Test could offer improved way to help assess risk of breast cancer recurrence
Better forecasting could help physicians opt for aggressive treatments
BY MITZI BAKER
A dilemma in breast cancer is how aggressively to treat patients to ensure that the cancer goes away—and doesn't come back. Currently, the best way to predict whether a breast cancer is likely to recur is to determine whether tumor cells have invaded the lymph nodes near the breast. But new research from the School of Medicine suggests that looking at the immune cells in those lymph nodes, instead of the tumor cells, will yield a more accurate forecast.
"Immune changes in the lymph node almost perfectly predict clinical outcome, much better than any other prognostic factor that is available today," said Peter P. Lee, MD, assistant professor of medicine and the senior author of the paper detailing the findings in this month's issue of Public Library of Science-Medicine.
In samples of breast cancer patients' lymph nodes, Lee and his colleagues identified unique patterns of immune cells. When the researchers compared the immune profiles to whether a patient's cancer returned within five years, they could divide the patients into two groups. The group with what Lee termed a "favorable" immune profile had an 85 to 90 percent chance of being disease-free after five years. The group with an "unfavorable" immune profile had less than a 15 percent chance.
The predictions could be made based solely on the immune cells, regardless of whether a lymph node contained tumor cells.
The origins of the study can be traced to about three years ago, when Lee began to question why the lymph node's immune cells didn't react to and destroy the invading tumor cells. He reasoned that tumor cells must do something to suppress the immune cells in the node. He wanted to see if he could identify changes in the immune cells in the lymph nodes of women who had breast cancer. If so, he wondered if it could predict how the women fared years later.
Lee enlisted Holbrook Kohrt, MD, a resident in internal medicine, to scour the Stanford pathology bank, looking for preserved samples of lymph nodes. They obtained lymph node tissue samples from 77 breast cancer patients taken more than five years ago. All of these patients had had cancer that migrated out of the breast. Each also had five-year follow-up information.
Using specific antibodies, they looked for three major types of immune cells: cytotoxic T cells, helper T cells and dendritic cells, as well as tumor cells. Using an automated imaging system, they collected up to 4,000 images per lymph node, allowing them to count cells from the entire lymph node. Previous studies have relied on fewer than 20 images.
Lymph nodes that were invaded by tumor cells showed dramatic decreases in helper T cells and dendritic cells. They also had fewer cytotoxic T cells.
"That finding was interesting, but somewhat intuitive," said Lee. He said it is logical to think that with tumor cells accounting for up to 80 percent of the cells in an invaded lymph node, there are bound to be some perturbations in immune cell populations.
"Then we found something more interesting and puzzling," he said. Some of the nodes were only minimally invaded by tumor cells—in some cases fewer than 10 tumor cells or even without a single tumor cell to be found. These lymph nodes also showed similar immune changes as nodes full of tumor cells. Statistician Susan Holmes, PhD, rigorously analyzed the data and found that strikingly, immune changes within these lymph nodes predicted clinical outcome even better than their tumor invasion status.
In other words, the numbers of immune cells alone seemed to predict whether a woman's cancer returned within five years. In the study, 33 of the 77 patients had their cancer return.
"It was a surprise to find immune changes in lymph nodes with no detectable tumor cells," said Lee. He added that their data support an intriguing theory: Perhaps tumor cells prepare the lymph node for invasion. "Even before it actually invades the node, it actually causes the node to change," he said. It might be a more sensitive and earlier method of detecting metastasis, or tumor spread, than actually seeing the migrated tumor cells themselves.
Such information could help determine which women could benefit from more aggressive therapy, and which could be spared undergoing costly and toxic treatments unnecessarily.
"The nice thing about this technique is that it could be applied to all women with breast cancer," said Kohrt, the lead author of the paper. "It's awesome that such a simple idea could affect more than 200,000 patients a year."
Lee envisions a simple clinical tool based on their discovery: it entails staining a lymph node biopsy for immune cells rather than for tumor cells. The researchers emphasize that their findings will have to be confirmed with larger numbers of breast cancer patients, in those with less advanced disease and with fresh samples rather than frozen.
From the broader perspective of cancer biology in general, the group's findings underscore the importance of immune response in determining the spread of breast cancer. A better understanding of these mechanisms may lead to novel treatment strategies for breast cancer specifically directed at modulating the immune responses within lymph nodes.
"This is a shot in the arm for the field," said Kohrt. "The immunology of breast cancer has not been very well explored yet, and these findings argue that the immune system is more important in cancer than previously thought."In addition to Lee and Kohrt, others involved in the study are research assistant Navid Nouri, associate professor of pathology Kent Nowels, MD, and associate professor of surgery Denise Johnson, MD. Kohrt receives funding from the Howard Hughes Medical Institute and the Donald E. and Delia B. Baxter Foundation. Lee's lab received funding for this research from the National Institutes of Health, the Damon Runyon Cancer Research Foundation and the American Cancer Society.