis to the lungs that MPA caused a 60 percent reduction in overt lung metastases by the end of the study. Kathy Miller, M.D., at the University of Indiana University’s Simon Cancer Center is currently leading a Phase II multicenter trial for the use of MPA in the treatment of metastatic breast cancer, a study that stems from Steeg’s preclinical work on Nm23.

Steeg and her colleagues are also looking for other targets in the Nm23 pathway that may influence metastasis. To find molecular targets that are suppressed by Nm23 and potentially involved in promoting metastasis, they have asked which genes are expressed in a pattern that inversely correlates with Nm23 expression. One promising candidate, EDG2 (endothelial differentiation, lysophosphatidic acid G-protein-coupled receptor, 2), appears to be sufficient to restore metastatic growth to cells in which Nm23 functions as a metastasis suppressor. Steeg’s team is currently asking whether EDG2 inhibitors will have anti-metastatic effects in preclinical models.

One Molecule, Two Different Effects on Cancer

TGF-β switches its role from suppressing tumors before malignancy sets in to promoting metastasis at later stages of disease. (Image: Adapted from A. Roberts and L. Wakefield; PNAS 2003; 100:8621-8623)

Around the same time that Patricia Steeg was embarking on her work with metastasis suppressor genes in the 1980s, Lalage Wakefield, D.Phil., now Head of the Cancer Biology of TGF-β Section in CCR’s Laboratory of Cancer Biology and Genetics, was also beginning her work as a Postdoctoral Fellow at NCI on another molecular player in metastasis. However, it took her a little while to realize where her research was leading.

"TGFs [transforming growth factors] had just been described in the literature," explained Wakefield, an echo of the excitement from those early days still in her voice. TGFs were secreted by cancer cells and were able to transform normal fibroblasts into a premalignant state. "It seemed to me that TGFs were going to be the answer to cancer. If we could purify and block these factors, then that would be cancer cured."

"You have to come to the conclusion that growth of a primary tumor is fundamentally different than the growth of a metastasis."

TGF-β was eventually discovered to have multiple roles in several different tissues and cell types. It was found to be a key regulator of immune system function, as well as a potent inhibitor of proliferation of normal epithelial cells. Importantly, the TGF-β pathway was genetically inactivated in a number of different cancer types and became known, paradoxically, as a tumor suppressor. Several preclinical studies and mouse models later, the dual role of TGF-β in cancer progression was finally revealed. In the early stages of cancer progression, TGF-β does indeed have tumor suppressor activity, inhibiting proliferation and maintaining genomic stability. As cancer progresses, tumor cells progressively alter their responsiveness to TGF-β. At that stage, TGF-β promotes cell migration, promotes invasion of cancer cells into different tissues, and becomes a pro-survival factor. Meanwhile, TGF-β acts on other cell types, such as fibroblasts, to promote angiogenesis, secrete different types of molecules into the extracellular matrix, and suppress immune surveillance. In short, TGF-β can promote metastasis through multiple routes.

"TGF-β is a master regulator that sits at the interface of the tumor [and its cellular environment]. It affects every cell that comprises that ecosystem," concluded Wakefield. A molecule with so many diverse effects, operating differently at different stages of cancer progression, would seem to be a pharmaceutical drug developer’s nightmare. No one was more surprised than Wakefield and her colleagues, therefore, when they were able to genetically engineer a mouse to encode an inhibitor of TGF-β in its genome and found that this inhibitor protected mice against metastasis in a genetic model of breast cancer. The team has since followed up this work with further preclinical studies that support the use of TGF-β inhibitors to treat metastatic cancer in the clinic. As a result, NCI investigator John Morris, M.D., is now leading a Phase I clinical trial to test GC1008, a human monoclonal antibody against TGF-β in patients with locally advanced or metastatic renal cell carcinoma or malignant melanoma. The trial is in an extension phase at the highest dose and appears to be showing some promising effects.

"It’s been an incredibly exciting story so far because I have seen this molecule go from its initial discovery and identification to clinical testing, and believe me, it was not a straightforward process," said Wakefield.