Expanding Therapeutic Options for Breast Cancer - Episode 2
Joyce A. O’Shaughnessy, MD: Hope, in the frontline setting, most of the studies have focused on aromatase inhibitors, but I guess there are some emerging data with fulvestrant here. Are we going to have a choice in the future?
Hope S. Rugo, MD: Because of FALCON, we have approval to use fulvestrant in the first-line setting, particularly in patients who are endocrine therapy—naïve. But the idea of what endocrine therapy is optimal is still a question. The data from MONALEESA-3, presented at the 2018 ASCO Annual Meeting by Dennis Slamon, are actually quite intriguing. That design allowed patients in the first- or second-line setting to be randomized to fulvestrant and placebo or fulvestrant and ribociclib. The randomization was 2:1, as we’ve seen before. The trial was fairly large—726 patients were randomized. What they saw was quite interesting. If you line up all of the studies on a table, all of the hazard ratios are the same. It’s the same in this trial population. Of course, there’s a big benefit. We already knew that this was the case in the second-line setting. What’s intriguing about this study is that the hazard ratio in the first-line setting is also identical. The trial wasn’t designed to say whether or not it’s better to give a CDK4/6 inhibitor or not in the first-line setting, specifically because both groups were included. But those patients who received a CDK4/6 inhibitor definitely did a lot better—again, with these hazard ratios of 0.56, around which we’ve seen across all of the trials.
I think that you could give fulvestrant. I’ve already done that in clinical practice. But somebody who has polymyalgia rheumatica can’t possibly take an aromatase inhibitor. I’ve used fulvestrant and a CDK4/6 inhibitor in a patient with de novo metastatic breast cancer who is premenopausal with extensive liver metastases. She’s been on therapy for 2.5 years. So, this is definitely a therapy that can be considered, and I think the data from this most recent MONALEESA trial will help to encourage that. Then we’ll see more data with fulvestrant with or without a CDK4/6 inhibitor versus an aromatase inhibitor in a European trial—fulvestrant versus aromatase inhibitor; both arms with a CDK4/6 inhibitor. Maybe that trial will actually help us to answer the question of which therapy is optimal.
Sara A. Hurvitz, MD: You brought up a really interesting point about the choice between fulvestrant and an aromatase inhibitor. I think we all feel like fulvestrant is probably the best drug in terms of frontline therapy, but the studies weren’t designed like this. What are you doing in your patients who’ve had 5 years of adjuvant aromatase inhibition, who come in a year or 2 or 3 later with a recurrence? Are you testing for ESR1 mutations to choose the endocrine partner? This is coming up in the clinic now. A lot of these patients will have ESR mutations, which would predict for a better outcome with fulvestrant.
Debu Tripathy, MD: It’s interesting. The trials that defined first and second line used a 1-year interval from having completed an aromatase inhibitor, but the Europeans have recently looked at the data to make a more objective definition of when your phenotype looks more like endocrine-resistant versus sensitive. They looked at all of the data and where the cut points were before starting first-line hormonal therapy, and they came up with a 2-year cut point. I tend to use that if I’m going to use aromatase inhibitors in the first-line setting. I’m going to expect at least a 2-year…but, to be honest with you, I feel like you do. I think for anybody who’s progressed on 5 years of an aromatase inhibitor—for example, now that we have the data from monotherapy, as well as a hint from CDK—I would prefer to go with fulvestrant. And yet, we all know that ESR mutations now, as more data come through, probably predict for some degree of resistance to aromatase inhibitors and sensitivity to fulvestrant. It may not be 1:1, and each mutation is different. It’s very complicated. But I agree with the general sentiment that fulvestrant is moving in.
Hope S. Rugo, MD: You have to keep in mind that women are getting an injection every single month. They have to coordinate it with their travel. It’s 2 injections. There’s always going to be a role for using aromatase inhibitors. More than 2 years out, I tend to retest with an aromatase inhibitor. Also, the idea is that if you save fulvestrant, you can combine it with another agent. There are lots of trials going on with fulvestrant with a variety of agents, and we’re going to see a lot of information on PI3-kinase inhibitors in combination with fulvestrant. Order and deciding what to do is going to become increasingly complex.
Joyce A. O’Shaughnessy, MD: The NCCN guidelines support fulvestrant and a CDK4/6 inhibitor as a frontline option, right? That’s very, very important for the practicality, obviously, and for being able to access it when you want to. Debu, is there any role for fulvestrant monotherapy in patients?
Debu Tripathy, MD: I do think there may be a minor role for endocrine monotherapy. In patients who have significant comorbidities or patients who are just not able to travel for the frequent blood checks, personal preferences may come into play. It may be reasonable to start with endocrine therapy. We don’t have many predictors of who can do well. The MONARCH study for first-line treatment did identify a subgroup of patients who were disease-free for more than 3 years. It was found that they have an excellent outcome either way. The event rate was low. Apart from that, I don’t think that we can identify someone up front. There are some studies planned for low-risk patients, where they will start with a hormone therapy versus both together and then have a CDK added to the monotherapy group. They are going to compare those outcomes. We may get some more information on that later.
Hope S. Rugo, MD: The subset analysis with abemaciclib—I haven’t put a lot of stock into it because they didn’t reach the endpoint. When you look at the confidence intervals, it wasn’t reached on the upper end. It was way too early to understand that those people didn’t benefit. We did an analysis in PALOMA-2. When we did the last progression-free survival assessment, there was no difference. We looked at all sorts of disease-free intervals. There was no difference in benefit. Sometimes I wonder if you take the patients who have the most endocrine-sensitive disease, are those the patients who are going to benefit even more and stay on therapy for a long, long time? This is a whole other issue, in talking about expense. I don’t know that it’s really an indicator right now for deciding on treatment.
Komal Jhaveri, MD, FACP: I would agree. The largest follow-up that we have is with the PALOMA studies, and those forest plots look identical for patients regardless of that.
Joyce A. O’Shaughnessy, MD: So, there are no clinical predictive subsets.
Komal Jhaveri, MD, FACP: No clinical predictive subsets.
Joyce A. O’Shaughnessy, MD: Every subset benefited. Are there any predictive markers at all that are emerging, Hope, in tissue or blood?
Hope S. Rugo, MD: There are predictive markers that are intriguing, but they aren’t perfect. One study that I thought was quite interesting was an analysis of PALOMA-2, which Richard Finn presented at San Antonio. We had looked at many different markers and had not seen anything that predicted, just as we have for all of these other targeted agents. But what was interesting, looking at intrinsic subtyping, was that the patients with luminal A and B disease benefited a lot. There was a small number of patients who had the bad intrinsic subtypes, and they did very poorly. They have up-front endocrine-resistant disease. It suggests that, again, the endocrine sensitivity is our best predictor.
There have been a number of different studies that have been interesting recently. One that’s intriguing is an abstract presented by Nick Turner at the 2018 ASCO Annual Meeting. This looked at, in PALOMA-3, the second-line study with fulvestrant and palbociclib, cell-free DNA as well as archival tumor tissue. It showed something that we really haven’t seen before—RB mutations emerged at the end of treatment, even though it’s extremely uncommon up front. I’ve seen a couple of patients with RB mutations—I’m sure we all have—who had up-front resistance to CDK4/6 inhibitors. It’s not something that we would look for, because it’s so uncommon. But what they saw was that a small number of patients who didn’t have the mutation in the beginning had it at the end. I think it’s encouraging for us that it’s under 5%. It was about 4.8%. We’re always going to see the emergence of resistance under the pressure of treatment. How rapidly resistance develops is important to us, so seeing that this occurs in a small number of patients is really important.
There’s always been a number of hypotheses suggesting that certain changes in cyclins will impact sensitivity to CDK4/6 inhibitors. Fabrice André presented data at AACR. Patients who were treated in MONALEESA-2 were looked at and, it was suggested that cyclin E1 might make a difference in terms of up-front resistance. But regardless of baseline factors, similar to other studies, with the sort of standard things we’d look at, like Ki-67, a doctor might say, “Oh, your tumor is highly proliferative. I should give you chemotherapy.” Those patients still benefited from a CDK4/6 inhibitor. The other thing I’ve noticed is that people worry about CCND1, or cyclin D1, gene expression levels. Again, similar to PALOMA-1, there was no difference at all in this really detailed analysis. Cyclin E alterations are, again, uncommon. It’s not definitive. That study also concluded that hormone receptor status and level of sensitivity to hormone therapy are really key in terms of sensitivity to CDK4/6 inhibitors.
Transcript Edited for Clarity