When breast cancer patients get chemotherapy before surgery to remove their tumor, it can make remaining malignant cells spread to distant sites, resulting in incurable metastatic cancer, scientists reported last week.
The main goal of pre-operative (neoadjuvant) chemotherapy for breast cancer is to shrink tumors so women can have a lumpectomy rather than a more invasive mastectomy. It was therefore initially used only on large tumors after being introduced about 25 years ago. But as fewer and fewer women were diagnosed with large breast tumors, pre-op chemo began to be used in patients with smaller cancers, too, in the hope that it would extend survival.
But pre-op chemo can, instead, promote metastasis, scientists concluded from experiments in lab mice and human tissue, published in Science Translational Medicine.
The reason is that standard pre-op chemotherapies for breast cancer — paclitaxel, doxorubicin, and cyclophosphamide — affect the body’s on-ramps to the highways of metastasis, said biologist John Condeelis of Albert Einstein College of Medicine, senior author of the new study.
Called “tumor microenvironments of metastasis,” these on-ramps are sites on blood vessels that special immune cells flock to. If the immune cells hook up with a tumor cell, they usher it into a blood vessel like a Lyft picking up a passenger. Since blood vessels are the highways to distant organs, the result is metastasis, or the spread of cancer to far-flung sites.
Depending on characteristics such as how many tumor cells, blood vessel cells, and immune cells are touching each other, the tumor microenvironment can nearly triple the chance that a common type of breast cancer (estrogen-receptor positive/HER2 negative) that has reached the lymph nodes will also metastasize, Condeelis and colleagues showed in a 2014 study of 3,760 patients. The discovery of how the tumor microenvironment can fuel metastasis by whisking cancer cells into blood vessels so impressed Dr. Francis Collins, director of the National Institutes of Health, that he featured it in his blog.
The new study took the next logical step: Can the tumor microenvironment be altered so that it promotes or thwarts metastasis?
To find out, Einstein’s George Karagiannis spent nearly three years experimenting with lab mice whose genetic mutations make them spontaneously develop breast cancer, as well as mice given human breast tumors. In both cases, paclitaxel changed the tumor microenvironments in three ways, all more conducive to metastasis: The microenvironment had more of the immune cells that carry cancer cells into blood vessels, it developed blood vessels that were more permeable to cancer cells, and the tumor cells became more mobile, practically bounding into those molecular Lyfts.
As a result, the mice had twice as many cancer cells zipping through their bloodstream and in their lungs compared with mice not treated with paclitaxel. Two other neoadjuvants, doxorubicin and cyclophosphamide, also promoted metastasis by altering the tumor microenvironment. “This showed that the tumor microenvironment is the doorway to metastasis,” Condeelis said.
The scientists also analyzed tissue from 20 breast cancer patients who had undergone pre-op chemo (12 weeks of paclitaxel and four of doxorubicin and cyclophosphamide). Compared to before the chemo, the tumor microenvironment after treatment was more conducive to metastasis in most patients. In five, it got more than five times worse. No patient’s microenvironment got less friendly to metastasis.
Pre-op chemo “may have unwanted long-term consequences in some breast cancer patients,” the Einstein researchers wrote.
That finding is “fascinating, powerful, and very important,” said Julio Aguirre-Ghiso, of Mount Sinai School of Medicine, an expert in metastasis who was not involved in the study. “It raises awareness that we might have to be smarter about how we use chemotherapy.”
Dr. Julie Gralow, a medical oncologist at the University of Washington, said that if pre-op chemo promoted metastasis, that should have shown up in studies that compared it to post-op chemo, but for the most part it hasn’t. However, that could be because only tumor cells containing certain proteins that make them especially mobile are affected in this way. “This is an interesting study, to say the least,” Gralow said. “I am willing to keep my mind open to the possibility that there are some breast cancer patients in whom things get worse” with pre-op chemo.
One reason to question the findings, however, is that if pre-op chemo promotes metastasis in some patients, that might be expected to have shown up in studies of the therapy. Overall, in fact, those studies show that “neoadjuvant chemotherapy does not seem to improve overall survival,” as the authors of an editorial in the Journal of Clinical Oncology wrote.
That’s not as bad as decreasing survival, of course. But Einstein’s Dr. Maja Oktay, a co-author of the new research, cautioned that the typical length of the studies — six or so years — is too short to assess the risk of metastasis, “which can take more than 20 years” to appear, she said. Such patients might never be flagged as having metastatic cancer, let alone having it linked to pre-op chemo decades earlier, said Aguirre-Ghiso.
On a brighter note, not all breast cancer patients have the kind of tumor microenvironment in which pre-op chemo can promote metastasis. Whether they do or not can be determined by a simple lab test, but one that is not routinely done, Condeelis said.
Serendipitously, an experimental compound called rebastinib, being developed by Deciphera Pharmaceuticals (DCPH), seems to be able to block the on-ramp to the metastasis highway. In a study currently recruiting patient volunteers, the Einstein scientists (who have no financial relationship with Deciphera) are studying whether rebastinib can improve outcomes in metastatic breast cancer.