Neratinib and HER2 Activating Mutations in Breast Cancer

Transcript:Adam M. Brufsky, MD, PhD: So, three quick things with HER2 before we move on. The first is, what do people think about HER2-activating mutations? Are they going to be an important issue? Because there are data actually with neratinib that when you have a mutation in the HER2 kinase domain that seems to be activated, then neratinib—at least in this phase I BASKET trial that they’re doing—seems to have activity. Do people think that it’s going to be an important thing to look at going forward?

Carlos L. Arteaga, MD: Absolutely, absolutely. And that’s one thing, actually, that you can also pick up in a liquid biopsy. So, the important thing also for people to realize is that these mutations happen in patients that are HER2-negative.

Adam M. Brufsky, MD, PhD: Right, that’s very important.

Carlos L. Arteaga, MD: This is another mechanism by which cancers are served, that HER2 locus, and mutated for a gain of function. Many of these mutations are activating, not all of them, and we now have data that some of them are drivers in that they respond very well to neratinib—which, to me, is a clinical proof of concept. They are driving the cancer at some level. So, I think they’re real. I think that not all mutations are going to be equally sensitive to neratinib or to other irreversible inhibitors of HER2. There may be tumors, like non-breast cancers, where the response may not be as great just because these cancers may occur in tumors that are a lot more complex and may not respond to single-agent neratinib. In those, we would have to think of combinations. And, in fact, in the breast cancers that are ER-positive, the majority of these mutations occur in ER-positive tumors. The combination with fulvestrant is showing great, interesting activity.

Adam M. Brufsky, MD, PhD: My understanding, in some data that I’ve seen, is that lobular cancers, for some reason, may seem to have an increased incidence.

Joyce A. O’Shaughnessy, MD: And this is unlike the ESR1 mutations; this is one of the things that I go after in the metastatic tumors, even, and especially in my practice with the lobulars in the peritoneum. The lobulars are highlighted in the Clinical Cancer Research, Jeff Roth’s paper; 27% of metastatic invasive lobular tumors had either activating mutations or amplicons, or a few truncations. I find that they respond in the metastatic setting to anti-HER2-based therapies, although fulvestrant/neratinib sounds really intriguing. But I’m also getting coverage for anti-HER2—based therapy, even for these activating mutations. But you also find amplicons. But the lobulars, in particular, are highlighted. They’re in ductal cancers, as well. If I need to go to general anesthesia to get a good peritoneal biopsy—and I’ve never done that before because you can’t really safely do CT-guided biopsies to easily get to the peritoneum—I get enough tissue.

Adam M. Brufsky, MD, PhD: So, instead of biopsying, would you then do a liquid biopsy instead?

Joyce A. O’Shaughnessy, MD: I’ve done that, but I’m having a hard time interpreting what these teensy little percentages are; it’ll come back saying a 0.6% of the cell-free DNA is a HER2 mutation. I’m not sure what that means.

Kimberly L. Blackwell, MD: I think the important point for the practicing clinician is that you have to do the testing on the metastatic sample.

Carlos L. Arteaga, MD: Yes.

Kimberly L. Blackwell, MD: I see a lot of people sending either the primary, because they don’t re-biopsy, or the first recurrence core. And then the other important point is some of these mutations will be missed on some of the more widely available commercial assays. So, if you really are interested in this, you either have to do a comprehensive exome sequencing through a commercial vendor or you need to enroll a patient—which there’s plenty out there—for the full sequencing of the HER2.

Carlos L. Arteaga, MD: That’s a very important point. In fact, some commercial vendors report some of these mutations as variants of unknown significance. It doesn’t mean they don’t have significance because they don’t occur necessarily in the same spot, so we’re discovering some of these reading errors.

Adam M. Brufsky, MD, PhD: Unknown significance of people interpreting them, but they’re significant to the patient.

Carlos L. Arteaga, MD: Exactly. And colleagues are throwing them away in some cases. In fact, we’re finding that some of these variants do make patients qualify for enrollment in some of these basket trials.

Kimberly L. Blackwell, MD: One of my colleagues actually had a patient in that situation. She’s one of the durable responders on a single-agent, neratinib study. So, I think, we’ve just got to be educated about what a mutation is. The point being is there’s lots of clinical trials where full sequencing is made available.

Carlos L. Arteaga, MD: For many genes, we are in the learning phase, so that’s the truth of the matter.

Adam M. Brufsky, MD, PhD: But, again, my own suspicion for the practicing clinician is probably going to be getting better liquid biopsy technology because that seems to be an easy thing to do. You can draw someone’s blood. You don’t have to worry about doing peritoneal biopsy under general anesthesia. I suspect—at least in my opinion, I don’t know what other people think—that’s where we’re going.

Carlos L. Arteaga, MD: We are going there. I think we have to be very careful of generalizing the approach. I believe that to be able to screen most somatic alterations that are actionable in a liquid biopsy would be great progress. Amplifications, we’re not there yet. Fusions, we’re not there yet.

Adam M. Brufsky, MD, PhD: I agree. We need better technology for that.

Carlos L. Arteaga, MD: Single nucleotide-variant mutations, yes.

Adam M. Brufsky, MD, PhD: Yes, I agree.

Carlos L. Arteaga, MD: In the SOPHIA trial, the mutations in ER correlated with a shorter PFS of patients on exemestane. And what was fascinating was that regardless of the allele frequency and the abundance, the percent that that mutation was in the total tumor DNA, regardless of that being itty-bitty or huge, the impact was the same.

Adam M. Brufsky, MD, PhD: It means that you had a clone that was resistant that eventually would do something.

Sunil Verma, MD, MSEd, FRCPC: And it’s really interesting when we were talking about the ESR mutations, data that were presented yesterday, they were able to predict 4 months earlier that these patients will progress by the occurrence of those mutations early on. So, to your point, if you can actually identify these patients early and project how they’re going to be doing, whether you then need to start monitoring these patients more closely to make those treatment decisions earlier may be very helpful. But I think, yes, there needs to be some kind of serial user-friendly component for us to be able to gauge treatment evolution, treatment resistance. And I think liquid biopsies are there but we’re not there yet, but that’s the way we’re going to get there.

Transcript Edited for Clarity