Now we’ll talk about some scenarios; we’re in a data-free zone because we don’t have randomized trials, except what we just went over. We don’t have any clear guidance on how to sequence. We have to go in part by the patient populations that were studied in the phase 3 trial. We have to go in part by antecedent therapy they had, maybe in the adjuvant/neoadjuvant, and what the disease-free interval might be. We have brain metastases, we have toxicities to consider. What are you doing? This might be a good time if you had any interesting observations or anything like that, it would be interesting and to hear about it. Who wants to start? How do you think about sequencing?

I can start. If you build off what is viewed as standard, and you’re more likely to use a CLEOPATRA-like regimen, then follow it by T-DM1 [trastuzumab emtansine]. The reality is that in the adjuvant setting some people get TH [paclitaxel, trastuzumab]. If they develop progressive disease, would that make you not give a taxane with HP [trastuzumab, pertuzumab]? It would be dependent on how recently the patient progressed and last received their therapy. I wouldn’t exclude the possibility in that kind of scenario of using HP with a taxane.

As for the newer agents, it’s to a large degree influenced by whether CNS [central nervous system] disease is present. That certainly would influence my thinking. We’ve used a lot of trastuzumab deruxtecan, and I’ve been impressed with its activity. Obviously if somebody had bad pulmonary disease for whatever reason, I’m not talking about cancer-related disease, it may be somebody who I would avoid it in, but not because they have cancer in their lungs. There’s also, obviously if somebody has active brain metastases, tucatinib would be very attractive. I’ll stop there so others can comment.

I’ll go. I do start with the CLEOPATRA regimen, and typically if somebody had a lot of CNS disease or something going on, I might think tucatinib, it’s approved as second line and beyond. There are instances where I might think a bit, but I would typically do T-DM1 next. It’s a really well tolerated drug, and we have good data for it.

In the third-line setting, I think it’s going to really depend on the patient. I have to say I was a little bit skittish about trastuzumab deruxtecan because of the interstitial lung disease and because it came just as coronavirus disease 2019 was hitting. Getting more used to everything, I’ve used it as well. I would probably think of tucatinib in the third line and then trastuzumab deruxtecan. Which way we do it, I don’t think we have evidence for; it ends up being individualized on the patient’s adverse effect profile, what’s more tolerable. It’s a lot of medication to take with the tucatinib and the capecitabine. There’s diarrhea with it, so it carries with a very different adverse effect profile. As usual, we now have 2 good, very solid options for third and fourth line, and it’s an individualized decision in my mind with the patient in terms of their preference, their adverse effects, their comorbidities, and other things that come into play.

Let me ask you Claudine because you make the point about the tucatinib being FDA approved and on National Comprehensive Cancer Network guidelines in the second line, which distinguishes it from the trastuzumab deruxtecan in the second line, which distinguishes it from the trastuzumab deruxtecan and the neratinib, not that we have to rigidly follow that. Sometimes we don’t for individual patients. Nonetheless, it’s nice to know where the guidelines and FDA stand. Having said that, is there any patient for whom you might consider the tucatinib in second line? Would you ever consider any kind of a brain MRI? We have a little ostrich effect here; we’ll look at the liver every day, but we’re not looking at that brain, we’re not going there. The question is, is there anybody we should, if you think they may be at particularly high risk for CNS metastases?

I’m not aware of what would predict for a higher risk of CNS metastases with HER2-positive disease. I don’t know if my colleagues are?

Yes, Nancy Lin, MD, taught me. I don’t know that it really pertains to the sequence of metastatic, but for those who recur in the brain, or who start off, it’s probably the same, it’s T3/T4 disease and ER [estrogen receptor]-negative. If you’re T3/T4, even ER-positive, you can be at risk, or if you have more sizable cancers, ER-negative—like node-positive, ER-negative. So, ER negativity and the bigger cancers would be the ones that would tend to seed back when these things were developing.

There you go. I always listen to what Nancy Lin says, so I’ll take that one home too. I’m debating whether we should be doing scans. I have to say, I have had that ostrich-like response. I didn’t want to know, and now that we have a drug that’s active, I think we need to rethink that strategy. I haven’t quite figured it out yet. I don’t know whether it’s something that we should be doing every X-period of time, whatever that might be, and to do it when it’s time to change therapies. It feels a little different when we know that we have something beyond repetitive local therapies for these patients to think about something. That could influence me if I had a patient who had CNS disease in the second line, and I thought I could spare and push off radiation if I chose a systemic therapy that had CNS penetration, that could influence things. I haven’t quite figured it out. I’d be interested to hear what my colleagues have to say about, if they’ve developed an algorithm for this. I’m still trying to get used to this playing field and figure out how to best strategize around that.

I don’t know that I have an algorithm, but I think it’s like alpelisib. We wouldn’t be looking for PI3 kinase mutations if we didn’t have something to do with the information. The practice is going to be as time goes on, we will be looking for more brain metastases, even in the absence of symptoms. But it does raise the issue, if there’s lead time that we’re creating, and we’re starting therapies that may or may not impact on very tiny or very indolent CNS disease. I don’t know the answer to that, but I bet we’re going to be looking more than we have in the past.

There was that subset analysis. It was Nancy Lin who did this one too, where she looked at a combination of patients who did have brain metastases and those who didn’t have brain metastases, and then looked on the tucatinib vs placebo for time to development of new brain metastases. A little preventive. There was a real nice separation in that curve where it certainly gives us hope in the curative setting. That’s helpful, but probably for right now I think if people have brain metastases, we’d all be thinking about tucatinib sooner than later, probably. Sometimes, even in my practice, leapfrogging over the T-DM1 if they had substantial brain metastases. I don’t mean necessarily 1 brain met. How about you, Lisa, how do you sequence these so far?

I have to say, I share the sequence that Claudine just outlined. I’m fairly conservative about that way. There’s some reassurance in that when they’ve done many of the trials, like EMILIA for example, they did include patients that had stable brain metastases. They did have some of them for whom they could see a response rate, and it didn’t look terribly different from the parent trial.

I suspect for a variety of reasons that tucatinib is going to outperform all of them, specifically in the CNS compartment, but I’m not so worried about it that I’m leapfrogging or rearranging my normal lines of therapy at this point. I do share Bill’s concern about routine screening in asymptomatic patients. If you look, you will find. Once you find, then you’re stopping and reconsidering, and you’re doing stereotactic radiosurgery here, there, and everywhere. I’m not sure that we’re helping patients if we do that. We might be, but I always get a little leery of it. You’re right, there is a disconnect because I’m imaging the systemic compartment all the time. But I typically don’t image asymptomatic patients for the CNS.

I echo what Claudine put out there as the framework. I am thinking about this disease now as a today problem and a tomorrow problem. The today problem is a lot of my patients have been through 1 or 2 lines of therapy. It’s a slam dunk to say I’m going to use tucatinib or deruxtecan, and maybe there are small, subtle reasons why I might pick one over the other, certainly brain metastases being present, one of the big ones. My concern is as we project forward, the patients who recur, those who had residual disease after THP [paclitaxel, trastuzumab, pertuzumab], and then we know that the patients who get T-DM1 in the adjuvant setting are still the ones having the events, right? Now they’ve taken out my first 2 lines of therapy, and if you’re recurring within 1 to 3 years, as many of these patients will, then I’m worried about going back to these agents.

Now we’re looking at in the first-line metastatic setting trying to get access to drugs like tucatinib and neratinib as well as deruxtecan. I don’t know how this is going to shake out. I’m just throwing out my hands with concern about what the near future is going to look like in these patients who are heavily pretreated in the adjuvant setting. Now we’re delivering them, and certainly if they get neratinib, like I said earlier, I’m a little worried about cross-resistance to tucatinib. Neratinib is a very potent HER2 inhibitor, it just inhibits a lot of other stuff as well. Now if you try to go back to the well and target HER2 kinases with tucatinib, will it work? We don’t have these data of course. It doesn’t exist. These are my fears and my worries for the coming days.

Can I ask VK a question on that? You go all the way out to; I’ve taken myself to a year, a year-and-a-half out and said after that I can think about reusing old agents. Does that?

I’m just worried. My anecdotal experience is that these drugs don’t maintain the same way as if it were a pristine patient in the metastatic setting who gets these drugs. I think that patient, you can very easily say, this is a person who’s going to be on these drugs for 2 years. But when those patients have been that experienced in that early phase setting and they have a metastatic event, I might, just my back of the envelope sense, is that they’re not getting that type of mileage.

Well the good news is that as we get better and better in this adjuvant setting, and we’re already seeing this in some of the modern things, they’re all de novos. All of these metastatic patients, there are like 45% in the most recent studies. That’s a testament to what we’re doing in the adjuvant setting. I share VK’s pessimism about patients with early relapse following poly-HER2 targeting, although I have to say I retry things unless there’s a compelling reason not to.

Videos

Sequencing Therapy in R/R HER2+ MBC

Let’s go on to the metastatic setting. Bill, you could lead us off with the standard of care. How do you see the standard of care, in the first and second line? 

It’s an area in flux based on some of the prior comments. There are a lot of trials that are trying to position the new drugs closer to the front than the back. Everybody wants to be up front. Today if you said, what do we typically do? We tend to give a taxane with trastuzumab and pertuzumab for most patients as the first line. We still follow that for most patients with T-DM1 [trastuzumab emtansine], and most of those data were generated from randomized trials.

In the last question, the issue is, can we move up trastuzumab deruxtecan? Can we move up tucatinib and replace T-DM1 as a standalone drug as a second-line best choice? Those trials are addressing that. For most patients, at least in our practice, we would follow the standard off of a trial, CLEOPATRA-like followed by T-DM1, and enroll in some of the trials we’ve been alluding to, to see if there’s an advantage.

Great, thank you. We’ll come back and talk about some of those different strategies for different patients. Let’s go through the data, a brief update from folks about what’s new that’s come out in 3 big new phase 3 trials that led to 3 new options in the metastatic setting—tucatinib, trastuzumab deruxtecan, and neratinib. Claudine, could you update us on what’s new with HER2CLIMB?

We have this abundance right now of options, which is great, as Bill was alluding to. For years we knew we had good data tell us what to do in first line and second line. Then third line was sort of a free-for-all, and it was really just choosing what chemotherapy partner with trastuzumab.

In the last year, we’ve seen the approval of 3 drugs in this space. Just to remind everyone about HER2CLIMB, because I stand in awe of people who do more than just breast cancer. I can barely keep up with breast cancer let alone trying to treat a multitude of different malignancies. The HER2CLIMB trial was a randomized trial looking at the role of tucatinib, which is a HER2-specific TKI [tyrosine kinase inhibitor], and was looking at it given with capecitabine, trastuzumab, looking what it added over and above that. It showed a benefit in terms of progression-free survival [PFS]. But the thing that we have now grown used to in the HER2-positive world and the metastatic HER2-positive world is that we expect to see survival benefits. We saw it with the addition of tucatinib. There was about a 4.5-month prolongation of overall survival.

The other thing about this trial is everyone who went on the trial had to receive prior trastuzumab, pertuzumab, and T-DM1. It’s asking what I think was practical for us as clinicians because that’s how we were treating most of our patients, and it showed the benefit of adding tucatinib on top of the trastuzumab and capecitabine with that survival advantage.

The other thing that we’ve talked about already is that we know that one of the issues for our patients with HER2-positive disease is that there unfortunately seems to be a predilection for the CNS [central nervous system]. In the HER2CLIMB study, they allowed patients who had stable CNS metastasis to enroll. It turns out they actually looked for it. They mandated a screening MRI to go onto the study, and what ended up happening is around half of the patients had CNS metastases.

Now to be eligible, you had to have CNS metastases that did not require immediate local intervention. These were patients who either had known CNS metastases that were stable, were found to have CNS metastases but they were small and did not require radiation, or they had had CNS metastases, and we saw they had a bit of progression, but they didn’t need immediate intervention.

The thing that came out in the trial is that there was benefit in the subpopulation, which was a pretty large one, about half of the trial altogether that had control and had the same benefit as the patients who didn’t have CNS metastasis. Then Nancy Lin, MD, presented at ASCO [the American Society of Clinical Oncology 2000 meeting], and showed that if you looked at CNS progression-free survival, it was also prolonged substantially in the group of patients who got tucatinib, cementing that this drug had activity in patients with CNS disease. So it clearly has CNS activity. It also had benefit in patients who did not have CNS metastases. It really positioned itself with this randomized trial as a drug that we need to figure out how we sequence in our armamentarium, but one that we really want to be using.

Great, thank you. We’ll come back. We’ll hear about the 3 trials, and we’ll come back and talk about sequencing strategies and your own experience with things. Thanks a lot. That was terrific. How about Lisa? What about trastuzumab deruxtecan? What’s new with that?

The most recent thing is the update of the DESTINY-Breast01 trial, which is the trial that generated the FDA approval for trastuzumab deruxtecan. That was updated at one of the recent meetings. Shanu Modi, MD, was presenting it, but I can’t remember where. I think it was recent, so it was at the San Antonio Breast Cancer Symposium. They looked at the outcome. This is a single-arm trial in heavily pretreated patients with HER2-positive metastatic disease at 20 months out. What they found was that, again, the progression-free survival was excellent. It was about 19 months. The range was still invaluable in the upper end. This is a very active drug with more than 50%, around a 60% response rate.

Counter to the enthusiasm is that it does have some significant toxicities, and in particular, the ILD [interstitial lung disease]. The numbers of the ILD were up a little. It was about 13% the first time it was presented, and it was about 15% this time, and they did have some deaths that were related to it, which we knew before. There’s some caution but it’s a wildly active drug, and we have to figure out if we can manage the ILD, who’s at risk for it, and trying to figure out where it goes.

Thank you, thanks a lot. VK, what about the NALA trial?

NALA you will recall was presented at ASCO 2019, but the paper is already out, and you can read that in the

. This is a funny trial and it’s a funny approval, and it’s hard to figure out how we’re going to use this in the modern world, so to speak.

This trial was designed as a comparator with lapatinib and capecitabine. This is a trial with neratinib and capecitabine. Keep in mind both molecules are small inhibitors of HER2 TKI, but also hit a lot of other tyrosine kinases as well. They’re not as clean and direct as tucatinib.

When they did the study, the trial had to be designed in a way that fully anticipated that the medians were not going to be very different. Because just inhibiting HER2 as neratinib does as opposed to the irreversible inhibition from lapatinib would mean that the biology of the disease could change over time.

What they did is they looked at landmark analyses at 12, 24 months, and so forth. Basically at those time points you see the curves separate and stay apart. When you integrate the area under those curves, and you look for a benefit for PFS, you see one. It’s modest. It’s only a couple of months, but it’s there and it’s statistically significant. This of course then gets registered with the FDA, leads to an approval, and now we have this molecule on hand.

The challenge is of course, you have now 3 HER2-targeting TKIs and how you position them. We’ll talk about this more. Tucatinib is probably a molecule that could be used first and maybe neratinib afterward. But I suspect if you get neratinib before and you’re resistant to neratinib, and this is just conjecture on my part, that you’re probably going to have cross-resistance to tucatinib.

Obviously the adverse effect profiles of these molecules are very different for that same reason, that they tend to hit multiple tyrosine kinases, so you’re going to have a lot more toxicity with neratinib. We’ll talk more about that, it’s an interesting study, and just out of necessity it had to be a funny looking trial.

Yes. Neratinib, capecitabine is in the NCCN [National Comprehensive Cancer Network] guidelines not only for HER2-positive breast cancer metastatic, but also for patients with CNS metastasis, as is paclitaxel plus neratinib. Just so we know because we don’t go necessarily looking at the CNS NCCN guidelines that much. Those are there, so that’s helpful to be able to get for patients. Thank you guys for updating us.

Newly FDA Approved Therapies: R/R HER2+ MBC

Taking out the crystal ball, how do you think the curative setting might change? What other newer agents do you think might make their way into the curative setting? Second, is there any role for the T-DM1 [trastuzumab emtansine]–pertuzumab KAITLIN regimen? Is there any patient who maybe that would help? Maybe they just have comorbidities or what have you. It wasn’t as effective as the TCHP [docetaxel, trastuzumab, pertuzumab, carboplatin] in the KAITLIN trial. It’s not a standard of care, but do you see any role for that for any patients? Do you think about that at all? What agents do you think will maybe move up in the future?

I do think that for our optimization, personalization, and tailoring efforts, we have a couple of trials that are actually trying to leverage this pCR [pathologic complete response] thing, right? There’s CompassHER2-pCR, which is an ECOG-ACRIN trial. In Europe there’s the DECRESCENDO trial, which is a very similarly designed trial in which they’re giving just a taxane and HP [trastuzumab, pertuzumab]. If the patient achieves pCR, they just get the HP and never get additional chemotherapy.

Those are maneuvers in which we serially try to eliminate the chemotherapy, right? You could argue what we should eliminate. Do you eliminate pertuzumab? Do you eliminate the duration of the 

 target? When you poll patients, they want us to get rid of the chemotherapy. One thing we did take away from KAITLIN is that although the pCR was different—and this is a small number, but it looked like it didn’t matter a huge amount whether you got a pCR through a T-DM1P [trastuzumab emtansine–pertuzumab] mechanism or a more traditional TCHP mechanism—they all had about the similar event-free survival. There is an argument, a testable hypothesis, that pCR achieved by whatever is pCR. Then we have to figure out who can we get away with just giving anti-HER2 therapy with endocrine therapy. There have been trials of this, and there’s always a low but real pCR rate. If we could figure who they are, that’s a way forward.

Because it’s a trial that’s about to open and we’re super excited about it for those who have residual disease—as we were just saying, there are groups in the residual disease setting who in spite of KATHERINE, in spite of T-DM1 still have a high relapse rate. You may remember that KATHERINE did not have an impact on brain mets. There’s an Alliance trial that is about to open that is with residual disease is T-DM1, or the same with tucatinib, and so it’s a very straight up. Can we build on that in high-risk residual disease?

That’s great. Hopefully it’s going to be a strategy for both cure and brain. What do you think VK?

 Lisa, I have 1 other thing. With these trials you mentioned, the 1 other neat feature is in the correlative work. They’re using circulating tumor DNA as a measure of deep molecular remissions. I hope that something like a deep molecular remission would track well with pCR. For those who didn’t, you can see the circulating tumor DNA that still suggests there might be cancer cells. Then having some adaptive therapy, based on that in some of these novel agents used, would be a future possibility. That’s really early science. I was heartened to see the collection of those specimens so we can assess for those things.

The things that were—the integrative biomarker that’s in these is actually the intrinsic subtype. Some are more 

 driven than others, so we may get a little more bang for the buck out of looking more closely at the tumor. Also, I would argue, although it’s a secondary thing: exploratory, immune microenvironment, getting back to Claudine’s comment about TILs [tumor-infiltrating lymphocytes].

Thank you. I want to mention the DESTINY-Breast05 trial too. We have tucatinib trying to improve overall outcome, but also brain, and then the DESTINY-Breast05. It’s just opened, T-DM1 vs TDXD—that’s trastuzumab deruxtecan. Just a head-to-head, and there’s a head-to-head going on in the metastatic, which we don’t know the answer to. It just opened in the curative setting as well. We don’t know much about TDXD [trastuzumab deruxtecan] in the brain. That’s still an unknown area, but another candidate isn’t it for the curative setting.

Moving Metastatic Therapy to Early Stage Setting

 Let’s say you have a patient who, for example, comes in and has some bone metastases, de novo metastatic disease. She’s young and triple positive, and she does really well with CLEOPATRA and with a THP [paclitaxel, trastuzumab, pertuzumab], and PETs [positron emission tomography scans] are now clear, with no FDG [F18-fluorodeoxyglucose] uptake. She goes in, gets her ovaries out, gets on an aromatase inhibitor, and continues the HP [trastuzumab, pertuzumab]. Maybe she even has local control because you’re hoping she’s going to be 1 of the ones who goes long term.

Just a show of hands in terms of the HP. The question is, we keep the endocrine therapy going, but how long do we keep the HP going? It depends. Someone’s got multiple mets, for example, or maybe just had like 1 or 2 bone mets or something like that. Who would like to speak from their experience? We obviously don’t have randomized data, but does anybody have any experience on that? What do you tend to do?

 It’s always easy to be an authority when we have no data. Everything is an anecdote, but the reality is the longer a patient maintains stable disease, the more likely I am to probably take 1 of the anti-HER2 therapies away. I don’t know when that is. It’s different with each patient. The younger the patient, if they’re tolerating it, fine. If they start having any gastrointestinal symptoms from the pertuzumab, I’d probably get rid of that first.

 And probably with lesser disease. This could all be individualized. VK, what were you going to say?

 I know it’s not data. It’s retrospective, and it’s from 2 centers. Rashmi Murthy published a paper not too long ago showing that if your best response to HER2-targeted therapy is a CR [complete response], most of the patients maintain that. They don’t speak to whether they de-escalate or get rid of drugs along the way. It certainly suggests that patients can do really well for a very long period of time, and some may actually be cured in that setting.

I’m just making that point, that we do have a little—not randomized controlled data, but there’s a little bit of—

I’ll have to take a look at that, yeah. How about Lisa or Claudine: Any experiences or sense from a practical standpoint of what you do every day?

 We grapple with that. It really is an individual case-by-case situation. The patient’s view of it is obviously of incredible importance as well. For those patients who have a great response—you’re still doing it 2 or 3 years out, and you’re still wondering what you’re doing—if they have adverse effects, I happily get rid of 1. I spread it out a little, so when they can go places, they can go more easily. I’ve had some patients who have been champing at the bit to stop. Others are saying, “There is no way I’m stopping this. I want to come in because it makes me feel better,” and I don’t have any data.

If you want anecdotal experience, I just had a patient who had a great response, and she really wanted to come off. She was convinced that if she did other things it would be a better way to go. She changed her whole diet, and she felt great. She came to see me, and everything was great. I was feeling I should do this more, and her disease just started to grow again. It was 2½ years later, and she got 2½ years without anything, and she has a little lung progression. She was probably right.

We don’t know the answer to that, and it’s obviously something we need to study. We all have these phenomenal responders, and we wonder what we should be doing in that situation. We really don’t know.

 I tend to agree. I have to say, this really comes down to the goals of therapy are to control the disease and not to make the patient sick. If you’re doing well 2 years into it, and they’re having diarrhea or they hate coming in—and you’re in the middle of a pandemic—it’s perfectly reasonable to take a break. You’re going to be following them anyway.

That said, I have had very little success in convincing patients who are doing well and having excellent tolerance of these drugs to stop them. I had 1 patient, and it took me years to get her to let me stop the Navelbine she was on. She was on Navelbine-trastuzumab. She was on that. She died of myeloma years later, still taking it. I might have gotten rid of the Navelbine, but it took a long time. She’d say, “I forget about my disease most of the time. I come in to see you guys intermittently, and you give me an infusion. I don’t get sick. I don’t get constipation.” She just happened to tolerate it really well. She said, “Why would I change everything? You know I have metastatic disease.” She had very diffuse visceral symptomatic disease up front. She remembered what it was like to have bad metastatic cancer. You have to individualize it. It wouldn’t surprise me if it never comes back for a few patients we’re able to take off it. We don’t know who they are. The anticipation is that they will, but a holiday isn’t necessarily a terrible thing.

I just wouldn’t count on it. You count on great longevity. I’d be very careful about the patients you choose it in. We know that with chemotherapy you’re better on maintenance doses than you are with intermittent dosing and coming off and on and stuff.

I did it in 2, but I won’t do it again. I took 2 off after zillions of years. One came back in the liver in the same spot. She had been on Navelbine—at least it took forever—and she was on it 8 years. Then she did great on trastuzumab for X number more years. It came right back in her liver. Then I got her right back, thank God, to a CR, but now she’s on HP for life.

I had another patient in the pre-pertuzumab, and we stopped after a couple of years. We had 1 solitary bone met. We radiated it, and she was NED [no evidence of disease] 2 years later, so we stopped it. She came back in her brain. She lived a long time, but we couldn’t salvage her back. You know what I mean? I’m like, “That’s it. I’m not doing it anymore.”

Hormone Therapy in HER2+ Breast Cancer

VK, we don’t know if we want to call it de-escalation, but personalizing the amount of treatment. People are calling it some response adapted as such. Say a little about where we are with that. What’s your take? Anything new there?

 I’ve heard the same feedback. De-escalation is so 2019. In 2020 we used optimized, personalized, and there are probably better words. I get the sentiment. It feels like we’re holding something back that might be important, whereas personalization and optimization makes more sense. These response-adapted strategies are fascinating.

There was a couple presented at San Antonio Breast Cancer Symposium in non-

 [human epidermal growth factor receptor 2–positive] disease this year as well. The idea that you’re letting the disease teach you which direction you go is very dynamic. What’s hard in practice is to get those strategies to a point where people are uniform and thinking about this instead of shooting from the hip. That’s going to be something we need to coalesce on—ideas on how to define these things.

For example, in recent meetings you’ve seen somebody use FDG-PET [fluorodeoxyglucose–positron emission tomography] imaging of the primary breast tumor to see if you are having a clinical response. If you’re having a clinical response, those patients might then go on to surgery right then, and if they don’t, they escalate to chemotherapy. There are a lot of trees and decision points here. At the end of the day, patients who got chemotherapy typically got the advertised pCR [pathologic complete response] rates of like 60%-plus. For patients who were randomized to the endocrine therapy plus HP [trastuzumab, pertuzumab] options up front, depending on what happens—maybe they got a pCR eventually, but that was a small group of patients. If they didn’t, then they got chemotherapy in the adjuvant setting and potentially 2 monoclonal antibodies in some of these trials for the fixed amount of time that we offer this.

It’s a little messy to look at these data and try to extract and come out with a uniformed message. When we don’t have that concretely defined for a community practitioner where breast is 1 of the many things they’re seeing, this becomes a challenging thing to memorize and to know. I don’t want to hamstring our doctors that way just yet until we have more mature data about these pathways. I agree, in principle, that if we can optimize our therapy and minimize toxicities, financial and physical toxicities, that’s very important. We’re still a little early for that outside the academic setting.

 Yeah. A lot of the response was adapted very early, before pilot feasibility trials, as opposed to randomized trials. It’s very interesting, though. People are clearly taking note of these high-level pathologic CRs and saying, “Wait a minute. What can we do a little differently?” That’s the major message.

De-Escalation of Adjuvant Anti-HER2 Therapy and Chemo

Let’s turn to Claudine. Maybe you could refresh and update for us about APHINITY, ExteNET, and KATHERINE. I think we all know what the data are, but is there anything new around those 3 particular topics?

Absolutely. For APHINITY we’ve already alluded that the differences were pretty modest. There were groups where we saw the differences on the patients with hormone receptor negative or node positive disease were where the differences emerged, but they weren’t huge. They were a few percentage points in terms of evasive disease-free survival.

There was a presentation at San Antonio Breast Cancer Symposium. There was a poster that tried to look at which populations were the ones that derived the greatest benefit and tried to get estimates based on treatment or patient subpopulations. What came out was very much in keeping with what we thought that the group that derived the greatest benefit were the patients who had higher clinical risk. Then they also looked at other parameters. They looked at TILs [tumor-infiltrating lymphocytes] and 

 [human epidermal growth factor receptor 2] fluorescence in situ hybridization copy number. They found that the patients with the highest TILs were also the group that seemed to derive the greatest benefit from pertuzumab. Something else to put in there, not that we’re necessarily getting TIL numbers when we’re getting path specimens on patients who are getting core biopsies on patients, but the concept that the patient was at higher risk, and still incorporating what we’ve been incorporating like the node-positive or the hormone receptor–negative patients.

I know I’ve heard others say that they started with TCH [docetaxel, carboplatin, trastuzumab]. I’d say in that the neoadjuvant setting, if I’m treating somebody, I do tend to give TCHP [docetaxel, trastuzumab, pertuzumab, carboplatin] except in an older patient or somebody where there would be a reason for me not to. But I realize I’m probably a little more conservative on that than I need to be, and I’m trying to figure out how to pull back on some of that.

That is what was new on APHINITY. On ExteNET, this is a study that we’ve seen a lot. This is a study that is looking at the role of neratinib, pan-TKI [tyrosine kinase inhibitor], in patients who had completed HER2-targeted therapy. This is a pretty old trial, so the patients typically got trastuzumab as their treatment, and it was a wide-open trial. We talked about that trial, and we thought about opening it, and it was a trial that basically allowed anyone at any time who had gotten trastuzumab. You could have been 2 or 3 years out, and then they shortened the time frame. It’s really a very wide-open trial.

What they showed us over the past couple of years is that the group that appeared to benefit from that year of neratinib after having completed trastuzumab and chemotherapy were actually the patients who had hormone receptor–positive disease. All patients with hormone receptor–positive disease also got endocrine therapy. We know there’s a lot of cross talk between those pathways, so you can think that there might be a rationale, particularly if you’re starting it later in the course.

That was really the subset that seemed to come out. The other thing is not surprising: The patients who were started more closely into completing their adjuvant trastuzumab were the other group that seemed to really get the benefit. The hormone receptor positive started within 1 year of completing their trastuzumab.

We hadn’t seen survival data and Frankie Holmes presented that at this past San Antonio Breast Cancer Symposium. Looking at that group, there was no benefit overall in terms of overall survival in the intent-to-treat population. These become exploratory analysis when they looked at the patients with hormone receptor positive. There emerged a survival benefit—about a 2% survival benefit—if you looked at the super-highest risk. The hormone receptor–positive patients who started within a year of completing their therapy, and the ones who didn’t achieve a pathologic CR [complete response]. We’re getting to really small numbers, so let’s be careful. There was actually a pretty significant benefit that emerged in that group in terms of overall survival.

It takes us back to the discussion of where we think about this drug now. What do we do with this drug in the era of KATHERINE and in the era of pertuzumab? That’s something that maybe we want to come back to as a group and talk about. That was really an interesting finding and it’s one that we debate in practice. I don’t know if you want to talk about it now, Joyce.

Let’s do it. What do you do? Do you use neratinib in some patients?

I’d be really interested to hear what my colleagues think. Like most of us, I was really impressed by the KATHERINE data. It showed this really big 3-year invasive disease-free survival benefit for the patients who just didn’t achieve a pathologic CR. It’s about an 11% absolute benefit. What also struck me is those patients weren’t doing that great—even the group that got T-DM1 [trastuzumab emtansine], right? At 3 years there was about 11% to 12% of them who developed a recurrence.

It was 3 years so it struck me that there was still room for benefit in that group. In those patients, particularly in the really high-risk patients, and particularly if they have hormone receptor–positive disease—and we all have some patients who have multiple nodes, 2 or 3 nodes—I tell them, and I do tend to give them neratinib.

I always tell people when we start on a drug, particularly 1 that has adverse effects: This is not a moving train; you can’t get off. Let’s try it, and let’s see how you tolerate it. If it’s tolerable, we may as well try to get benefit from it. If it’s not, we try it, and obviously they go on endocrine therapy and everything, but that’s the group for whom I still think about using neratinib. I recognize that we have a paucity of data to support my recommendation, except that there’s still a place for improvement in that group of patients.

I agree with everything that Claudine said, and the 1 other element that has to go into the mix is that it has to be a highly motivated patient partnered with a highly motivated oncologist because it’s going to take a lot to get them through it. I frankly don’t use it a lot. It would be in those patients that Claudine described. Of the handful of patients who did go on, half discontinued, and that’s even with the antidiarrheal maneuvers. It’s not for the faint of heart oftentimes.

Yeah, and I just want to make the point about how important it is to escalate the agent. In case somebody hasn’t seen the results of the CONTROL trial, a lot of things were looked at. Loperamide, which we certainly want to use for more prn as needed with the escalation strategy, plus budesonide and colestipol. Starting at 3 or 4 pills, from 120 to 150 mg, and then every 1 or 2 weeks as tolerated, increasing it by 1 pill has made a substantial difference. In the CONTROL trial, only 2% or something stopped for toxicity. I’ve really been pleased with that. That’s really the only way to go. People are going to try to give patients a shot at this, and using the escalation strategy is really important. How about you, VK? What do you think about neratinib?

I’ve actually found myself using it increasingly of late. That’s because of the cross talk issue. I believe that ER [estrogen receptor] disease is fundamentally different from the kind that’s ER negative. If we don’t address that cross talk, and we don’t do that with T-DM1 [trastuzumab emtansine], I’m still worried about these patients having events down the road.

The other thing, and we learned this from tucatinib in the metastatic setting, having that molecule that can cross the blood-brain barrier even in this context may actually have benefit. I know that trials are underpowered for CNS [central nervous system] as a first site of recurrent type event. It’s my suspicion that those are patients who are vulnerable to those events when you have residual disease, have done T-DM1, have ER-positive disease. I’d like to give them that extra bit of coverage for that as well as for prevention for those CNS events. I’ve gradually shifted: I was not at all using it, and now I find myself more often using it. Now that we have the ability to give it, start with a few pills, work our way up…. It’s not for the faint of heart, to steal Bill’s words.

I’ll be honest: I know practices that don’t let their doctors use it in off pathways and stuff. I’d love Bill to change this for us, the NCCN [National Comprehensive Cancer Network] has it like it’s a footnote—on the top, Bill, just as an option. It’s down in the fine print, it’s the argument that the ExteNET was done with pertuzumab and T-DM1. That’s the argument, and I totally get that.

It’s a non-cross-resistant agent, and you’ve got the cross top coming. To me, it’s like a totally different drug, and I don’t know why antecedent pertuzumab and T-DM1 make a difference. What do you guys think? Lisa, do you ever use this drug?

Rarely. The thing I struggle with is that the cross talk is certainly something that’s preclinically important. We haven’t seen that direction of effect by hormone receptors with any of the other small molecules or, in fact, with neratinib itself in the metastatic setting.

I struggle a little with it because it didn’t quite seem to hold true. That said, the ExteNET data are what they are. If I have a very high-risk patient, I have occasionally reached for it, and I’ve been happier with the option of dose escalation. In the CONTROL trial they did quite well. They escalated relatively quickly. I was actually surprised. It seemed like the low hanging fruit of managing your diarrhea, and it was the most effective 1. If I was going to do it, I would do that.

We’re going to talk about it in a little bit, but NALA and HER2CLIMB are the only studies we have in the metastatic setting. Both did not permit endocrine therapy usage. Personal opinion, it would have been nice to have studies that just said, “We are just managing 

+, ER-positive patients with the anchored regimens with endocrine therapy.” We do have studies in the early stage metastatic setting, like in the PERTAIN trial, and you see that synergy there. I think there’s still room.

Adjuvant Systemic Therapy Data Updates

Bill, how do you define patients at high risk of recurrence after their neoadjuvant therapy? What do you do with them? Who is high enough risk in your practice to get a pertuzumab up front? Do you know what I mean? When you start it, do you ever stop it? Do you just keep it going because you feel patients are high enough risk? Then in the adjuvant setting for people with residual disease.

Right. To piggyback on some of the other comments, we’ve more or less exclusively got away from anthracyclines preoperatively. There may be exceptions where that’s an appropriate thing to do. For most patients, they are getting TCHP [docetaxel, trastuzumab, pertuzumab, carboplatin] upfront, particularly if they have bigger disease at the outset. We do use TCH [docetaxel, carboplatin, trastuzumab] for those where we’re trying to avoid some of the excess toxicity.

In my mind, the fundamental thing that defines risk after receiving neoadjuvant therapy is what the residual disease burden looks like. If there’s significant disease, particularly in nodes, that would be very worrisome, as it would be if there’s a significant amount of disease in the breast. I would also say as a caveat, quite frankly, I haven’t seen that many patients in my own personal experience getting TCH or TCHP who actually progressed on that preoperatively. I know it happens, but I don’t see it very often.

For those that have significant residual disease, and Lisa was already alluding to KATHERINE with those patients switching to T-DM1 [trastuzumab emtansine] and good results, even in those patients where the subsequent 

 [human epidermal growth factor receptor 2] assessment is negative, we continued the idea to switching to T-DM1 in those with residual disease makes sense.

That’s the strategy we employ. With respect to those who get a pCR [pathologic complete response] at the outset, for most patients—not all—we’ve continued or at least started with the idea that they would complete HP [trastuzumab, pertuzumab]. Who would we not do that in? Certainly, anybody who’s having excessive toxicity. Anybody who’s ER+ [estrogen receptor positive] I would be reluctant to do that in because I don’t think they’re getting as much benefit. There are some nuances to how it would be used.

Bill, do you think that if somebody has residual disease after a TCHP or occasionally a AC-THP [doxorubicin, cyclophosphamide, paclitaxel, trastuzumab, pertuzumab], that T-DM1 is really the real standard of care? I say that because not long ago, say a few weeks ago, I saw a patient who had some residual disease and didn’t get T-DM1. Frankly, I was surprised, you know what I mean? It’s become a must-do standard. What do you think? Do you think there’s some room for maneuver with a tiny amount of cancer left?

I think it is the standard. Just as we can find exceptions to everything we do if we look hard enough, I suppose if somebody had a millimeter of disease, would I feel obligated that they must get T-DM1? Maybe not. Maybe if they’re a little older. For most patients, it really is the standard. Most patients do OK with it, so I would try to leverage that benefit that we saw in KATHERINE and confer to whoever has residual disease.

Yeah. Does anyone else want to make any comments about KATHERINE? We’re going to come back and go into it a little more deeply, but any other issues around the adjuvant?

I’ll share a struggle. In our patients who are node-positive preoperative therapy and get a pCR, what’s the thinking around continuing HP? Even in those pCR patients who were node positive, knowing that there’s a little benefit there when they’re node positive. I found a presentation at San Antonio Breast Cancer Symposium by a very young scientist from Europe looking at pCR rates across a lot of different diseases and saying, “pCR equals pCR.” These are people who have good outcomes. You look at some of the trials, and you start picking at them a little. There seems to be a suggestion that maybe you should continue the HP. I have to be honest. I don’t find that data compelling, but I’m always curious what others think.

I have to say, I don’t think we know the right thing to do. I stay with whatever brought you. You know? If I got a pCR out of an HP-based regimen, I tend to continue the HP. Your point is well taken. The patients can do very well. In our experience, CALGB-40601—which was designed deliberately—is very similar to the European analog Neo ALTTO; the pCR is different. It’s about a 10% absolute benefit difference between the 2 of them in the pCR group, with the event-free survival around 90%-plus for the American trial. But it’s in the 80s for European trials, without a particularly good explanation if you dive into it. I’m not sure that I have, because I don’t feel I can take a pCR to the bank. And so if I gave HP up front, I’d give HP out back, but I don’t know if that’s right.

I tend to continue HP as well if somebody is tolerating it and there’s no compelling reason to stop it. I have Lisa’s take on it that. If it’s not causing much harm—and I’m not sure what it would do for me to take it back—I’ll continue it. I agree. I wonder about de-escalating from the patient’s perspective and then the time and the inconvenience and everything that goes with it to continuing the therapy.

I also think that the other thing that figures into it is APHINITY, though we can’t jump up and down about this huge gap between the lines. That gap was identified with a full year of therapy, particularly in node-positive patients. That has to figure into it at least a little. It is a struggle for the patient.

Yeah, I did also want to point out that the FDA label for pertuzumab talks about the neoadjuvant. In the label there’s a section called neoadjuvant, and talks about pertuzumab preoperatively. What it says is interesting. It says that if you start it preoperatively, you should finish the year of it. It’s really interesting. Even though we honestly don’t know. If somebody has a pCR, do they really need it, or you don’t even know. If they’re clinically node negative, they start off with it, but they’ve got a big cancer. You decide to use it; we just don’t know this so much.

Bill does the NCCN [National Comprehensive Cancer Network] have a statement about this anywhere, if somebody starts with pertuzumab preoperatively and gets a pCR, are there any recommendations in NCCN Guidelines?

No, we haven’t gone to that granular level. We wanted to let doctors be doctors.

That’s a beautiful thing. I love that. Thank you. It’s refreshing.

Joyce, I’ll point out 1 thing that does make giving both therapies in the adjuvant setting now—and we’re going to talk about this later this evening—is the ability to give both as a single injection.

Yeah, there you go. That might help too, VK.

Adjuvant Systemic Therapy Decision Factors

® titled “Refining Treatment Approaches for 

+ Breast Cancer.” I am Dr Joyce O’Shaughnessy from the Baylor University Medical Center, Texas Oncology, and US Oncology in Dallas, Texas. I’m delighted to have 4 friends and longtime colleagues with us to debrief after San Antonio Breast Cancer Symposium and find out what people thought about some of these data.

I have with me tonight Dr Lisa Carey from the University of North Carolina Lineberger Comprehensive Cancer Center in Chapel Hill, North Carolina; Dr VK Gadi from the University of Illinois College of Medicine in Chicago, Illinois; Dr William Gradishar from Northwestern Memorial Hospital in Chicago, Illinois; and Dr Claudine Isaacs from the Lombardi Comprehensive Cancer Center at Georgetown University in Washington, DC.

Today we are going to highlight a number of topics pertaining to systemic treatment of 

+ [human epidermal growth factor receptor 2–positive] breast cancer and the impact of recent clinical trial data on clinical decision-making. In addition, we’ll be covering key clinical data from ESMO [European Society for Medical Oncology Congress] 2020 and San Antonio 2020. That’s what we’re going to be talking about over the next about hour and a half or so.

+ early stage breast cancer, looking at the neoadjuvant and adjuvant settings. We’re going to start with early stage and go to metastatic next. First, it’s probably going to be helpful to set the stage looking at the standards of care—where are we today—and then we can look at where we’re going. Lisa, can you give us an overview of where we are today?

What’s changed and what is certainly true nowadays is that for the majority of 

+ breast cancer patients, we’re treating them in the neoadjuvant setting. They’re getting their chemotherapy and anti-HER2 targeting, for the most part, preoperatively, and then HER2 targeting postoperatively is predicated on extent of disease that’s found at surgery. There’s a reason for that. It’s better for surgical end points. Clearly you can create more lobectomy candidates, and you can reduce the need for axillary dissection. Moreover, from the medical oncologist’s standpoint, we all have the great data from KATHERINE with the shift to T-DM1 [trastuzumab emtansine], improving outcomes in those with residual disease.

The group that’s still treated with adjuvant therapy is stage I, T1N0. Many of us, in order to confirm that they would fit into these—we call it de-escalated, although that’s a term I’m trying to train myself not to use because the patients hate it. The APT regimen was just 12 weeks of paclitaxel plus trastuzumab in continuation of that. That’s the standard general framework. The debates that are going on is what’s the appropriate chemotherapy backbone, and do you need an anthracycline, for example? What’s the extent to which we need all those drugs and whether we can rationally optimize therapy even in the higher clinical risk patients?

 What do you tend to use, Lisa, in the neoadjuvant setting for preoperative treatment?

 I typically use TCH [docetaxel, carboplatin, trastuzumab]. I will add pertuzumab if the patient is clinically node positive.

 What if they have a tumor like a T2 or T3 but are clinically node negative as far as you can tell?

 I still use TCH. The APHINITY data that we’ll talk about a little later. I’ve been fairly convinced that the marginal benefit of pertuzumab is relatively small in the node-negative setting.

 How about other people? Any other points you would like to make around the state-of-the-art therapy for patients? What about anthracyclines? Does anybody still use anthracyclines? Somebody could just mention with the TRAIN-2 data, for example.

 I’ll jump in on the anthracycline piece. For patients who have substantial burden of disease—maybe somebody who has not screen detected but walks in with a big mass—I frequently start with a taxane portion, and then I’m on the fence a little on the anthracycline. Most of those patients still end up seeing the anthracycline in my practice.

 I shifted a little while ago more toward TCHP [docetaxel, trastuzumab, pertuzumab, carboplatin], although I do find that it’s a difficult regimen. From a symptom-control perspective, it’s 1 of the toughest regimens we give. ACTH [doxorubicin, cyclophosphamide, paclitaxel, trastuzumab] or AC-THP [doxorubicin, cyclophosphamide, paclitaxel, trastuzumab, pertuzumab] is an easier regimen to take.

But the TRAIN-2 data were further evidence for us that the anthracyclines really didn’t add much. Because you minimize the cardiac toxicity risk again, that risk is not sky-high, but it is marginally higher with the AC-THP than it is with the TCHP. I’ve predominantly shifted to TCHP unless I’m treating a patient on a clinical trial. We all have a few patients here or there who have synchronous tumors, and 1 is not 

+ and you feel that both of them need chemotherapy in that situation. If the chemotherapy is needed, I tend to give the ACT [doxorubicin, cyclophosphamide, paclitaxel] backbone.

 How do you think about the TRAIN-2 data, VK? It wasn’t the hugest trial in the world—maybe that’s something that gives people a little pause—but the randomization to preoperative, 6 cycles of TCHP vs the equivalent of AC-THP. The pathologic complete response rates were really good from both of them: 60%, 78%. Then the disease-free survival was superimposable; albeit not the biggest trial in the world, but what do you think about that trial?

 You hit the nail on the head, which is that they are still small experiences in the realm of things. But it is really encouraging, this ability to get away from the anthracyclines. Long term, that should be our goal. There’s always going to be, as Claudine pointed out, patients whose circumstances are different, where you think you need to escalate in that direction and offer more with an anthracycline or what have you. Certainly, if you have started a taxane-based therapy and you are watching the tumor grow on you, then that shakes your confidence about relying on the taxane-based therapy.

There are always going to be some circumstances in the preoperative setting in which you’re watching the tumor and you may have to change course. But as a rule, I agree with everybody here. We’re migrating away from anthracyclines in this place, but we’ll see what the long-term survival data are, even though it’s the small experience that can also be informative over the long haul.

 Great. We’re going to come back. We’re going through APHINITY and the T-DM1, and we’re going to touch on neratinib too, which may be helpful for some patients. Thank you. That’s a good place for us to start.

Vijayakrishna Gadi, MD, University of Illinois College of Medicine

Articles