What is your take on the clinical relevance of the APHINITY trial evaluating adjuvant pertuzumab?

DR BLACKWELL: APHINITY is 4,800 patients with a median follow-up of about 45 months; 1:1 randomization. Patients had all had their surgery and then were randomized to standard chemotherapy plus trastuzumab or standard chemotherapy/trastuzumab plus pertuzumab. So really asking the important question, all the clinical activity we saw with pertuzumab in the preoperative space and in the metastatic space, can we use it postsurgically to improve outcomes for women facing HER2-driven breast cancer?

The answer is the study was a positive one. It met its primary endpoint. There was a 1.8% absolute difference between the 2 arms of the study in terms of invasive disease recurrence. And the primary signal was driven by the lymph node-positive population of patients.

Lots of discussion around was there additive toxicity when the pertuzumab was added to pertuzumab? And the reality is, I didn’t really see a safety signal. You see diarrhea. You see rash. There was nothing else. There was no increased cardiac toxicity. So at least we learned that adding pertuzumab didn’t increase significantly the toxicity of the regimen.

I think there’s been even more discussion about do we give it for this really 2% absolute difference in invasive disease recurrence? In my practice, at least, I’m going to offer it to all the patients that I’m seeing in the adjuvant setting in combination with trastuzumab, with the exception of, probably, ER-positive Stage I breast cancer. I think a lot of us have talked about which patients we’re going to give it to, which ones we’re not.

There’s really not a set group of patients that you can look at and really say, you know what, I feel like the study definitively proved or disproved that there’s a benefit for adding pertuzumab. Lots of discussion around the lymph node-negative subgroup, where there was really no benefit. Lots of people have countered, saying basically, “We need longer follow-up. This was an extremely low-risk population of patients.”

So my take on APHINITY is that pertuzumab is here to stay, that it benefits patients postsurgically. The absolutely magnitude of that benefit was disappointing, but it was still a positive study and showed a documented benefit to adding pertuzumab in the adjuvant setting.

DR LOVE: What about the issue of survival?

DR BLACKWELL: So there’s no survival benefit. And I think you and I have had this discussion over the years, which is, we hold cancer to a different standard than any other disease we treat in human medicine, which is, antihypertensives, diabetic medicines, lung medicines, none of those medicines have been required to improve overall survival in potentially curable or long-term chronic diseases. I don’t personally need to see a survival benefit to offer medications to my patients, as long as there’s not a strong toxicity signal. And we still need more events within the APHINITY study to really demonstrate either a presence or absence of a survival benefit.

DR LOVE: So could I put, like, a little bit of a different spin on what you just said, because — certainly what you said is correct. But kind of thinking about the history of how we’ve looked at the effects of adjuvant therapy, I’m kind of more oriented not so much to the absolute benefit but the relative risk reduction. And then apply that to what the risk is. And what you see here is a modest risk reduction in recurrence. I think overall it was around 20% —

DR BLACKWELL: Nineteen percent.

DR LOVE: Nineteen percent.

DR BLACKWELL: Yep.

DR LOVE: There’s lots of drugs approved by the FDA with those kind of minimal benefits. But in terms of clinical use, if you take the relative risk and you look, if they’re node-positive, sure, they’re going to get a little bit bigger of an absolute benefit. If they’re node-negative — I mean, I don't know. And then, of course, as you say, with such really mainly, I guess, financial toxicity purposes, how much benefit do you really need in order to justify doing it? But do you think that the relative risk concept can be applied here clinically?

DR BLACKWELL: Yes. I think you’re right on target. So we give 4 cycles of paclitaxel for a hazard ratio of 0.81 in ER-positive breast cancer. So that 19% proportional reduction that we saw in the overall intent-to-treat analysis from APHINITY, the APHINITY study, was exactly the same primary endpoint that was met in 9344, which set a standard of care for the past 15 years.

So I think what’s happened, at least for me, personally, is that the financial component of adding this drug has really blurred what we think is a meaningful hazard ratio/proportional reduction. And when I started my career 20 years ago, an 18% proportional reduction or 19% proportional reduction, a hazard ratio of 0.81 was like stand up at ASCO and applaud.

And now that we have many other treatment options, I think it becomes the art of oncology to help our patients understand what the benefit is and put that in the context of lifestyle toxicity, which is, for the case of pertuzumab, diarrhea and rash. And I set an expectation with my patients that this is probably going to happen. We’ll be happy if it doesn’t happen. And if it does happen, we’ll figure out how to manage it.

And then the financial component of it. And again, you and I have had this discussion before, which is, my job as a practicing oncologist is to take the population-based data and say to the patient sitting in front of me, “I’m here to help you make a good decision. And here’s a drug that we can add on top of standard chemotherapy and trastuzumab that will offer you a 20% reduction. That means we’re going to help 2 more out of 100 women just like you not have their cancer come back.”

And I think in general, my patients are willing to accept that the absolute benefit is small. And I think from a practicing oncologist, I’m willing to accept that the hazard ratio or the proportional reduction in the chance of cancer coming back is consistent with many of the things that we do in the adjuvant setting.

How, if at all, did the data with capecitabine/neratinib influence your treatment approach to patients with HER2-positive brain metastases?

DR BLACKWELL: The study looked at the combination of capecitabine and neratinib for patients who had had previously treated brain metastases that had then subsequently progressed. Two thirds of the patients had received whole brain radiotherapy as local-regional therapy, so that left about 1 out of 3 of these patients having only seen focal radiation or SRS, stereotactic radiosurgery, for their brain mets.

And what was amazing is, if you looked at volumetric measurements, which is not what we typically measure treatment response with, 50%, 48% of these patients had a volumetric response to the combination of capecitabine and neratinib. If you look at the RANO BM, which is a little bit more strict criteria and really looks at the actual unidimensional measurements, which is how we’re used to measuring cancer, which was really a high bar to meet in many studies, 1 out of 4 of the patients had an objective response.

So I thought this was actually one of the more impressive presentations at ASCO in the breast cancer space. It leaves a lot of questions of how much of this is neratinib and how much of it is capecitabine. I’m thoroughly convinced that capecitabine has activity in the brain. We’ve seen that with the lapatinib studies where we added capecitabine.

But a fairly impressive response, and it will become my go-to agent in patients who have brain mets, maybe had 1 or 2 SRSs, one was left that was small. If they’ve not seen a small-molecule HER2 inhibitor and they’ve not seen capecitabine, that will be my next therapeutic choice based on the results of that study.

DR LOVE: So did you actually put patients on that study?

DR BLACKWELL: Yes. We did. It was done through the TBCRC.

DR LOVE: And I’m just kind of curious in terms of tolerability. Any problems when you combine capecitabine/neratinib? Neratinib itself has issues in terms of diarrhea.

DR BLACKWELL: Yes. Sure. They both cause diarrhea. If I was going to use it in clinical practice, I’d probably start at 1,600 mg/m² total of the cape a day. And if patients didn’t get diarrhea on that, then I’d probably bump it up to 1,800, although I don’t feel really compelled to bump it up. So you have to be very, very aware of that and obviously be very aggressive with your loperamide prophylaxis.

DR LOVE: Did you see any patients that had objective responses and significant clinical benefit from your point of view?

DR BLACKWELL: Yes. So I had 1 patient that did have a clinical benefit on the study. And she actually did well for a while. What people I think recognize in day-to-day practice and wasn’t probably reflected in the presentation at ASCO is these people were really, really sick. I mean, two thirds of them had had whole brain. These were not the hardiest patients. So I think in that context, knowing, kind of, the poor functional status of patients with progressive brain mets, to see this kind of response and really not any unexpected toxicity is quite impressive.

DR LOVE: So it’s interesting that now that neratinib is actually available clinically — I’ll ask you whether or not you have or will plan on using or recommending to your patients to use it as postadjuvant therapy. And also, you’ve already mentioned 1 situation with brain mets, but another one would be HER2 mutations. Are there other — would you consider neratinib (a), HER2 mutations or (b) in the clinical indication as postadjuvant therapy?

DR BLACKWELL: Yes. So let me start with the brain mets and the HER2 mutations. So for brain mets, I think this will become my preferred agent over lapatinib. When we studied capecitabine and lapatinib in a similar study performed by Nancy Lin, and we put patients on that study as well, we only saw an objective response rate of about 20% to 25%. Here we’re seeing a response rate in the 48% to 50% — I think it was like 48%. So if you do an unfair cross-trial comparison, it does appear that in a similar population of patients with progressive brain mets you see a higher response rate, which really does matter in progressive brain mets. So I will be using this.

The question will be, do you use it in combination with capecitabine in patients who’ve already seen capecitabine? I’d have a tough time going back to capecitabine. I might actually use it like I use lapatinib, in combination with trastuzumab, and try to avoid some of the diarrhea.

DR LOVE: Neratinib and trastuzumab?

DR BLACKWELL: Yes. And so we actually have a paper that’s being prepared of the safety and tolerability of trastuzumab and neratinib. And it looks very, very well tolerated. So again, for patients facing brain mets, given Dr Freedman’s data at ASCO, I think this is going to be a very active agent.

DR LOVE: So how about HER2 mutations?

DR BLACKWELL: Yes. So as you know, there are 2 large studies looking at the use of neratinib in patients who do not have a classically defined HER2-positive breast cancer.

So we have a paper that’s actually been submitted looking at the activity of neratinib in a study led by Cynthia Ma at Washington University, specifically asking the question, does neratinib have activity in patients facing HER2-mutated not classically amplified or -positive tumors? And some of this data’s already been presented.

And then there’s a second neratinib study. And it looks like the drug does have activity, if the tumors have specific HER2 activating mutations. So if I get that result back on my next-gen sequencing assay, I certainly will try to get the patient on study. As we know, sometimes that’s challenging. And now that neratinib’s commercially available, I think I would certainly have no hesitation in using it.

In what situations, if any, would you offer neratinib as postadjuvant therapy for patients with HER2-positive disease?

DR BLACKWELL: I’m going to offer it to everybody. So maybe I don’t like to stop and think too much in the middle of a busy clinic day, but the approval is for patients who’ve completed a year of trastuzumab. And that’s it. There’s no “only in ER-positive,” “only in lymph node-positive.” Will I have a much, much lower threshold to stopping the drug if they’re ER-positive and lymph node-negative? Yes, I will. But I think at this point, we have an approved drug after a year of trastuzumab that’s been reviewed.

And we have raked the ExteNET study over the goals. We’ve chopped it up in little bits. And as much as we didn’t want to believe that there would be this effect, there is an effect. It does meet its endpoint. It actually — again, we were just talking about hazard ratios and proportional reduction. This fits into that category. Here’s another, layered on top of 20% proportional reduction in the chance of your cancer coming back if you take this pill for the next year.

So am I going to be overtreating some people? Absolutely. Can I figure out which of those patients who I’m their advocate? I can say, “You don’t need neratinib”? I can’t at this point.

I mean, again, I think that there are patients who might choose not — at a year, patients are ready to be done. But there are patients, perhaps because they have an excessive risk, maybe perhaps because I’ve scared the bejesus out of them, I don't know — who say, “What else can I do?”

And for that patient who says, “What else can I do,” especially node-positive, ER-negative — especially ER-negative — even though the subgroup was less impressive in ExteNET for the ER-negative, if my patient asks me or if I say there’s 1 other thing we can do at the end of a year of trastuzumab, I think we’re obligated to say, “There’s this study and this drug. And you can take it. And it might cause diarrhea, but it’s been shown to further reduce your risk of your cancer coming back.” So yes, I feel obligated to offer it to everybody.

What are your thoughts about the Phase III ALTERNATIVE study of lapatinib, trastuzumab and endocrine therapy for ER-positive, HER2-positive metastatic disease?

DR BLACKWELL: So I thought this was in the top 5 ASCO presentations, maybe in the top 2. The study’s kind of designed with a strategy that’s near and dear to my heart, which is, I do think there’s synergy between small-molecule inhibitors of HER2 and trastuzumab. And as you know, we did studies that showed that this combination improves survival over lapatinib alone. So this is a combination I’ve been using for 10 years in my day-to-day clinical practice without any real, large-scale clinical data to support it, but it made sense to me, which is, we’ve seen lackluster responses.

So let me tell you the study design. So what Bill did was, he took about 380 patients who were facing ER-positive and HER2-positive metastatic breast cancer and he randomized them to 1 of 3 arms, an AI plus lapatinib, an AI plus trastuzumab or an AI plus lapatinib and trastuzumab. So it was this idea of lapatinib/trastuzumab, is that better than single-agent HER2 inhibition in combination with an aromatase inhibitor?

The study basically showed a 6-month absolute improvement in overall survival for the combination of lapatinib/trastuzumab and AI over AI/trastuzumab alone. So that’s pretty impressive, if you think about it. In these patients — and I see them probably once a week in my practice — who come in de novo ER-positive, HER2-driven breast cancer — whether or not they had trastuzumab in the adjuvant setting probably to me doesn’t determine my thinking about this patient.

So patient comes in. And I think the biggest decision you have to make for ER-positive, HER2-positive first-line metastatic breast cancer is, do you give them a taxane/pertuzumab/trastuzumab? Maybe they have bone-predominant disease. Or do you give them endocrine therapy plus a HER2-targeted agent? And I have to say, that’s a very tough decision, right, because we have a survival benefit with adding pertuzumab to trastuzumab. So you can do that and then add endocrine therapy and continue the pertuzumab. That’s the PERTAIN study.

Or, you can spare that patient the up-front chemotherapy by starting them on endocrine therapy. And I think Dr Gradishar’s study really suggests that if you’re going to go along that “no chemo at the beginning and try endocrine therapy plus HER2-targeted therapy,” you’re obligated at this point to add a small-molecule inhibitor. I feel pretty strongly about that, given what we saw in terms of progression-free survival and overall survival from his study.

And the p-values, I should point out — the study was stopped early, in part because there was T-DM1 that was introduced. There was pertuzumab. So there were some accrual limitations. I think it was initially designed to enroll 700 patients. It ended up enrolling about 380, and so some of the p-values in this 3-arm comparison, pair-wise comparisons, don’t reach statistical significance. But the absolute differences in terms of adding lapatinib on top of trastuzumab and endocrine therapy are quite meaningful in my mind.

DR LOVE: Could I say, with all due humility, that to me, like, it seems like this study, the ALTERNATIVE study and the PERTAIN study actually didn’t address the question you said, which is chemo or not.

DR BLACKWELL: Yes, it — they didn’t.

DR LOVE: So how do you — why does that help, making the decision? To me it’s just like, if you’re not going to use chemo, okay, use this approach. But it doesn’t really address whether to use chemo or not.

DR BLACKWELL: Yes, but — and so it doesn’t. But it makes me feel a little bit better that if I’m not going to start someone on chemo/pertuzumab/tras, they’re still going to get the benefits of a dual HER2 blockade. And for me, this combination of AI plus trastuzumab alone always feels a little skimpy, knowing what we’ve seen with the survival benefits from pertuzumab and trastuzumab. So it does not help in any definitive way. Should I give chemo/pertuzumab, or should I go with endocrine therapy?

But it sure as heck makes me feel more supported — because I was doing this for a long time, which is, “Okay. I’m going to give you endocrine therapy, whatever you haven’t had, and I’m going to double down on HER2, because we know that’s important.” And you can’t get it — honestly, you can’t get it covered, the pertuzumab. And we have no data with pertuzumab and up-front endocrine therapy, removing the chemo. So this makes me a feel a heck of a lot better about doubling down on HER2 and giving endocrine therapy for those patients that are minimally symptomatic, ER-positive, HER2-driven metastatic breast cancer.

How do you generally care for women with small, node-negative, ER-negative, HER2-positive breast cancer?

DR BLACKWELL: I follow the APT eligibility criteria. And so 3 centimeters or less, ER-positive, I will give them 12 weeks of paclitaxel and trastuzumab, no pertuzumab.

Where I kind of flip-flop is the ER-negative population of patients, so a 2.8-cm, ER-negative, HER2, those were not the patients — we participated in the APT study. And the reality is, those are not the patients that we put on APT. There were some of those patients. They typically tended to be older patients with ER-negative, HER2 driven. But I usually like to see the tumor size in ER-negative at least less than 2 centimeters, preferably even lower than that. Two centimeters or greater, ER-positive in a relatively healthy patient, I’m not — I don’t look at age. I look at do you go to the grocery store? Do you wash your car? Do you vacuum? Those kinds of things. In a relatively healthy patient, 2 centimeters, ER-negative, I’ll give them TCHP.

Would you order a genomic assay for a 65-year-old woman status post-surgery with a Grade II, 0.8-cm, ER-positive, HER2-negative, node-negative breast cancer?

DR BLACKWELL: So in this case, I would order a 21-gene assay. And maybe that’s just because I tend to be a lumper, meaning if you have ER-positive breast cancer where I’m not clear in my head that there’s going to be an absolute benefit from chemotherapy, which is the majority of lymph node-negative, ER-positive breast cancer — I don’t tend to pay too much attention to size — I order the assay and see.

For me, in these smaller tumors, I mean, you do get surprises. And the literature would support that there are some very aggressive ER-positive small breast cancers that were probably picked up early. And I don’t particularly trust the ER staining, honestly, especially if it’s anything less than 100% or Allred 9. We see all these varieties of quantitation of ER on our immunohistochemistry. And I don’t think it’s as precise as the 21-gene assay in helping me not just the ER but just really figuring out how much can we rely on adjuvant endocrine therapy to prevent that patient’s cancer from coming back.

I actually tend to think of it in the converse a lot, which is, does it help me determine whether or not there’s going to be a chemotherapy benefit? The other component that we forget a lot about with the 21-gene assay is it really helps us understand how much the endocrine therapy is going to help. That’s the converse of really — in fact, that’s probably the more precise question that was asked in the retrospective studies of endocrine therapy, because all of the patients got it.

So I think, independent of the size, I would order the assay, because you can get surprised. That’s the first thing. And (2), it gives you a real marker of how much you can rely on the endocrine therapy to prevent the cancer from coming back. And final thing, 8-mm cancer that comes back with an intermediate Oncotype, those are challenging. But again, I go back to the patient needs to be informed. “This is your risk if you don’t take the chemotherapy. Here’s the risk if you do.” And I think they need to be — it’s 2017. We are involving patients in this decision-making more and more.

DR LOVE: It’s interesting, your point about the ER with the 21-gene assay. Have you had patients that you went in thinking, quote, they were ER-positive and you came out saying, “I’m not even going to give them hormone therapy”?

DR BLACKWELL: No. I mean, the default tends to be to initiate endocrine therapy even when the ER is even borderline on the Oncotype. I have had patients who I thought were ER-positive, perhaps were sent from a smaller community hospital and you got a path report that says, “ER-positive, 3+,” and then the Oncotype comes back at borderline. We repeat it, and it’s not really ER, 3+. It might have looked brown under the microscope, but quantitatively, it’s certainly not 3+.

Would you order a genomic assay for a 65-year-old woman status post-surgery with a Grade II, 1.5-cm, ER-positive, HER2-negative breast cancer with 1 positive lymph node?

DR BLACKWELL: So an ER-positive breast cancer, I use them routinely. And again, you’ve heard my bias, which is, small tumor, node-negative or large tumor, node-positive, there’s a whole spectrum there. And I think we have a single point in time where we’re diagnosing breast cancer. And we don’t know if that cancer’s in the lymph node because it’s aggressive or because it just took us longer to diagnose it. And that’s where the genomic predictors really help me in thinking through the benefits of chemotherapy.

A low 21-gene assay, a low 70-gene assay in node-positive, to me that’s kind of the IQ test of tumors. That’s what I explain to patients. And it’s in the node not because it’s smart but because it just took us longer to make that diagnosis.

DR LOVE: We asked about what assays people were using. And at least at this point, we didn’t see many people who are using the 70-gene assay. Are there situations where you utilize it?

DR BLACKWELL: I utilize it when it’s been sent at an institution outside of my own. We’re not mandated to use one assay over the other. I think again, in a day-to-day practice, what’s more important is that you have a knowledge base around the assay as a practicing oncologist and can explain it to the patient. It’s not black or white. It’s a tool in our toolbox of figuring out basically who needs chemotherapy and who doesn’t. So I tend to use the 21-gene assay, mainly because I’m most familiar with it, to be quite honest, and (2), I think in general I’ve had the most experience with it.

DR LOVE: So — but you said you do it when somebody else orders it. Does that mean you don’t order it?

DR BLACKWELL: If it’s already been sent on the tumor, I certainly don’t do 2 genomic assays. I use it. Honestly, I think it has the same information that we get within the Oncotype. I know that both commercial parties that make these are saying, “Mine’s better because of this,” “Mine’s better because of this.” “This is retrospective.” “This is prospective.” I mean, I think there’s a lot of small differences, but at the end of the day, I think either one can help me help the patient make a decision about the benefits of receiving adjuvant chemotherapy.

So I thought the updated ASCO recommendations for biomarker use in breast cancer were very appropriate, which is, just like we’ve said probably for 10 years with the 21-gene assay, which is, “These are guidance assays.” And for the patient that has 3 positive lymph nodes, that has this kind of — 70-gene assay, you still have to use your clinical gestalt a little bit. And I think that’s what they are trying to hint at, which is, this is one of multiple things that you need to look at in making the decision about whether or not you’re going to recommend chemotherapy or not for patients facing lymph node-positive breast cancer.

To what extent, if any, are you using genomic assays for decision-making in the neoadjuvant setting for patients with ER-positive, HER2-negative breast cancer?

DR BLACKWELL: I use genomic assays in the setting of neoadjuvant therapy planning, because the most common scenario that I get in my clinic is, the surgeon sends me an ER-positive patient — let’s say 4 centimeters — clinically node-negative, maybe has a small- to medium-sized breast. We’ll say a C-cup bra size. And they send them to me and say, “Shrink this tumor.” Right? The patient highly desires breast-conservation therapy.

And again, I think there is still a persistent perception that if you give chemotherapy, cancers melt away, and maybe more so amongst the surgeons than the oncologists. So they send the patient to me, 4-cm tumor, “Shrink the tumor.” And I look at it and say, “It’s strongly ER-positive, strongly PR-positive. I might not give this patient chemotherapy if I order a genomic predictor in the postoperative setting.” So that’s a situation where I’ll use it.

And (2), I think you really have to set an expectation that that tumor’s not going to melt away, especially if it has a low Oncotype score. So it’s too bad it takes a couple of days to a couple of weeks to get these assays. I think that’s the other kind of confounder is that frequently the frenzy of a newly diagnosed breast cancer patient, we want to do something, right? And frequently what happens is the surgeon’s already set an expectation. “Go. Go see Dr Blackwell. She’s going to melt the thing, and then I can save your breast.”

And then we have to do a lot of backtracking, which is, “There’s only a 10% chance that we’ll have a complete pathologic response. I don't know if you even need chemotherapy for long-term disease-free survival benefits” — I don’t use those words, but that’s kind of the gist of what I talk to patients about. And so I will use it in that scenario.

I don’t use it in the patient where there’s not really an indication for preoperative therapy. In fact, I think we do a lot of patients — and I think we talked about it at the symposium at ASCO. We’re doing some patients a disservice by cookie-cuttering and saying, “everyone needs neoadjuvant therapy.” I just don’t think that’s the case anymore, especially with ER-positive breast cancer.

So I’m constantly having to have a discussion with the surgical teams about what are the indications for neoadjuvant therapy. If you know the patient’s going to have a mastectomy to begin with and you know that they’re clinically node-negative, there is no indication to do anything in the preoperative setting. That’s the reality. So why not take that patient to surgery? Because they have to have a mastectomy no matter what. They have multicentric disease. And then get the genomic predictor off a larger piece of tumor in a more — kind of, the tumor’s out. The patient’s calm. Let’s have a discussion with all the information.

So I think we are tending to think a little bit more about who really does need preoperative therapy. And the patients who do need it, we have to set an expectation of how well it’s going to work in ER-positive breast cancer. And that’s where the genomic predictor really comes in.

DR LOVE: It’s funny. In one of the surveys we did — and we’ve done this many times in meetings — we asked this question: Do you use genomic assays? And typically half or more people say, “No.” And we present a clinical situation like you described, the patient needs to have tumor shrinkage. “What would you do?” They say, “Give chemotherapy.” And then we change the case and we go, “Same exact case, what would you do, except the surgeon sent a 21-gene assay and it’s low.”

DR BLACKWELL: Yes.

DR LOVE: And then they don’t give chemotherapy. So it’s kind of — I don't know, like, not quite logical to me.

DR BLACKWELL: No. And I think in general, we’re still as oncologists caught up in this idea that we give chemo and we’re melting cancers away. And then it’s always amazing to me when I see a patient for a second opinion that has a strongly ER-positive breast cancer who got, let’s say, dose dense, and then they still had 5 positive lymph nodes. And everyone’s freaking out. And there was a radiologic response, but it’s because the tumor was slow growing and ER-positive to begin with. It’s not because it’s an aggressive cancer, it’s just that chemotherapy doesn’t melt those kinds of tumors. And that’s where I think the Oncotype can really help you.

In what patients, if any, do you use genomic assays in making decisions about offering extended adjuvant endocrine therapy?

DR BLACKWELL: So I’ve only sent a couple of them. And the scenario was something along the lines of, “I’ve finished 5 years of an aromatase inhibitor. I had lymph node-negative breast cancer. I’m barely tolerant of the medicine. It’s not making my life miserable, but show me where it’s going to help me.”

And in one patient, she had a history of blood clots. So I couldn’t necessarily switch her to tamoxifen, which I think we’d all agree is a better-tolerated medicine, especially if you’re going to take it for another 5 years.

And so I’ve sent it, probably, maybe 10 people at most. And it’s really the patient that comes in and says, “Do I really need to take it?” In my experience, there’s 2 groups — the majority of patients fall in 1 of 2 groups. Either, “If you try to stop this medicine, I’m going to kill you. And there’s no way that I’m going to stop it. And I’m going to find another doctor. And I’m doing just fine. I don’t even know that I’m on it.” I don’t need to send a predictor there.

And the other population of patients is, “If I have to take this medicine another day and all my friends have taken it for 5 years and you told me I need it for 5 years, I’m going to find another doctor who tells me I can stop it.” So if you have a patient that falls in one of those 2 categories, I don’t think there’s any test in the world that’s going to change.

And there’s no test that really says at the end of 5 years, there’s 100% certainty that you will benefit from either continuing it or stopping it. If we had that test, I’d be sending it on everybody. But what this is really telling us is what the risk for recurrence is after 5 years of endocrine therapy. And it leaves a lot of kind of nonspecificity for the patient sitting in front of you.

DR LOVE: Yes. Hal Burstein has a good way of saying that. He says the patient who has a tamoxifen T-shirt and the patient who has a calendar that has the fifth-year day marked out on it.

DR BLACKWELL: Yes, I get those invitations quite a bit. And so it’s a weird dynamic with newly diagnosed patients, because you kind of have to say — especially with newer evolving data — when you put them on it, “For right now I’m going to recommend 5, but that might change.” Because, the worst thing in the world is to have a patient get that number in their head and then you see them at 5 years and they’re planning their party, and you say, “Oh, by the way, I think you might need to continue it a little longer, because you had 5 positive lymph nodes,” or something like that.

Do you routinely use CDK4/6 inhibitors in combination with first-line endocrine therapy for patients with ER-positive, HER2-negative metastatic disease?

DR BLACKWELL: In my own practice, I’m utilizing the CDK inhibitors in the first-line setting or, less commonly, in the second-line setting for patients who are getting endocrine therapy and they’ve been on it for first-line metastatic. They’ll definitely get it in the second-line setting. To me, that’s kind of a no-brainer. I hear probably once a week, “Why would you give it to a patient that has 3 bone mets?”

But the reality is that the bone-only population, at least from the PALOMA studies, derived as much, if not more, benefit in terms of hazard ratio for the addition of the CDK inhibitor. It does not look like burden of disease predicts the proportional or absolute benefit from adding the CDK inhibitors. And I have not seen a forest plot yet where the CDK inhibitor did not benefit a subgroup of patients. So at least in my practice, if you’re facing metastatic ER-positive breast cancer, you’re getting a CDK inhibitor.

I think at least in my second opinion referrals, I’m pretty shocked at how many patients are getting initiated on endocrine therapy, either in the first- or second-line setting, and they’re not being started on a CDK inhibitor.

DR LOVE: Wow! That is surprising.

DR BLACKWELL: I don’t really understand it. I would say probably twice a month, I see a patient who comes in, has been started on endocrine therapy, where I’m having to actually call the referring physician and say, “Oh, by the way, I think you should add a CDK inhibitor.” I have to say the number of this happening has dropped over time. So I think people have become more aware of it.

I think there’s a population of providers out there that say, “If there’s not a survival benefit, I don’t know if it’s worth the hassle and the cost and the toxicity,” but the PFS benefits are so significant across the board with the CDK inhibitor. When I say to a patient, “I can give you an endocrine therapy by itself or I can give you 2 pills and this is going to increase the time of the drug working by about 10 months, but you’ve got to come see me once a month, we’ve got to do blood work,” they’re fine with that. Ten months is very meaningful for patients facing metastatic breast cancer. So bottom line, I’m using it across the board. And I’m pretty surprised they’re not routinely incorporated at a fairly high level.

DR LOVE: What about in the patient who has disease relapse or metastatic disease that’s not occurring on an AI? Either they present up front or, for some reason, they’ve had endocrine therapy, it’s been stopped. How do you approach that situation, particularly in terms of selection of the endocrine therapy?

DR BLACKWELL: Yes. So both MONARCH 2, PALOMA-3 and I think it’s MONALEESA-3, those are the 3 CDK inhibitor trials that look at fulvestrant as the endocrine therapy backbones. And from my perspective, I give them the endocrine therapy they’ve not had. So if they had adjuvant AI and it’s been more than a year, I’ll put them back on an AI and a CDK inhibitor. If they relapse on an adjuvant AI or they’ve progressed on an AI — those are going to be the patients that have been on an AI in the first-line metastatic setting for a couple of years. Then I’ll use fulvestrant.

But I don’t perceive that you have to use the CDK inhibitors with any particular endocrine therapy backbone. I’d pick your best endocrine therapy backbone, and I would incorporate a CDK inhibitor.

How do you decide which CDK4/6 inhibitor to use in combination with endocrine therapy?

DR BLACKWELL: That becomes more of a practical issue, right? So we have approval for ribociclib in combination with letrozole. I’m a big fan of their blister pack, because what you see with all these CDK inhibitors, for at least the first 2 months, is a lot of confusion. They have to be taken for 3 weeks, and then 1 week off. But the majority of women, at least in the US, have utilized oral contraceptives at some point in their life and they understand the blister pack. And it makes their life a little easier, and I think it helps compliance. So in general, I think that’s a real positive.

The current approval in combination with fulvestrant is really only with palbociclib, so I’ll use that. We don’t have data with ribociclib. So a lot of it depends on the endocrine therapy backbone. A lot of it depends on patient assistance program and really which drug can I get for the patient the fastest. Because remember, these women have just been diagnosed with metastatic breast cancer.

And sometimes there’s a lot of hassle in getting the CDK inhibitor. So you see them one week, and then you have to see them back the next week. They haven’t gotten their CDK inhibitor. And there’s a lot of kind of very practical staff issues that I think we’ve worked through, at least in my own practice. But for practices that are just kind of starting to routinely incorporate them, it’s good to think about what the system’s going to be.

DR LOVE: But just kind of putting that aside, at least reimbursement aside, am I hearing you say that in general if you are going to use an AI, you’re going to use the ribo, because you like the blister pack?

DR BLACKWELL: That’s not the only thing that drives it. I mean, if I look at MONALEESA-2 and I look at PALOMA-2, which are the 2 letrozole-based studies of a CDK inhibitor, the hazard ratios appear very similar.

At ASCO 2017, we did see the updated overall survival from MONALEESA-2. And I think the p-value is like 0.056 or something. So it does look, in terms of survival, that MONA-2 — and again, it’s one point of time, but it was a planned analysis. If 1 of these 2 drugs creates a survival advantage in the first-line setting in combination with letrozole, then I think you’re going to be obliged to use the one.

But for right now, I don’t see huge, at least, effectiveness differences across the trials.

How do you generally care for patients who experience disease progression while receiving endocrine therapy and a CDK4/6 inhibitor?

DR BLACKWELL: I use a lot of everolimus. And I know people perceive that as a fairly hard-to-get-patients-through drug, but the hazard ratio, last time I checked, for adding everolimus to exemestane was 0.4-something. It’s actually one of the smallest hazard ratios we’ve seen for adding a targeted agent on top of the endocrine agent. So I do routinely incorporate that.

I think the most common scenario in my practice is, “I was on an adjuvant AI. I either stopped it or I relapsed on the adjuvant AI.” Then I’ll use fulvestrant plus a CDK inhibitor. And then I’ll use exemestane plus everolimus, because I do think these targeted agents — again, it’s all about kicking the can to having to give the patient chemotherapy further and further down the road by incorporating targeted agents.

And I think we’re going to start seeing some data in the large Phase III alpha-specific PI3 kinase inhibitors. They might take the place of everolimus in my practice if they become commercially available. But I’m pretty excited about those drugs as well.

DR LOVE: Is that monotherapy, or combined with hormones?

DR BLACKWELL: Combined with endocrine therapy, in that case fulvestrant.

DR LOVE: Hmm. And incidentally, the assays, if that does come into practice, is that like an NGS assay, or what?

DR BLACKWELL: Yes.

DR LOVE: How do you do it?

DR BLACKWELL: So both of the studies have 2 cohorts. And they’re both fulvestrant alone versus fulvestrant plus. And both of those studies have — I think they’re cohorts. I don’t think it’s stratification. So there’s like 2 arms, right? So it’s I have a known PI3 kinase mutation or I don’t.

And so they’ll get an approval either on the whole population if it works in the nonmutated —

DR LOVE: Wow!

DR BLACKWELL: — or just approval in the mutated. That’s how the studies are designed.

DR LOVE: When do you think those trials are going to report?

DR BLACKWELL: I think that the PI3-specific arms are fully accrued in both studies.

DR LOVE: Wow!

DR BLACKWELL: And I think SANDPIPER is actually more fully accrued than SOLAR-1.

DR LOVE: That’s fascinating.

DR BLACKWELL: Yes. So we’ll hear something in the next year, only because if there’s a signal, we should be able to see it.

In general, what is your treatment algorithm for patients with a germline BRCA mutation and ER-positive, HER2-negative, metastatic breast cancer?

DR BLACKWELL: So in an ER-positive BRCA-mutated tumor, I would utilize a CDK inhibitor/endocrine therapy first, in part because of the activity and in part because of really what I consider a fairly favorable side-effect profile. And then in the second-line setting, I would use olaparib.

I think the data presented in the OlympiAD study is quite impressive. It’s at least as good as chemotherapy. But there’s a fair amount of toxicity, lots of transfusion, lots of anemia, lots of fatigue. I’ve been using olaparib in my BRCA mutation carriers really since it was approved for ovarian cancer, because it made sense to me and a lot of these patients really had very limited treatment options. And it’s not a drug without side effects.

The good news is, it’s active. And I’ve had a couple of patients — I have 2 or 3 patients with metastatic disease who’ve been on it for, like, 2 years, because — I know that because I just had to fill out the preauthorization for their third year on the drug to get it covered. But there is a fair amount of fatigue. And you have to watch the counts. And in general, I think endocrine therapy plus CDK inhibitor is better tolerated.

You have to kind of set an expectation that GI side effects, poor appetite, fatigue are common with this drug. Then I think the patient’s at least expecting it. I made the mistake of just handing it out like it was candy when it was first available. And my patients came back and said, “This certainly isn’t candy and I don’t know if I” — and with dose reduction and some modification of things, it’s much better tolerated.

What are your thoughts about incorporating olaparib into the care of your patients with a germline BRCA mutation and triple-negative metastatic breast cancer?

DR BLACKWELL: I might use it in the first-line setting for triple-negative. And I think what we see is progression-free survivals with standard chemo of somewhere between 3 to 5 months. I wouldn’t start it and have my triple-negative breast cancer patient come back in 8 weeks. I mean, I wouldn’t have them come back in 8 months or even 4 months. I’d probably start it and rescan in 8 weeks.

The one thing that we saw in the OlympiAD data from ASCO is that the responses occur fairly early, just like chemotherapy. So I would monitor the patients on this. People I think think just because it’s a pill, it takes longer to kick in. I would start it in the first-line setting and monitor very closely.

DR LOVE: So when you say “first line,” are you saying first line is somebody who’s had adjuvant anthracycline/taxane?

DR BLACKWELL: Yes.

DR LOVE: What about somebody who’s never had chemo?

DR BLACKWELL: I would have a tough time. I would give them some type of a taxane, at least. Maybe I could argue my way out of giving them an anthracycline. And then I would certainly try it. And a lot of it depends, Neil, on the burden of the disease they’re facing, too. If I need to really debulk, I don’t think that the single-agent PARP inhibitors, even in a mutation carrier, is going to be as active as gem/cis or a taxane/carbo kind of thing.

DR LOVE: Yes. That was my next question in terms of the patient who’s symptomatic or has bulky disease, and even in a patient with a prior anthracycline/taxane, in a situation like that would you be tempted to use a platinum-containing regimen? Would you ever use platinum and then PARP maintenance? I don't know if you can even get it paid for.

DR BLACKWELL: I think that’s interesting, given the NOVA data in the ovarian world. Right now, I wouldn’t necessarily take someone off of treatment and put them on a PARP inhibitor if they’re benefiting from it.

Although I will say that I and Rebecca Dent out of Singapore have a study asking that exact question in non-BRCA-mutated breast cancer, which is, we’re giving patients facing triple-negative breast cancer who respond to carboplatin, we’re randomizing them to a PARP inhibitor versus a PARP inhibitor plus a checkpoint inhibitor, kind of —

DR LOVE: Wow!

DR BLACKWELL: — the maintenance. It’s the NOVA equivalent but in breast cancer. So outside of the setting of a clinical trial, I would not do that. But I think it’s a good clinical trial question.

DR LOVE: Yes. It’s really weird that in ovary, they just went right into this idea of maintenance.

DR BLACKWELL: Yes.

DR LOVE: Like chemo and then followed by a PARP inhibitor. And yet in breast, it was just monotherapy.

DR BLACKWELL: I think our practice styles are different. I think they give big doses of carbo/paclitaxel —

DR LOVE: Absolutely.

DR BLACKWELL: — and in 6, and say, “You know what? Let’s give you a break.” In breast, we tend to treat until progression or toxicity.

In general, how do you care for patients with de novo asymptomatic ER-positive, HER2-positive metastatic disease?

DR NANDA: We definitely see a fair bit of de novo metastatic HER2-positive, hormone receptor-positive disease. And I do consider hormonal therapy with HER2-directed therapy in some of these patients, particularly in those with bone met-only disease. So in patients who’ve got bone metastases only, low-volume bone metastases who are asymptomatic — although some of these patients do present with a little bit of bone pain, where I’ve given some radiation to palliate their symptoms, and I do consider them on antiestrogen therapy in combination with HER2-directed therapy.

I think with the CLEOPATRA data, we know that there’s certainly a survival advantage with giving dual HER2-directed therapy and chemotherapy. But what we don’t know is, do you need to start with that in patients with low-volume disease? So it’s a frank conversation with the patient to think about what her wishes are. Is she hoping to maybe delay starting chemotherapy?

And I’ve had a number of patients who I’ve given hormone therapy and trastuzumab to alone who’ve done very well for years. And so I certainly think there is a segment of the population that can benefit from HER2-directed therapy, maybe monotherapy with trastuzumab and endocrine therapy. And I think I would probably reserve starting chemotherapy and dual targeted therapy for those patients with bulky, very symptomatic visceral metastatic disease.

DR LOVE: What about the choice of endocrine therapy in this situation, de novo disease in a postmenopausal woman, both in the HER2-positive and even in the HER2-negative situation?

DR NANDA: I think for endocrine therapy, I typically would use aromatase inhibitor therapy. So I think you can certainly choose. I don’t think one’s going to be any better than the other. If someone’s premenopausal, I’d probably consider giving ovarian suppression along with aromatase inhibitor therapy.

DR LOVE: What about fulvestrant or fulvestrant and an AI?

DR NANDA: I think we certainly have data about fulvestrant plus AI being better than an AI alone in the randomized Phase III setting. But again, I don't know that I would do dual hormone receptor blockade therapy. I’d probably just favor an aromatase inhibitor and trastuzumab, potentially alone.

How does time since relapse after adjuvant therapy for HER2-positive disease influence your choice of initial therapy for metastatic disease?

DR NANDA: I certainly do take that into account, because obviously if patients can potentially still benefit from trastuzumab/pertuzumab and chemotherapy, you certainly don’t want to miss the opportunity to have a therapy that patients can benefit from. And a lot of times with insurance companies, it can be challenging to get pertuzumab for patients down the road. In the front-line setting, it’s much easier. So I certainly would consider using chemotherapy with pertuzumab and trastuzumab in patients who’ve been quite some time, at least a year, out from their adjuvant therapy completion.

Now, I think there’s not a hard and fast rule. If someone’s relapsed very quickly after dual HER2-directed therapy and chemotherapy, I probably would favor moving on to something like T-DM1. But I think in patients who’ve had a nice period, at least a year or 6 months of disease-free interval, I would certainly reconsider rechallenging them with trastuzumab and pertuzumab with chemotherapy.

What are your thoughts about continuing indefinitely versus stopping “maintenance” HER2-directed therapy in patients with metastatic disease?

DR NANDA: It’s not that common, but I certainly have seen it. And as an example, I have a patient who had hormone receptor-positive, HER2-positive disease and, after a period of 5 years, I actually stopped the HER2-directed therapy and have her on hormonal therapy alone. And she’s been doing well for over a year.

So I certainly think there are patients. Now, the concern is, if you stop suppressing the HER2 signaling pathway, are you going to then have a recurrence? But then there’s nothing to say that those patients can’t benefit from reinstituting HER2-directed therapy at the time of disease progression. So I do consider stopping it. But I think it’s an individual discussion with a patient. Some patients want to remain on that therapy, and others are very happy to stop and just be followed.

Would you consider administering a CDK4/6 inhibitor to a patient with ER-positive, HER2-positive metastatic disease outside of a trial setting?

DR NANDA: I think that’s a great question. And you could argue that, for my patient who has hormone receptor/HER2-positive disease, at the time of disease progression, would I perhaps want to consider a CDK4/6 inhibitor with her aromatase inhibitor therapy, as opposed to restarting HER2-directed therapy? I don’t think we have data on that, but certainly it’s a discussion that can be had with the patient.

I think there are trials ongoing now, so hopefully we’ll have some data there. But I don’t think there’s any reason to expect that CDK4/6 inhibition wouldn’t work for patients with HER2-positive disease.

Would you comment on the OlympiAD study results and how you think it will affect treatment for patients with BRCA-mutant metastatic breast cancer?

DR NANDA: There have been a number of studies that have looked at PARP inhibitors in breast cancer with varying degrees of success, but certainly the OlympiAD trial is a practice-changing trial, in my opinion. We’ve seen some patients that have been heavily pretreated, treated with olaparib, for example, who’ve had some responses, but not very long lasting.

So in the study that Dr Robson presented, the OlympiAD trial, this was a randomized 300-patient trial. Patients were randomized to olaparib versus chemotherapy of physician’s choice.

And what we found was that those patients who received olaparib in the setting of a BRCA1- or 2-associated advanced breast cancer had a longer progression-free survival with olaparib versus chemotherapy, an improvement in almost 3 months. So that was quite impressive and was statistically significant. And I certainly think that trial was practice changing.

So I think, while some of the earlier studies haven’t really shown a significant improvement, it’s really promising to see a progression-free survival in patients and to finally have a targeted therapy to offer these patients.

DR LOVE: What have you been hearing from your colleagues about this trial, about this strategy? What have you been doing, what have your colleagues been doing since you saw these data? Of course we have the issue that at the moment it’s not approved in breast cancer. Obviously it is in ovarian cancer. I don't know how easy or difficult it might be to actually access it. But what are you doing nowadays?

DR NANDA: Yes. I mean, there’s still a number of studies ongoing. So obviously if I have a study for a patient, that’s an easy way to potentially get access to a PARP inhibitor for those with BRCA1 and mutations. I have previously used olaparib, compassionate use, in patients with BRCA1- and 2-associated advanced breast cancer and have been able to get compassionate use through the company. So I believe that is a program that’s still ongoing, probably made easier by these data.

Now, the dose that’s been used for the OlympiAD trial was a little bit different than the dose that’s approved for ovarian cancer, but it’s certainly commercially available. And I suspect many insurance companies will allow us access to it even commercially.

DR LOVE: You were mentioning the improvement of several months’ progression-free survival. But also, when you look at the hazard ratio, it’s 0.58, which seems pretty significant. What about survival? It looks like it’s not really statistically significant, but it seems to be trending around in the same area of 50% or so.

DR NANDA: Yes. There was a trend to improvement in survival. It wasn’t significant. I think we need more follow-up. But the study wasn’t powered to detect a significant difference in overall survival, too. So even with longer follow-up, we may not quite get there.

DR LOVE: The other thing that was interesting was the response rate was much higher. It was just about double, almost 60% versus 30%. What are your thoughts about that? I would think the 30% in the control arm actually sounds pretty good for a late-line chemotherapy. Any comments about response rate?

DR NANDA: Yes. I think 60% is a very impressive response rate. And my understanding is that the responses were very quick. So when you think about patients that have bulky disease and they’re symptomatic from their disease, it’s really reassuring to think that you can get a quicker response rate and potentially have a better improvement in quality of life in terms of symptoms from disease burden.

What do we know about the tolerability and side effects of olaparib compared to chemotherapy from the OlympiAD trial?

DR NANDA: The investigators did study quality of life. They used a survey instrument and surveyed patients at different time points across their disease course. And what they found was that those patients who were assigned to the olaparib arm actually had a better quality of life than those patients who received chemotherapy. So that was pretty reassuring.

Not only is this drug more effective with a higher response rate and an improvement in progression-free survival, it’s also more tolerable.

The main toxicities that were seen were cytopenias — and these were managed with dose reduction — and some low-level nausea, which can be managed with antiemetics. So again, overall, great efficacy and great tolerability.

DR LOVE: Can you talk a little bit more about the cytopenias? At least in ovarian cancer, where it seems like these drugs have been used a lot more, with olaparib you hear about anemia. There’s niraparib, where you hear about thrombocytopenia. What do we know in general about these cytopenias?

DR NANDA: Yes. I mean, they’re certainly related to the PARP inhibition’s effect on the bone marrow. And we definitely did see anemia and neutropenia and thrombocytopenia. But again, they can be managed with dose modifications. And that’s certainly what we do in the clinical setting when we see these patients in the clinic.

DR LOVE: One of the issues, again, that we heard about in ovarian cancer — there’s a lot less data in breast cancer — is the question of whether or not there’s an increase of MDS. These patients have had a lot of chemotherapy, a lot of alkylating agents. What do we know right now about PARP inhibitors and MDS and AML?

DR NANDA: Yes. I think, at least in the breast cancer population, these drugs have primarily been used in patients with metastatic disease. So they’re not really living long enough for us to see an MDS. It’s certainly a possibility, absolutely, that you could see MDS in relation to PARP inhibition. And some of the adjuvant and neoadjuvant studies might help us gather some information here. But in the metastatic setting that hasn’t really been a concern, at least in the context of breast cancer.

DR LOVE: And you also mentioned GI tolerability issues. And again, we’ve heard that from the gynecologists. Actually, gynecologists sometimes use preemptive antiemetics, or at least they give the patients antiemetics and say, “Take it if you have any problems.” Again, what was seen in the trial? And what’s your own clinical experience?

DR NANDA: Mm-hmm. Clinically, having used olaparib and other PARP inhibitors in clinical trials and for compassionate use, a little bit of low-level nausea is certainly common. And I encourage my patients to use antiemetics as needed to manage this.

DR LOVE: Any other tolerability issues that docs should be concerned about with olaparib?

DR NANDA: No. I mean, it’s primarily the cytopenias and the nausea. And I think to be aggressive about dose modifications and instituting antiemetics would be the key here.

What are your thoughts about BRCA germline testing, particularly in patients with metastatic disease?

DR NANDA: I think that’s a great question. And it’s going to be a question that we’re going to all wrestle with in our clinics. The NCCN Guidelines are very clear in terms of family history and who would qualify for testing. But what I will say is when we have a therapy that we know those who have mutations can benefit from, I think we need to have a lower threshold for who we’re going to test.

I think certainly anyone with triple-negative disease should — there should be a consideration for testing, even in the absence of a family history, because this is the only targeted therapy that is potentially going to be approved for these patients.

For other patients with hormone receptor-positive disease who’ve presented with metastatic disease or first diagnosed at a later age, I think it’s a little bit more complicated. It’s not as easy as ER/PR and HER2 testing, because it has implications for family members. It’s genetic testing. But I certainly think for those patients with hormone receptor-positive disease where there’s even a hint of a potential family history of familial syndrome ongoing, whether there are other patients in the family with ovarian cancer or prostate cancer, pancreatic cancer, there should be a strong consideration given for testing. And this is something we’re going through not just for breast cancer and ovarian cancer but prostate cancer and a lot of different types of cancers.

DR NANDA: I think we’re definitely going to test a lot more people. And there’s definitely going to be a lower threshold for it. But I don’t think that we’re going to start broadly testing everyone. And part of that is because the more patients you test, if there’s not a family history or a compelling reason to check for it, we’re going to be left with a bunch of variants of undetermined significance. And we’re really not going to know how to interpret that information.

But I also think as we’re testing more patients, we’re going to gather more information about variants and hopefully be able to better categorize variants versus actual deleterious mutations.

Do we have data about other germline mutations, besides BRCA, and responsiveness to PARP inhibitors?

DR NANDA: There’s no data that I’m aware of. But there’s certainly other mutations such as PALB2, where it would make sense that if a PARP inhibitor works for a BRCA1- and BRCA2-mutated tumor that PALB2 would also be a potential target there.

There is going to be a trial that’s opening up through the Translational Breast Cancer Research Consortium. That’s going to be looking to address this question of patients who have other mutations in DNA repair genes and also somatic mutations in DNA repair genes to see if PARP inhibition may be of value. So we should get some information in the not-too-distant future about that.

What do you view as the treatment options for a postmenopausal woman with ER-positive, HER2-negative disease who had a BRCA2 germline mutation and developed minimally symptomatic lung metastases 1 year after starting anastrozole?

DR NANDA: For a patient like this, while we don’t have data comparing fulvestrant, say, and a CDK4/6 inhibitor versus a PARP inhibitor, what we do know is, from the general hormone receptor-positive metastatic population, the progression-free survival with fulvestrant and a CDK4/6 inhibitor is reasonably long. So I would probably, in this patient, favor fulvestrant plus a CDK4/6 inhibitor and then reserve the PARP inhibitor for later down the line, say before I were going to start this patient on chemotherapy.

DR LOVE: I guess the other issue is, even though PARP inhibitors seem to be generally well tolerated, you maybe would expect that the patient might tolerate endocrine therapy and CDK4/6 a little bit better. Is that part of your thinking also?

DR NANDA: I think so. I mean, certainly having given patients PARP inhibitors, that low level of nausea can be annoying. And to not have that with fulvestrant and a CDK4/6 inhibitor is certainly, I think, a little bit associated with a better quality of life.

DR LOVE: What do we know about the combination of hormone therapy even with a CDK4/6 inhibitor and a PARP inhibitor?

DR NANDA: I’m not familiar with any data in that setting. I think it’ll be interesting. But again, it’s probably a pretty small population, so I don't know that we’ll really get head-to-head comparison data.

What’s your take on the effectiveness of PARP inhibitors after disease progression on platinum-based therapy for patients with triple-negative metastatic breast cancer?

DR NANDA: I think that’s a great question. And the OlympiAD trial doesn’t really address how effective are PARP inhibitors going to be after a platinum-based therapy or if it’s better than a platinum-based therapy. I’ve certainly had patients with triple-negative breast cancer and a BRCA1 and 2 mutation who’ve responded very well to platinums for months. And I think it’s unclear. And that’s certainly one of the criticisms of the OlympiAD trial, that it didn’t go head to head against a platinum. And we don’t know if it’s more effective than platinum-based therapy.

It’s an option. And I think there’s also a suggestion from ovarian cancer that PARP inhibitors may not be as effective after progression on a platinum-based therapy.

DR LOVE: And I guess the other issue is, what do we know about combining a PARP inhibitor with chemotherapy? And what about the strategy, again, that you see in ovarian cancer, what they call maintenance, where they give platinum therapy, stop it and then use a PARP inhibitor as maintenance? What about those two strategies in breast cancer, combining and sequencing?

DR NANDA: Yes. There have been a number of trials that have looked at combining a PARP inhibitor with chemotherapy, primarily a platinum. And there’s also a SWOG study that’s ongoing now of cisplatin plus or minus the PARP inhibitor veliparib. And that trial’s ongoing now and we don’t have data from it.

But what I will say is, both platinums and PARP inhibitors can lead to cytopenias. And that’s been the challenge of combining chemotherapy with PARP inhibitors, are the cytopenias. We already see cytopenias with platinums. We already see them with PARP inhibitors. And when we combine the two, the question is, what are we going to dose reduce, the PARP inhibitor or the platinum, and is dose reducing one better than the other, because we certainly want to make sure that we get an effective dose of a PARP inhibitor.

So I think there is still data that we’re going to need to generate there to help us figure out the best strategy moving forward. We know that PARP inhibitors by themselves are better than a handful of chemotherapies from the OlympiAD data. What we don’t know, the questions that remain to be answered, are platinums going to be better than PARP inhibitors or not? Is the combination going to be better than either therapy alone? And I think we really still need some data there.

What are some clinical scenarios in which you would consider using a PARP inhibitor for women with breast cancer?

DR NANDA: I think for hormone receptor-positive HER2-negative disease, for me, I would probably use it prior to chemotherapy, so after I’ve exhausted hormonal therapy options prior to chemo is probably the place that I would use it. But again, it’s a conversation with the patient and pros and cons. And it’s very easy to transition from hormonal therapy to pills, right? That’s why a lot of times we’ll go to capecitabine first as the first-line chemotherapy. But I think in that setting, it’s pretty easy for me and I would probably go to a PARP inhibitor.

I think for triple-negative disease as well, why wouldn’t you want to use an oral agent that’s well tolerated before we start IV chemotherapy or even other chemotherapies? So I certainly would consider it up front right away. I think the earlier you use it, it’s more likely to be beneficial and offers patients a great quality of life over chemotherapy. So I think in that setting, based on the OlympiAD data, I’d probably use it before chemo.

What are your thoughts about using anti-PD-1/PD-L1 checkpoint inhibitor therapy in breast cancer outside of a clinical trial?

DR NANDA: I have not used a checkpoint inhibitor outside of a clinical trial. That’s largely driven by the fact that I have a lot of clinical trials available to me. So I certainly think for patients who have triple-negative disease who don’t have access to a trial, it’s certainly a reasonable option.

However, what we know from the trials that have been presented to date in the advanced cancer setting — and there have been a couple of larger, or large-ish, trials that have been reported — the KEYNOTE-086 cohort A trial that was presented by Sylvia Adams at ASCO this year as well as the atezolizumab Phase IB cohort study that was presented at AACR by Peter Schmidt earlier this year, have both shown that those patients who get checkpoint inhibition as their first line of therapy for advanced triple-negative breast cancer have the highest response rates, in the mid-20% range.

And the response rate 5% to 10% range. And it doesn’t really seem to matter if it’s second line or greater than second-line therapy. So I think what we know from the trials that have been presented to date is really, we want to try to give these therapies early on in the disease course, so potentially front line. And I’m very hopeful that we’re going to have some combination data that’ll be presented in the not-too-distant future.

The IMpassion study, which is nab paclitaxel with or without atezolizumab, has completed accrual. And hopefully within the next year, we’ll get some preliminary data there in terms of efficacy. But at least what we know now is that response rates seem to be highest in the front-line metastatic setting, and then it drops off thereafter. But I certainly think for patients who have the ability to get a checkpoint inhibitor outside of a trial who have triple-negative disease, I don’t see the downside to trying it.

DR LOVE: And you mentioned the idea of combining with chemotherapy. Of course now we see an approval in lung cancer with that strategy. Any other combination strategies that you think are exciting that are being looked at specifically in breast cancer?

DR NANDA: Yes. So as I alluded to, the IMpassion study is looking at nab paclitaxel with atezolizumab. There’s also data with nab paclitaxel and pembrolizumab in the neoadjuvant setting, also with paclitaxel. And Sara Tolaney presented at San Antonio last year eribulin plus pembrolizumab. So we are gathering some combination data. The drugs, the combinations, appear to be safe with no increase in toxicities that we would expect. And I certainly think, given the very low monotherapy response rates, that the future is going to be in combination, either a combination with chemotherapy, other checkpoint inhibitors, other targeted agents, radiation therapy. So moving forward, that’s really where the field is heading.

What do we know about estrogen receptor mutations and their predictive utility?

DR NANDA: I think we need a little bit more information. Estrogen receptor mutations are more common than we previously thought. They occur in about 20% of hormone receptor-positive advanced breast cancers. And there are a bunch of SERDs that are in development, right? So the serum estrogen receptor downregulators are being studied in this population, and I’m hopeful that we’ll have some more data soon.

I think that aromatase inhibitors and tamoxifen are less likely to work in those patients who have estrogen receptor mutations, but we need more data before we really have clinical decisions that are made based on mutation status.

What do we know about androgen receptor positivity in patients with triple-negative breast cancer and response to antiandrogens, such as enzalutamide?

DR NANDA: So there have been a handful of studies that have looked at antiandrogen agents in triple-negative breast cancer with androgen receptor positivity. So about 10% of triple-negative breast cancers are driven by androgen receptor signaling. And there have been a number of studies looking at enzalutamide among them in patients with triple-negative, AR-positive disease. And there have been some clinical benefits and responses seen in these patients.

These drugs aren’t approved yet in the study. There was a randomized Phase III study that was supposed to be starting soon called the ENDEAR study, that was going to look at chemotherapy with or without an androgen receptor blocker, enzalutamide, versus enzalutamide alone. Unfortunately, that study’s really been put on hold. But I think we need randomized Phase III data in this setting to really broadly use these agents for AR-positive, triple-negative breast cancer.

But I think it’s certainly reasonable to check for androgen receptor status in triple-negative breast cancers. And it may be possible to get some of these agents, bicalutamide in particular or even enzalutamide, outside of a trial, because there are certainly a subset of patients who can do well on androgen receptor blockade. But it doesn’t appear that AR positivity is really the best biomarker, and so there certainly needs to be more work done in that setting.

Would you comment on the different tests for BRCA that are available and which test you think is most appropriate for a patient with HER2-negative metastatic breast cancer?

DR NANDA: One of the most commonly used is the Myriad Genetics panel. And Myriad Genetics certainly has the most data in terms of helping us discriminate between variants of undetermined significance and deleterious mutations. And I would say that’s the testing that I primarily use.

Obviously, most of us have gone to panel testing now, not just testing for BRCA1 and 2 but PALB2, p53, among other genes that may have implications for other family members and hereditary risk in the family members of mutation carriers. Unfortunately, though, a lot of the testing that we do isn’t going to be covered in patients without a family history. I suspect insurance companies aren’t going to cover it. And there are a bunch of other tests that are much cheaper that can test for just BRCA1 and 2, where patients may be able to afford paying out of pocket for testing.

Some of these tests can be about $250 out of pocket. It doesn’t get billed to insurance. Patients just have to pay for it. And for those who can afford to pay for it, that may be a more reasonable, economical alternative to panel testing or testing more formally at some of the other companies that have insurance coverage.

Would you comment on the ABRAZO study with the PARP inhibitor talazoparib that was presented at ASCO this year?

DR NANDA: Preclinical data suggested that talazoparib is actually one of the most potent PARP inhibitors. And at ASCO this year, Dr Turner presented data on the ABRAZO trial. And the ABRAZO trial was a single-arm multicohort study in patients with BRCA1- and 2-associated metastatic breast cancer. The ABRAZO trial enrolled patients with all forms of metastatic breast cancer, including HER2-positive disease.

And the different cohorts studied the response of talazoparib monotherapy in patients who’d previously been exposed to a platinum agent and those who had not previously been exposed to a platinum agent but were heavily pretreated and had had at least 3 prior lines of chemotherapy for advanced-stage disease.

And responses were seen in both cohorts. In Cohort 1, the cohort that had previously been exposed to platinum-based therapy and actually had had documented responses to platinum-based therapy, the response rate was about 21%. And in those patients who had received other chemotherapy agents but not yet a platinum, the response rate was 37%, so a bit higher in those patients who’d not previously been exposed to platinum-based therapy.

But I think what we are seeing is that at least with talazoparib, some patients can go on to receive a response to talazoparib after exposure to platinum-based therapy. So that’s pretty exciting, because I think to date we haven’t really seen that with other PARP inhibitors.

DR LOVE: What do we know about niraparib and rucaparib in breast cancer? Niraparib in ovarian cancer is really very impressive. They saw positive results not just with germline patients but people with somatic mutations and even without that. What do we know about niraparib and, for that matter, rucaparib in breast cancer?

DR NANDA: Yes. So there are clinical trials that are ongoing now, but we don’t have any great data yet. To date, most of the data has been with veliparib, olaparib, through the OlympiAD trial and some of the earlier-phase trials in veliparib.

DR LOVE: What do we know about veliparib? The one thing that I’ve heard about is that it you can combine it with chemotherapy, which seems like a very attractive strategy.

DR NANDA: Yes. There have been a number of trials that have looked at combining veliparib with different forms of chemotherapy. Many of them have looked at combinations with platinum-based therapy. And one of the concerns about combining PARP inhibitors with chemotherapy is because chemotherapy and PARP inhibitors can both cause cytopenias, that the dose levels of the PARP inhibitor that we can achieve with the combination might not be the most effective doses that might give us true PARP inhibition.

So I think we certainly have trials that are ongoing now. And the trials have been completed. But that’s the concern. And the question is, what is the appropriate dose of a PARP inhibitor and are we going to be able to combine that dose with chemotherapy and effective strategies?

Are results from next-generation sequencing useful in determining whether or not a patient has a BRCA mutation?

DR NANDA: I think that’s a very reasonable strategy and very appropriate. I will say, in my patients who have mutations that I’ve sent whole genome sequencing on, I’ve certainly seen their mutations show up in their tumor. So that may be a backend way to get at those patients who may benefit from PARP inhibitors, and we may be unaware that they actually do harbor a germline mutation in BRCA1 and 2.

And then once that’s identified, that may be a way to go about doing germline testing and offering that information to the family members if there is a germline mutation.

Would you comment on the I-SPY 2 study that evaluated pembrolizumab in the neoadjuvant setting?

DR NANDA: So I-SPY 2 is an adaptively randomized neoadjuvant Phase II trial. And it’s been ongoing for about 7 years now. There have been over 1,000 patients who’ve participated in this trial and gone on to surgery. And in this trial, the control arm is a taxane, so paclitaxel given weekly for 12 weeks followed by 4 cycles of AC.

And as part of this trial, there are a number of investigational arms. And one of the arms — and this was the arm that I presented at ASCO this year — looked at the addition of pembrolizumab given for 4 cycles during the paclitaxel portion of the neoadjuvant therapy in patients with HER2-negative disease. So those patients with hormone receptor-positive high-risk disease and those patients with triple-negative were eligible for randomization to the pembrolizumab arm for the trial.

And what we found, because this is an adaptively randomized trial, we don’t report the results as PCR rates, although the primary endpoint is pathologic complete response. And that’s defined as the absence of invasive cancer in the breast and the axilla. And what we found was the estimated PCR for patients with hormone receptor-positive disease and triple-negative disease was much higher in patients who received pembrolizumab in combination with standard neoadjuvant chemotherapy as compared to chemotherapy alone.

There was a tripling in the estimated PCR in triple-negative breast cancer patients, an increase from 20% to 60% with those who received pembrolizumab and more than doubling in the estimated PCR for those with hormone receptor-positive disease. And this estimated PCR increased from 13% to 31% in those patients with hormone receptor-positive disease.

Success is defined in the I-SPY 2 clinical trial as the predictive probability of an 85% or greater likelihood of success in a randomized Phase III trial. And pembrolizumab is predicted to be better than standard chemotherapy for both hormone receptor-positive and triple-negative disease. So that’s pretty exciting.

And there are neoadjuvant studies already ongoing, randomized Phase III neoadjuvant studies looking at pembrolizumab with standard chemotherapy in the neoadjuvant setting in triple-negative disease. So hopefully in the next year or so we’ll have some preliminary results there.

The beauty of the I-SPY 2 clinical trials, that we’ll have serial biopsies and the surgical specimen to examine. And we have plans to look at TILs, PD-L1 expression, a variety of other immune markers as well. So I think it’ll be really interesting to see. We also have serial MRI data in these patients. And so we’ll also be able to look at MRI predictors of benefit. And I think that’ll be pretty interesting.

But right now what we know from the metastatic setting, from the atezolizumab data that was presented at AACR this year, it doesn’t appear that PD-L1 is a great biomarker to predict for success for immune checkpoint inhibition therapy, at least in metastatic triple-negative breast cancer. There was an association, however, with TIL infiltration, and the greater the level of TIL infiltration in the tumors, the higher likelihood of success of an immune checkpoint inhibitor. And that’s really preliminary data from 1 study, but we certainly have a number of other studies that we’ll be able to validate that data in.

DR LOVE: I’ve got to say, in a way it is kind of surprising if this really holds up about the path CR rate increasing. Even in lung cancer, you kind of look at it, it seems like you’re just adding 1 and 1 to get 2. But if these numbers hold up, it seems like maybe there’s something more going on here.

DR NANDA: Yes. No. I think it’s really exciting. And what’s even more exciting is these patients only received 4 cycles of pembrolizumab. We are incorporating an 8-cycle arm of pembrolizumab into I-SPY 2. And that arm has just started enrollment now. So we’ll be able to look at whether 8 cycles of pembrolizumab is even better than 4, so it’s pretty exciting.

Are we seeing prolonged duration of responses to immune checkpoint inhibitors in patients with breast cancer like we have seen in other solid tumors?

DR NANDA: While the response rates are modest in the metastatic setting in the front-line setting, in the mid-20s and a little bit under 10% in the subsequently treated metastatic triple-negative patients, the duration of response is really quite remarkable. In the KEYNOTE-012 study that I presented at San Antonio a few years ago and updated this past year, the patients, the 3 patients who were responders, responded for more than 2 years and were off of therapy. So we’re certainly seeing the same efficacy and the prolongation of response in metastatic triple-negative breast cancer that we’ve seen in other tumor types.

Would you comment on the KEYNOTE-086 trial and what was observed in patients with previously treated and untreated metastatic triple-negative breast cancer who received pembrolizumab monotherapy?

DR NANDA: So the KEYNOTE-086 trial is a follow-up to the KEYNOTE-012 study. And this was a single-arm multicohort study in advanced triple-negative breast cancer. And Sylvia Adams presented both the cohort A and preliminary results of cohort B at ASCO this year.

Cohort A was a previously treated metastatic triple-negative breast cancer population. They had to have received, these patients, at least 1 prior therapy for advanced-stage disease. And what we found was the overall response rate in these heavily pretreated patients was about 5%. So that was lower than what we’ve seen before. These patients were not selected based on PD-L1 status. They could be PD-L1-positive or -negative. But when we went back and looked at response rates by PD-L1 status, it really wasn’t any different. It was about 5% for both of those cohorts.

Cohort B was a previously untreated metastatic triple-negative population. And these patients were required to have PD-L1-positive disease. And the response rate in that setting was 23%, so significantly higher than what was seen in the previously treated cohort.

So again, I think just going back to what we know from the atezolizumab Phase IB study, it really appears that the first-line setting is associated with a much higher response rate than the subsequent-line setting for metastatic triple-negative patients.

What do we know about the autoimmune toxicities associated with checkpoint inhibitors in breast cancer, in both the metastatic and earlier-stage disease setting?

DR NANDA: So we know that adrenal insufficiency and thyroid abnormalities can be seen with immune checkpoint inhibitors, but the levels are really quite low in the metastatic setting, usually under 1%. And that’s not any different for metastatic triple-negative breast cancer either.

But in the I-SPY 2 clinical trial, we did see an increased incidence of adrenal insufficiency. So 6 of the 69 patients who were randomized to the pembrolizumab arm experienced either a primary or secondary adrenal insufficiency. And this was a little bit unexpected. I suspect it may be related to the fact that these patients are not previously treated and they have much more intact immune systems and are able to potentially mount an autoimmune disease.

But I think we still have more follow-up that we need to do on these patients to try to understand, are there certain predictors of who’s going to have an autoimmune disease from immune checkpoint inhibitors?

So I think that that still remains to be seen, but we are seeing more of it in the neoadjuvant setting than we would have suspected based on the metastatic data.

DR LOVE: That's interesting. Is the thinking that when you see adrenal insufficiency, it’s from the pituitary or the adrenal?

DR NANDA: I suspect most of them are from the pituitary, but these patients, many of them didn’t have a workup that was conclusive. And so it was hard to distinguish primary versus secondary. And these patients just needed to start on hormone replacement therapy. And interestingly, all of these women are doing very well on their hormone replacement therapy. And there did appear to be an association between response in those patients who also developed an autoimmune disease, which is consistent with what’s been seen in the metastatic setting as well, although we’re still teasing that information out.

How do you envision the immune checkpoint inhibitors playing out in breast cancer?

DR NANDA: I certainly think there will be a role for checkpoint inhibitors in breast cancer. I suspect they’ll be approved in the metastatic setting in the not-too-distant future. But I think what we’re seeing from the I-SPY 2 trial and some of the other very small studies that were presented at ASCO this year of 10 to 20 patients is these promising PCR rates, particularly in triple-negative disease, and also maybe even in hormone receptor-positive, high-risk disease. So I think there will likely be a role in the neoadjuvant setting as well, but that’s still some time away.

What’s your perspective on the role of capecitabine for patients with HER2-negative breast cancer who have residual disease after neoadjuvant chemotherapy?

DR NANDA: So the CREATE-X trial was a trial that randomized patients to capecitabine versus observation if they had residual disease after neoadjuvant chemotherapy and had HER2-negative disease. So this enrolled both hormone receptor-positive and triple-negative patients who’d had residual disease after neoadjuvant chemotherapy. There were about 900 patients that were enrolled in this study. Half were assigned to capecitabine and the other half observation.

And what we found was a significant improvement in disease-free survival for those patients who received capecitabine versus those who were observed. These were high-risk patients. The majority of them had lymph node-positive disease, T2 or larger tumors at the time of initial presentation. Ninety-five percent of them had received anthracycline and taxane-based therapy. And only those patients who had residual disease were randomized to capecitabine versus observation.

And there was a significant 6.5% improvement in disease-free survival for the overall group. And interestingly, those with triple-negative disease actually had an even more substantial benefit. So this study has certainly been practice changing. I see a lot of patients with triple-negative disease who don’t have a PCR to taxane and anthracycline-based chemotherapy. So this is certainly something that has changed my practice and the practice of many physicians around the country.

DR LOVE: What about patients with ER-positive, HER2-negative disease?

DR NANDA: I think it’s unclear how much benefit these patients may receive from capecitabine. The trial certainly showed a trend to improvements in disease-free survival and overall survival in those patients with high-risk, hormone receptor-positive disease. But we do know that these patients also benefit from adjuvant endocrine therapy. And they go on to receive adjuvant endocrine therapy as well.

So I think we’ll need some more follow-up to really understand, is there a subset of patients with hormone receptor-positive disease who are going to benefit from capecitabine in addition to their endocrine therapy? I suspect there will be a cohort. But for me, it’s primarily changed practice in the triple-negative population.