Study Suggests Broader Chemotherapy Attack in Breast Cancer

In the first comprehensive survey of gene activity in each cell type composing normal and malignant breast tissue, scientists at Dana-Farber Cancer Institute have identified genes in non-cancerous supporting cells that can spur the growth of breast cancer cells.

The findings suggest that aiming chemotherapy at both cancer cells and their genetically normal cellular “microenvironment” might improve the success of breast cancer treatment.

In the July 20 issue of Cancer Cell, the researchers report that genes in these “stromal” cells were overactive in specimens taken from women with ductal carcinoma in situ (DCIS) – an early, precancerous condition – and in full-blown “invasive” cancer.

“By finding factors released by surrounding stromal cells that support the growth of the tumor and targeting these components with cancer drugs, it might be more effective than targeting the tumor cells alone,” says Kornelia Polyak, MD, PhD, of Dana-Farber and the study’s senior author. This suggestion is based on basic research findings and physicians are not directing therapy at stromal cells currently – except for antiangiogenic drugs that target the blood vessels that surround and nourish tumors.

The scientists singled out two such genes as potential targets for therapy. The genes, CXCL12 and CXCL14, contain the genetic code for proteins called chemokines that can prompt cancer progression.

Chemotherapy is designed to kill malignant cells in tumors that develop from the lining, or epithelium, of the breast’s milk ducts. But under prolonged assault by chemotherapy agents, these cancerous epithelial cells often undergo genetic mutation that makes them resistant to the drugs that previously were killing them. Resistance to drugs is the bane of successful treatment for breast cancer.

The researchers led by Polyak studied the gene activity in all types of cells known to be or suspected of being involved in normal breast development and breast cancer. They found that the epithelial cells of the ducts’ lining had many mutations in genes and damage to chromosomes – the hallmarks of cancer.

In the cells of the stroma – including fibroblasts, myofibroblasts, leukocytes, and myoepithelial cells – genes were intact and the cells weren’t cancerous. Yet their abnormally high activity levels indicated they were sending signals causing epithelial cells to grow abnormally and burrow through barriers that are meant to contain cancer, explained Polyak, who is also an assistant professor of medicine at Harvard Medical School.

By sampling tissue taken surgically from normal breasts and in early and late stages of breast cancer, the researchers traced changes in gene structure and activity in the different cell types.

Until now, most research has focused on the epithelial cells inside the milk ducts. Breast cancer begins with abnormal growth of epithelial cells due to genetic damage, and in many cases the cancer remains confined harmlessly within the ducts for years or decades. Some, but not all DCIS lesions penetrate and escape from the “basement membrane” that separates the ducts from the stromal tissue. When this happens, DCIS has progressed to full-fledged “invasive” breast cancer that can spread widely and become life threatening.

One potential benefit of the new gene survey could be a way to detect a gene activity “signature” in DCIS cells to predict how likely they are to progress to invasive cancer. Such a test, not available now, might save some women from needlessly aggressive treatment.
Before they could study gene activity in breast cells and make a “profile” of each gene, the researchers first needed to make very pure samples of each of six cell types in the breast tissue specimens. (These were taken from women without cancer undergoing breast reduction surgery, and from women with DCIS or invasive breast cancer.)

Next they used a powerful laboratory tool, called SAGE (serial analysis of gene expression) to measure the amount of activity of each gene in the different cell types. These activity profiles changed depending on whether the cells were from a normal specimen or were from DCIS or invasive cancer.

From the genes’ activity in tissues of different types, the scientists were able to deduce which ones were involved in the beginning or progression of cancer. Some of the genes were previously unknown. In other instances, the genes were known but their involvement in breast cancer hadn’t been recognized.

For example, the Polyak team found in the cancer samples the two abnormally active genes that make chemokines, molecular messengers implicated in cancer. Polyak said work is in progress to determine whether blocking the overactive chemokine genes might be an effective therapy.

Mina Bissell, a noted breast cancer scientist at the Lawrence Berkeley Laboratory in California, is the author of a commentary accompanying the article in Cancer Cell. She termed it “an important proof-of-principle study” for still more comprehensive efforts to profile the genes involved in breast cancer progression.

Noting the cast of supporting molecular players that is now being implicated, Bissell quipped, “It is time to accept that treating cancer will also take a village!” The paper’s first author is Minna Allinen, PhD, of Dana-Farber.

The study was funded in part by the National Institutes of Health.

Dana-Farber Cancer Institute is a principal teaching affiliate of the Harvard Medical School and is among the leading cancer research and care centers in the United States. It is a founding member of the Dana-Farber/Harvard Cancer Center (DF/HCC), designated a comprehensive cancer center by the National Cancer Institute.