Transcript:Adam M. Brufsky, MD, PhD: Switching to a few more topics before we move on, the FDA now approved palbociclib with just about every hormonal agent. Is that correct, every AI? Does that have any clinical meaning or benefit to us, Debu?

Debu Tripathy, MD: Well, I think it gives us more flexibility. Every now and then, we have patients who may tolerate one AI better than another. Obviously, they have a lot of overlapping toxicities, but I think it makes it easier if some health plans make it harder to get one AI versus another. I think it makes things simpler. When the MONALEESA-7 trial data come out, we will actually have some data with tamoxifen, since that is allowed on the MONALEESA-7 trial, which is dedicated to premenopausal patients who all get gonadotropin analogs—so they do get ovarian suppression. But if they have not seen either endocrine agent, they could elect to take tamoxifen.

Adam M. Brufsky, MD, PhD: It’s physician’s choice in MONALEESA-7, correct?

Debu Tripathy, MD: Yes.

Adam M. Brufsky, MD, PhD: It will be very interesting to see because as I said before—and I think we all probably agree—it doesn’t matter what scenario you have, it doesn’t matter what endocrine agent you have, but it appears that if it has a hazard ratio of 0.5, you double it. That seems to be the rule of thumb. We’ll see if that ends up happening. Presuming all these patients are going on these agents—say you put someone on ribociclib and letrozole—but now it’s 2 years later and they progress. Kim, what do you do?

Kimberly L. Blackwell, MD: Well, I put them on a clinical trial, Adam.

Adam M. Brufsky, MD, PhD: If you can’t get a clinical trial.

Kimberly L. Blackwell, MD: Yes, because Carlos mentioned there are lots of trials that will help us understand: Should we continue, should we not? I think that’s one of the big unanswered questions here. Is it a similar phenomenon as we believe HER2-positive disease to be, which is that you’re taking your foot off the brake by stopping the CDK inhibitor? And that, we don’t know. In the absence of a clinical trial, I use single-agent fulvestrant after an AI or CDK inhibitor.

Adam M. Brufsky, MD, PhD: Carlos?

Carlos L. Arteaga, MD: Well, before choosing a therapy, I think that we should probably re-biopsy that patient and, in all possibility, re-interrogate that tumor.

Adam M. Brufsky, MD, PhD: For?

Carlos L. Arteaga, MD: Well, for a number of somatic alterations that may be associated with resistance or not. Think about it. These tumors have never been subjected to this Darwinian block in the history of the planet, and we know they’re going to progress. And they won’t do it by magic. They have to develop something, so I think we have an obligation as physicians to at least make an attempt to re-interrogate that disease, which is probably going to look a little different from the diagnostic biopsy and the primary metastatic biopsy. I agree with Kim after that: I think fulvestrant is a good choice. Fulvestrant/everolimus would also be a good choice or a clinical trial informed by whatever alteration we find when we biopsy.

Adam M. Brufsky, MD, PhD: I agree. Our lung colleagues have great, genomically driven therapy. Could this be the first instance where, potentially, we as medical oncologists who treat breast cancer do genomically driven therapy—particularly for ESR1 mutation? Around 30% of patients likely will have an ESR1 mutation on long-term AIs, right?

José Baselga, MD, PhD: Yes.

Adam M. Brufsky, MD, PhD: You were talking about interrogating the tumor. Should we just interrogate the plasma DNA or sequence?

Carlos L. Arteaga, MD: Both. Sometimes you may only be able to do plasma testing, but at this point, we cannot equate plasma testing with a tumor biopsy.

José Baselga, MD, PhD: So, I think that’s something that we should clearly do. ESR1 mutations—we now have a data set of over 2000 patients whom we had biopsied at progression of disease. In patients who have an ESR mutation, the progression-free survival to any AI is 3 months; if they don’t have an ESR mutation, it’s 14 months.

Adam M. Brufsky, MD, PhD: This is at progression?

José Baselga, MD, PhD: At progression. So, today, I think that there is a bounty of information that you need in order to make decisions. I still see, and we all still see, some patients being reexposed to an AI. Well, if they have a mutation of the ER gene, they’re not going to respond, so I think that’s something we need to take into account. Carlos is right: We have ER mutations, and we have transcriptional factor mutations that also influence endocrine resistance. We’ll talk about these later, but we have PI3-kinase alpha mutations that also play a role. We have something very interesting, which is ERBB2 mutation. And we have therapies for that. We have clinical trials going on, which are very exciting, with neratinib and other drugs. So, I think this is important, and it ought to be brought into clinical practice.

Adam M. Brufsky, MD, PhD: The question is, can we convince insurers to pay for this test, whoever’s guarding this $3000?

Debu Tripathy, MD: I don’t think we’re ready to apply it universally right now. Certainly, we are all using it for assignment to clinical trials. I totally agree with what you’ve said about the natural history of these patients, but we have to get some numbers. In other words, if you have an ESR1 mutation, what do you do?

Adam M. Brufsky, MD, PhD: He has data from…

Debu Tripathy, MD: No, I understand that. We know that AIs are not as effective, but we haven’t done controlled studies to see if we should move them over to fulvestrant. We’re doing one, too—a blood biopsy-based triage study, where ESR mutations may steer you toward a SERD versus not. I think that’s coming very soon, but we have to do our homework.

Adam M. Brufsky, MD, PhD: Right. It’s not standard of care by any means.

Kimberly L. Blackwell, MD: I’d like to offer a word of caution, which is strange coming from me. Because first of all, I’m all about sequential biopsies, profiling, and understanding what’s driving that tumor at the time of the progression. But we have to be a little careful, in that I’ve seen patients who’ve experienced this: I’ve done the Guardant360 test, done a biopsy, got a Foundation Medicine after patients have progressed, and then they see an ESR mutation and they switch the patient to chemotherapy. And I don’t think that that is the right thing to do.

Carlos L. Arteaga, MD: That’s not.

Adam M. Brufsky, MD, PhD: An ESR mutation and switching to chemotherapy? No, I wouldn’t do that.

Kimberly L. Blackwell, MD: Correct. But I’m just saying that it’s not like a mutation we typically think of as “You will not respond to endocrine therapy.” So, if you are going to be using these genomic assays, you have to be very cautious about how you’re going to use them, and the implications are really at a research level at this point. They’re informative, but I don’t think they should drive major treatment decisions.

José Baselga, MD, PhD: I think today, with the data that we have—and there are maybe 5 or 6 good papers published already—I agree with you, that with multiple genes, you have to be very careful. But in the case of ESR1 mutations, we have data that fulvestrant works better in these patients. And I think at this point—that particular gene—it’s enough information, at least for me, to dictate practice. I look at these data. I agree with you that there are many other mutations for which we don’t have enough data, but for that particular one, the data look pretty solid to me.

Kimberly L. Blackwell, MD: Yes. But I think clinically, though, in a patient you just described who progressed on an AI…

Adam M. Brufsky, MD, PhD: After 2 years.

Kimberly L. Blackwell, MD: We know that the benefits of fulvestrant are as good if there’s an ESR mutation, so I would switch the patient to that. What I’m saying is that there are so much data, and for the practicing oncologist, they see ESR mutation—and I just want to make certain we’re being very precise about this—but it doesn’t mean that they won’t respond. It just means that if they have the mutation, fulvestrant might be a more preferred agent. It doesn’t mean switching to chemotherapy.

Carlos L. Arteaga, MD: Maybe. I agree with what you’re saying, but there are 2 aspects of this. One is the decision for the patient. There is also this disease we’re creating: An ER mutation is a disease we created with aromatase inhibitors and estrogen deprivation. The point is that we’re going to be creating new diseases, and we’re going to have to write that book. And we don’t profile that disease. Now, that patient with the ER mutation may well respond to capecitabine, and I’m happy with that. But at least I know that the mutation occurred in that setting.

Adam M. Brufsky, MD, PhD: Well, I’ll tell you, the PEARL trial—which is a tough one—design is people who have progressed on an AI getting exemestane and palbociclib versus capecitabine. They may or may not be stratified for ER mutations, but 30% of the patients in that trial may not respond to the exemestane because of that very reason.

Carlos L. Arteaga, MD: The other thing is that CDK4 inhibitors are not the solution to ER mutations either. It was published that 20% of those patients, post-CDK4 inhibitor, will have ER mutations.

Transcript Edited for Clarity