As part of our coverage of the Miami Breast Cancer Conference, held March 8–11 in Miami Beach, Florida, we spoke with Elizabeth A. Mittendorf, MD, PhD, about the latest research on immunotherapy for breast cancer, which she discussed during a talk at the meeting. Dr. Mittendorf is a breast cancer surgeon at Brigham and Women’s Hospital and Director of the Breast Immuno-Oncology Program at Dana-Farber Cancer Institute, both in Boston.
—Interviewed by Anna Azvolinsky
Cancer Network: Several tumor types, including melanoma and renal cell carcinoma, have been known for a long time to be particularly amenable to immunotherapy. However, preclinical and clinical studies indicate that, in comparison, breast cancer appears to be less responsive to the immunotherapy approaches that have been tried so far. How do you view breast cancer in the context of other malignancies, with respect to tumor responsiveness to various immunotherapy agents?
Dr. Mittendorf: As you mentioned, a number of tumor types that are considered prototypical with respect to being responsive to immunotherapy. To just take a brief step back, immunotherapy is actually a very broad term; it really is defined as treatments that modulate the patient’s immune system to fight disease. There are a variety of relevant treatment modalities—vaccines; adoptive cellular therapy; and, more recently, chimeric antigen receptor, or CAR T-cell, therapy and of course antibody checkpoint therapy, using agents that target the inhibitory molecules of T cells.
When we say that tumor types such as melanoma are more responsive to immunotherapy, what we are really talking about is the observation with respect to checkpoint blockade. The thought is that checkpoint blockade agents take the “brake” off T cells, and so one might very simply hypothesize that this type of therapy would be most effective in tumors infiltrated by a lot of T cells. In fact, that is the case with melanoma. It’s thought that melanoma might be highly infiltrated by T cells because this disease results from mutations caused by exposure to ultraviolet light. These mutations result in what is referred to as neoantigens and these are viewed as foreign by the immune system and so the T-cells recognize them and come into the vicinity of the tumor.
Now if we think about breast cancer, this tumor type is not as commonly caused by mutations. So we don’t have as many neoantigens and we certainly know that breast tumors don’t have as many of these T cells present within the tumor microenvironment. To your point, there are data suggesting that breast tumors have less of a T-cell infiltrate. However, as I mentioned in the talk I gave here in Miami, there are signals or suggestions that breast tumors are, in fact, immunogenic. Breast tumor–associated antigens have been identified [in cases where] there are not a lot of T cells in these tumors, but there are some T-cells present, and there are strategies to augment those T cells. In fact, one of our primary therapies in HER2 [human epidermal growth factor receptor 2]-positive breast cancer is the monoclonal antibody trastuzumab, and that is an immunotherapy because one of its mechanisms of action is antibody-dependent cell-mediated cytotoxicity. So I think there is growing preclinical and clinical evidence to suggest that breast tumors, although perhaps [characterized by] less immunogenic endogenous [ligands], can be made more immunogenic and therefore susceptible to these treatments.
Cancer Network: What have we learned in the last few years about the biology of breast cancer and its interaction with the immune system that has been helpful or is potentially helpful in the development of immunotherapies for breast cancer?
Dr. Mittendorf: In my talk here I focused on something we learned in 2017. I began by summarizing findings from what now a fairly large set of data looking at checkpoint blockade as monotherapy; if we look at the results in total, we see that the rate of response to this treatment is only about 10%. To your question, the other thing we learned in 2017 is that we can do better than a 10% response rate if we use combination strategies. Specifically, can we make these tumors more immunogenic to render them more susceptible to checkpoint blockade.
So, what have we learned? We’ve learned is that there are forms of chemotherapy, including agents commonly used in breast cancer, such as anthracyclines and taxanes, that can stimulate an immune response. Therefore, we can combine chemotherapies together with checkpoint blockade to get a response, and several large trials are now looking at this question. A very large phase III trial is addressing this question in triple-negative breast cancer, investigating the anti–PD-L1 [programmed death ligand 1] antibody atezolizumab. That study has completed accrual, and I think most of us in the field anticipate that we will learn the results within the next year or so.
Another really nice advance in the field is improvements in our pre-clinical models. One very specific example is that from a study by my colleagues at the Dana Farber, Shom Goel and Jean Zhao. In a very elegant paper published in Nature they demonstrated the impact of CDK4/6 [cyclin D–dependent kinase 4/6] inhibitors (abemaciclib and palbociclib) on [increasing] the immune response in a mouse model of breast and other solid tumors. CD4/6 inhibitors are now becoming commonly used in the treatment of patients with metastatic hormone receptor–positive breast tumors, and I will point out that these HR-positive tumors are the ones that are most immunologically cold, meaning they have the lowest levels of immune infiltrates. What these investigators showed is that CDK4/6 inhibitors enhance antigen presentation, so if more antigens are being presented to the immune system, there will be more of an immune response. They also demonstrated that these CDK4/6 inhibitors decrease the level of regulatory T cells, which are the immunosuppressive T cells—“bad” T cells that are actually more protumor.
Based on these data there is an ongoing trial at Dana Farber studying the combination of these CDK4/6 inhibitors and immunotherapy. Data from another small study were presented at the San Antonio symposium in 2017 by Hope Rugo, of the University of California San Francisco, which suggested that the combination is safe and may have activity. Another example is nice emerging data on PARP inhibitors, which are now approved for use in breast cancer patients with BRCA mutations. Just saying the word “mutations” makes me think of neoantigens, so perhaps BRCA-mutant breast cancer is a type that will become more immunogenic [with the right treatment]. There was a nice study presented by Susan Domcheck of the University of Pennsylvania, also at San Antonio in 2017, that showed signs of efficacy from combination of PARP inhibitors with checkpoint blockade. So I think it’s an exciting time in that we are getting more models that we can use preclinically to inform our strategies, and as we are able to look at these combinations in patients, [we will gather] more clinical evidence of potential strategies that could be effective.
Cancer Network: Are there any additional recent preclinical or clinical immunotherapy approaches for the treatment of breast cancer that you feel are particularly innovative or promising?
Dr. Mittendorf: Another subtype of breast cancer that we deal with a lot is HER2-positive breast cancer, and there have has been some data recently suggesting that there might be a benefit to giving these patients immunotherapy. This is something that has been “near and dear” to my heart, since my own academic program has largely been investigating vaccines that target HER2 tumors. Our group recently published some nice data demonstrating a potential mechanism of synergy between HER2-encoding vaccine—meaning one that will elicit a CD8+ T-cell response against HER2—in combination with trastuzumab. That leads me to other exciting strategies that are being investigated in clinical trials. In addition to multiple combinations of our known and standard chemotherapy or targeted-therapy agents, there are other trials coming online that are looking at vaccine studies to try to make “cold” tumors “hot”; that is, to stimulate an immune response that could then be augmented with checkpoint blockade. Another investigator, Heather McArthur from Cedars-Sinai Medical Center in Los Angeles, published a study that builds on preclinical work from Jim Allison showing that if you cryoablate, basically make a tumor into an icicle, that that is immunogenic and the effect can be augmented with checkpoint blockade. So Heather is currently working toward a phase II study based on her successful phase I study showing that the combination is safe. There are a number of investigators interested in looking at something referred to as the abscopal effect, meaning could we use radiation in combination with some of these immunotherapies? So those are forms of combination therapies that incorporate augmentation of checkpoint blockade, which I referred to in my talk here as trying to make a “cold” tumor “hot.”
Cancer Network: Lastly, what are the big questions in the field regarding the biology of breast cancer as it relates to the immune system and responsiveness to immunotherapy?
Dr. Mittendorf: The biggest question is one that we have already been discussing: How do we make immunotherapy a more effective treatment for breast cancer? I spoke broadly about current efforts to turn “cold” breast tumors into “hot” tumors, but I think it will be a much more intricate story than this. I think the greatest challenge will be to determine, after we stimulate an immune response, how we can get that immune response to actually be specific for the tumor and then how we get the T cells to track into the tumor. I think the other real challenge is the issue of how to address the multiple immune-suppressive mechanisms that are present in the tumor microenvironment. Even if we get T cells to the tumor, the microenvironment may be a hostile one for these cells. There is tremendous opportunity, and to advance this field will require taking observations made at the lab bench and bringing them into the clinic. In addition, when we conduct immunotherapy clinical trials, the responsibility is ours to collect biospecimens from patients and take these back to the laboratory, so that we can learn as much as possible to inform the next round of studies.