Side effects associated with PARP inhibitors

DR ROBSON: There clearly is myelosuppression. And it appears to be that anemia and thrombocytopenia are the most clinically relevant. I think you do see sometimes a little bit of leukopenia and neutropenia, but febrile neutropenia is vanishingly rare. And it’s also extremely uncommon to have to delay or reduce because of leukopenia.

The challenge has been that many of these patients groups have been pretty heavily pretreated. And particularly in the ovarian space, they’ve been heavily pretreated with platinum, which, of course, as essentially a hematopoietic stem cell toxin is kind of a little bit of an issue. And so I think one of the things that moving forward we need to sort out is, what are the predictors of clinically significant anemia, clinically significant thrombocytopenia? Is prior exposure to platinum, particularly in the breast space, is that a relevant predictor or not?

In OlympiAD, yes, we had a fair amount of anemia. We had a significant amount of Grade 3 anemia, like 16% Grade 3 anemia. About 18% of the patients wound up getting a transfusion at one point. Whether that was truly because it was clinically significant anemia or because there was a hemoglobin requirement to continue on trial and people were just trying to push people up so that they didn’t have to interrupt the drug, I don’t think we know that yet. And it will be interesting to see whether or not, especially in the breast space, there’s an overlap with prior platinum exposure.

DR LOVE: What about these GI issues?

DR ROBSON: Yes. So, I mean, I can only speak to it from the breast space, but most of the nausea and vomiting was actually Grade 1. And, I mean, you’ve done clinical trials like I’ve done clinical trials. If you’re sitting there doing your tox assessment and have you felt for the last 4 weeks have you ever been nauseated, yes, I had some nausea. And they’re Grade 1 nausea.

I have personally had some patients, though, that had pretty significant nausea and required antiemetics. It tended to get better with time. And dose reductions and dose interruptions were relatively uncommon on the OlympiAD trial. And dose discontinuations for nausea were essentially nonexistent.

DR ROBSON: There are certainly patients who have nausea and need antiemetics. But it wasn’t that much of an issue for most of my patients.

DR LOVE: That’s good to know.

Efficacy of olaparib for patients with BRCA germline-mutant metastatic triple-negative breast cancer (mTNBC)

DR LOVE: Why don’t we just walk through some of what we know right now in terms of clinical research with PARP inhibitors, and particularly with the OlympiAD trial. And maybe you can just summarize what we know about olaparib, specifically in breast cancer, BRCA germline and other, prior to the OlympiAD data.

DR ROBSON: Yes. The answer is, not much. There were 2 basically small trials. There was the ICEBERG trial, which was Andrew Tutt’s study. And that was a Phase II study looking originally at 400 mg twice a day, which was the now FDA-approved dose in patients with germline BRCA mutations. And that had about a 40% response rate, just overall response rate. And then there was an additional cohort that was added. A 100-mg BID cohort was added. It was not a randomized Phase II dose comparison trial, but the response rate was lower in that arm with, again, a relatively small number of patients.

Then Karen Gelmon did a study in triple-negative breast cancer, including BRCA mutation carriers. And in that study, the patients who did not have a BRCA mutation essentially did not respond. And the patients who did had a decent response rate, but it wasn’t particularly well maintained in that study. And that was, aside from some Phase I data and then Study 42, which was Bella Kaufman and Susan Domchek’s study of patients who had been heavily pretreated, that was pretty much the sum total. So certainly there was indication of response, but again, heavily pretreated cohorts by and large.

DR LOVE: Of course, ovarian cancer is a very different disease than breast cancer, but really the incredible finding that came out last fall with niraparib, that basically people who are HR even negative benefited from the use of a PARP inhibitor in that situation really talked about the biology of ovarian cancer. Is there any reason to believe that PARP inhibitors would be helpful outside of people with BRCA germline or maybe other germlines, similar, in breast cancer?

DR ROBSON: So far the only data that really looked at that was Karen’s study. And it was a small trial that was not encouraging. I think the challenge with HR is measuring it. And so although I agree with you that the NOVA trial did show benefit in patients who didn’t have that particular biomarker, it was pretty clear that the benefit was greater, substantially greater, in people who had BRCA mutations, who kind of have the ultimate knockout of their HR capability, and was more intermediate in patients who were HR deficient by the assay.

So while yes there was a benefit, it was much more marginal. And you wonder if that was more a failure of the definition of HR capacity than it was necessarily benefit outside of that space.

DR LOVE: Although, I mean, the hazard rate was, like, 0.5, so — but, I guess, you could imagine within that group, there might be people who didn’t benefit whatsoever. But anyhow, I guess, again, high-grade serous is not the same as infiltrating ductal cancer of the breast.

DR ROBSON: No. And also, you have to remember that all those trials, all the ovarian trials really have been maintenance trials, right? So they’ve all been patients who are largely — I mean, rucaparib study, ARIEL, is a little bit different. But the other two, both NOVA and the olaparib trials, were patients who were predefined as being essentially platinum sensitive. So they were really selected for being — or at least not platinum refractory. Let’s put it that way. So they were kind of preselected for potentially being responsive.

DR LOVE: Oh. That’s a really good point. Like, maybe if you took breast cancer patients who responded to platinum, you’d see the same thing. I don't know.

DR ROBSON: That trial is being done, actually.

DR LOVE: Really?

DR ROBSON: Yes. Kim Blackwell and Rebecca Dent are doing that study.

DR LOVE: What’s the design?

DR ROBSON: To try to give a run-in of carbo. And then for responders, randomize to placebo versus continuing a PARP inhibitor.

DR LOVE: Wow!

DR ROBSON: So following a similar thing. And that’s a trial design that Andrew Tutt actually has been pushing for years. So I’m glad to see it’s finally getting done.

DR LOVE: Is that in triple-negative?

DR ROBSON: I believe it is in triple-negative.

DR LOVE: What would you guess just you would see in terms of response rate to a platinum, triple-negative?

DR ROBSON: I think Steve Isakoff’s data suggested around 20%, 15% to 20%.

OlympiAD: A Phase III trial of olaparib monotherapy versus chemotherapy for patients with HER2-negative metastatic breast cancer (mBC) and a germline BRCA mutation

DR LOVE: So anyhow, let’s talk about the OlympiAD trial. Maybe you can talk about the design of the study and what you found.

DR ROBSON: So this was a pragmatic trial, just because back when we were trying to figure out how to study PARP inhibitors in breast, we had a very, very difficult time coming up with something that would be acceptable to everybody. And certainly we couldn’t replicate the ovarian experience.

So we took essentially the EMBRACE design and took folks who had anthracycline/taxane-treated disease, HER2-negative, germline BRCA mutation and had 2 or fewer prior lines of chemotherapy, randomized 2:1 to the tablet formulation of olaparib versus physician’s choice of capecitabine, eribulin or vinorelbine, with the primary endpoint being PFS and targeting a hazard ratio of about 0.63. So we were looking for an improvement from a baseline of about 4 months. And the control arm was the predicted PFS.

And we had to get 230 events, and so we randomized 302 patients and wound up with what you saw, which is a hazard ratio of about 0.58 as a blinded independent central review.

DR LOVE: I think that one of the numbers that surprised me the most was the response rate.

DR ROBSON: Yes. Sixty percent response rate overall versus about 20% in the patients who were getting conventional chemotherapy. But I think when you look at the control arm, I don’t think any of us who treat a lot of breast cancer would be particularly surprised.

DR LOVE: No. Absolutely. No. Sixty percent — and were you expecting that?

DR ROBSON: No. That was better than we expected. I mean, remember, in the ICEBERG it was about 40% for the 400. So that was probably a slightly more heavily pretreated group of patients. But it was quite nice to see, 9% complete response rate and, importantly, no difference in response rates between the patients who had either triple-negative or hormone receptor-positive disease, which was a nice backup to the subset analyses.

DR LOVE: How about adverse events?

DR ROBSON: Largely the ones that we’ve been talking about. I mean, really the main issues were nausea/vomiting, anemia. Those were the big ones. And in chemotherapy, as you would have expected, it was neutropenia, a little bit of nausea, LFT abnormalities and hand-foot syndrome from the capecitabine, so no unanticipated toxicities. You had mentioned earlier MDS/AML, no cases of MDS/AML. And I think that the MDS/AML story hopefully is going to become not exactly moot, but I think we’re going to get more and more reassurance as more and more patients are treated.

DR LOVE: So one of the questions that this leads to of tremendous practice importance relates to mutation testing.

Somatic alterations in BRCA1/2 genes and response to PARP inhibitors

DR LOVE: And before we get into kind of what you think might be standard at this point and would make sense, I’m curious about another issue that, again, we’ve been hearing about in ovarian cancer. And you mentioned rucaparib, the approval for rucaparib actually, if I understand it correctly, is to just germline but also somatic BRCA mutations. And there’s an assay coupled with it.

You had a paper at San Antonio last December on somatic gene alterations, BRCA gene alterations. What did you see and what do we know right now about somatic mutations and, of course, the issue of do they respond — or people benefit from PARP inhibitors?

DR ROBSON: Yes. So it’s a great question. I think we need to figure out exactly what the proportion of patients is who have somatic mutations in BRCA1 or BRCA2. We have an abstract from our group that’s going to go in from our IMPACT data suggesting that it’s relatively rare, certainly much less common in breast than it is in ovary, part of that, of course, being the overall heterogeneity of breast cancer. And there may be subsets of breast cancer that have slightly more. But it’s not going to be 10%, 15%, 20% of patients with somatic mutations in BRCA1 or BRCA2.

Will they respond? Well, I mean, we’ll find out. There’s a couple of trials that are getting ready to start, looking specifically at the question of responsiveness both in the setting of BRCA mutation and, as in the prostate cancer experience, in the setting of germline mutations in other genes like ATM, PALB2, RAD51C, RAD51D. So Nadine Tung is going to be leading, hopefully, a TBCRC study to address that specifically in breast cancer. And there are other efforts underway in other diseases.

BRCA testing for patients with BC

DR LOVE: And obviously there’s a lot of discussion in ovarian cancer about what kind of testing to do. I’ve heard some people say, do somatic testing first and then back into germline, if you want to. From a practical perspective, certainly a lot of people in practice are sending off next-gen sequencing in all kinds of patients, including breast cancer, particularly, I think, triple-negative when they start running out of options. You’re saying it’s not that likely, but if you see in this testing that there’s a BRCA germline mutation, would you attempt to utilize olaparib?

DR ROBSON: What you’re going to find on tumor-only testing is, if you do find a mutation, you have to figure out whether that’s germline or somatic. And so you would then need to reflex into the germline testing to define. And until there is a kind of label, if you will, for somatic mutation in the breast space, I’m not sure that that’s necessarily going to be the best strategy. On the other hand, it may be more feasible to get reimbursement for somatic testing than germline testing of a low-probability patient, for instance, some older patient with hormone receptor-positive breast cancer who’s going to have a low prior probability of germline mutation, doesn’t fit NCCN criteria, whoever’s criteria, for germline testing. It might be more feasible to test them somatically.

I think the real key there, though, is going to be to make sure that people follow it up, because if you don’t resolve the germline issue, you could potentially leave something on the table for the family, which is something that wouldn’t be ideal.

DR LOVE: So you mentioned the NCCN and other guidelines about BRCA testing, but to me, that’s, like, history now. I mean, you tell me. It seems like any patient who’s got metastatic HER2-negative breast cancer needs BRCA germline testing regardless of age, regardless of anything, I would guess, at least from a — I’m not sure exactly when, but certainly at some point, I would think. Would you agree with that?

DR ROBSON: I think that if somebody coming up as a candidate for therapy, it’s going to be difficult not to do it. I mean, this is a very challenging space right now, because this collision of germline predisposition with therapy and how do you decide who’s right to do testing and what’s the appropriate process for that, as well, is something that’s pretty disruptive. And I think that that’s a good thing, I think, to take it back out and look at it.

But we’re working right now with what we call mainstreaming approaches, which is ways of providing clinicians with educational materials and, like, videos and consent materials up front so that they can do the testing. And then we, in the genetic service, are hooked in on the back end so that we can follow up if they’re positive, so getting rid of — “getting rid” is unfair, but turning the pretest process into more of an education process than a choice process. But how to do that at scale across the country is going to be interesting.

Importance of genetic counseling for patients with germline mutations

DR LOVE: Can I drill down on what you’re trying to get at here a little bit, because I’m not entirely clear. It kind of seems like there are 2 issues. One is the therapeutic issue from the point of view of the patient, that there’s essentially a biomarker to predict a treatment that’s got a 60% response rate that they would want to know about, but then the implications to the family. And is there a way that you can divide that up a little bit? In other words, I would assume there may be situations where family or patients don’t want to know from a counseling perspective, but clearly they would want to know from a treatment perspective.

DR ROBSON: Yes. And so that may be the case.

DR LOVE: I mean, is that what you’re trying to get at with the disruptive —

DR ROBSON: No. What I’m trying to get at is that all of genetics has been set up that this is personal utility information, right? I mean, the whole model that we start with has this idea when we first began testing that the information was just for people to know for personal reasons, not necessarily for medical reasons. And that model has obviously changed over the last 20 years. And this is kind of now the peak of that, right? Now it’s a, as you said, a therapeutic biomarker.

So we have to readapt our model so that people get this information, but also we can’t neglect the implications for the family. And we have to make sure that that information gets out to the family, because yes, they’re all focused on therapy and that’s totally appropriate, but on the other hand, if their daughter now is at risk for developing a bad disease, they also need to know about that.

DR LOVE: So are you saying that you think maybe in clinical practice there are people who are getting BRCA tested, they’re found to be BRCA-positive and they’re not getting genetic counseling?

DR ROBSON: I don't know that to be the case, but my concern is a little bit of what you kind of alluded to earlier, is that when people are getting this for therapy in the middle of, obviously, the devastation of having metastatic disease, if they deprioritize the implications for the family, then their family members might not get the information the way they need to. If the patient’s sick, busy dealing with being sick, it’s like, “Yes. Okay. Later on I’ll tell my family.” But they need to know this.

So I’m just worried about how we’re going to set up the — “worried” is not fair. I think we have to think about how we’re going to set up the model to make sure that people don’t fall through the cracks.

DR LOVE: I’m going to say telemedicine/teleoncology could probably help a lot, but just my thought about that.

I was just thinking. I’ve got to tell you about a great case that we had at ASCO in our GI symposium. These were cases from practice. So we had a general medical oncologist from Pennsylvania, gets this patient, comes in, 60s, who’s got metastatic pancreatic cancer. So he actually gives us this detailed pedigree that he wrote up. And his patient’s got all these cases of cancer in the family.

So he sends off, and the patient’s got a BRCA germline mutation. And, of course, the patient’s got metastatic pancreatic cancer. He’s going downhill. The patient has a 35-year-old niece who is BRCA-positive, is now being considered for prophylactic surgery. And just the astuteness of this oncologist, because this patient got pancreatic cancer, perhaps the life of this woman was saved. I mean, it’s an amazing case.

DR ROBSON: Yes. No. And I think we’re seeing more and more of that. I think there’s more and more understanding that these germline mutations in BRCA1/BRCA2, Lynch mutations and certainly others, are being found in people who would not have been tested based upon existing criteria. And how we work that into our model is going to be very interesting.

DR LOVE: Incidentally, the last 2 prostate cancer symposia that we’ve done — one was the AUA. The other was at GU symposium — I asked the faculty, do you do BRCA testing on your patients with castrate-resistant prostate disease? I think Howard Scher was one of them. Every single one of them is doing it. And every single one of them is using a PARP inhibitor if they’re positive. And almost all of them have cases of patients who responded, which really leads into my question.

You mentioned the MSI thing, which is really, I think, the first approval that was kind of based on a biomarker and not on a tissue type. Do you think that’s where we’re heading with BRCA and PARP inhibitors? Again, you’re seeing people use it in prostate cancer.

DR ROBSON: Yes. That’s a really interesting question. And I think that that is going to be a complicated regulatory/science issue. And how to get the FDA on board with that is — I don’t know who would take point on that. But it makes sense. We’re seeing responses to PARP inhibitors in individuals with germline BRCA mutations in many different settings. And you hear anecdotal ones in even more. So it would be logical for this to be a pathway-driven type of approval. But how you set up the design so that the FDA approves that is not clear to me.

Clinical implications of the OlympiAD study results for patients with HER2-negative mBC

DR LOVE: So I’m wondering if we’re going to get beyond the topic of PARP inhibitors and BRCA before the end of this interview. But there’s so many other things I want to ask you about, but one is the practical implications of this OlympiAD study to patient care. You would think that this is changing standards. So, I mean, I guess you would have the issue of the patient with ER-positive, HER2-negative disease and where hormone therapy would fit in, CDK4/6 inhibitors. How do you see this actually coming into practice? How does it come into your practice?

DR ROBSON: I mean, I think that the results for hormone receptor-positive patients have been so stunning with the CDK4/6 inhibitors, particularly in first line, that I don’t think there are data to throw that away for this yet. And so I would start there. We’re talking about developing some studies to look at combinations of antiestrogen and CDK4/6 inhibitors and PARP inhibitor to see if that’s tolerable, something we can drive forward.

But for now, I think the results have been good enough that I would do that and then think about a PARP inhibitor instead of chemotherapy when you get to the point that you’re starting to think about chemo.

In triple-negative disease, one of the things that was very gratifying to see in OlympiAD was that the time to response for the patients was similar in the patients who got olaparib compared to those who got conventional chemotherapy. So this wasn’t something that worked slowly. If it was working, it was working quickly.

And so the big question, I think, in triple-negative disease is going to be the role of PARP inhibitors versus platinum-based agents in the BRCA group. If it was up to me, I would feel like a PARP inhibitor was kind of a lot more tolerable than a platinum and would go that way. But whether or not one is going to be better than the other is going to require a trial that I’m not sure is ever going to get done. But numerically, the responses are pretty similar in the TNT trial and the OlympiAD trial, so I think it’s a reasonable option to try to avoid IV chemo for those patients as well.

DR LOVE: Do you think it would be a reasonable option in a younger patient who wants everything possibly to be done, to give platinum therapy short-term and the olaparib as maintenance outside a trial setting?

DR ROBSON: Again, it’s an intriguing idea. The cross resistance is incomplete between platinum and PARP inhibitors. There’s definitely overlapping mechanisms of resistance, particularly in the BRCA mutation carriers.

But conversely, I think sensitivity to platinum has, in other settings, overlapped with sensitivity to PARP inhibitors. So that would not be an unreasonable thing to do, a trial of platinum followed by a maintenance with olaparib. Of course that’s going to be so completely off label that you’re not going to know what to do with it. But it’s a sensible clinical approach.

On the other hand, I wouldn’t feel like I necessarily had to start with a platinum. I’ve felt like the results are similar enough between OlympiAD and TNT that I would feel pretty comfortable in most patients with triple-negative disease, beginning with olaparib. And if you get a quick response, then you’ve done that with an oral agent. And that’s a good thing to do.

TNT trial: Results of a Phase III study of carboplatin versus docetaxel for patients with metastatic or recurrent locally advanced triple-negative or BRCA1/2 mutation-associated BC

DR LOVE: Could you just briefly summarize what the TNT study looked at?

DR ROBSON: So TNT study was — again, Andrew Tutt did the study over in the UK looking at patients with triple-negative breast cancer, randomizing to either get carboplatin or docetaxel as first-line therapy for metastatic disease. And the responses overall between the patients who got docetaxel and carboplatin were pretty similar in the triple-negative group as a whole. But in the group of patients who had germline BRCA mutations, relatively small preplanned subset, there was an advantage with the overall response rate, was 68% in the carbo arm and down in the 30s in the docetaxel arm. And there was also a difference in PFS. And all of those numbers were pretty similar to what we saw in OlympiAD.

DR LOVE: Any rationale to continue a PARP inhibitor on disease progression and add something else in?

DR ROBSON: I don’t think so. I think the mechanisms of resistance to PARP inhibition, there are several. There are a number of different ones. But all of them would compromise this synthetic lethality. Basically, they all reestablish homologous recombination. And given that, I think the likelihood that you’re going to derive benefit from continued therapy is pretty small.

We did clinically, again — and completely anecdotally — there were patients who technically progressed according to RECIST criteria and yet were felt by their investigators to be continuing to derive clinical benefit, so either very slow progression or progression in one site with continued disease control in other sites. So I think in that setting, it might be okay to, as we all do in clinic under other circumstances, kind of hold off from changing until it was pretty clear that there was significant progression.

DR LOVE: What about — you see this with targeted therapy, doing local — you have 1 lesion progressing. You radiate it, you take it out, whatever. Everything else, is that a strategy you ever use?

DR ROBSON: Yes. I mean, I think that that’s, again, a perfectly reasonable strategy. Now, we’re just talking about kind of clinical oncologic medicine. I mean, one of the most common mechanisms of progression or resistance to PARP inhibitors appears to be reversion of the mutation. In other words, there is a clone of cells that has somehow managed to reestablish an open reading frame to reestablish homologous recombination. And if that only occurs in 1 site or 2 sites, then directed therapy towards those sites might be fine.

The problem is that these cells are so genomically unstable that it tends to happen in multiple different sites, and so multiple different clones arise with different revertant mutations.

So evolution is tough to circumvent.

DR LOVE: That’s really interesting. Do you see, if you can get rid of those cells with, say, chemotherapy or whatever, do you get clinical sensitivity to PARP inhibitors back?

DR ROBSON: I don’t think anybody’s actually tried that. Nobody’s tried, that I know of, to rechallenge. But, unfortunately, that’s also a setting where there’s frequently resistance to platinum-based agents as well. So if you have somebody who’s anthracycline/taxane-resistant, progressed through platinum, progressed through PARP inhibitor, unfortunately the likelihood that they’re going to respond to other therapies is going to be —

Management of nausea and anemia associated with olaparib

DR LOVE: I want to find out a little bit about how this plays out clinically. I want to hear about at least 1 of your cases here, maybe this 49-year-old lady. But just any pearls of advice to somebody who maybe has not used the drug, olaparib, and about — you’re about to start it in a patient with metastatic HER2-negative breast cancer, particularly about how to deal with anemia. I’ve heard the gynecologists say, “If the patient’s responding, we transfuse and just keep going.” What do you do about dose reductions? What are some of the common clinical issues that come up?

DR ROBSON: So as we talked about a little bit earlier, the 2 most common issues are nausea, which is often worse at the beginning. And I think that giving people appropriate antiemetics and trying to kind of carry them through that and, hopefully, with some encouragement, that it will generally get better with time. Sometimes it takes a couple of months, but it does get better.

And then with regard to the anemia, my sense is that if the anemia develops, it develops quite quickly. And so I think a little bit of earlier monitoring for anemia, even though it’s an oral drug, perhaps checking at 2 weeks, 4 weeks, 6 weeks just to kind of make sure that the bottom doesn’t drop out. And again, my hypothesis, although without any supporting data, is that that might even be something you want to particularly do in people who’ve received prior platinums.

And then if it drops, I think you would use your conventional transfusion threshold and your judgment about whether or not somebody is receiving clinical benefit. And if they are and you just need to give them a unit every now and then or even put them on an ESA to kind of try to keep them propped up, then you’re still deriving the other benefits that are avoiding chemotherapy.

Case: A 49-year-old woman with BRCA-mutant mTNBC whose disease progresses through several lines of systemic therapy

DR LOVE: So when you ask people to bring cases in, I mean, it’s natural to try to bring in people who’ve done really well. But sometimes you can learn a lot. You don’t have to see too many people do really well, particularly on a monotherapy, to get my attention. And this 49-year-old lady sounds amazing. What happened with her?

DR ROBSON: Ah, she’s my poster child. She really is. So she was on Andrew’s original ICEBERG study. She was actually the first patient that I put on it. And actually, she’s from Miami. She flies up to see me from Miami.

She didn’t have high-volume disease. She had relatively low-volume pulmonary disease but clearly had progressed through anthracyclines/taxanes, had had a wedge resection, progressed through K, progressed through vino.

And she has tolerated the drug incredibly well, has never had any significant toxicities and is just chugging along, essentially NED at this point. I mean, every time I see her, I’m, like, knocking wood. But she’s doing great.

DR LOVE: I’m not sure we mentioned yet that this was in December 2007.

DR ROBSON: Right. Yes. I mean, it’s 10 years with clearly documented metastatic disease, refractory to multiple chemotherapy regimens.

So it’s interesting. If you look across all the trials, there are a few patients like this treated with olaparib. I mean — and, of course, there are a few patients like this with any chemotherapy drug, so you can’t make too much of it. But there’s even been an effort to pull together these extreme responders and see if we can find any predictors. I did exome sequencing on this lady’s primary tumor, and I couldn’t find anything that was different than what you would normally see in a triple-negative patient.

DR LOVE: Did you look at her immune system?

DR ROBSON: No, I haven’t actually done that. But she was just a straight-up conventional BRCA1 founder mutation, triple-negative disease. There was nothing about it that was unusual except for her dramatic and prolonged response.

Mutational landscape of BC

DR LOVE: I was going to ask you before, because I see you’ve done a lot of work, really — I see this paper in Nature Medicine just recently, “Mutational Landscape of Metastatic Cancer, 10,000 Patients.”

DR ROBSON: That’s been Mike Berger’s work. I’m just along —

DR LOVE: What do we know at this point, though, about the — I guess the mutational landscape of breast cancer?

DR ROBSON: I think we’re learning that there are certain recurrent mutations in different disease types. There’s a little bit of correlation with histologic subtypes. You see p53 mutations in patients with triple-negative disease, of course, PI3 kinase in ESR mutations in hormone receptor-positive disease. I think that breast cancer is not as hypermutated, by and large, as some of the other types of cancers. You don’t really see the kinds of things that you do with Lynch syndrome or mismatch repair-deficient colon or endometrial, their hypermutated phenotypes.

And so we’ve found a few pragmatic targets, and people are certainly trying to leverage those and, at the same time, looking at more complicated types of genetic abnormalities like fusions, rearrangements and some interest now developing in epigenetic types of abnormalities to see whether or not those might provide something that’s a little bit more tractable/druggable.

Ongoing trials of PARP inhibitors in the (neo)adjuvant setting

DR LOVE: Just to finish out with PARP inhibitors, what other trials are out there right now, particularly Phase III trials, that we’re going to be looking to see some results on, particularly in the earlier stage, neoadjuvant and adjuvant setting?

DR ROBSON: So I mentioned the EMBRACA trial, which is the talazoparib Phase III and essentially exactly the same design as OlympiAD, except it’s a little bit larger. It’s trying to power it for an overall survival advantage.

Then there is the carbo/paclitaxel with or without veliparib study that’s being done in triple-negative disease with veliparib. And then I think in the early-stage setting, there’s OlympiA, which is an adjuvant trial in relatively high-risk patients who have germline BRCA mutations, so either residual disease after neoadjuvant therapy or a significant number of nodes involved.

And then a number of Phase II trials, kind of a let-the-flowers-bloom type of approach for using in combinations with IO and in combinations with other targeted therapies. There are some studies that are being done in combination with conventional chemotherapeutic agents, but I think by and large, those are perhaps a little bit less exciting. And then some correlative studies trying to understand the biology of response and resistance a little bit better.

DR LOVE: So in the OlympiA trial, they can get adjuvant chemo, adjuvant hormones?

DR ROBSON: Conventional adjuvant chemo and plus-minus hormones, if that’s appropriate. And then the randomization is for the PARP inhibitor on the back end.

DR LOVE: But HER2-negative?

DR ROBSON: Correct.

DR LOVE: And then how long do they receive the PARP inhibitor for?

DR ROBSON: It’s a year.

DR LOVE: A year.

DR ROBSON: Mm-hmm. That’s right.

DR LOVE: How do you think that’s going to go after maybe having adjuvant chemo?

DR ROBSON: Let me say, I mean, they’re accruing, and it doesn’t look like they’re having a huge falloff. So I hope it goes well. And again, most people, after perhaps some initial nausea and then depending on what happens with their counts, a little bit of anemia that declares itself pretty early, most people tolerate the drug very, very well.

DR LOVE: I mean — do you put patients on that study?

DR ROBSON: No, I haven’t, actually, not yet. But I know everybody who’s enrolling on it.

DR LOVE: Just wondering, for example, so they would, for example, be taking tamoxifen and olaparib at the same time?

DR ROBSON: Yes.

DR LOVE: I guess it would be just completely independent toxicity.

DR ROBSON: Yes. Yes.

DR LOVE: Although, who knows?

Case: A 47-year-old woman with a 0.9-cm, Grade 2, ER/PR/HER2-positive invasive ductal carcinoma (IDC) receives adjuvant trastuzumab/paclitaxel

DR LOVE: Why don’t we talk about a couple other of your cases just real briefly? I was trying to predict how some of these patients got treated. Okay. So 47-year-old woman, 0.9-cm, Grade 2 IDC, ER 95%, PR 50%, HER2 1/2+, FISH 2.2, status postbilateral mastectomy. I don't know. I’m going to guess you gave her trastuzumab/paclitaxel.

DR ROBSON: You got it. Exactly. APT trial. I mean, the results from that have been so stunningly great for the T1Bs that it’s kind of hard to make a case for giving much more than that, so been a fairly standard approach right now for us.

DR LOVE: How did she do?

DR ROBSON: She’s doing great. I mean, just got started, so, but as you know, the overall disease-free survival from that study, especially breast cancer-specific survival, is very, very high.

DR LOVE: Right. I think they just reported their 7-year results or more follow-up.

DR ROBSON: Yes. Sara Tolaney reported it at ASCO. Yes. So yes. I feel it’s going to be hard to prove anything to be better than that.

DR LOVE: What other regimens do you use in the adjuvant setting for higher-risk patients in the HER2 setting?

DR ROBSON: Yes. We are still using anthracyclines, so ACTHP is kind of our high-risk approach, kind of the APHINITY approach. Whether or not that changes after APHINITY, I think we’re going to see. I mean, I think there’s going to be a little bit of risk stratification that takes place, but —

APHINITY trial: Results of a Phase III study evaluating the addition of pertuzumab to chemotherapy and trastuzumab as adjuvant therapy for patients with HER2-positive early BC

DR LOVE: That was going to be my obvious next question, which is your thoughts about APHINITY. Maybe you can summarize from your own perspective what you think from your point of view were the most important findings and how it affected your practice, if at all.

DR ROBSON: I think the most important finding is that there wasn’t an enormous benefit. I mean, it’s a positive trial, of course, so it’s hard to be overly negative. But I do think that there’s generally a sense of perhaps a wish that things had been better, because it now puts us into this box of trying to decide, is the risk enough to justify the potential incremental toxicity and, of course, financial toxicity?

So certainly for higher-risk HER2-positive patients continuing to do that, because there is an incremental benefit and you’ve got to assume that the absolute benefit is going to be proportional to the baseline risk, but —

DR LOVE: What were you doing prior to APHINITY? The former head of Breast Cancer, Cliff Hudis, who now moves over to ASCO, was kind of, I think, very much involved with the NCCN idea of if they qualified preop it’s okay or maybe should be done postop. Were you giving pertuzumab adjuvantly prior to seeing those data?

DR ROBSON: Yes, we were. Yes, we were doing that. I mean, we’ve done a lot of neoadjuvant therapy for HER2 patients. And so anybody who kind of was obviously clinically high risk, big tumors, clinical and positive nodes, we were getting ACT/HP up front and then continuing the HP out back. But there was some variability in reimbursement about that. So that often dictated things as well.

Adjuvant pertuzumab for patients with HER2-positive, node-positive mBC

DR LOVE: Would you say that in general you’re going to be continuing to use or using adjuvant pertuzumab for node-positive patients?

DR ROBSON: I imagine we will. I mean, I think especially in younger patients who tolerate it. We certainly do other things for similar degrees of absolute benefit. I mean, breast perhaps is a little bit different than some of the other diseases that way, but I think many patients would consider a 2% to 3% invasive disease-free survival benefit to be worth it. And if they started having substantial toxicity, we’d have to readdress. But at least as an initial plan, I think it’s hard to back away.

DR LOVE: When we were looking at APHINITY in the node-positive patients, the hazard rate’s a little bit better. And it kind of comes together a little bit better, but I don't know if that’s the way it’s going to actually shake out in practice.

Case: A 65-year-old woman with ER-positive, HER2-negative mBC receives fulvestrant and palbociclib after disease relapse on exemestane

DR LOVE: I want to ask you about a couple of your other cases just really briefly, your 65-year-old lady with metastatic ER-positive, HER2-negative disease. And maybe you can talk a little bit about kind of what happened at the point when you saw her. It looks like she was on exemestane, and then you switched her to fulvestrant/palbo. Can you talk about that situation, kind of what you were thinking?

DR ROBSON: So this is a lady who had kind of a long disease-free interval. And also, a lot of what was going on with her at the time of relapse was kind of before we had much data about CDK4/6 inhibitors, and so started her on an AI, just kind of fairly conventional, and then when she progressed changed her to exemestane. She had very minimal progression and was feeling extremely well. And I personally found it hard to swallow the idea of putting her on everolimus and exposing her to the potential toxicities of that.

But then when the exemestane didn’t work, palbo was coming online. Fulvestrant was going to be the natural next thing for her to get anyway. And so we did the two of them together. And so far, she’s tolerated it. She had some cytopenias, some neutropenia. I had to drop her dose a little bit. And she’s had, perhaps, some slow progression but been on it for nearly a year now, I guess, and has gotten clinical benefit for sure.

DR LOVE: Yes. I wanted to ask you about that dose reduction. How often do you end up having to do dose reduction because of neutropenia? When do you monitor patients? And what actually happened with her?

DR ROBSON: Yes, so the first time I monitor them at 2 weeks, 4 weeks, 6 weeks, 8 weeks and just kind of get a sense of what’s happening. And she really went in the dumper. I mean, she went down to like 0.1 or something like that. So she clearly needed a dose reduction. She went really low. Other people, I think, have had much less of an issue.

But at 100, she’s been able to kind of hover her ANC around 1.2/1.3 and tolerating it fine. I don't know if she had something intercurrent that led her to really go down low at the beginning, but she also actually travels back and forth from Florida. So I don’t feel like having her hanging out there with an extremely low white count without me knowing about it, and especially since she’s doing well, has low-volume disease. I don’t really feel that I need to push the dose that hard.

Therapeutic options for patients with ER-positive, HER2-negative mBC after disease progression on a CDK4/6 inhibitor

DR LOVE: So I’m getting a real flair for this situation. So she’s had exemestane. Now she’s on fulvestrant/palbo. If you become convinced that she’s progressing and you’re still okay with hormone therapy, what do you think would be next?

DR ROBSON: Yes. That’s interesting. I mean, I haven’t really been — the couple of times I have tried everolimus-type combinations after CDK4/6 progression, I have not been very impressed. I get the feeling that by the time somebody’s kind of gone through a SERD and a CDK4/6 inhibitor that the HR axis is no longer that critical to their tumor. And I’d be starting to think about capecitabine or something like that, if she was in a safe enough place to give an oral agent. I’ve already tested her, so I know she’s not a BRCA carrier.

DR LOVE: You know what? Oh, she’s BRCA-negative.

DR ROBSON: I know she’s not a BRCA carrier because I’ve already — yes.

DR LOVE: And I see she — I mean, she had tamoxifen, I guess, as adjuvant therapy. I don't know if that could potentially be a consideration in the future.

DR ROBSON: We could recycle. I mean, I — just — again, I think a lot depends on exactly what is going on with her. If progression is just a little bit more growth of her intermammary node or a little bit more growth of her pleural nodule, I mean, I probably would just continue her on what she’s doing.

If she started developing other sites or if something felt like the pace of her disease was changing, then I might not feel so good about just going back to tamoxifen and wanting to switch over. So it’s almost like a catch-22. If her disease is bad enough to require me to change her therapy, then I kind of really want to change her therapy. I don’t want to try something that is less than likely to work.

DR LOVE: So speaking of somatic mutations, any comments about estrogen receptor mutations? Have you seen any patients with it? Does it change what you do?

DR ROBSON: Yes, we’ve definitely seen patients with it. It’s fascinating. Does it change what I do? No, because usually what winds up happening at our place is we wind up profiling people kind of in the context of their metastatic disease when we’ve already made our decisions. Personally, I’m pretty impressed with the data and if I had somebody who had significant ESR mutation, I might be leaning more towards SERDs than to SERMs, but I haven't been using that as a biomarker for treatment determination.

Targeting the androgen receptor in patients with mTNBC

DR LOVE: So I think I heard that there’s been a — I don't know, the whole thing with androgen receptors and antiandrogens. And Tiffany Traina’s work — I actually interviewed her a while back — is not proceeding. Is that the case?

DR ROBSON: I don't know. I mean, again, getting people interested in it, it turns out that a lot of triple-negative breast cancer have androgen receptor alterations. I mean, it’s not a small subset of patients. And it’s been a little hard to get the appropriate trial designs to figure out, how do you prove the right way to treat those folks compared to, for instance, the conventional therapy? So I think abandoning that would be a disappointment, just because it’s such a potentially significant pathway.

And many of those folks have fairly indolent disease, and if we could treat them with an oral agent, oral antiandrogen as opposed to knee-jerking into chemotherapy for everybody who has triple-negative, I think it would be a service to them.

DR LOVE: I think I just heard that mantle cell lymphoma has androgen receptors, too.

DR ROBSON: There you go.

DR LOVE: I know they’ve tried CDK4/6 inhibitors with mantle cell, but I don't know if that’s connected.

DR ROBSON: Anything that has an RV pathway alteration is something you want to look at.

DR LOVE: Right.

Case: A 54-year-old woman with bilateral ER/PR-positive, HER2-negative BC, 2 positive sentinel lymph nodes and a high genomic risk by the 70-gene assay

DR LOVE: So maybe we can also hear about your 54-year-old lady, get into another issue, which is the issue of genomic profiling in the ER-positive, HER2-negative situation. What happened with this lady?

DR ROBSON: Yes. So she was an interesting lady, because she had tiny primaries, and then she had axillary nodal metastases. And also, there’s an overlay of patient preference here, where she really didn’t want to get chemotherapy unless she absolutely had to.

And so this was shortly after the 70-gene assay had kind of had its publication and been approved. And so I thought, “Let me see if, just by chance, she has a low genomic risk score, in which case perhaps we can make a little bit more of an argument or provide her some data to help her make her decision about chemo.”

And she didn’t. I mean, she had a high score, which made me more confident in the conventional recommendation to treat her. But it was just a way to essentially help inform patient decision-making in that setting.

DR LOVE: So she decided to get chemo?

DR ROBSON: Yes, she did. She was fine. I mean, she’s a sensible lady. She wanted to avoid it, if at all possible, but wasn’t refractory. And if she’d been negative — I mean, if she’d had a low genomic score, it would have been tough, I mean, because micrometastatic disease in 1 node, yes, she had a macro met in the other, but it was only 9 millimeters, and then these tiny primaries within the breast. Her pretest risk prediction would have probably been pretty small, and so committing her to certainly anthracycline-based chemotherapy in that setting would have been tough in terms of risk/benefit.

Use of the 21-gene assay for patients with ER-positive, node-positive BC

DR LOVE: What do you think would happen if we presented this case to 25 breast cancer investigators and said, would you order a genomic assay in this patient and, if so, which one? What do you think they would say?

DR ROBSON: You’d get 3 different answers. I mean —

DR LOVE: No, I can tell you, you’d get 2 answers, neither of which is what you did. Incidentally, just want to point out, which is — because we actually did that survey. And we had an abstract at ASCO on this.

So what we found is that half of the investigators were using 21-gene Recurrence Score in node-positive, particularly low-volume axillary, 1 node in particular. And the others don’t use anything. But none of them use the 70-gene assay, which you used in this case. What was your thinking in doing that as opposed to the 21-gene?

DR ROBSON: I just feel like the 21-gene assay in the node-positive disease just has so much less in the way of validation. I mean, we’re still waiting for RxPONDER. And until that comes out, just the data hooking onto the SWOG trial just doesn’t feel good to me. And sure, if they have a low score, then that’s potentially reassuring, but so many of these wind up being intermediate. And then that wouldn’t really have gotten me anywhere.

And so the MINDACT data, for me, I understand the concerns that it hasn’t excluded a benefit of chemotherapy in low genomic, high clinical. I get that. But on the other hand, it does suggest that the benefit of chemotherapy in that setting is going to be less. And if that’s the case, in this particular scenario where the lady was already kind of biased against it, I personally would have felt pretty comfortable just giving her hormone therapy alone in that setting.

DR LOVE: Interesting. I thought globally the thing that I found most interesting about the survey — and I don't know. Did you take it? Because I know we send you stuff all the time, but —

DR ROBSON: I don’t remember.

DR LOVE: You must not have been in there, because nobody answered 70-gene, although maybe it was individual for this patient. But the thing that I thought —

DR ROBSON: Yes. I’m okay with the 70-gene assay for node-positive disease. I know that the vibe has not been particularly enthusiastic. But it seems like at least it was validated in a big data set.

DR LOVE: Absolutely. And I think the vibe is just more, “We’re used to the 21-gene. There’s not a gigantic reason to change” kind of thing rather than it’s not valid. But kind of like, “We’ve already gotten used to using palbo.” I don't know what.

But in any event, I thought globally the thing that was most interesting, though, about it, the survey that we did — and we knew this was going to happen, because we ask people all the time. We know people are split about it. But the thing that was amazing is you have this ASCO guideline statement out there saying, “Do not do a genomic assay in a patient who has a positive node.” And then you’ve got half the investigators who are doing it. And it’s just — you read that paper, you don’t get — we did a piece about this. And actually a very prominent member of that committee contacted me and gave me a long list of why they put that in there. He didn’t want it published, but I totally understand it. But I just think it’s amazing that there’s a guideline out there, and yet a lot of investigators are not following it, including you.

DR ROBSON: Yes. Yes, including me. I mean, I think this is a case where you’re trying to give patients information to help them make decisions. And would it have been wrong to have given this lady chemotherapy if she had a low genomic risk score or even if she had a low Oncotype score? No, I don’t think it would have been wrong. But certainly in terms of her own decision-making, having an additional test to support the idea that she needed to be treated was helpful to her. It gave her that additional confidence that this was the right decision and the motivation to go through the toxicity of what she’s going through.

Case: A 49-year-old woman with BRCA mutation-positive mTNBC experiences rapid disease progression through several lines of therapy, including olaparib

DR LOVE: So let’s finish out with your last case, getting back to PARP inhibitors. And maybe you can just briefly talk about this 49-year-old lady.

DR ROBSON: This is just to show that this is unfortunately not a panacea, right? This is a lovely lady. She really — just a lovely, lovely lady who had metastatic triple-negative breast cancer and came to see me after progressing through cape to go on OlympiAD and got started on olaparib and had a pretty decent initial response and then lost it pretty quickly. I mean, she started and then failed and unfortunately failed platinum after that and then failed a series of other things and passed away relatively quickly.

So I think it’s more just that we still — as enthusiastic as I am, as exciting as it is, that to have this therapy that makes biologic sense and does provide benefit and does provide benefit sometimes that’s quite prolonged, it still means that we’ve got a long way to go. And we need to continue to do research. We need to continue to do the biologic research that allowed us to develop the therapy in the first place and to understand the mechanisms of resistance and to try to improve the outcomes.

Case: A 68-year-old woman with ER/PR-positive, HER2-negative, moderately differentiated IDC and 5 of 20 positive axillary nodes

DR KROP: So I saw a 68-year-old woman who comes from New Hampshire, who presented with a palpable mass. She was biopsied and found to have a strongly estrogen and progesterone receptor-positive, HER2-negative, moderately differentiated breast cancer. And unfortunately she also had some axillary adenopathy. And she went on to have a lumpectomy with axillary dissection, which demonstrated about a 4-cm cancer in the breast with negative margins, but 5 of 20 lymph nodes that were involved.

And given that anatomic risk that she had, significant nodal involvement, I think her primary oncologist was thinking that adjuvant chemotherapy prior to hormonal therapy would make the most sense. And she placed a port and was getting her set up to start chemotherapy. And the patient came down to see me for a second opinion.

And while certainly chemotherapy is a very reasonable approach for someone with multiple lymph nodes involved, I was struck by the biology of her cancer, which, based on the strong hormone receptor expression and lack of high-grade histology, I was feeling that this was very likely a cancer that chemotherapy was not going to make a big impact. And in addition, this particular patient, who was 68 and she has significant obesity. She has pretty bad osteoarthritis. And she’d already had 1 hip replaced and needed another hip and really wasn’t very mobile. I was also worried she would be more at risk for the complications of chemotherapy.

So based on the presumed biology and the fact that — which would, to me, imply that chemotherapy may not have much of a significant impact and the fact that she was at increased risk for complications of chemotherapy, I did think she was a patient who performing an Oncotype DX assay would be helpful. We don’t have as much data on multiple lymph node-positive patients as we do node-negative patients with Oncotype, but the data we do have would seem to suggest that the node-positive biology is fairly similar to node-negatives, in that patients with low-risk Oncotype scores really don’t benefit much from chemotherapy.

So we had this discussion. The patient was very concerned about the toxicities of chemotherapy and, therefore, really did feel that trying to get a better sense of the biology of her cancer was worthwhile. So we did send an Oncotype. I talked to her primary oncologist, who was completely on board with that. We got an Oncotype, and it came back with a score of 14, which is squarely within the low-risk range.

And if you look at the data from the Oncotype analysis of Kathy Albain’s study, this is a population that really does not seem to benefit from chemotherapy. They overall have a higher risk of recurrence, but chemotherapy doesn’t seem to impact that. So based on that Oncotype score, I did recommend that she proceed with an aromatase inhibitor and not do chemotherapy.

DR LOVE: And I see that she actually ended up on a clinical trial.

DR KROP: Right. So these are patients in whom we don’t think chemotherapy is likely to be helpful, so we like to maximize the benefit of hormonal therapy. And we’re trying to explore ways to do that. Right now, in terms of standard of care for a postmenopausal patient, there really isn’t anything else besides aromatase inhibitor therapy that’s going to make an additional difference, but, of course, we’re exploring additional types of targeted therapies to try to improve the efficacy of hormonal therapy.

So one of the ways that we think may be possible to do that is by adding a CDK4/6 inhibitor. Obviously, we’re all aware of the data in the metastatic setting, that these drugs seem to improve the efficacy of hormonal therapy. So the natural next step is to see whether that also holds true in the adjuvant setting.

PALLAS: An ongoing Phase III trial evaluating the addition of palbociclib to adjuvant endocrine therapy for hormone receptor-positive, HER2-negative early BC

DR KROP: So we currently have a trial called PALLAS, which is exploring the benefit of palbociclib in addition to hormonal therapy in higher-risk ER-positive patients. There are other trials getting underway looking at the other CDK4/6 inhibitors as well. So I do think right now, we obviously have no data that this is going to be a beneficial approach, but it’s certainly a question that we need to ask.

DR LOVE: But this is Phase III. And is it blinded? Do you know what she’s getting?

DR KROP: So this is a Phase III trial. This particular trial is not blinded. There’s no placebo control. And the rationale for that was that, given the incredibly high rate of neutropenia with palbociclib, that there really would be no purpose in doing a placebo control, because everybody’s going to know which arm you’re on.

DR LOVE: Yes. I was thinking the same thing.

DR KROP: And it adds extra monitoring for the patients who are — if you’re on a placebo, you don’t need that. So this particular trial is not blinded. I think some of the other adjuvant trials may be blinded. And I think that’s just a decision that different companies are making.

DR LOVE: And is she getting the palbo?

DR KROP: Oh. So she is. She just started. So she just finished her postlumpectomy radiation and was just put on the study. So I don’t have much experience yet with her on how she’s tolerating the palbociclib.

DR LOVE: In other patients that you’ve seen — I don't know if you’ve had other patients who got palbo — what do you typically see in terms of the counts? And is there any kind of quality-of-life thing that you can pick up in terms of toxicity or side effects?

DR KROP: Yes. So, I mean, I have a reasonable amount of experience with patients in the metastatic setting getting palbociclib. And it’s variable. I mean, certainly neutropenia is very common. And many patients end up needing a dose reduction to deal with the neutropenia.

But in terms of quality of life, I actually find that the things that we see from the clinical trial data would suggest that fatigue and diarrhea and some hair thinning are common toxicities. Certainly, neutropenia is present. I haven’t had much trouble with my patients getting diarrhea, for whatever reason. People do notice fatigue. And fatigue is one of those things that I think is very variable from patient to patient. Some patients really have a lot of trouble with it. Others don’t at all. And I have no idea why there’s this variability.

Role of the 21-gene Recurrence Score^® (RS) in the neoadjuvant setting

DR LOVE: What an incredible case, 5 positive nodes. And amazing that she ends up on palbo. I mean, if I were the patient, kind of like I think that’s what I’d want. But how about if I go back to you and say maybe, ideally, this lady should have had a genomic assay? And we’ll talk about which one she had, the 21-gene. Ideally, she should have had it preop, when she was first diagnosed, and it should have guided preop therapy.

DR KROP: Yes. I think that’s an interesting question. And we actually — it’s funny you mention that, because we actually are starting to do preoperative genomic testing, but we’re doing it in a specific population, which is, I think, the one you’re alluding to, which is patients in whom you’re considering doing preoperative therapy, postmenopausal or — not — pre or post but with hormone receptor-positive, nonhigh-grade tumors that you’re really not sure what the benefit of chemotherapy is. And what you’re doing in the preoperative setting is, you’re trying to cytoreduce the cancer to potentially change someone from needing a mastectomy to being a breast conservation candidate.

And you want to do that with the most effective therapy you can. And with these hormone receptor-positive cancers, certainly chemotherapy is not the answer to all of them. And there are older data randomizing preoperative patients to chemotherapy or hormonal therapy, which seem to suggest that actually hormonal therapy led to better clinical responses than the chemotherapy. And I think that just goes along with what all we’ve learned in the last 10 or 20 years, that these nonhigh-grade ER-positive cancers really are fairly unresponsive or insensitive to chemotherapy and that hormonal therapy is really the best way to attack them.

DR LOVE: Although, I mean, there are people who do have ER-positive, HER2-negative tumors that, for example, have high 21-gene Recurrence Scores. And maybe that wouldn’t be predicted in this situation, but it certainly could occur.

I think if this lady had either HER2-positive or triple-negative disease, she would have gotten neoadjuvant therapy. My question is, if she had a high Recurrence Score, would you have preferred giving her neoadjuvant chemo?

DR KROP: Yes. So I think that’s exactly how we’re using this. So if I would have met her on day 1 rather than after she’d had her surgery, I would have offered her preoperative therapy. I do agree with you that I think that that would — in someone who has multiple positive lymph nodes, especially palpable lymph nodes, that the surgeons prefer neoadjuvant.

And before you do that, you kind of want to know what’s likely to be the most effective treatment. And so doing a genomic assay in that population, I think, is a very reasonable thing. And that’s actually what we’re doing.

We actually have preoperative clinical trials now for both chemotherapy-based trials as well as hormonal therapy-based trials for ER-positive patients. And we’re using genomic assays to help steer patients to which one of those we think is most appropriate, as well as nontrial options.

DR LOVE: It kind of seems like this is an emerging trend. When we survey investigators, maybe half or less do that. So not everybody’s doing it, but we do hear a substantial number of people who do.

Comparison of the 21-gene RS versus the 70-gene assay to determine benefit from chemotherapy in patients with ER-positive BC

DR LOVE: The other question I have, particularly in view of your recent publication that we want to talk about, is your choice of genomic assay. You chose in this patient with 5 positive nodes to use the 21-gene. And the obvious alternative — there’s a number of alternatives, but particularly the 70-gene. And, of course, I wanted to ask you about the publication of the paper that came out looking at that from the MINDACT trial.

DR KROP: Right. So you’re right that we now have different options. I think from my personal standpoint at Dana-Farber, where I work, we have a lot of experience with Oncotype. We’ve been using it now for a number of years. And it’s something that we’re comfortable with. But now that there are starting to be more data with the 70-gene assay, or MammaPrint^®, and this no longer requires frozen tissue, that that’s also an appropriate test.

I mean, the interesting thing is, MammaPrint gives you a yes-no result, whereas Oncotype gives you a continuous score. In some ways, there’s pros and cons. In some ways, when you’re making decisions for patients, there’s something inherently nice about having a dichotomous score, because then you don’t really have to waver about what to do with the intermediate scores. But I think we have enough experience with Oncotype that we are very comfortable with that, but I think this is an evolving area.

Our general sense is that all of these tests, the MammaPrint, Oncotype, PAM50, are probably measuring similar things. And which particular test you use may not matter. And I think it’s just something that you want to be comfortable with understanding the scores, understanding the report and how that influences your decision-making.

Updated ASCO clinical practice guidelines on the use of biomarkers to guide decisions on adjuvant systemic therapy for women with early-stage invasive BC

DR LOVE: So that does tie into some of the papers that have come out. There was an ASCO Guidelines paper that came out previously about the use of genomic assays that I was surprised pretty much said, “Don’t use genomic assays in patients with node-positive tumors.” Dan Hayes was on that paper.

And then — which was interesting, because we know that a lot of, if not the majority of the investigators, as you demonstrated in this case, were actually doing it. They said, “Don’t do it.” And then more recently, just very recently actually, you had a paper, it was an ASCO Guideline paper specifically addressing the issue of the 70-gene assay. Can you talk about kind of what the point was that you were trying to make in that paper and, globally, what the guideline process is about, the whole issue of the issue of using genomic assays in patients with node-positive tumors?

DR KROP: Yes. So, I mean, I think that’s a good question. And again, it’s an evolving field. I mean, the original ASCO Guidelines, which, as you said, Dan Hayes was heavily involved in, felt, as I also mentioned, that the data for Oncotype in node-positive patients is limited. We have 1 retrospective analysis of a prospective study. And the number of node-positive patients was not — excuse me — large that were evaluable.

And because of that limited data, I think the panel at that point felt that it wasn’t yet appropriate to use that assay in most patients, knowing that you may be telling a patient that they don’t need chemotherapy, when the benefits of chemotherapy may be still significant. And I think that that’s kind of a philosophical question of how much data does one need before you put something into clinical practice.

The updated ASCO Guideline Panel, which you mentioned, happened because there were new data. And the new data came from this prospective study of the 70-gene assay. And so based on these new data and the way the ASCO Guidelines are set up now is that we wanted to keep this kind of an evolving process and not something that becomes obsolete, because you wait 2 or 3 years between each change.

So the way the process is set up is that if there is a new data set that we think may impact clinical decision-making, that is then reviewed by the ASCO expert panel and used to reevaluate the guidelines. And so that’s what happened when the MammaPrint data came out from the MINDACT study. So this was a very specific change to the guidelines that only pertained to MammaPrint, the 70-gene assay and MINDACT.

MINDACT trial: Utility of the 70-gene assay in selecting patients with BC and 1 to 3 positive nodes for adjuvant chemotherapy

DR KROP: The trial included a pretty wide range of tumors, both node-negative and node-positive. The lack of utility of MammaPrint is not in node-negative, per se. It’s in patients who are determined to be low clinical risk. And that included a lot of node-negative cancers. It actually included a few node-positive cancers. And so the determination of low clinical risk is done using an Adjuvant! Online-based type of categorization.

DR LOVE: Which is not online, so basically, clinical staging.

DR KROP: Yes. Right.

DR LOVE: Of course, you have a 4-cm node-negative tumor — you’re talking about smaller node-negative tumors.

DR KROP: Right.

DR LOVE: I guess.

DR KROP: No. That’s it. Right. But you’re right that Adjuvant! Online, unfortunately, is not online right now. But we did include the way you can do it on your own in the manuscript.

But essentially the idea is, low clinical risk patients aren’t going to need chemotherapy no matter how biologically sensitive they are to chemotherapy, because they’re going to do well regardless.

DR LOVE: But that’s with 70-gene.

DR KROP: Correct.

DR LOVE: Okay.

DR KROP: I think — right. That’s with the data we have from the 70-gene assay in the MINDACT trial. But I think what we’re trying to say is that if you have a patient who has a high clinical risk as defined by that trial — and obviously that’s an argument where you draw the line between low and high clinical risk. But if it’s a patient who has a high clinical risk as defined by the trial, that the MammaPrint assay, if you come back with a low score, a low genomic risk score, would suggest that that patient’s going to do well and that it’s unlikely that chemotherapy is going to make a big impact on further reducing their risk.

So the major data set from MINDACT — and it was somewhat of a complicated design, in my opinion. But the major point was not so much the predictive value, it was more the prognostic value, meaning if in a patient population who has a high clinical risk if you get a low genomic risk score from MINDACT, that patient population is likely to do well with hormonal therapy alone. It doesn’t specifically exclude a benefit of chemotherapy, but it says that they’re likely to do well without it anyway.

There is a randomization in that trial. And in that randomization, it did not appear that chemotherapy offered additional benefit either in the node-positives or in the node-negatives, about a 1% difference.

DR KROP: You can’t just look at the cancer. You have to look at the patient as well. I would say in a 70-year-old woman, a 1.2-cm ER-positive, node-negative cancer, I’m not going to do a genomic assay on that patient, again, because their risk of complications from chemotherapy are higher and competing risks of mortality. But in a 32-year-old, I am going to do a genomic test for that 1.2-cm cancer.

DR LOVE: And we tease this out all the time with our surveys. But just out of curiosity in terms of — yes, you have to individualize it. But 2.2 centimeters, node-negative, the 70-year-old woman?

DR KROP: No. I wouldn’t do it in that patient either.

DR LOVE: That’s interesting.

DR KROP: But I would in a 60-year-old. I mean, at Dana-Farber now, we reflexively do Oncotype for 1-cm cancers for patients who — 1 centimeter or greater cancers, who — patients under the age of 65 or so.

But I think everybody has to be comfortable with where they draw that line. And I think, as you said, you have to tailor it to the patient as well. It’s not just the biology of the cancer that matters.

DR LOVE: And it appears, at least from this case — and I guess there are other cases — that you will consider a genomic assay. It sounds like you prefer a 21-gene in some node-positive patients.

DR KROP: Yes, I do. And I agree that right now, that’s counter to the current ASCO Guidelines, but I think that’s going to evolve as well. And, obviously, we will have randomized data from Oncotype once the TAILORx trial comes out and eventually in node-positive patients as well.
 
I think it’s worth, to me at least, highlighting the fact that I really do think that these genomic assays have been — next to HER2-directed therapy have probably had the biggest impact on our care of patients in breast cancer of anything in the last 20 years. I mean, I think the fact that most oncologists — and you have experience with this from your surveys — are getting comfortable using these assays. And we’re cutting down the use of chemotherapy substantially.

There are now studies showing that oftentimes these assays can help change decision-making in almost half the patients in whom they’re used. And I just think these have really been a tremendous improvement in our care. And I’m glad that people are now comfortable with them.

DR LOVE: Yes. It’s interesting. I don't know whether people who are newer to oncology realize that. I mean, this has really maybe been 10 to 12 years that the data came out. And prior to that time, there was a consensus conference — I think it was in 2000 — that said everybody with a tumor over 1 centimeter gets chemo.

DR KROP: Yep. Yes. No. That’s exactly right. And that seems so archaic right now. And you’re right. It’s not very long ago. And it just shows you that we all kind of knew that the chemotherapy was not likely to be very beneficial in those patients. But having that objective genomic score makes it a lot easier to talk to patients and makes us feel a lot more comfortable deferring chemotherapy in that patient. And I think it’s been a big deal.

ABC trials: TC versus anthracycline/taxane-based chemotherapy for high-risk HER2-negative BC

DR LOVE: The other paper I wanted to ask you about relating to the issue of adjuvant chemotherapy is, we had seen the data first presented — now it’s been published in the JCO — from the so-called ABC trials, looking at anthracyclines. Seems like in a lot of ways we’ve kind of gotten beyond some of these questions. But some of these trials are still reporting. What did they look at here? And from your point of view, what did it mean?

DR KROP: Yes. So I actually thought that was an impressive data set. So we have these old data that we’re comparing anthracyclines to nonanthracycline chemotherapy. And some of them did show a slight benefit to anthracyclines. And if you look at the Oxford Overview, which did a meta-analysis of many of those trials, there was a slight benefit to anthracyclines. But a lot of people weren't quite sure about that.

A lot of those older trials included HER2-positive patients not treated with trastuzumab, so we know that the anthracyclines are an effective drug in that population. So I do think that despite the fact that this is 2017, that the anthracycline question was still a little bit up in the air, especially given that there was this randomized trial comparing AC versus TC, which seemed to show a little benefit against anthracyclines.

So again, I think it was an open question. I think the investigators of the ABC consortium did a nice thing. They took 3 different trials, all of which were relatively small, and they did a combined analysis. And by doing that, we got enough patients — I think a little bit over 4,000 patients — to be able to make a conclusive result. And rather than 3 inconclusive results, which I think is probably what would have happened.

So what they did is, they combined 3 trials that all compared a TC, or docetaxel/cyclophosphamide, regimen, 6 cycles, versus various anthracycline/taxane regimens. And in a mix of ER-positive and triple-negative breast cancers. And what they found was that — and this was a noninferiority analysis, trying to show that a nonanthracycline regimen was noninferior to an anthracycline/taxane.

And what they found was that the TC regimen did not meet the noninferiority boundary. In fact, it was statistically inferior. But interestingly, essentially all of that benefit was in triple-negative patients and high-risk ER-positive patients, which, again, I think is consistent with what some of the previous studies have shown. But it really was nice to have it fairly in concrete terms. So I think the bottom line is, if you have a high-risk patient, particularly a triple-negative patient, it makes sense to still use an anthracycline. For lower-risk patients, there does not seem to be an added benefit to using anthracyclines, and using a nonanthracycline regimen is very reasonable.

DR LOVE: I don't know exactly what the statistical rules were here in this study, but I’ve been trying to see whether there’s a hazard rate. I’m not sure if they were allowed to get one. But, I mean, the absolute difference was actually extremely small. In your mind — and maybe it’s in the paper and I don’t see it here — what’s the hazard rate for improvement when you add an anthracycline?

DR KROP: So again, I don’t remember exactly the number, but the hazard ratio was around 1.2-something, I think. But I think what’s important is that you have to look at the subsets. And although there was not a strict interaction term, if you looked specifically within the ER-positive population, all the benefit was in the node-positive, ER-positive, not in the node-negative, ER-positive, and the benefit was in triple-negatives. And within those groups, the benefit was more substantial than in the overall population, which, as I said, the hazard rate, I think, was around 1.2 or 1.22 or something.

DR LOVE: So yes. I mean, it kind of seems like it’s falling into that modest improvement that you can barely detect. But from your own global sense, without giving me a scientific answer, it sounds like you think that maybe — because, kind of, 20% reduction is about as low as you can get and still be there. But it sounds like you believe that it’s greater with, for example, triple-negative.

DR KROP: Right. No, I think — and again, we’re always, I think, more comfortable accepting something that confirms what you already believe, which probably isn’t the best way to look at data. But you’re right. I think the overall difference was very modest, as you said. But within the high-risk patients, the benefit seemed to be more substantial.

Role of anthracyclines in patients with HER2-positive BC

DR LOVE: As long as we brought up anthracyclines, what’s your view in terms of the issue of anthracyclines in HER2-positive disease?

DR KROP: Yes. So that’s a little bit more complicated. And I still think there’s no right answer, because we have limited randomized data. But I think that right now, it’s reasonable to use either an anthracycline regimen or a nonanthracycline regimen, using TCH.

And I don’t think we have a conclusive data set that would tell us that anthracyclines are not needed anymore. And we don’t have one that says that docetaxel-based treatment is completely the way to go for everybody.

DR LOVE: So I’d like to hear about your Case 2. We were talking about CDK4/6 inhibitors with your first case that the lady got on this adjuvant trial. Incidentally, I was going to ask you before, when this lady went on the adjuvant trial, 5 positive nodes — if a patient with 5 positive nodes like this came to you, and particularly if it was a low 21-gene score, and said, “I’ll pay for the palbo,” will you write for it adjuvantly?

DR KROP: I would not do that.

DR LOVE: Really? Interesting.

DR KROP: As you know, there have been a number of studies that have looked at drugs that looked very promising in the metastatic setting that did not pan out in the adjuvant setting.

DR LOVE: Absolutely.

DR KROP: And this is a novel class of drug. It causes a cell cycle arrest. It is possible that this drug for 2 years — these trials are using the drug — or at least the PALLAS trial uses the palbociclib for 2 years.

It’s very possible — I mean, I hope this is not the case, but it’s very possible that you can arrest the micrometastatic disease for 2 years and then, as soon as you stop the palbo, their cells are going to start growing again and you’ll have no long-term benefit. Again, I hope that’s not the case, but it’s certainly a possibility.

So I think this is very much of an unproven hypothesis, that palbociclib is going to meaningfully impact recurrence over time. Again, we’re all optimistic, but it’s certainly far from proven. And that’s why this trial needs to be done and that’s why it’s randomized and not just everybody getting the drug. So I don’t think we should be using CDK inhibitors in the adjuvant setting outside of a clinical trial.

Case: A 56-year-old woman with ER/PR-positive, HER2-negative invasive lobular BC and bone metastases

DR LOVE: I do want to get to HER2, because there are a couple of things going on there. But maybe you can just briefly talk about your next case, the 56-year-old woman, in terms of the issue of CDK4/6 inhibition. It looks like — maybe you can talk about this lady when she first presented with metastatic disease and what happened since that point.

DR KROP: Yes. So this is a very nice woman who initially presented with what we originally thought was a primary localized breast cancer. It was strongly estrogen receptor-positive, progesterone receptor-positive, HER2-negative. And during her workup, she mentioned that she’d been having some back pain for the last 6 months or so. And even though it didn’t look like she was that high risk, because of the pain I did perform staging studies. And, unfortunately, she had some bone disease.

DR LOVE: And she had lobular?

DR KROP: She had a lobular breast cancer. I think if this was a ductal ER-positive breast cancer, it would be the same situation. But she had clear concerning findings on scans, and so we did biopsy the bone. And it did come back conclusive for metastatic adenocarcinoma that was estrogen receptor-positive.

So based on that, we did not move ahead with standard adjuvant treatment and, instead, put her on an aromatase inhibitor, a bisphosphonate and — I can’t remember whether ovarian suppression at that time, because I think —

DR LOVE: Yes, she did.

DR KROP: — can’t remember whether she was premenopausal at that point.

DR LOVE: Yes. She had anastrozole/ovarian suppression.

DR KROP: Yes. And her pain got a little better. Her scans, over time, showed sclerosis. It looked like she was responding. Her tumor markers came down. And I think she started hormonal therapy about 4 or 5 months before the unexpected accelerated approval of palbociclib. And so when that happened, I think a lot of us talked to our patients about this new development. And we talked about potentially adding palbociclib to the aromatase inhibitor, but I suggested that she didn’t do that. And she was comfortable with that.

And my rationale was, we had, at that point, no data on adding a CDK inhibitor to a preexisting hormonal therapy regimen. We still don’t have any data on that. And she was doing well. And since she had de novo disease, we had every expectation that she would have a long disease-free interval, or progression-free interval, on the hormonal therapy alone. So adding palbociclib adds toxicity. It adds monitoring. And she was doing well on this therapy alone.

And that changed just a few months ago. She started having a little bit more pain. We did imaging. She had a single new bone lesion in the femur. And her tumor markers were up. So she was clearly progressing at that point. And so we switched her to fulvestrant and palbociclib. And since then, her pain’s gotten better and it looks like she’s responding.

DR LOVE: She’s having an objective response?

DR KROP: I mean, she doesn’t have measurable disease, so she can’t technically have an objective response, by RECIST anyway. But her pain’s better. Her scans looked a little better. And she didn’t have a whole lot going on to begin with. We could have a whole conversation about how to follow patients with bone-only disease. But clinically, I think she’s improving.

Emergence of ESR1 mutations in patients with ER-positive mBC

DR LOVE: So I think there’s a couple of clinical questions this case brings up. I am kind of curious, because I know being in your center, people get a lot of assays. I don't know if she’s had any kind of genomic assays. You’ve written about estrogen receptor mutations. Did you look for that in her?

DR KROP: So we have. But it’s a little difficult, because the only tissue we have from her is her primary biopsy.

DR LOVE: Right. Right. In the bone.

DR KROP: In the bone. And there wasn’t enough tissue there to do DNA analysis on the bone.

DR LOVE: Right. Hmm. Interesting.

DR KROP: So we don’t know. On her primary cancer, she doesn’t have any obvious actionable mutations. We’ve looked. We do that in all of our metastatic patients.

DR LOVE: As long as we brought it up — and again, you’ve looked at this issue — could you talk about what these mutations are and, at this point, kind of what they mean?

Clinical significance of ESR1 mutations in patients receiving fulvestrant for ER-positive mBC

DR KROP: Yes. And of course then the next question is, how do you find them?

DR LOVE: Right.

DR KROP: So what they are are mutations in the ligand binding domain of the estrogen receptor, so they’re a specific area of the gene. And what they do is, they basically lock the estrogen receptor into an active confirmation. And so in the laboratory, if you take these mutated estrogen receptors, they still have their normal activation function but even in the absence of estrogen. So essentially, this is an estrogen receptor that’s on regardless of whether there’s estrogen around. And, of course, that’s a problem, because at least aromatase inhibitors and tamoxifen work by binding to the estrogen receptor in one way or another.

And so if the estrogen receptor doesn’t care about estrogen anymore, then that’s going to make the cancer somewhat immune to the effects of these kinds of antiestrogen drugs.

DR LOVE: And particularly those that try to take away the ligand.

DR KROP: Right. So you’re right. So that’s with —

DR LOVE: AIs.

DR KROP: — tamoxifen and aromatase. So that’s why there was hope, at least, that drugs that degrade the estrogen receptor, like fulvestrant, would still be active in patients who have an estrogen receptor mutation in their cancer. And that does appear to be the case. If you look, they’ve done some retrospective analyses of some of the randomized trials comparing aromatase inhibitors versus fulvestrant. And the effectiveness of the aromatase inhibitors definitely seems to be compromised, if you have an estrogen receptor mutation, whereas fulvestrant still seems to be relatively active. So it does look like degrading the estrogen receptor still, at least, moderately active even in this presence of an ESR1 mutation.

DR LOVE: So — and I guess if you have someone who hasn’t gotten anything, if you knew they had a mutation, then I guess you would start with fulvestrant?

DR KROP: Yes, except here’s the thing. These estrogen receptor mutations don’t happen in primary cancers. They only seem to occur as a way of becoming resistant to an aromatase inhibitor. So — at least the larger data sets so far would say that you don’t find the estrogen receptor mutations in patients who haven’t had an aromatase inhibitor.

DR LOVE: So actually, because as we were talking, I thought of a question I haven’t asked anybody before. And I’m dying to know the answer, because we’ve worked a lot with your colleague Jeff Oxnard in lung cancer.

DR KROP: Yes.

DR LOVE: He’s had a leading role in looking in lung cancer for mutations in the serum, in the plasma assays.

Detection of ESR mutations in the plasma of patients with ER-positive BC

DR LOVE: What about looking in the plasma for ER mutations?

DR KROP: Yes. So as usual, you’re always ahead of the game. So you’re right. So that does look like the way to find these estrogen receptor mutations.

DR LOVE: Wow! Interesting.

DR KROP: And so, in fact, most of the large data sets that have been analyzed have been analyzed using circulating DNA. A lot of our patients have these bone-only cancers. And so it’s hard to do these assays on the DNA from the bone. So most of the data has been with circulating DNA.

DR LOVE: I didn’t even know that. So, for example, like, does Guardant 360 pick this up?

DR KROP: Yes.

DR LOVE: Really?

DR KROP: Yes.

DR LOVE: Huh.

DR KROP: So yes. So that does seem like that’s the way to go. And all the correlative studies that have been done on most of the metastatic ER-positive trials lately, looking both for ESR1 mutations and PI3 kinase mutations, have been done using circulating DNA.

I think that we need to do a better job of getting this to be both clinically available, and we need to have clear — at the same time, we need to be able to tell doctors what to do with that information, because right now, we don't have any prospective studies that would say, “Oh. If you have an ESR1 mutation, you should definitely avoid an aromatase inhibitor,” or, “You should only use fulvestrant.”

And we think that may be the case, but we don’t really know that. So we need to do trials to tell oncologists what to do with the information that they’re going to start getting from ctDNA assays.

DR LOVE: So I don't know if you view this as, quote — again, I’m thinking about the lung model. I don't know if you view this as a driver mutation, but, I mean, tamoxifen’s not a TKI. Are there TKIs that theoretically could hit these ER mutations?

DR KROP: So probably not, because ER is not a kinase. So, I mean, the best approach right now looks like just get rid of the estrogen receptor. And so we already have fulvestrant, but fulvestrant is not an ideal drug because of pharmacokinetic issues.

So I think a lot of interest right now is developing new estrogen receptor degraders. And there’s a series of what we call SERDs. So kind of like tamoxifen is a SERM, which is selective estrogen receptor modulator, these are selective estrogen receptor degraders. So they’re oral drugs that both block estrogen binding as well as degrade the estrogen receptor.

And there was some interesting data from one of them presented at ASCO this year, which showed some decent amount of clinical activity in pretreated patients. That Maura Dickler had presented some data. So it does look like these drugs are active. And there’s a number of them kind of making their way through the development process. So that could be one way to overcome these estrogen receptor mutations.

DR LOVE: That’s fascinating. I always thought fulvestrant was a really cool drug, but it just came in and sat there for a long time.

CDK4/6 inhibitors as first-line therapy for patients with ER-positive, HER2-negative mBC

DR LOVE: This case, getting back to your case, does bring up a couple of other issues. First of all, looking at this retrospectively — I mean, she was diagnosed before — or she had mets before palbo was even approved. But would you have given her palbo up front? And are there patients right now who you’re giving first-line endocrine therapy where you don’t use a CDK inhibitor? And I’ll just throw in, how do you choose between ribo and palbo?

DR KROP: Yes. So those are all not necessarily easy questions, and they’re important things right now as we start learning how to use these drugs. So does everybody need to go on a CDK inhibitor in the first line? I don’t think so. This patient I just talked to you about is a perfect example.

She has de novo disease. We know that patients with de novo metastatic disease — so they never developed any resistance yet. We know that those patients have a better prognosis than patients who have recurrent disease. So it’s very likely that a patient with a biology such as this, strongly estrogen receptor-positive breast cancer, is going to do well with her first-line hormonal therapy, regardless of what that is.

So do you need to be using a CDK inhibitor in a patient who’s probably going to be on their first-line therapy for 2 or 3 years without that CDK inhibitor? What are you bringing to the table by adding a CDK inhibitor to that person? You’re obviously adding extra toxicity. Are you meaningfully going to improve their long-term outcome by adding the CDK inhibitor from the get-go? We know the CDK inhibitors work in the second-line setting. Is it appropriate to hold off until that situation?

And I personally think it is. The hazard rates for CDK inhibition are the same in first line versus second line. We know that they’re active drugs. And for a person like this, who I think is going to do well on an aromatase inhibitor or tamoxifen for a while and doesn’t have a big burden of disease, so you don’t really — you’re not trying to maximize response rate, I think it’s okay to defer adding the CDK inhibitor until the next line of therapy.

But that all being said, for most of our patients who present with recurrent disease, given this very strong data from PALOMA-2 and from the MONALEESA trials in the first-line setting, I think most of our patients, we should be adding the CDK inhibitor in the first line, but not everybody. The people who you think are going to do well without — on hormonal therapy alone, people who you’re worried about toxicity because of other problems with CDK inhibitors, I think it’s perfectly fine to start out with hormonal therapy alone and use the CDKs later.

Efficacy and tolerability of CDK4/6 inhibitors for patients with ER-positive, HER2-negative mBC

DR LOVE: So what about choice? How do you decide between palbo and ribo? What do we know about abema? And if that becomes approved, then how are you going to make that decision?

DR KROP: Yes. So it’s already hard enough with 2. And then we’re going to have 3. I mean, based on the data that George Sledge presented at ASCO about second-line abema, which looked very good — we already know from press releases that the first-line abema study also, I think, is going to be positive. So presumably, it will eventually be approved. So we’re going to have 3 CDK4/6 inhibitors. How are you going to pick which one to use?

It’s remarkable that the hazard rates for these drugs are essentially identical for most of the studies in first and second line. They’re all around 0.5. And the absolute benefits have been a little different from one trial to another, but I think that’s probably because of the populations that each trial used. So it looks like the efficacy is pretty similar.

The toxicity between ribo and palbo also look incredibly similar in terms of percentages of significant neutropenia. The only difference I see comparing palbo to ribo is that ribo does have these rare instances of LFT abnormalities and potential for QTC prolongation. Again, they’re quite rare. So I’m not sure that should be a driving decision for us. But I think clinicians should probably think of these drugs as being fairly identical right now, unless something in longer-term follow-up comes out to suggest that they’re different.

Activity and tolerability of abemaciclib

DR LOVE: What about abema? Of course, you guys have done a lot of the research on the drug. And the thing that I thought was fascinating was — I don't know. People didn’t seem to be that excited, but I thought it was cool that it caused monotherapy responses, which you didn’t see with the other two. What do you think about that? And how would you integrate that if it became approved?

DR KROP: Yes. So you’re right that we did participate in some of those early studies with abema. My colleague, Sara Tolaney, led a few of those. And unlike the other 2 drugs, abema does have a single-agent response rate of 20-something percent in pretreated patients. And the other ones don’t. Abema is given continuously, whereas the other ones are intermittent schedule. But we actually thought that abema actually may distinguish itself.

And it does in terms of toxicity. It doesn’t have nearly as much neutropenia. It’s got less than half the rate of serious neutropenia as ribo and palbo. But it does have more diarrhea. In the data that Dr Sledge presented, the Grade 3 diarrhea rate was something like 13%, which the other drugs don’t have. It’s, like, 1% in those. So it’s a different toxicity profile.

But I think, to be honest, I was a little disappointed in that the abema data looked so similar to the ribo and palbo data in terms of hazard rates. I kind of thought it might be a little better. So I really don’t think, despite the fact that there is this monotherapy activity, I think from the large data sets we have, it’s hard to get too dogmatic that one drug should be used versus another, unless somebody’s not tolerating one. You could consider switching to the other, abema, once it gets approved.

Case: A 52-year-old woman with a Stage I, ER/PR-negative, HER2- positive, poorly differentiated IDC

DR LOVE: So I don't know if we’re going to get to your PARP case, but, I mean, there are 3 PARP inhibitors out there, too. So that just became approved in the last year or so, a number of situations kind of like that, where multiple drugs are available.

DR KROP: Yes.

DR LOVE: Let’s talk about, though, first a little bit about HER2-positive disease. And you had a 52-year-old lady maybe we can take a look at in terms of the way you approached her.

DR KROP: Yes. So she actually was kind of very much of a bread-and-butter type of patient that we’ve seen a lot over the last few years — this was a 52-year-old woman who, very healthy, had a mammographically detected Stage I. It was a T1. I think she was, like, 1.4-cm, HER2-positive, hormone receptor-negative cancer who had a negative sentinel lymph node, did not have lymphovascular invasion, but it was a poorly differentiated cancer. So it was a typical HER2-amplified breast cancer, but Stage I.

And this is a relatively common scenario. And in the past, these cancers kind of were a little bit of a black box, because the adjuvant trastuzumab trials all mainly focused on high-risk patients. They didn’t include these lower-risk patients. So there really wasn’t a standard of care for this population. But on the other hand, we know from retrospective data sets that these still do have a significant risk of recurrence in the absence of treatment.

So you can’t blow them off, but giving them TCH or ACTH always seemed like overkill.

APT trial: Results after a 7-year follow-up of adjuvant paclitaxel/trastuzumab for lower-risk, HER2-positive BC

DR KROP: So based on that, Sara Tolaney and Eric Winer in our group led this 400-patient trial where they took patients with largely Stage I — although there were a few patients who had cancers between 2 and 3 centimeters, but all node-negative — and treated them with 12 weeks of paclitaxel and a year of trastuzumab, with the idea that really what we need to do is give them the trastuzumab. And a little chemotherapy is probably going to be enough. And you didn’t really need to give the multiple chemotherapy agent type of regimen.

And the trial was designed to show that this regimen was associated with a low risk of toxicity. Sara presented the data a few years ago and published it, showing incredibly good outcomes after roughly 4 years of follow-up. And now at ASCO this year, she presented the 7-year follow-up data. And we always had the concern that they’re small cancers, so maybe they need longer to recur. So maybe we’ll start seeing higher recurrence rates over time. But we did not see that.

So at 7 years, the rate of distant recurrence risk was 1%, which I think is quite impressive. And it really says that this regimen, which is very well tolerated, really should be the standard of care for most of our Stage I HER2-positive patients. It’s really hard to argue that they need more treatment with a 1% distant recurrence rate at 7 years.

DR LOVE: And we certainly have seen, though, widespread adaptation of this so-called APT regimen to practice.

Results of the APHINITY trial evaluating adjuvant pertuzumab

DR LOVE: However, there are a couple of things that have come up recently related to higher-risk patients, particularly node-positive disease. At least, I think people are viewing this. And obviously, one is the APHINITY trial that was presented and then published. It was presented at ASCO, looking at adjuvant pertuzumab, and then was the approval of postadjuvant neratinib. And I’d like your take on both of those developments.

DR KROP: Yes. So this is a — yes. So after not much change in a long time in the adjuvant setting, we have a couple of new developments. And so — right. So going from the low-risk patient population of APT, now we’re going the opposite with ExteNET and APHINITY. So we all had very high hopes for pertuzumab in the adjuvant setting. The data in the metastatic setting are as good as any trial we’ve ever seen with the CLEOPATRA trial in terms of a 16-month improvement in survival. The neoadjuvant setting, pertuzumab doubles the path CR rate when added to trastuzumab.

So again, we thought this really was going to be a very positive trial. And I’m on the Steering Committee for APHINITY, so I’m biased. But even still, I think we thought the data were a little disappointing. It was a positive study, but the hazard rate was 0.8 or 0.81. And the absolute differences were small.

So just to remind everybody, so APHINITY was a randomized trial of mostly higher-risk patients, but a decent number of smaller node-negative patients were included. And the trial randomized patients to your choice of several different — either anthracycline or nonanthracycline regimen with trastuzumab or trastuzumab and pertuzumab. And these were the first data that were presented. And as I said, the hazard rate was 0.81. It met statistical significance, but just barely. And the absolute benefit was small, really no benefit in node-negative patients. And in node-positive patients, the absolute benefit at 3 years was a couple of percent, or not even quite 2%.

DR KROP: I guess we’ve just come in a world where the last few targeted therapy trials have all come in with hazard rates of 0.6 and 0.5, that 0.8 doesn’t seem quite as impressive.

But I think the biggest problem, which is not a problem, is that the control arm did incredibly well. And that’s just the state of the affair right now in HER2-positive disease, is that trastuzumab-based chemotherapy works very well. And so it’s hard to improve upon that substantially when your control arm patients are fortunately doing well.

DR LOVE: So let’s think about a patient, for example, who has, say, 2 positive nodes. Let’s say they’re ER-negative and HER2-positive. And you’re going to give them adjuvant chemo and trastuzumab. Okay? So once you do that, what would you estimate to be their residual risk of recurrence, globally? What’s your own clinical ballpark guess? And how do you think that would be affected by — we could say ER-positive or ER-negative, incidentally.

DR KROP: Okay.

DR LOVE: How do you think that would be affected by either pertuzumab being added to adjuvant or neratinib used in the back end after adjuvant?

DR KROP: Yes. And so let me just — I just want to be clear for people who may not have seen or read the APHINITY data. Some people took from that that the benefit was all in the ER-negative patients in APHINITY. And really, I think the biggest thing was that the ER-positives just had a low event rate, because the hazard rates between the ER-positive and the ER-negative really wasn’t that different. It was like 0.84 or 5 for ER-positives and 0.76 for ER-negatives. So I don’t think ER status seemed to predict benefit of pertuzumab very much. It was mostly just risk. Low-risk patients didn’t get a benefit, at least so far.

So in your patient with two nodes in terms of recurrence risk, I think if you have to — HER2-positive patients recur earlier than HER2-negative patients, but there’s still a rate of recurrence over time. So you have to look at what time frame you’re talking about. But I think over 10 years, her risk of recurrence is probably, with a trastuzumab-based regimen, it’s probably 10% or 12%. That’s an estimate.

So in that patient, I actually would give her pertuzumab. In general, I think it’s reasonable, based on the APHINITY data, to use pertuzumab in most of our node-positive patients, certainly with multiple nodes positive.

DR LOVE: But then I guess you would assume that it’s only going to bump it down a couple of percent, absolutely.

ExteNET: Results of a Phase III trial investigating neratinib after trastuzumab-based adjuvant therapy for patients with HER2-positive BC

DR LOVE: And what about post-trastuzumab/pertuzumab/neratinib to bump it down a little more?

DR KROP: Right. So let’s talk about the ExteNET. So this was an interesting trial design that looked at patients who’d completed a year of trastuzumab and then randomized patients, mostly node-positives, to either a year of neratinib or placebo. And the data that’s come out over the last few years show that there was a substantial improvement by hazard rate with the addition of neratinib. Absolute differences were not huge but definitely there.

And, interestingly, unlike APHINITY, in which hormone receptor status didn’t seem to make a difference, or a large difference, in neratinib essentially all the benefit was in ER-positives, which is counter to what we’ve seen in some other HER2-directed therapy studies. But the difference is quite striking. It’s pretty much black or white there. And as you mentioned, based on the ExteNET data, neratinib was just approved by the FDA.

So who would I use neratinib in? I think, given its toxicity profile and the fact that most patients do pretty well on trastuzumab — and pertuzumab is going to knock that down a few points as well — I’m really going to reserve it for the highest-risk patients, patients with many nodes positive and probably mostly ER-positives. Although, my sense is the label was in both, ER-positive and ER-negative.

DR LOVE: Right. Yes, it didn’t specify.

DR KROP: Right.

DR LOVE: I think it didn’t even specify the time, although I assume you wouldn’t want — I don't know — how long after somebody’s stopped, if they came to you for a second opinion and it was a year later, would you do it? You have to look at the numbers, I guess.

DR KROP: So you have to look at the numbers. And I think the ExteNET group is looking at those kinds of things, whether there was a differential benefit in patients who had a lag or didn’t have a lag, or other kind — we don’t have any biomarker data yet. I think we’re all expecting that to come out soon. So hopefully we’ll have a little bit more information. But I think for the average person coming through now, I think neratinib does offer a benefit for high-risk patients.

DR LOVE: I was just thinking. I was going to ask you, if you were going to use neratinib, what would you do in terms of prevention or detection or management of diarrhea? And I was just wondering. I wonder what it’s going to be like to deal with the diarrhea of neratinib after the diarrhea of pertuzumab. But anyhow, what would you do?

DR KROP: Yes. So, I mean, I think that the neratinib investigators have done a lot of work in trying to prevent the diarrhea, because we all know that it can be a problem for patients and they — so now there’s a very strict primary prophylaxis with loperamide, and we’re also talking about using a steroid, an oral steroid as well, which also may help. So I think something that’s continuing to be worked on, but certainly a primary prophylaxis with loperamide is really important. And that’s now built into all of the neratinib studies and should be part of clinical use as well.

ATEMPT: An ongoing Phase II trial evaluating T-DM1 versus trastuzumab/paclitaxel for Stage I, HER2-positive BC

DR LOVE: So I was going to ask you before, because we were talking about APT. You now have the son of APT, so to speak, trial, where you’re comparing it to T-DM1. I’m just kind of curious clinically. Do you find that patients who go on that trial in the adjuvant setting do better with T-DM1 or APT?

DR KROP: So the trial you’re mentioning, which is called the ATEMPT study, is a study that randomized patients, similar population to APT — so Stage I, HER2-positive patients — randomized them to T-DM1 for a year or the same paclitaxel/trastuzumab. So we finished enrollment. We’re just waiting for the data to mature. The idea was to see if by using T-DM1, could you get rid of all of the standard chemotherapy toxicity? So patients don’t lose their hair.

It was a very well-received trial. Patients really liked not having hair loss, which is a big concern for many women in the adjuvant setting. And so I think patients felt that it was a very reasonable approach. And we’re now waiting for the data to mature. So I can’t tell you efficacy yet, because it takes time. But I think it was part of this idea of further deescalating therapy for these HER2-positive patients.

DR LOVE: That was really my question, which is, if you do believe, after seeing the efficacy data, that they’re the same, would you choose T-DM1? Do you think it’s easier on the patient?

DR KROP: And I should clarify that the actual trial design is not to compare the efficacy of the two, because we expect that the event rates are going to be very low. So you’re not going to be able to statistically compare. The actual comparison is for toxicity, hoping to show that T-DM1 is less toxic than paclitaxel. The major point of the trial, though, is to show that in a large population of patients that T-DM1 is associated with a very low recurrence rate, similarly to what we did with APT.

DR LOVE: But again, if that’s what’s seen, do you think you would prefer T-DM1?

DR KROP: I probably would. I think patients would — more importantly, I think patients will prefer it. I think there’s — obviously, there’s the reimbursement. And this is not a registration trial, so if T-DM1 isn’t approved in the early disease setting based on some other trial, several of which are underway, I think it’s going to be hard to be able to give patients T-DM1 in that setting. But I do think if that shows a very low recurrence rate that it would be something that would be appealing to patients.

Efficacy of enzalutamide in BC

DR LOVE: So another thing I saw that I guess a trial, single-arm Phase II study that’s being done, that’s kind of surprised me. I’ve been trying to figure out where enzalutamide and androgen receptors fit in nowadays.

But I see that you’ve got this study looking at enzalutamide with trastuzumab, which I think is really interesting.

DR KROP: You know of the data with triple-negative AR-positive patients. Clearly, enzalutamide had some activity in those patients. We presented data with — as you mentioned, of enzalutamide with trastuzumab. And there was some activity there as well. And there’s another trial that’s been completed, that’s randomized, of an aromatase inhibitor with or without enzalutamide in ER-positive patients, because AR is very prevalent on ER-positive breast cancers.

At the same time, there’s actually drugs stimulating AR, that’s a drug called GTx-02 — I’m sorry — 024, enobosarm, which is an agonist of an androgen receptor that’s also being tested in ER-positive cancers. So the bottom line is, AR is probably fairly complicated in its mechanism, particularly in ER-positive patients, but it’s commonly expressed across breast cancers, less so in triple-negative, but certainly in some. And I think it’s an open question right now, whether there’s really going to be enough of a signal there to make it worth further pursuing.

And as I said, I don’t have the keys to enzalutamide. But there will be more data to come. And so we hope to present the ER-positive data set soon. And maybe that will change people’s mind. Again, I don't know what the situation is.

DR LOVE: That’s really interesting.

Case: A 40-year-old woman with mTNBC and a BRCA1 mutation receives olaparib on the OlympiAD trial

DR LOVE: So I want to finish out with your thoughts about, again, the OlympiAD data. I see that you have a patient who actually — a 40-year-old woman who actually participated in the study. It looks like she was on it for about 9 months, had, I guess, initially stable disease, minor response.

DR KROP: Yes.

DR LOVE: I was curious, though, what happened with her tolerability wise — I see she had some GI problems, some neutropenia — and what your experience is, if — I don't know how much experience you have right now with olaparib or PARP inhibitors in breast cancer in terms of tolerability.

DR KROP: This particular patient, as you said, had a recurrence after chemo. She was triple-negative, with a classic Ashkenazi BRCA1 mutation and had a recurrence a few years after adjuvant chemotherapy. And as first line, she was on that trial and got randomized to olaparib and had a little bit of a response but on it for 9 months.

She had some mild diarrhea and fatigue, but she continued to work. And the neutropenia really wasn’t a clinically significant problem. I mean, this is triple-negative breast cancer. The alternative is chemotherapy or a trial. In terms of chemo — compared to chemotherapy, the olaparib was quite appealing. So I thought it was very much of a win for her. And I’m glad the trial was positive and the drug will be approved for breast cancer.

I mean, unfortunately the hope had always been that PARP inhibitors will work outside of the BRCA1 and 2 mutation carriers, but at least so far that doesn’t seem to be the case.

You know clearly, for this particular subset of breast cancers, it’s an active drug. And I think the tolerability is fine relative to what else we use in the metastatic setting.

So I think it’s a good thing for our patients that that drug will become available for BRCA carriers. And we’re still — I think many investigators are still working hard to try to figure out can you use these drugs in combination with other agents to work outside of the BRCA subsets.

DR LOVE: Yes. It’s interesting. I was telling Dr Robson of our experience in doing CME in ovarian cancer. We’ve worked with your colleague Ursula Matulonis a lot, for example. And, of course in ovary, it’s a very different story. It looks like maybe everybody benefits. But it’s a completely different cancer. I’m curious right now, the way you’re thinking through your clinical algorithm, if olaparib is approved — as you say, seems like it should be — for both triple-negative and ER-positive, HER2-negative, where do you think you’re going to be thinking about olaparib?

DR KROP: In terms of line of therapy?

DR LOVE: Yes, with BRCA germline patients.

DR KROP: I think it’s probably going to be first line, at least that’d be my — I mean, we always want to use the least toxic therapy that is active. And none of these drugs have shown survival benefit, so I don’t think the schedule is so important or the sequence is so important. So for triple-negatives, I would probably use it first line. For ER-positives, the benefit actually was a little less, it looked like in OlympiAD. And so I still would stick with hormonal therapy first and use that as a way to stave off chemotherapy once patients have progressed on hormonal therapy.