Why Cisplatin Kills Breast Cancer Cells When Other Drugs Fail

"This paper describes a specific molecular pathway that makes these tumors sensitive to a therapy infrequently used for breast cancer," says Leif Ellisen, MD, PhD, of the MGH Cancer Center, senior author of the study to appear in the May 2007 Journal of Clinical Investigation. "We're excited that this work has led to the design of a clinical trial for women with a very difficult to treat form of breast cancer."

About two thirds of breast cancers contain receptor molecules for the hormones estrogen or progesterone, and in recent years antiestrogen drugs like tamoxifen have improved outcomes for women with those tumors. About 20 to 30 percent of tumors, some with hormone receptors, have elevated levels of a growth-promoting protein called HER2, and those tumors are candidates for treatment with the monoclonal antibody Herceptin. The third major subtype is the 15 to 20 percent of breast tumors that have neither estrogen nor progesterone receptors and also do not overexpress HER2.

Since these so-called "triple-negative" tumors are treatable with neither Herceptin nor antiestrogen drugs, the prognosis for patients with the tumors has been poor. Triple-negative tumors are the most common subtype found in patients with mutations in the BRCA1 gene, but they also appear in women without alterations in the so-called "breast cancer gene." There have been reports that BRCA1-associated, triple-negative tumors might be sensitive to cisplatin, a drug used to treat several other types of cancer, but whether the more common sporadic triple-negative tumors shared that sensitivity was unknown. The current study was designed to answer that question and to investigate the mechanism underlying cisplatin sensitivity.

The research team focused on the function of p63, a protein that plays a role in normal breast development and is related to the common tumor suppressor p53. They analyzed tissue samples from triple-negative breast tumors and normal breast tissues for the expression of several forms of p63 and another related protein called p73, known to promote the cell-death process called apoptosis.

The researchers found that a significant number of triple-negative tumors overexpress particular forms of p63 and p73, a pattern not seen in other types of breast cancers. Using an RNA interference system to inhibit the action of p63, they showed that the protein stimulates tumor growth by interfering with p73's normal ability to induce cell death. Cisplatin was found to break up the binding of p63 to p73 and reactivate the cell-death process.

"The most important finding was that, if the tumor cells did not express both p63 and p73, the cells were not sensitive to cisplatin," says Ellisen. "These results suggest that testing p63 and p73 levels in patients' tumors might help predict whether they would benefit from cisplatin therapy." Ellisen is an assistant professor of Medicine at Harvard Medical School.

The clinical trial to investigate the role of p63/p73 expression in determining cisplatin sensitivity will be led by MGH researchers through the Dana-Farber/Harvard Cancer Center. Starting in Boston in the coming weeks, the trial will be open to patient with advanced triple-negative breast cancer and eventually will be offered at other U.S. cancer research centers. Patients or physicians interested in the trial should call Karleen Habin at (617) 726-1922 and ask about the cisplatin trial for breast cancer.

Additional authors of the Journal of Clinical Investigation paper are first author Chee-Onn Leong, PhD; Nick Vidnovic, Maurice DeYoung, and Dennis Sgroi, MD, all of the MGH Cancer Center. The study was supported by grants from the Mary Kay Ash Charitable Foundation, the National Institute of Dental and Craniofacial Research, the Tracey Davis Memorial Fund, and the Avon Foundation.