Vijayakrishna Gadi, MD, PhD: That first patient set the case for a hormone receptor-negative patient. Let’s talk about hormone receptor positive. We have 2 targets there, and they seesaw back and forth. One of the things I did at the San Antonio Breast Cancer Symposium this year 2021, that was kind of pleasant because very few people were there, I was able to really walk through the posters and read them. One that caught my eye was an interesting one, I think from Wisconsin, looking at patients who had seen preoperative endocrine therapy for hormone-positive, HER2-negative disease. And during that portion of time, which we all did with COVID-19, because we wanted to bridge people so that they could get to surgery safely, they started looking at HER2 expression after that exposure. They saw that in half the patients, HER2 expression got upregulated pretty meaningfully to 2+ and in some cases 3+. It makes me wonder when we have poor responses to endocrine therapy, is it because they are recruiting HER2 and could you actually target that? Obviously, they didn’t address that, but it just raises the question of how intricately tied in some patients, perhaps as many as half the patients, how ER [estrogen receptor] and HER2 are tied together. With that kind of framework, Chuck give us a quick overview of how you see hormone receptor positivity and HER2 interacting.

Charles Geyer, MD, FACP: Clearly, when one goes up the other goes down, and so if you inhibit one you may release the other. There clearly is that issue in the double-positive, triple-positive patients particularly. That’s very interesting to start out patients who are zero and 1+, to see a little more staining. But in the card-carrying, HER2-positive with hormone receptor positivity patients, I think that’s an aspect of therapy we’re going to have to think more about, particularly in more advanced lines of therapy. The triple-positive and double-positive—we probably need a better term for it—tumors are a much more heterogeneous group than their ER-negative counterparts. The intrinsic subtyping shows you that two-thirds of these things are luminal, split between the As and the Bs. As long as HER2 is overexpressed, it’s clear that it is the dominant driver, and we need to target that. We are not really seeing evidence that there is differential response or benefit based on hormone receptor status. The hazard ratio when you look at all these agents, you don’t really see reduced efficacy of HER2-directed therapy in ER-positive patients. You see that prognostically the ER-positive patients have a bit of a better prognosis, so to me it’s going to be where does one start trying to intervene with that additional question? As it gets more into, like I say the heterogeneity issue, more advanced disease, that’s where I use it because I tend to, still with metastatic patients I am treating the HER2. When I stop my cytotoxic, then I am thinking about should I as a maintenance try to modulate my endocrine receptor with something I haven’t done yet? That is what I do in my practice, but you really get off the evidenced-based reservation very quickly when you start doing those sort of things.

Vijayakrishna Gadi, MD, PhD: Chuck, it’s interesting you made that point about the data being thin about the differential responses in ER positive versus ER negative, but there is one set of trials, the ExteNET and NALA, where you see an interesting phenomenon around that. Obviously in ExteNET, it’s the combination being really effective in hormone-positive disease, and then in NALA where that wasn’t permitted, you lose that signal. So I think there are some indirect data out there about how to co-target this, especially with small molecules. Michelle, you brought it up earlier, so I am going to come back to you, which is this HER2 low and HER2 maybe heterogeneous. Tell us the modern thinking on that phenomenon and how we use these drugs, or maybe not use these drugs.

Michelle Melisko, MD: We’re very excited. As Chuck alluded to, trastuzumab deruxtecan is really disruptive in this space because it seems that it can be active across the spectrum of HER2 expression, and a couple of trials address that. The DAISY trial, a relatively small phase 2 study, looked at this agent across the spectrum of HER2 expression. And consistent with other trials, the response rates for the HER2 amplified, or 3+, were in the 70% range, but the magnitude of the benefit in the HER2 low, or 1+ and 2+, was about 35%. Then most interestingly, the HER2 zero, or HER2 null population, they even saw some responses in the 25% to 35% range, and the duration of response across these groups wasn’t tremendously different. I think it speaks to the point of the definition of HER2 positivity, that’s based on 1 biopsy at 1 point in time. Most trials demand that the biopsy be done either at the time of the enrollment, or maybe some archival tissue that was present in the last 12 months, but there may be evolution even among sites of disease for a particular patient, with one metastatic site being more HER2 positive or less HER2 positive.

I think that these antibody-drug conjugates, where they let go of their payload more easily, you won’t need to see as much of the particular target, if you will, of HER2. It goes across to other antibody-drug conjugates in the triple-negative setting, drugs targeting Trop-2, which gets back to this idea, are these actually super-targeted therapies anyway if they are not necessarily requiring expression of that marker? But going back to the question about different heterogeneity and looking at that ER-positive,HER2-positive population, and the sort of yin and yang, if you target one, the other one might become more important.

A trial we haven’t mentioned yet is the PATINA study, which is looking at patients who have received chemotherapy plus a HER2-targeted monoclonal antibody regimen, achieve stability, then are randomized to get hormone therapy alone or hormone therapy plus palbociclib. That will address the question I think about, if you’ve really tamped down the tumor and gotten control of the disease in someone stable, then perhaps allowing maintenance with HER2-targeted antibodies plus hormone therapy plus a CDK4/6 inhibitor, may put us in a different space, where those patients who got a CLEOPATRA-like regimen might end up, instead of having stability for 2.5 years, might have stability for 5 years. Again, this is not the purpose of this, but all of these regimens, these are all costly. You are adding, on top of the antibodies, on top of the hormone therapy, another drug, and the toxicity associated with it, so I think it is going be very interesting, that trial, to see how much adding a CDK4/6 inhibitor in this space adds.

There are other trials, we are focusing on the metastatic setting. But in the HER2-positive, ER-positive setting, looking in the neoadjuvant setting, can we get away from chemotherapy and use either an antibody-drug conjugate plus hormone therapy, or...just attacking the hormone pathway rather than using chemotherapy? The data that we have to date on a number of these trials, like the ADAPT trial, suggest that the path CR [pathologic complete response] rates are lower without chemotherapy, but those patients who achieve a path CR do very well. I think we might end up in the ER-positive, HER2-positive population using less chemotherapy in the neoadjuvant setting, and then trying to figure out how to add the hormone therapy back into the HER2-targeted therapies. We use a lot of cell-free DNA, as Nancy mentioned, in the ER-positive population to identify markers, but ideally we need to find the best way of measuring HER2 expression in circulating tumor cells, and really understanding how high does the HER2 need to be where you need to target that more than the endocrine? That’s a question that needs to be answered in the biomarker space, I think.

Transcript edited for clarity.