Research involves doing experiments in different conditions and making observations. Eventually the data points us to an explanation of the mechanism of the drug and sometimes even some insight into how cancer works. Then clinical trials are done to help us decide whether the drug is effective and which people can benefit from taking it.
In this regard, bevacizumab (Avastin) is still a work in progress. To understand how Avastin works, you need to know that in order to grow, all cells require a blood supply to bring them oxygen and nutrients. This includes cancer cells. And because cancer cells have an inherent need to grow, they need a larger blood supply than typical cells. To get all the oxygen and nutrients they need, cancer cells secrete a protein called vascular endothelial growth factor, which stimulates more blood vessels to grow and feed them.
We think that Avastin works by blocking this signal—almost like an embargo—depriving the cancer cells of the blood they need to grow. If that was all Avastin did, then you would expect that the cancer cells would all shrivel up and die. Yet when it was given to women with metastatic breast cancer, it showed some initial benefit but then did not prolong lives. Now we have some more evidence to help us sort out how Avastin works and why this happened .
One of these studies, which was published online in the Proceedings of the National Academy of Sciences Early Edition, looked at how cancer cells themselves, respond to anti-angiogenesis like Avastin and sunitinib (Sutent). These experiments were done on mice with breast cancer by a research group at the University of Michigan Comprehensive Cancer Center. They found that when the tumors were deprived of their blood supply by these drugs, their response was to make more cancer stem cells. These cells are thought to be the ones that fuel a cancer’s growth and spread. It is also believed that they are often resistant to standard treatment.
In sum: The anti-angiogenesis drugs backfired. Instead of stopping the tumor growth they actually stimulated more bad cells. Could this be what happened when the drugs were given to women whose breast cancer had spread? We don’t know, because women are not mice. But it may be one reason why Avastin didn’t work as well as anticipated.
The researchers at U-M believe this suggests that the way to make these drugs effective is to combine them with a drug that can stop the growth of cancer stem cells. There are a number of potential cancer stem cell inhibitors now being studied, and the U-M Comprehensive Cancer Center has already begun lab studies that are investigating whether combining them with Avastin or Sutent is effective.
The other two studies, which were published today in the New England Journal of Medicine, were done in women newly diagnosed with breast cancer. They both investigated the effectiveness of adding Avastin to chemotherapy given prior to surgery (called neoadjuvant treatment) to women with HER2-negative tumors.
Neoadjuvant therapy is typically used to shrink a large tumor, so that it is easier to remove surgically. It also has the benefit of allowing doctors to see if the tumor is responding to the treatment. Researchers are now using this practice to investigate how well different combinations of drugs work by looking at how much cancer is left after a defined period of neoadjuvant treatment. If no tumor is found in the breast at surgery, it is called a pathological complete response.
The first study, which was done in the U.S., enrolled 1200 patients, and looked at whether adding capecitabine (Xeloda) or gemcitabine (Gemzar) to docetaxel (Taxotere), followed by doxorubicin (Adriamycin) plus cyclophosphamide (Cytoxan) would increase the rates of pathological complete response. It also looked at whether adding bevacizumab (Avastin) to these regimens would further increase the rates of pathological complete response. The study showed that the exact combination of chemotherapy was not so important; however, adding Avastin increased the pathological complete response 5% (it was 28.2% in the group that did not receive Avasin, and 34.5% in the group that did). This means that in a group of 100 women you would expect an additional 5 to have no tumor left at surgery.
In the second study, which was done in Germany, 1948 patients were randomized to receive epirubicin and cyclophosphamide (Cytoxan) followed by docetaxel (Taxotere), with or without bevacizumab (Avastin). This study found that Avastin increased the rate of a pathological complete response by about 4%, from 14.9% to 18.4%. However, when the researchers looked more closely, they found that the pathological complete response was 12% higher (27.9% vs 39.3%) among the 663 patients with triple-negative tumors yet basically the same among the women with hormone-receptor–positive tumors.
So what does this tell us? It says that there is a small but real benefit of using one of these anti-angiogenesis drugs in combination with chemotherapy prior to surgery and there is a suggestion that this benefit may be in women with triple-negative tumors. But because the other study did not have the same finding–in that study, triple-negative tumors did not do better–it’s not really clear.
The two clinical trials described above were done in women whose tumors were large, but had not yet metastasized. Could it be that the size of the tumor has something to do with the response? Also, don’t forget that these studies were done in the neoadjuvant setting. These women haven’t been followed long enough for us to know how having the tumor in the breast disappear after neoadjuvant therapy relates to survival. And of course we did not have a chance to look for the increase in cancer stem cells seen in the mice. Lots of food for thought, but nothing that is ready for prime time!
So, does this mean the U.S. Food and Drug Administration (FDA) was wrong to have revoked Avastin’s approval as a breast cancer treatment? Some people who are angry about the FDA’s decision may spin it that way. But the reality is that even if these studies had been presented prior to the FDA’s decision, they should not have made a difference. The FDA decided to revoke approval for Avastin because the studies showed that, overall, the drug did not help patients with metastatic breast cancer live any longer. And these new studies don’t dispute that.
The problem is that science rarely develops in a straight line. And this is a great example! In reality, science zigs and zags. Right now, we know that Avastin does not effect overall survival in women with metastatic disease, although it’s possible that there might be a subgroup in which it does. It does seem to improve local control in early disease. But then there is that worrisome mice data, in that first study. Now we need to go back to the drawing board and see if we can answer some of these questions. If we can figure out who should get Avastin, when, and for how long, we may find it has a place in breast cancer. But until we answer those questions, we should not be giving it to everybody because we hope it will help someone!