Research ArticleMelanoma

Human Melanoma Metastasis in NSG Mice Correlates with Clinical Outcome in Patients

Science Translational Medicine  07 Nov 2012:
Vol. 4, Issue 159, pp. 159ra149
DOI: 10.1126/scitranslmed.3004599

You are currently viewing the editor's summary.

View Full Text

Log in


Melanomas in Miniature

No two patients are exactly alike, so studying cancer biology in humans is complicated. There is no way to standardize experimental conditions or to analyze tumor biology in isolation. To overcome these difficulties, researchers use model animals such as mice, but these are never a perfect model of human physiology. For example, the process of cancer metastasis in mice frequently does not reflect findings in human patients. Some metastatic human cancers do not grow well in mice or do not metastasize. Now, Quintana et al. have developed a new mouse model of melanoma that provides a more faithful reproduction of human tumor metastasis and offers insights into the biology of circulating tumor cells.

Quintana and colleagues used highly immunosuppressed mice to implant tumors from patients with stage III melanomas—those that have spread locally (typically to lymph nodes), but not beyond. All of the patients’ tumors were successfully implanted in this mouse model, but not all metastasized. Pathological analysis and bioluminescence imaging demonstrated that the spread of melanoma in the mouse model correlated with eventual spread of the disease in human patients. The same tumors that would later metastasize and kill their human hosts also metastasized widely in the mice. Conversely, melanomas that did not spread after surgical removal of the parent tumors from patients also metastasized slowly or inefficiently in animals, even after repeated passages of tumor cells through several generations of mice. To explain this phenomenon, the authors searched for human cancer cells in the blood of tumor-bearing mice and found a correlation there as well—the mice whose tumors metastasized efficiently had easily detectable tumor cells in their bloodstream, whereas the mice with poorly metastatic tumors did not.

In this study, tumors were removed from their normal microenvironment and transplanted into mice; thus, this protocol permits only a partial view of the metastatic process. The new mouse model can be used to study intrinsic factors in cancer cells that contribute to metastasis but not immune factors or other patient-specific characteristics that may contribute to survival. Nevertheless, this report nicely demonstrates that at least some of the key factors that regulate melanoma metastasis are intrinsic to the tumors. The new mouse model will allow the characterization of metastatic processes and possibly point to markers that predict metastasis or therapeutic targets for human melanoma interventions.