Targeted therapies, like trastuzumab, are supposed to work only in patients whose tumors express the drug's target. Thus, when Soonmyung Paik, MD, Director of the Division of Pathology at the National Surgical Adjuvant Breast and Bowel Project, and colleagues published data showing that adjuvant trastuzumab prolonged disease-free survival in patients with HER2-negative breast cancers as assessed by central lab review, many oncologists and researchers cried foul, even questioning the central lab’s techniques. Shortly thereafter, though, researchers found a similar trend in a North Central Cancer Treatment Group trial, raising the possibility that the underlying biology might be more complex than anticipated.
Now, Max Wicha, MD, and colleagues think they can explain those controversial data. If correct, their new results suggest that trastuzumab may have a role in controlling HER2-negative tumors -- and that the traditional approach to adjuvant therapy development may not be the right one.
HER2 Regulator in HER2-Positive Cancers
Several years ago, Dr. Wicha, Professor of Internal Medicine and Director of the University of Michigan Comprehensive Cancer Center in Ann Arbor, and colleagues found that HER2/neu is an important regulator of mammary stem cells in both normal and cancerous tissue. Overexpression of HER2 increased the proportion of stem-like cells in the population, as detected by expression of the stem cell marker ALDH, and increased the cells' ability to form tumors in mammary fat pads of immunocompromised mice.
Moreover, if they transfected luminal- or basal-type breast cancer cells, which normally lack HER2 amplification, they found that the proportion of ALDH-positive cells in the tumor increased, as did their tumor-initiating potential in animals.
"Therefore, we think HER2 is an important regulator of HER2-amplified cancers," Dr. Wicha said during a presentation here at the AACR-NCI-EORTC Molecular Targets and Cancer Therapeutics Conference. "That all seemed to make sense.
"What didn’t really make sense was the very controversial paper published in NEJM [2008;358-1409-1411] by [Dr.] Paik and colleagues," he continued.
Stem Cell Marker in HER2-negative Cells
To try to make sense of Dr. Paik's clinical observation, Dr. Wicha and colleagues examined breast cancer cell lines that are classified as HER2-negative, such as MCF7 cells, where they found that a subpopulation of the cells did express HER2.
"MCF7 cells are said to be completely HER2 negative," Dr. Wicha explained. "But what we mean by HER2 negative is that the cells are negative for HER2 gene amplification or negative if you look by immunohistochemistry. But we can see in samples that rare cells do express HER2 by fluorescence microscopy."
The HER2-expressing cells frequently expressed the cancer stem cell marker ALDH as well. Furthermore, they saw this pattern in several luminal breast cancer cell lines. (They did not find a HER2-positive subpopulation in basal cancer cell lines, however.)
When the team sorted the HER2-positive cells from the marker-negative ones, they found that trastuzumab reduced ALDH staining in the HER2-positive cells and limited their ability to form spheroids in culture, which is a cancer stem cell characteristic. Based on these data, it appears that HER2 might help regulate cancer stem cells in luminal breast cancers -- even though they are traditionally considered HER2-negative disease.
If that hypothesis is accurate, then trastuzumab should have an effect on luminal tumors in animal models, and that is what Dr. Wicha's group saw, but only if they perform the experiment in a particular manner. If the team injects MCF7 cells into the mouse, lets a tumor grow, and then treats the animals with trastuzumab, the drug has virtually no impact. By contrast, if they inject MCF7 cells into the fat pad and then start trastuzumab treatment the next day, the effect was dramatic. "You completely block the growth of the tumor -- and remember, these are the cells that everyone uses as the negative control for HER2,” Dr. Wicha said.
The team saw a similar pattern of response in a bone metastasis model. If they initiate trastuzumab treatment shortly after introducing MCF7 cells into the bone, no tumors form. If they delay treatment until the tumor is established, the drug has no impact. Finally, if they knock down HER2 expression in MCF7 cells using short-hairpin RNAs, they find that the cells can no longer establish bone tumors and they have fewer ALDH-positive cells.
"HER2 is absolutely necessary for cells to grow in metastatic sites like the bone," Dr. Wicha concluded.
HER2 on the Margins
When Dr. Wicha's group examined tumor samples from two patients with luminal-type breast cancer, they found that a subset of cells express HER2 and many co-expressed ALDH. The location of the HER2-positive cells was remarkable as well.
"What we find in primary luminal tumors is the distribution of HER2, and the stem cell marker is not uniform throughout the tumor," Dr. Wicha said. "But on the invasive edge of tumor you see induction of the HER2-positive cells....[Where you see] invasion into fatty tissue, you see cells expressing both markers."
When the researchers examined matched pairs of primary breast cancer tumors and bone metastases from 10 patients, they saw an upregulation of HER2 expression in the bone microenvironment in luminal tumors. There was no change in HER2 expression in basal-type tumors between the primary and metastatic sites.
Dr. Wicha said they now have evidence that RANK ligand induces HER2 expression in the cancer cells in the bone microenvironment and that the cells become sensitive to both trastuzumab and the RANK ligand inhibitor denosumab.
"This tells us several interesting points," he said. "It provides a biological explanation for the puzzling clinical result for adjuvant efficacy of trastuzumab in patients who are not HER2-positive, and it supports the B47 trial." The B47 trial is an ongoing prospective test of adjuvant trastuzumab in patients with HER2-negative breast cancer.
Dan Von Hoff: Explains Discordance in New England Journal Study
"I think he is definitely on to something there," said Daniel Von Hoff, MD, Physician in Chief and Distinguished Professor at The Translational Genomics Research Institute (TGen) in Phoenix, who attended the talk. "It does explain the discordance in the New England journal work. But the nice thing is, we've got the clinical trial to test it, to follow-up."
A Problem with Adjuvant Trials
During his presentation, Dr. Wicha emphasized repeatedly that while these new data raise questions about adjuvant trial design, "it provides evidence that we probably develop adjuvant therapies in the wrong way.”
Shrinking tumors in the advanced disease setting makes sense, because the tumors are already established. But in the adjuvant setting, physicians want to reduce the bulk of the primary tumor and prevent disease recurrence and metastasis. Therefore, if the drug used to shrink the tumor has no impact on the subpopulation of cells that drive invasion and metastasis -- as Dr. Wicha hypothesizes the HER2-positive subpopulation does in luminal breast cancer -- then the adjuvant therapy will have little impact on survival.
Therefore, if one wants to prolong survival, one has to target the cancer-driving cell population and tumor shrinkage is not the right response measure for that, he said. "Tumor heterogeneity is not only a genetic heterogeneity, but also a developmental heterogeneity in which cells with self renewal capacity are driving tumor initiation," Dr. Wicha said.
"Early tests suggested that trastuzumab worked only in HER2-amplified cancer, but that is because the tests were set up to evaluate response using classical tumor shrinkage criteria, rather than stem cell criteria."
Dr. Von Hoff was a little more cautious about criticizing the adjuvant drug development model across the board. "I understand what he is saying, but we have gotten some spectacular results," Dr. Von Hoff said in an interview with OT.
"But many people are starting to believe that it is the few cells that are going to kill you, so I think he has an important point. I see a lot of patients, and we can shrink tumors down, but that shrinkage doesn't always translate to improved survival."