Ahead of the annual American Society of Clinical Oncology (ASCO) meeting, held May 31–June 4 in Chicago, we speak with Angela DeMichele, MD, associate professor of medicine at the Hospital of the University of Pennsylvania, Philadelphia, about neoadjuvant therapy options and neoadjuvant clinical trials in breast cancer. Dr. DeMichele treats patients with breast cancer and participates in clinical trials. At this year's ASCO meeting she is chairing a session called "Pushing the Limits of Upfront Care and Drug Development: Neoadjuvant Opportunities in Breast Cancer."
—Interviewed by Anna Azvolinsky, PhD
Cancer Network: Dr. DeMichele, let's start with the available neoadjuvant agents for breast cancer? What are the different types of agents and which patients are eligible?
Dr. DeMichele: We typically use neoadjuvant or preoperative therapy in patients who are candidates to receive chemotherapy. And instead of receiving chemotherapy after surgery, as a way to prevent the cancer from coming back, we are giving the chemotherapy before surgery. We know that it is safe to do that. A series of studies compared giving the same therapy before and after surgery and found that it was equally effective. So, any agent that we would use after surgery as an adjuvant chemotherapy can be used as neoadjuvant chemotherapy. But, you have the added benefits in the neoadjuvant setting of actually down-staging the tumor. This means that tumors that previously weren't operable could now be resected if they respond.
Other situations or benefits include that patients who would need a mastectomy may, through shrinkage of the tumor, now have the opportunity for a lumpectomy and breast conservation. Those are really the advantages. The drugs are the same whether you use them before or after surgery. But the other real benefit that we think of when we think of neoadjuvant therapy is the opportunity to see if drugs are really working on the tumor. If they are working on the tumor in the breast, we have good evidence that they are also working on the microscopic cells that may be in the bloodstream. We know that women who have an excellent response to chemotherapy given before surgery have good prognosis, and in fact a better prognosis compared to women whose tumors don't respond to the chemotherapy.
Cancer Network: What about neoadjuvant agents that are currently in development for breast cancer. Are there any you can highlight?
Dr. DeMichele: I think that there has been a lot of interest in looking at new agents in this setting because we get a real-time readout on whether a drug is active or not. More recently there has been a push to actually use this setting as a way to really quantify whether drugs are going to be beneficial when added to standard therapy. Drugs that come to mind that have been studied more recently in this setting include drugs that are targeted to HER2—for example the drug lapatinib has been extensively studied in the neoadjuvant setting in an effort to figure out if giving it before surgery would be beneficial.
Other drugs that are of interest and have been tested in this setting include PARP inhibitors in women who have triple-negative breast cancer. We want to understand who the patients that respond to the drug are, and who are the patients that do not respond, rather than treating a large number of patients after surgery with no way to target the therapy. Those are just a couple of examples, but there are really numerous new therapies that are being tested in the neoadjuvant setting, and the criteria for determining whether a drug can be safe in that setting really has to do with how much experience we have with the drug to date.
When we think about testing new drugs in this setting, we need to be mindful of the fact that patients may be cured with our standard drugs. We really need to know that the drugs we are adding and testing for effectiveness are actually safe. We need to have a track record with those drugs. We need to know what the side effect profile is and know that they are not so dangerous that we are exposing patients to undo risk. So, you will see the drugs that are actually tested in the neoadjuvant setting are not the newest of the new—they are drugs that we already have some hint of activity about, and we have adequate knowledge that they are safe for patients. Now we really want to get a handle on which patients are going to respond and how long the response will be, and which patients won't respond.
Cancer Network: You mentioned some of these already, but are there any other goals of treating patients in the neoadjuvant setting, before surgery?
Dr. DeMichele: I think that we really do think about the surgical outcomes as being the most clinically relevant thing. We do neoadjuvant therapy even without testing new drugs, just standard neoadjuvant therapy, for women in whom we really want to try to improve surgical outcomes—they couldn't be resected before or they couldn't be candidates for breast conservation. Then there is the additional benefit of being able to see if a patient will respond or not in that setting. Having a sense of whether this is a drug that works on that patient's particular tumors is really the ultimate in personalized medicine, of really being able to see if your tumor will respond to this therapy rather than waiting until after surgery when the tumor is gone, when you don't have any indication of whether or not the chemotherapy is working.
Cancer Network: The road to a cancer drug approval in the United States usually starts with clinical trials in patients with the most advanced forms of cancer, and only then are these drugs tested in earlier treatment settings as part of clinical trials. Recently though, the FDA issued a draft guidance on methods to accelerate drug approval through neoadjuvant therapy clinical trials. Can you describe the interest in accelerating these types of trials?
Dr. DeMichele: As you said, the typical approach is that we will test the drug in patients with advanced disease, starting with patients who have had a lot of prior therapies—so their tumors are very resistant to almost anything we give them—and working up to patients who have advanced disease but may just be getting their first or second therapies, and getting a sense of whether the drug is active. Typically what would happen is that once we have that information, we would put the drug into a very large adjuvant trial, a trial for drugs given after surgery. And again, because you cannot see if the drug is working in real time, you have to wait to see if patients getting the new therapy relapse more or less frequently compared to patients who are getting standard of care. This takes many thousands of patients and often upward of 5 to 10 years to really see whether the drug is effective or not. That has led to a situation where for even the most effective drugs, it can take upwards of 15 years and millions of dollars and thousands of patients to figure out if a drug works or not.
So, the idea here is that we can get a sense in the neoadjuvant setting of whether or not the drug is shrinking the tumor over and above the shrinkage we would see with standard therapy alone. We get a sense of whether that shrinkage will be partial or complete, and how best we can give this therapy in combination with our other standard drugs. In that way we can learn in a trial with just a few hundred patients, what would have taken us many more years and thousands of patients to learn in the older strategy. So, what the FDA has recognized is if we can come up with appropriate surrogate markers, (early indicators that really accurately reflect whether a patient's tumor will relapse, whether a patient will die of the disease), then we could use these smaller trials to at least give us a sense of whether or not a drug is promising. The FDA has issued a pathway or an approach whereby this kind of trial could be used to get a provisional approval. There would need to be a confirmatory trial later but while that trial is going on, this drug could be available to patients. So rather than having to wait 5 to 10 years for the final results of one of these 3,000-plus patient studies, after a few hundreds of patients have been treated, a promising drug could get provisional approval, and while it is being confirmed, patients would have the opportunity to use that drug. It really could shave off many years and probably millions of dollars in the process.
There are risks to doing this. The FDA is well aware of this and has sought a lot of input from the scientific community about the right way to go about it. We don't want to prematurely throw out drugs that look like they aren't effective when in fact it wasn't the right setting in which to test them. So, we have to be careful and choose the drugs correctly that could be appropriately tested in this setting. The FDA is really involved in thinking through all of the different aspects of this process, and it is really a dynamic process of setting up this approval strategy that has been going on over the past year.
Cancer Network: The idea is not to use an overall survival endpoint, as with advanced cancer trials, but a surrogate endpoint; in this case it would be the pathological complete response to identify drugs that can benefit patients?
Dr. DeMichele: Pathological complete response is really the surrogate that has been the most extensively studied and looks promising, but it is only really useful in certain kinds of breast cancers. So, what is pathological complete response? When we talk about this, pathological complete response, it is really what we see in the specimen that has been removed from the breast. The patient has completed their therapy, and the surgery goes in and removes the area where the tumor was. The pathologist looks at it under the microscope and makes a determination of whether or not there are any residual cancer cells there. If there are no residual cancer cells seen, then that is called a pathological complete response. It is really the eradication of all of the tumor cells that we can see as a result of the therapy that was given before surgery. It has been shown that this outcome has been highly associated with a good long-term outcome—that is, freedom of relapse and longer survival, but only in certain kinds of breast cancer.
For triple-negative breast cancer and HER2-positive breast cancer, this seems to be an accurate predictor of whether or not a patient will do well. In estrogen-receptor–positive breast cancer, however, tumors paradoxically are slower growing and tend to be less responsive to chemotherapy, and yet patients do very well. The correlation with attaining a pathological complete response and actually having a good outcome doesn't actually hold very well for most patients with estrogen-receptor–positive tumors. So, we see the drugs that are most likely to take this neoadjuvant approval pathway are really drugs that are targeted to patients with HER2-positive or triple-negative breast cancer, and we still have more work to do to figure out what are the appropriate surrogate endpoints for women who have estrogen-receptor–positive breast cancer.
Cancer Network: Are there other issues to consider when testing a new drug in neoadjuvant trials that we haven't talked about yet?
Dr. DeMichele: Well, I think that the other piece that we are starting to recognize now is that if we do more neoadjuvant therapy, and we learn that a tumor is responding or not responding, we have a whole new problem on our hands in patients whose tumors don't respond. So, a patient receives this therapy, they go for surgery, and there is actually a lot of tumor left. The question then is what to do after surgery? What you are seeing now—and what you will see in the next few years—is the development of what we call post-neoadjuvant trials (strategies to try to add additional therapies after surgery for patients who didn't get a very good response to the treatment that was given before surgery).
This is a whole other challenge that we are facing now: How do we design these post-neoadjuvant trials? I don't think those trials will really be focused on standard types of chemotherapies—because the data to date suggest that if the tumor was not particularly sensitive to chemotherapy before surgery, giving more chemotherapy after surgery is not going to be the answer—but because we have targeted therapies now, the efforts are really focused on looking at that residual tumor, figuring out what the targets are, and matching additional therapies to what is most likely to work in that tumor.
Cancer Network: Thank you so much for joining us today, Dr. DeMichele.
Dr. DeMichele: Thank you!