Unmet Needs in HR-Positive Metastatic Breast Cancer - Episode 4

HR+ mBC: Treatment After Progression on CDK4/6 Inhibition


Andrew D. Seidman, MD: So, this also begs the question of, do we have molecular markers that can help us decide when endocrine resistance has occurred? Can you talk a little bit about what we know, what’s the state-of-the-art right now in terms of thinking about endocrine resistance at the cellular level? And are there tests that doctors should think about, or is it they’re not ready for prime time?

Joyce A. O’Shaughnessy, MD: So, there’s certainly de novo and primary and then acquired resistance to endocrine therapy. Unfortunately, there is a subset of these aggressive ER-positives, luminal B. We can see them because sometimes they’ll be either low PR or they’ll be PR absent, and they’ll be more highly proliferative. A lot of times they’ll be grade 3 on the primary. If people get Ki-67s, so if you see a Ki-67 getting up there—25% to 30% or more—you’re really into a breast cancer that’s not likely to have a durable benefit from endocrine therapy alone. The underlying molecular alterations are varied, like P53. FGFR1 amplification is a very poor. That has been really proven by Nick Turner and others, that with predictions in the clinic for endocrine therapy resistance, cyclin D1 amplifications are some of the key ones we see. There are a number of others. There’s no way that we don’t assay for those, as you know, up front, Andy. We kind of go by natural history, the short disease-free interval, just the tempo of disease. And the progesterone receptor can be somewhat helpful, and Ki-67, although in the metastatic setting, we don’t typically get the Ki-67s.

Andrew D. Seidman, MD: Are your patients in Dallas asking you to send their blood for circulating tumor DNA to look for ESR1 mutations, as they sometimes do in Manhattan?

Joyce A. O’Shaughnessy, MD: No, come to Dallas. No, they don’t. I will tell you what I do is look for HER2 mutations. ESR1, the estrogen receptor, mutations that really are, unfortunately, quite common in the setting of resistance to aromatase inhibitors, are typically in the metastatic setting under that selective pressure. We don’t have any clinical good algorithms that follow from an ESR1 mutation, in my opinion. I don’t think the data are firm to tell us what to do.

Andrew D. Seidman, MD: They’re early.

Joyce A. O’Shaughnessy, MD: They’re early yet, and we know mTOR inhibitors work, we know CDK4/6 inhibitors work. But the HER2 mutations are the thing that I will get either a biopsy and look for with next-generation sequencing or a ctDNA, and I’ll periodically look for HER2 mutations because they are not 100%, but they are largely mutually exclusive with ESR1 mutations early on. Later in the course of the disease, when you have multiple, multiple subclones, you can see different subclones with HER2 mutations and ESR1 mutations but not in the same cell, because they’re not necessary. But if somebody does not have an ESR1 mutation, I tend to keep looking for it, because it can lead you down a whole other pathway of therapy for patients.

But in the acquired setting, there’s remodeling of the cells. When you take away the estradiol ligand, the cells can make heregulin, which then, in an autocrine fashion, signals through HER2, HER3. The whole HER family comes upregulated; IGF1R, the insulin-like growth factor receptor, comes up. They all cross-talk with each other. They’ll use whatever ligand is around, basically. So, the cells just remodel to signal down, and then they’re all impinging on the cyclin D1. They all impinge down on the CDK4/6 proteins, which then drive the cell cycle. It’s very remarkable. Whether it’s a primary resistance with the FGFR1 amplification or upregulation of the HER family in response, it all comes down to central nodal point of cyclin D1, CDK4/6, which is why these agents are so useful for patients, basically.

Andrew D. Seidman, MD: Yes. It certainly will be interesting if, and perhaps when, PI3-kinase inhibition is ready for prime time, because it’s going to expand how we interrogate breast cancers, for sure.

Joyce A. O’Shaughnessy, MD: Yes, yes, the whole PIK3CA mutation, which is 40% of ER-positive/HER2-negatives. Interestingly, in the adjuvant setting, that predicts for a favorable prognosis, because it’s a real luminal A—type biology. But in the metastatic setting, when you’ve taken away estradiol as a ligand, the PIK3CA-mutant cancers can make heregulin, and then that’s how it remodels up through the HER2, HER3. And so, it would be lovely to have a specific inhibitor of a PIK3CA, which we can talk to. So, here we are with endocrine therapy resistance.

Andrew D. Seidman, MD: Right, and still we’re just testing for ER and PR at this moment in time, and everything else is…

Joyce A. O’Shaughnessy, MD: Right. Watch that space closely.

Andrew D. Seidman, MD: Right.

Joyce A. O’Shaughnessy, MD: Right, exactly, exactly.

Andrew D. Seidman, MD: Right.

Joyce A. O’Shaughnessy, MD: So, what do we do after CDK4/6 inhibitor progression?

Andrew D. Seidman, MD: This question makes me think back to, what do we do after progression on trastuzumab in HER2-positive breast cancer? And I remember once upon a time my colleagues said, “Well, we’ve just got to continuously suppress HER2,” and I said, “Well, show me the data,” and there weren’t any data. And I tend to be more conservative in my oncology and maybe more liberal in my politics. But I will say that I personally am happy to wait for the data. My practice is, in the absence of data—and there will be data from trials—I tend not to continue CDK6/4 inhibition beyond progression and, for example, combine it with yet another antiestrogen therapy. I think it’s a great question for a clinical trial. There is a certain toxicity, and I’m not so sure what the benefit is, and maybe the believers will say I told you so when they have the information, but currently, we don’t.

The 1 scenario where I think it’s fair—and granted, MONARCH 1, the abemaciclib monotherapy trial, did not include CDK4/6 inhibitor—experienced patients—if I have a woman who has had a CDK4/6 inhibitor with their initial endocrine therapy, perhaps has progressed, has been on other therapies, I am inclined to use abemaciclib monotherapy. I don’t think that’s crazy, and, again, those patients weren’t enrolled; they weren’t allowed to enroll in MONARCH. There probably weren’t that many of those patients around at that point, anyway.

Joyce A. O’Shaughnessy, MD: Right. Have you seen any responses or stable disease?

Andrew D. Seidman, MD: So, I think my sample size is 2 or 3, and actually 1 of them is a male breast cancer patient who has actually had a very nice durable response.

Joyce A. O’Shaughnessy, MD: With abemaciclib?

Andrew D. Seidman, MD: With abemaciclib monotherapy, yes.

Joyce A. O’Shaughnessy, MD: Wow, interesting; cool, interesting.

Andrew D. Seidman, MD: So, yes.

Joyce A. O’Shaughnessy, MD: I do exactly the same thing, Andy. I’m not continuing the CDK4/6 inhibitor where there has been progression on, even though there has been a lovely response. I’m not continuing it. I think probably more than anything because of what we’re kind of learning about some of the mechanisms of resistance, it doesn’t exactly make sense to me biologically to continue it. Of course, we have no clinical data, as well—for example, upregulation of CDK2 and cyclin E, for example, or loss of RB. Sometimes though there are preclinical data of upregulation of the HER family pathway. And then you say, “Well, maybe I’ll keep that going then and add something else.” So, for example, the triplet of ribociclib/letrozole/everolimus. You can just get a small dose of everolimus in with that triplet.

Andrew D. Seidman, MD: Right, because the PK interaction.

Joyce A. O’Shaughnessy, MD: Yes, PK interaction, and they are just early phase I/II data, but the waterfall plot looks pretty nice. But those patients—now they’re studying patients who are progressing on a CDK4/6. The phase I was in a more heavily pretreated population that not necessarily had prior CDK4/6 exposure. So, that’s a triplet of great interest, but I haven’t done that in practice; but I think that’s an interesting evolution.

Transcript Edited for Clarity