Among women, breast cancer is the most common cancer worldwide and is the leading cause of cancer mortality. Of the more than 5 million women currently living with a diagnosis of breast cancer, recurrence represents the most common cause of death from this disease. Remarkably, recurrences can appear up to 20 years following surgery, although most occur within the first two years. "Up to 40 percent of women thought to be cancer free following surgery, radiation, and chemotherapy still have tumor cells in their bodies in a dormant state. As such, approaches to prevent cancer recurrence in these women would be broadly applicable," says senior author Lewis A. Chodosh, MD, PhD, Vice Chair of the Department of Cancer Biology and Director of Cancer Genetics at the Abramson Family Cancer Research Institute at Penn. The researchers published their findings in the September 2005 issue of Cancer Cell.

"To this point there are extraordinarily few targets that have been causally implicated in breast cancer recurrence. Consequently, there are few treatments available to offer women who are at risk for recurrence once they have received standard treatments," says Chodosh.

The Penn team of researchers induced breast cancer in the genetically engineered mice by giving doxycycline to turn on the oncogene HER-2/neu. This oncogene is commonly amplified in human breast cancers and is associated with aggressive disease and poor clinical outcome. The researchers then induced these tumors to regress by turning off the HER2/neu oncogene in fully formed tumors. This mimics important aspects of molecularly targeted therapies and leads to the dramatic regression of tumors to a clinically undetectable state. Nevertheless, residual tumor cells lie in a dormant state and later grow out after a month to a year in the mice.

Using microarrays, Chodosh's team compared recurrent tumors with the original tumors from which they arose. They found that a variety of genes were turned on in recurrent tumors that were not on in the original tumors, including the transcriptional regulatory protein, Snail, which was induced ten-fold. The Penn team also identified changes in the microscopic appearance of the cells in recurrent tumors, which had transformed from a cuboidal, epithelial shape to a spindle, fibroblastic shape - a change associated with more aggressive tumors in humans.

Snail was first identified in fruit flies and later in mice based on its essential role in embryogenesis during a developmental transition in which normal cells undergo a similar change in shape. "Snail controls a complex set of cellular functions that cancer cells appropriate by turning on this master regulatory gene," explains Chodosh.

To prove a cause-and-effect, the researchers added Snail back to the original tumor cells in mice and showed that Snail increased the rate of recurrence.

But could Snail expression play a similar role in women with breast cancer? When the Penn team delved into public databases of breast cancer tissue data, separating cases into those with high levels of Snail and those with low levels of Snail, they found that women whose original breast cancers expressed high levels of Snail were twice as likely to experience a recurrence within five years following surgery compared to women whose cancers expressed low levels of Snail.

The magnitude of risk associated with high Snail expression is comparable to standard prognostic factors such as estrogen-receptor status, HER-2/Neu amplification, tumor size and grade, and lymph node status and - after correcting for the effects of these factors - Snail expression was shown to predict a woman's risk of recurrence independent of these factors.

Currently, Chodosh and colleagues are exploring the precise molecular mechanism by which Snail triggers breast cancer recurrence, as well as ways of targeting Snail's signaling pathways as a possible therapeutic approach to prevent recurrence.

med.upenn/