HER2-Positive Breast Cancer — Ian E Krop, MD, PhD

DR KROP: I’m Ian Krop. I’m a medical oncologist at the Yale Cancer Center, and I appreciate the opportunity to talk about some of the new findings in HER2-positive breast cancer. I think as many of us know, this has been a fast-moving field for years now, and 2023 really is no exception, so there’s a lot to talk about.

But I wanted to just first level set and remind people that HER2-positive early breast cancer now has quite favorable outcomes, fortunately. That’s exemplified by this adjuvant KAITLIN study that was recently published. And if you look at even the highest-risk subgroup in this trial, so the patients who had 4 or more positive nodes, those patients treated with standard AC-T, chemotherapy with trastuzumab and pertuzumab, still had a 94% 3-year iDFS, so people are doing very well these days, which is of course is great for patients. But it also means that we have the opportunity to really think about de-escalating therapy for our lower-risk patients to cut down the toxicities of treatment, at the same time still thinking about escalating treatment for the minority of patients who are still at high risk despite our best treatment.

So this here, we had 2 de-escalation trials that were presented or published, and one of them was updated data from this APT trial, which asked I think the very important question of whether we can leverage the effectiveness of trastuzumab to de-escalate how much chemotherapy we give to low-risk HER2-positive patients. So this trial enrolled over 400 patients who were node negative, had cancers less than 3 cm, almost the vast majority had cancers between 1-2 cm, and were treated with just 12 weeks of weekly paclitaxel and a year of trastuzumab.

And the trial was designed to ask whether this type of low-toxicity treatment is associated with a low risk of recurrence at 3 years, and low risk was defined as 5%. And the study clearly met that endpoint. The 3-year iDFS was over 98%, and now with 10-year follow up the disease-free survival is still very good at 91%.

But I think if you look on the right, where we’re looking at recurrence-free interval, which doesn’t include things like new cancers, which probably aren’t so relevant for this type of question, the 10-year RFI is over 96%. And importantly there were only 6 out of more than 400 patients who had a distant recurrence event, so 1.5%. And it’s hard to imagine that giving more chemotherapy would meaningfully reduce that distant recurrence rate.

And it didn’t matter whether the cancers were less than 1 cm or greater than 1 cm. Both groups did extraordinarily well. And hormone receptor status also didn’t matter. The hormone receptor-negative, HER2-positive patients did just as well as the hormone receptor-positive patients.

So the conclusion is that this paclitaxel/trastuzumab regimen is associated with excellent outcomes and really should be the standard of care for most patients with Stage I HER2-positive breast cancer. And again, it’s not just for the hormone receptor-positive patients, the ER-negative patients did just as well. And it’s not just for the tiniest cancers, T1c cancers did very well.

So having this regimen, though, available doesn’t mean that we should be giving chemotherapy or HER2-directed therapy to all of our Stage I patients. I’m particularly talking about the T1a, node-negative patients, where for the hormone receptor-positive patients certainly can get by with just hormone receptor — endocrine therapy alone. And I think many of us for the ER-negative, HER2-positive patients start thinking about giving chemotherapy for those cancers that are 3 or 4 mm or higher and not giving chemotherapy for the 1 and 2 mm cancers.

And I didn’t have time to talk about data using T-DM1 in Stage I patients, but a year of T-DM1 is also associated with very good outcomes and can be considered as an alternative approach for patients who may be at particularly high risk for the toxicities of paclitaxel.

So the next study I wanted to mention was PHERGain, but maybe we can stop for a question if you had one.

DR LOVE: Yeah. I was just curious in general what your clinical preference is in that low-risk situation, PT or T-DM1?

DR KROP: Yeah. I mean I do think that right now the trastuzumab/taxane regimen is really the standard, and that’s consistent with guidelines, and T-DM1 I think really is an alternative regimen for select patients. And that’s the way I use it, usually use TH in patients who already have neuropathy or already have — are violin players or somebody who neuropathy is going to be a big problem for them or have other reasons to avoid conventional chemotherapy. I think T-DM1 is certainly a very effective regimen in that situation, but for me it’s a minority of the patients.

DR LOVE: So I’m just kind of curious, though, what do you think the major reason is that it’s evolved towards the paclitaxel/trastuzumab as opposed to T-DM1?

DR KROP: Well, I mean I think it partly is that the TH regimen, the APT trial, came out first, we have the very long follow up data now available. And taxane/trastuzumab was already an approved part of neoadjuvant or adjuvant treatment for HER2-positive disease, where T-DM1, particularly back then, was not a standard — did not have a standard role in early breast cancer. Now with the KATHERINE trial we use T-DM1 in the post-neoadjuvant setting for patients with residual disease, but I think that’s part of that, is kind of the historical way that the data evolved.

DR LOVE: And from your point of view, in terms of quality of life, do you think they’re pretty much equivalent? Because I would have thought that maybe T-DM1 was a little better tolerated.

DR KROP: Yeah. So that’s a very good question. And Sara Tolaney designed a trial, this ATEMPT study, where not only was it shown that T-DM1 had very good outcomes, but we had a comparison arm of TH to look specifically at toxicity and quality of life, and the differences weren’t quite as dramatic as I think we had assumed they would be, meaning we didn’t see a huge improvement in terms of toxicities with T-DM1 compared to TH. And partly that’s because TH was only given for 12 weeks, whereas the T-DM1 was given for a whole year, and so over time you start getting more toxicity with T-DM1 as you get out towards the end of the year of treatment.

I mean I think if you look at quality of life certainly when the patients are actually getting the chemotherapy the quality of life is better for T-DM1 than it is for TH, and there’s less work that’s lost, and patients generally have better quality of life. It’s just the whole year of T-DM1 that started to become an issue, and that’s why Dr Tolaney and colleagues are now doing a follow up study where they’re doing a much shorter course of T-DM1, and then patients are moving on to trastuzumab. And I think that very well may show that kind of very favorable toxicity profile that I think you’re alluding to and still maintaining very good outcomes. So I think when the data from that trial come out hopefully that will allow us to feel much more confident using T-DM1 knowing that there’s both good efficacy and favorable quality of life.

DR LOVE: Okay. Please continue.

DR KROP: So the other de-escalation study that I think is worth talking about is a study called PHERGain that was presented by Javier Cortez at ASCO last year. And it’s a relatively complicated study design, but I wanted to focus on just part of the trial, which is a study that enrolled patients with Stage I-III early breast cancer that was HER2-positive.

And 285 of the patients were randomized to a regimen where they just got 2 cycles of pertuzumab and trastuzumab, so HP, they then had a PET scan, and if they had a favorable response to that HP regimen they were then continued on the antibodies alone for another 6 cycles. If they didn’t have a favorable response they would get conventional TCHP. And then at time of surgery if they had a pathologic complete response to the 8 cycles of HP they would then just continue to complete a year of HP; so no chemotherapy for those patients. And patients who had residual disease would then get conventional TCHP chemotherapy.

And so what they found was that about 80% of the patients who got the HP up front had a PET response and therefore continued on the antibodies, and of those about 38% had a pathologic complete response, so that’s a pCR to just HP alone for 8 cycles. And now with long-term follow up we see that overall the patients treated with this guided — PET-guided de-escalation approach had a favorable 3-year disease-free survival of over 95%. This met the primary endpoint of the study. And if you look specifically at the patients who had a path CR and therefore never got chemotherapy their disease-free survival was over 98%, consistent with a lot of data showing that if you have a pCR essentially no matter how you get there you still have a favorable long-term outcome.

So the conclusions, at least in my opinion, are that this PET-guided de-escalation approach certainly yields favorable outcomes, with a disease-free survival of over 95%. It allowed about 30% of patients to omit chemotherapy. I do think you have to point out — we have to remember that this is only 3-year data. We know that particularly for the hormone receptor-positive patients, which made up about two thirds of this population, that you can have later recurrences, so we do need longer follow-up.

I think another question to ask is how does this approach to de-escalation compare to just a short course of chemotherapy, like 12 weeks of paclitaxel with HP, which has a pCR rate of about 55% and presumably doesn’t need additional chemotherapy for those patients. So in other words, it’s a tradeoff between giving a little bit of chemotherapy but getting a substantially greater percentage of patients having a pCR and therefore no need to escalate further versus no chemotherapy at all but having a smaller percentage of patients have a pCR and therefore are able to skip more intensive chemotherapy.

I think another question is can we make the approach a little bit more savvy by using up-front biomarkers, such as looking at a HER2-enriched subgroup, which we know have higher pCR rates when treated with HER2-directed therapy, and that would potentially allow more patients — a greater percentage of the patients to get through this de-escalation approach.

So I think it’s very interesting data. I think it’s where the field is moving, but exactly how we do our de-escalation I don’t think we’re quite to a consensus yet.

DR LOVE: So it’s interesting, I hear from people outside of breast cancer, a lot about the issue of using trastuzumab and pertuzumab as a therapy for metastatic HER2-positive, in this case colon cancer, and like this patient that they presented had a great response in metastatic disease, and I think to myself have I seen that in breast. Like because I kind of had the impression that just trastuzumab and pertuzumab by itself was not very effective in metastatic. I know you’re talking about neoadjuvant here, but the combination of pertuzumab and trastuzumab in terms of metastatic disease. Do you ever use that? I kind of felt like it doesn’t work in breast. That’s why I was surprised when I saw it in these other cancers.

DR KROP: Well, I mean it’s a good question, and I think you do have to distinguish early stage from metastatic. So early stage we have this trial showing a pretty substantial, almost 30% pCR rate, higher than that with just antibodies. Again, this was 8 cycles, so this was a prolonged course. We had the previous NeoSphere study, which did the same thing with just 4 cycles, and they still had a pCR rate, but it was substantially lower than this.

DR LOVE: Right.

DR KROP: So time does matter. But again, these are treatment-naïve tumors, and they’re much more sensitive to HER2 therapy than in the metastatic setting, where you’ve had multiple lines, oftentimes patients had recurred after already essentially “failing” HER2 therapy in the adjuvant or neoadjuvant setting, so they already have some resistance to HER2 therapy. And there we generally don’t use HP alone.

We do know that in triple-positive patient that HP plus endocrine therapy is an effective regimen, but it’s just you get so much more benefit when you add chemotherapy in the metastatic setting that it’s hard to pass that up for most patients. So we generally don’t just give antibodies alone in the metastatic setting, not because it’s not effective, it’s just so much more effective when you give at least some chemotherapy or give an antibody-drug conjugate. But in a de novo metastatic patient you still would get benefit to HP, and again specifically if you add endocrine therapy you can really get some long-term disease control in the triple-positive patients. But that’s really the difference, pretreated versus naïve patients’ tumors respond differently.

DR LOVE: Yeah. That’s a really great point. Yeah. Because I keep hearing people presented from this basket trial I think the NCI did where they — and every now and then, you see these great responses. Interesting.

Anyhow, please continue.

DR KROP: No, but you’re right. But our expectations are just so high with HER2-positive disease now, you can get a 25% response rate with HP in the metastatic setting, particularly in less pretreated patients, but you get an 80% response rate with adding a little chemotherapy.

So let’s switch to escalation. We’ve talked about de-escalation. Escalation. So we have long-term results of the ExteNET trial, which was a study that asked whether extended adjuvant therapy with a panHER TKI, neratinib, can you improve outcomes in early-stage, high-risk HER2-positive disease. This trial looked at patients who had already had a year of trastuzumab-based therapy and then randomized to an additional year of neratinib or placebo.

And we heard some time ago that there was benefit, interestingly only in the patients that were hormone receptor positive and HER2-positive, with the addition of neratinib. It led to an improvement in disease-free survival that was statistically significant. There was not a statistically significant improvement in overall survival. And this came, unfortunately, at a cost of substantial diarrhea, 40% Grade 3 diarrhea. That trial did not use any antidiarrheal prophylaxis.

There also seemed to be a numeric modest improvement in risk of CNS recurrence with neratinib as well. And now we have long-term follow up, the final overall survival analysis, which still shows a slight numeric improvement in survival, but this was not statistically significant.

So we’re now, again, asking the question of who should receive adjuvant neratinib. There is clearly benefit in the hormone receptor-positive, HER2-positive patients that has to be balanced against the potential for toxicity, and we have to remember that this ExteNET trial was done at a time where we didn’t use pertuzumab or T-DM1 in the early disease setting. So we don’t know whether neratinib still has benefit in patients who receive pertuzumab and T-DM1 in the adjuvant or neoadjuvant setting, which of course virtually all of our high-risk patients will have received.

So who should receive it? Well, again, I don’t think there’s a right answer here. My personal opinion is that it’s reasonable to consider this in hormone receptor-positive, HER2-positive patients with substantial residual disease after neoadjuvant therapy.

And while we’re talking about this, certainly if we could reduce the toxicity it would be easier to advocate for the use of adjuvant neratinib. And we have this CONTROL study, which has looked very comprehensively at ways to reduce neratinib-induced diarrhea. A number of different approaches were tested. It really looks like the way to go is with a de-escalation approach where you start out at a reduced dose of 120 mg a day for a week, you then move up to 160 mg for a week, and then you go to full-dose neratinib. And with that approach you get a Grade 3 diarrhea rate of about 13%, substantially lower than what we see without that.

And interestingly if you look at when patients discontinued their neratinib because of diarrhea you can see it’s almost all in that first cycle. So what we’ve learned is that a dose escalation approach of neratinib, along with loperamide, is associated with a relatively low rate of Grade 3 diarrhea and that most patients who get through the first cycle then seem to do quite well with the drug. So letting patients know that I think is helpful.

And lastly, and a little bit of a variation on the theme of neratinib, does neratinib have activity in breast cancers with somatic HER2 mutations? So this is not amplification of HER2, this is just activating mutations within the HER2 gene itself. This was evaluated in the SUMMIT trial. We know that these mutations are relatively rare, but they happen in about 3-5% of hormone receptor-positive breast cancers, they’re more common in lobular than ductal cancers, and they may be acquired to some extent as cancers become resistant to endocrine therapy.

And so the SUMMIT trial was a basket trial that evaluated either neratinib monotherapy or in various combinations, specifically in cancers that had one of these HER2 mutations. There was some initial monotherapy activity, about a 17% response rate, and in combination with neratinib with fulvestrant in the hormone receptor-positive, HER2-mutant cancers the response rate went up to 30%. And there was some correlative data that suggested that HER2 amplification may be one of the ways that cancers become resistant to neratinib.

And so there’s a third cohort which is now being presented — or being published looking at the combination of neratinib/fulvestrant plus trastuzumab in ER-positive, HER2-mutant cancers that had previously had a CDK4/6 inhibitor. And also within the 71 patients they had a small cohort that looked at fulvestrant and fulvestrant/trastuzumab alone to look at the specific contribution of neratinib.

And what was found in this SUMMIT study was that the objective response rate to the triplet was almost 40%, 39%, clinical benefit rate was over 50%, and the median PFS was over 8 months. Interestingly, none of the arms that did not include neratinib had any objective responses, so it’s clearly neratinib that’s playing the major role here. You saw more activity in specific HER2 mutants and more activity in patients who had more than 1 HER2 mutation. Interestingly, HER2 expression as measured by protein or FISH didn’t associate with activity.

Unfortunately, the Grade 3 diarrhea rate in the triplet was over 50% despite loperamide prophylaxis.

And so the conclusion here is that neratinib and fulvestrant does have clinically meaningful activity in patients with hormone receptor-positive, HER2-mutant breast cancer. The addition of trastuzumab may further increase the benefit but also may further increase the diarrhea. So I think whether you need to have — you should include trastuzumab is really an open question at this point.

But neratinib-based therapy I do think now is clearly an effective treatment option for ER-positive, HER2-mutant breast cancer, it’s now included as an option in NCCN Guidelines, so we should think about that. And it also provides yet another motivation for sequencing our patient’s cancer either by tissue or ctDNA in the metastatic setting.

DR LOVE: So a couple of questions. Obviously the 2 drugs have been developed very differently, but I was just thinking like how would you contrast what you understand, what we understand about the pharmacology and clinical data with neratinib to tucatinib. Are they very similar or is there something different about them that they’ve kind of gone in different directions?

DR KROP: Yeah. So there’s 2 major differences between the 2 tyrosine kinase inhibitors. So neratinib, as I mentioned, is a panHER2 kinase inhibitor. So it blocks HER2 very well, but it also inhibits EGFR and HER4. And it’s the EGFR, blocking that, is what leads to a lot of toxicity, particularly diarrhea and rash. In contrast, tucatinib is the only kinase inhibitor that I know about that is virtually completely specific for HER2. So it doesn’t hit EGFR at all, and so you don’t have those off-target kind of effects.

Neratinib is an irreversible inhibitor. It’s very potent. It’s just you probably can’t use it at the doses you’d really like to because you just start ramping up the toxicity. And that’s really the big difference, and I think that’s what led to the tucatinib-based treatment that we’ll talk about later being so appealing, it’s because it really is much more manageable than neratinib-based treatment for most patients. And probably because you can push the dose. It seems to have more efficacy as well.

DR LOVE: That’s really helpful. Age-old question. Anything new in terms of why people seem to benefit more in the post-adjuvant setting with ER-positive?

DR KROP: No. I mean I think I wish we had an answer. In the past we’ve speculated that, and I think maybe some of these data support that, that neratinib in combination with antiendocrine therapy is beneficial because upregulation of HER2 signaling and other signaling of receptor tyrosine kinases can be one way that cancers get around the effects of endocrine therapy, so having neratinib there helps prevent that type of resistance developing.

I don’t think that’s been proven in any situation, but it is one potential mechanism why we see that neratinib benefit in the ER-positive patients. We don’t see that in the metastatic setting, though, where neratinib seems to work well in both ER-positive and ER-negative cancers. But that’s a different setting, and the cancers are resistant in that setting, so it may explain it.

DR LOVE: Interesting. The one thing we’re not going to talk about in this webinar, because we had to spend an entire hour talking it about it, was all the new things going on in ER-positive metastatic disease, the SERDs, capi, et cetera.

DR KROP: Yeah.

DR LOVE: Are we going to have to continue to wait until you have HER2-specific trials of these things? Because I mean that’s like moving along, and yet the HER2 patients I guess are not getting any of that, right?

DR KROP: Correct, and as always you ask very, very salient questions. Because we know that the PI3 kinase pathway is incredibly important for HER2-positive, or HER2-driven cancers. We know that the CDK4/6 pathway is very active in HER2-positive cancers. And it’s just I think in large part just due to the complexities of drug development and adding a separate subtype of breast cancer just complicates everything, and there’s so many other things going on in HER2-positive breast cancer that I think it’s hard for companies designing trials to feel like they’re competitive in the HER2-positive field.

But I think that drugs like capivasertib could be very active in HER2-positive disease. We’ve shown that PI3 kinase inhibitors can be effective in HER2-positive disease. We know from the monarcHER trial that CDK4/6 inhibitors can be effective in HER2-positive disease. It’s just to make a registrational trial that’s going to lead to an FDA approval is just — it’s a big lift, and I think people are hesitant to do that, unfortunately.

DR LOVE: I mean I just wonder whether — I mean do you think it actually is fundamentally different in terms of ER? I mean do we really need to do all this stuff?

DR KROP: No. I mean I do think that ER and HER2, because of the interaction they clearly have, there are some differences, but I don’t think they’re differences that would make these kinds of drugs not work. In fact, they may make them work better. And as I said, Dr Tolaney’s monarcHER trial clearly shows that a CDK4/6 inhibitor has activity, and there’s other studies looking at PI3 kinase inhibitors also showing activity. It’s just, again, whether somebody is going to be able to do trials that are definitive enough that would lead to labels for those drugs.

But I think off label there are situations where many oncologists, myself included, have used a CDK4/6 inhibitor in HER2-positive cancers. It’s just it would be nice to have something so that all oncologists can just look up in NCCN Guidelines or other venues and say okay, clearly I have the approval to do this kind of thing.

DR LOVE: That’s really great points. Please continue.

DR KROP: So let’s switch and now talk about antibody-drug conjugates, obviously an important field in HER2-positive disease. We have updated data from the DESTINY-Breast03 trial, which you may remember was the study that looked at a head-to-head comparison of T-DXd versus T-DM1 in largely second-line patients. And we saw updated survival data showing now highly statistically significant reduction in death with the use of T-DXd compared to T-DM1.

In the same analysis we saw updated progression-free survival, which showed a really impressive fourfold increase in PFS with T-DXd over T-DM1. So now we have a median PFS of 28.8 months, a number that we’ve never seen before in any trial of breast cancer that I can think of. So we’re really kind of moving into unchartered territory with T-DXd. Response rate was about 80%, and in terms of toxicity we saw the same types of profile that we’ve seen in the past, with nausea being the most common adverse event. In this trial 15% of patients had ILD. Fortunately there were no high-grade ILD events because I think we’re getting better at managing and monitoring for ILD.

So conclusions here I think are very straightforward. This trial really establishes T-DXd’s role as the preferred second-line standard of care for most patients based on these kind of really unprecedented levels of clinical activity and the reassuring ILD data.

To me the only real question that remains is should there be a different standard of care for patients with progressive CNS metastases. And we’ll get back to that in a little bit.

We also had data from the DESTINY-Breast02 trial, which looked at T-DXd versus chemotherapy and HER2 therapy in third and later-line patients, so this is after patients had already had T-DM1. This also was a positive trial. It showed a big improvement in progression-free and overall survival with, again, similar toxicity, in this case 10% ILD. Unfortunately there were 2 deaths in this trial, about 0.5%.

And the conclusion would be that we now have once again confirmed that T-DXd is highly active in later-line patients, but in the context of the DB03 data this does not really change practice; we should still be using T-DXd in the second line. But if you have a rare patient who received T-DM1 in the second line certainly these data would support using T-DXd after that.

But I do just want to reiterate that the DB02 data do not imply that just because T-DXd works later line that you should reserve it for later-line use. And that’s because the benefits are substantially higher in earlier-line therapy, as we just saw from DB03, but also unfortunately there’s significant attrition in each line of metastatic therapy for our patients. Patients develop catastrophic complications of their treatment or of their disease, they lose their performance status to the point where they’re not really candidates for further therapy, and that happens in about 10% or even 15% of each line. So if you hold back your best therapy for later line there’s going to be patients who never are able to receive that therapy, and that’s not great.

And lastly, this kind of is an academic point in some ways. DB02 was the first randomized trial that showed that there’s benefit of one antibody-drug conjugate after a patient had already progressed on another antibody-drug conjugate. And this is important because it provides support for this paradigm that we’re going to start treating patients with sequential ADCs, and I think you’re going to start seeing more and more data and trials looking at that question, not only in HER2-positive disease but other kinds of breast cancer and other cancer types as well.

We had some more data on T-DXd this past year. We looked at this question of whether age impacts efficacy or tolerability of T-DXd. We know from multiple trials that older patients experience higher rates of toxicity from virtually all types of chemotherapy.

So to see if this was also true with T-DXd we did a pooled analysis of patients from all the 3 large DESTINY trials and looked at patients younger than 65 versus those that were 65 and older. As you can imagine, there were higher rates of comorbidities in the older patients, particularly vascular disorders, hypertension, and renal function, but despite that the efficacy of T-DXd was essentially the same in older versus younger patients. And that’s true for progression-free survival, shown here, as well as overall survival, shown on the right. And response rates were virtually identical depending on younger versus older patients, and that probably is because the dose intensity that was able to be given of T-DXd was essentially the same for younger and older patients.

Now the 1 difference was that there were greater rates of Grade 3 or higher adverse events in older patients. It was about 44% compared to 54%. This was largely due to neutropenia and anemia. There also was a greater rate of ILD but no greater rate of fatal ILD.

So I think the conclusion here is that T-DXd is an effective treatment option regardless of age. The safety profile was manageable both in younger and older patients, although there were higher rates of Grade 3 adverse events in the older patients. But I think the take-home message is that T-DXd is an effective option for patients across all age groups.

DR LOVE: So I’m just kind of curious, at San Antonio last year, a year ago, we actually had 2 90-year-olds presented in 2 different conferences that got T-DXd. You might have been on the panel of one of them. I don’t remember. But both of them responded. Both of them had no toxicity. It was incredible. They both had recurrent chest wall disease. The same doc, actually, in both patients, pre-emptively dose reduced them. If you have an older patient like that do you use pre-emptive dose reduction?

DR KROP: Honestly, I can’t remember the last 90-year-old I’ve treated with T-DXd, but I think you have to look at the performance status of the patient. I think more importantly than age is look at their comorbidities, look at their performance status. But I think these data would say that there’s not a strong reason outside of — that age alone should not be a reason to pre-emptively dose reduce patients. But clearly T-DXd has activity at lower doses, as well, so I think you have to take it on a patient-by-patient basis.

DR LOVE: So another question. You mentioned this issue of sequencing ADCs, and of course we’re hearing about this in other cancers. And one of the things I hear people talking about is this I guess early question of whether or not you want to use 2 ADCs in a row with either the same target or the same payload. Any thoughts? Do we have any preliminary data about, again, target versus payload in terms of sequencing?

DR KROP: Yeah. I think these are early days because up until pretty recently we didn’t have multiple ADCs to play with in the clinic. I think ideally it’d be great to be able to switch payloads, as well as switch targets. I think in practice if we can switch one or the other that has utility.

At San Antonio this year there were some very small datasets oftentimes from single institutions looking at how people did who had progressed on one ADC and then got a second ADC. The data weren’t terribly encouraging, in my take on it. The PFS’s in the second ADCs tended to be short, 2 and 3 months. But I think we really need more prospective data, and there are trials getting underway looking at that.

And again, here in this trial we saw, in DB02, quite good activity of T-DXd after progression on T-DM1 with progression-free survivals in the 16, 17-month range. So it certainly can work, but we need some prospective studies. It’d be great if we had biomarkers that say hey, you should start with a TROP2 ADC for this patient and then more to HER2 or vice versa, depending on how much TROP2 and how much HER2 they have, or some other biomarker. And we’ll get that as we do some of these sequencing trials, but they’re just getting underway now.

DR LOVE: I mean if you think about T-DXd versus T-DM1, there it’s switching the target — I mean there it’s switching the payload not the target.

DR KROP: Correct. Correct.

DR LOVE: And that totally worked. Anyhow, please continue.

DR KROP: And then let’s get back to this, what I think is kind of one of the more clinically important and hot topics right now in HER2-positive breast cancer, is managing how we best manage patients who have brain metastases, which unfortunately is quite common. Up to 50% of patients with HER2-positive metastatic disease will develop CNS metastases. And I think we had this big breakthrough that, Neil, you had mentioned earlier with tucatinib, which is not only HER2 selective but also clearly had some activity in the brain in early studies.

And that led to this HER2CLIMB trial, which looked at patients who had already had trastuzumab/pertuzumab and T-DM1 and randomized them to trastuzumab and capecitabine with either tucatinib or placebo. And because of that early data showing activity in the brain this trial, unlike any other large, randomized trial in breast cancer that I can think of, allowed not only patients with stable brain mets but progressive brain mets, and about 50% of patients on this trial had brain metastases at baseline.

And this trial showed not only that the addition of neratinib improved overall survival but particularly in this difficult to treat patient population with progressive brain mets there was a large improvement in survival, 48% reduction in death, with the addition of tucatinib. And there was an intracranial response rate of 47%. So clearly this demonstrated the activity of tucatinib-based treatment in patients with HER2-positive brain mets.

So the conclusion of this study was that the addition of tucatinib improved overall outcomes with manageable toxicity. It’s the first study to include active brain metastases as a substantial population, and tucatinib was particularly effective in this subset. It established, as I mentioned, that tucatinib-based treatment was the standard of care for progressive brain mets, but it left this question out, or maybe led to this question of should T-DXd be the second-line standard of care for these patients when we knew that tucatinib was so active in patients with progressive brain mets.

So are there other options for treating progressive brain mets besides TKIs? And I think the big question is does T-DXd have activity in these patients with brain mets. The dogma had been that large biologic molecules like antibodies or ADCs don’t cross the blood-brain barrier so therefore aren’t good to use for patients with brain mets.

So we were very fortunate to have this very nice pooled analysis that Sara Hurvitz presented at Madrid at ESMO this past year which looked specifically at this brain met question in the DESTINY HER2-positive trials. And what I think was surprising and encouraging was that there was substantial intracranial response in patients with brain metastases treated with T-DXd. And interestingly it didn’t matter whether patients had stable treated brain mets, on this first column on the left, versus untreated progressive brain mets. The response rate was exactly the same, 45%, and both were substantially better than the control arm response rate.

And if you look at progression in the brain, particularly for the progressive brain metastases patients on the right here, huge improvement with T-DXd compared to the control arm with a hazard ratio of 0.19, so an 80% reduction in PFS events in this population.

We also have had some small studies of T-DXd in progressive brain mets, the TUXEDO study and a case series from Dana-Farber and other institutions, showing response rate above 70% in the brain. And then in this updated data that was presented in the DEBRA trial we saw intracranial response rates of 40-50% in HER2-positive breast cancer.

So I think the conclusion from both the pooled analysis, as well as these small studies is that we now have I think fairly compelling data that the activity of T-DXd works in both stable and progressive brain mets, and together I think this question about whether there are — a biologic like T-DXd can have activity in the brain has now been answered. It’s most likely because the blood-brain barrier breaks down when there’s brain metastases present.

So how do you put all of this together knowing we now have multiple approaches that work in progressive brain metastases? This is the way I think of approaching patients with HER2-positive metastatic disease. We still are using the CLEOPATRA regimen in the first line, but in the second line we have to start thinking about the presence of brain metastases. So patients who don’t have brain metastases or stable brain mets clearly T-DXd is the standard of care given its very high rate of efficacy and durable activity.

But then in patients who have progressive brain mets I think we’re at the point where we have to start parsing this out a little bit more and looking at the patients who have CNS-dominant disease, those patients who you’re convinced or you’re worried that are really going to get into huge trouble with their progressive brain mets, more so than their extracranial disease, and here I think we have to rely on the HER2CLIMB data with the very tucatinib activity in this population. However, patients with progressive brain mets but a smaller amount, where their extracranial disease or liver disease or lung metastases are more dangerous for that patient, then I would use trastuzumab deruxtecan in that patient population, again, taking advantage of its 80-plus-percent response rate and durable activity. And then in later lines you’re just going to use what you haven’t used previously.

And then lastly, just one more. We have new data with tucatinib from the HER2CLIMB-02 trial. So this looking at this question of should you combine T-DM1 and tucatinib, because we know that there are laboratory studies suggesting synergy between T-DM1 and tucatinib, the idea by combining them we also potentially can get better brain met coverage than we’d have with T-DM1 alone.

So this trial enrolled patients who had progressed on trastuzumab and taxane and randomized them to T-DM1 with placebo or T-DM1 plus tucatinib. And these patients had not had prior HER2 TKIs or T-DXd. So this is not exactly the same population that we would be seeing now in the clinic.

There was a benefit to adding tucatinib. Progression-free survival was about 2 months greater, so it was a pretty modest difference. Hazard ratio was 0.76. There was a little bit more activity in the patients who had baseline brain metastases, still about a 2-month improvement, but a hazard ratio of 0.64. Response rate was pretty similar between the 2 arms, and survival, while immature, shows no evidence of benefit with tucatinib.

And adverse event profile overall showed a modest increase with adding tucatinib, but there was — I should call attention to the fact that there was about a 15% risk of Grade 3 transaminase elevation, a 7% risk of Grade 3 thrombocytopenia, so there was increasing toxicity in that regard.

So in conclusion, I think the addition of tucatinib to T-DM1 does modestly improve PFS. The benefit’s a little bit greater in CNS-positive patients, but there’s no survival benefit, and there is increased toxicity. And again, as I said, these are patients who had not yet had prior T-DXd, which would be the patient population that we would typically think about this question.

So in my mind it’s unclear whether tucatinib should be used with T-DM1 or should you use it with trastuzumab and capecitabine. And it begs the question of is there a role for using it in both combinations. So do you use tucatinib twice, first with T-DM1 and then with trastuzumab/capecitabine or the other way around? We don’t know whether there’s still benefit to tucatinib in patients who previously had it, but it’s possible that the resistance could be mediated through T-DM1 or through capecitabine, and therefore there could be activity in multiple lines. But I think that’s a question that we really would like to answer with some clinical trial data.

So I’m happy to stop there, and if you have any additional questions.

DR LOVE: Yeah, just a couple. I see that you had I guess a Phase II trial of tucatinib plus T-DXd at ASCO 2022, but that’s like a trial in progress. I mean that seems like a logical thing to do. Any data from that?

DR KROP: No. So it’s still underway, and that’s not a randomized trial, so that’s a single-arm study.

DR LOVE: Right.

DR KROP: But again, the same kind of rationale that there is preclinical synergy between TKIs and antibody-drug conjugates, and the hope was that by combining them together you’d not only have better efficacy, you also might have better brain met therapy. So we’ll have to see how that — when that trial reads out how things look.

DR LOVE: It’ll be interesting to see how they tolerate it. I mean I guess my take in general is like T-DXd is a little more complicated or more tolerability issues than T-DM1, I think.

DR KROP: Yes, no. That’s a correct perception, and that’s something that we’re going to be careful about in that trial because, as you pointed out, T-DXd does have more toxicity in general than T-DM1, so adding another drug on top of it could be a problem. So we’ll have to see.

DR LOVE: So one thing I was hoping you could try to straighten out for me is like patritumab and like how it works and like how it like — in terms of like what pertuzumab does and its target and what patritumab does. That whole biology kind of confuses me. Can you help me out?

DR KROP: So pertuzumab is a HER2-directed antibody but does block the interaction of HER2 and HER3. Patritumab is a HER3-directed antibody, and its main role, we think, in antibody-drug conjugate is providing another target as a way of delivering the deruxtecan payload to those cancer. Whether it has a signaling effect I think it’s hard to tease out at this point.

But the data that we presented with patritumab deruxtecan and which subsequently has been shown has been replicated in other studies is that it is an ADC that’s quite active in all subtypes of breast cancer because HER3 is expressed in all subtypes of breast cancer. And so there’s been activity particularly in hormone receptor-positive, heavily pretreated patients, response rates over 30%, with substantial progression-free survival. And interestingly in a small population of HER2-positive patients, where we know that HER3 can be upregulated, there was a response rate of I think 44% in a heavily pretreated population, but small numbers.

So it’s one of those situations where it’s a drug that seems to work everywhere, including in triple-negative cancers, which makes it a little bit hard to develop. It also has the somewhat Achilles heel of having the same payload as trastuzumab deruxtecan —

DR LOVE: Right.

DR KROP: — datopotamab deruxtecan, both of which are active in HER2-positive or HER2-low and triple-negative cancers. So where do you fit in — where does patritumab deruxtecan fit into there? I think it’s a little complicated.

DR LOVE: So yeah. Can you actually target HER3? For example, if you look at let’s say patritumab deruxtecan in triple-negative breast cancer, are some patients with triple-negative breast cancer that you can actually identify HER3 or you’re just giving it to everybody, and you study triple-negative separately? Or can you actually measure HER3 in a patient?

DR KROP: So you can measure HER3. In the trial that we presented you had to be HER3-positive, triple-negative to go on.

DR LOVE: Huh. Wow.

DR KROP: In retrospect in the hormone receptor-positive population, we included both HER3-low and HER3-high populations, and it really didn’t make any difference, which is maybe because we saw that HER3 levels seemed to go up and down over time, so it’s a dynamic biomarker, so it may be the reason why we’re not seeing strong correlations between level of expression and activity is because the level of expression changes pretty rapidly.

DR LOVE: Right.

DR KROP: But the bottom line is that HER3 is overexpressed in the majority of all 3 breast cancer subtypes, similar to TROP2, and we haven’t been able to show a strong relationship between level. That doesn’t mean we can’t do it. It just may mean that we need better testing. But I think it gets back to what we talked about earlier, which is as we’re trying to sequence these various ADCs could you do a biomarker test beforehand to say this is going to be most sensitive to a TROP2-targeted drug versus HER3.

DR LOVE: Right.

DR KROP: This one’s HER2. And hopefully some of these trials that are getting underway will — the translational data will be able to start answering some of those questions.

DR LOVE: So another question. I don’t know if you ever see in your inbox an email from somebody you don’t know, and you see where it’s coming from, and you know exactly why — you haven’t thought about it, but you know why they’re contacting you. And that just happened to me, which is I got contacted by a person who runs an advocacy group for people with lobular carcinoma of the breast. I don’t know if you’re familiar with that advocacy group or that idea. I had never heard or even thought about it, but as soon as I thought about the idea of an advocacy group for lobular carcinoma I immediately started to get it, I think.

And I’m curious, in your own mind, like when you think about lobular — I mean obviously you have the clinical things that we know, the pathologic things, but is it really a different disease? Has there been anything targeted that’s specific to lobular?

DR KROP: Yeah. So I mean I understand the motivation of this advocacy group, and we hear it a lot from other patient groups and from many physicians, is that it is a different disease, it’s just in the clinic we don’t really treat it differently for the most part. But it definitely has some very interesting, as you said, kind of pathologic differences and molecular differences, but for the most part we don’t treat it differently, and that’s because so far we haven’t really been able to find big differences in how lobular cancer responds to treatment than other ductal cancer for example.

There were some data presented a number of years ago suggesting that lobular breast cancers respond better to adjuvant therapy with aromatase inhibitors than it does to tamoxifen, and I don’t know whether clinicians are using that as a way to determine what treatment they use in the adjuvant setting. But as I said earlier, they have higher rates — lobular breast cancers have higher rates of HER2 mutations, so using neratinib in those — thinking about that, testing lobular breast cancer for HER2 mutations, and using neratinib-based treatment in those cancers is very appropriate and supported by the data.

But yeah, in terms of in general we really haven’t been thinking of lobular cancers differently than ductal cancers for most patients, even though we do know that there are clearly some differences. They have more leptomeningeal disease, they have more peritoneal disease compared to ductals, but outside of that I think we’re kind of — we don’t really pay attention to that, and that is an area that is ripe for research.

DR LOVE: So on the other hand I really immediately empathized, and I just hadn’t thought about it. I imagine you see this clinically with the patients. They’ve got a label. And it’s all well and good to say that it’s very similar to this other thing, but clearly it is different. We just don’t know the biology. So I totally understand why it’s maybe concerning or stressful or uncomfortable for patients. And I don’t know, I just thought it was a very interesting thought, and I imagine the day that anything comes out, if it ever does — I don’t know. Cadherin? Is that a target?

DR KROP: Yeah. I mean it’s a loss of cadherin, so that’s harder to target losses than it is gains.

DR LOVE: Right. Right. Right.

DR KROP: But your point is well taken, and again, I appreciate where the advocacy groups are coming from because, as you said, it’s not — I’m sure it’s not a good feeling to be labeled something and then say, “Well, we don’t know what to do with that.”

DR LOVE: Or there’s no research on it.

DR KROP: Right. Right. Well there is research. It just hasn’t really been as fruitful as we’d like.

DR LOVE: Right. Not specific. Yeah.

DR KROP: And there’s room for more research, and I think there are people who are studying that.

But I think we can reassure our patients that even though there are these molecular and pathologic differences people with lobular breast cancer still do very well. There’s no data that they do worse than other breast cancer histopathologic types. So it doesn’t say that there’s not room for improving our treatment, but they don’t seem to be at a disadvantage. There’s cancers respond well to our current armamentarium of treatments right now. But certainly there’s more work to be done.

DR LOVE: So yeah. I mean I appreciate the issue, but also if you think about it, or actually I looked the numbers up, there are actually more patients with lobular cancer of the breast than patients with myeloma diagnosed every year, and if you think about research specific to lobular compared to myeloma, I mean in myeloma we talk all day long about the research. So I think it’s an interesting question. I don’t know what the answer is though.

DR KROP: Yeah. No.

DR LOVE: But anyhow, I never thought about it, and then as soon as I saw that email I’m like I get it. Anyhow, we’ll see where that heads.

One more question, more global, again. There was a paper that came out I think fairly recently stating that deaths from cancer in general are going down, including in breast cancer. I’m curious if you have any sense of the magnitude in breast cancer specifically about let’s say change in age-related mortality now let’s say compared to 20 years ago, and what is responsible for it.

DR KROP: So I can’t quote you the specific number. I did see that paper, but off the top of my head I don’t remember the percent. But I think it’s clearly multifactorial. Obviously we’re screening patients much better and identifying cancers earlier than later, which of course is a huge benefit.

We have much better treatments for the most lethal breast cancers now. We’ve just talked about the incredible transformation of how we treat HER2-positive disease now, where very few people fortunately are recurring with HER2-positive breast cancer. We now have multiple targeted therapies for triple-negative disease where we had virtually none just a few years ago. And our hormone receptor-positive patients now we have targeted therapy beyond endocrine therapy for early stage patients, which should really help the higher-risk ER-positive patients now that we can use CDK4/6 inhibitors.

So we’re making a lot of progress on all fronts in terms of detection, early stage treatment, and later stage treatment, and all of those are contributing. And prevention we’ve also made — we’ve kind of made prevention benefit in that we’ve now learned that exogenous hormone replacement is probably not a good thing for postmenopausal, and I think people — that message has gotten out, and the amount of postmenopausal hormone replacement has been minimized, and that’s led to decreases in incidence of breast cancer. So I think all of those things together are translating into kind of meaningful survival benefits, which is great to see, but we still have a ways to go.

Triple-Negative Breast Cancer — Priyanka Sharma, MD

DR SHARMA: Hello. I’m Priyanka Sharma, a Professor of Medicine at University of Kansas Medical Center, and I will be discussing updates in triple-negative breast cancer today.

So the first update is from KEYNOTE-522 trial. We’re all aware of the seminal study that established the role of neoadjuvant and adjuvant pembrolizumab for patients with high-risk, early-stage, triple-negative breast cancer.

This update was presented at ESMO last year, and this was interim analysis 6, which continued to demonstrate the superiority of pembro plus chemo over chemo alone in terms of event-free survival, with an absolute difference of 9%, as you can see on the curve on the left-hand side. On the right-hand side the EFS curves by pCR are shown, and again, improvement in patients with both pCR and residual disease is seen. The delta of improvement in patients with pCR is much smaller at 4%. And you can see overall outcomes of patients who achieved pCR with either chemo or chemo plus pembro is much better than those that have residual disease. Again, this is something that we have known about triple-negative breast cancer for quite some time.

We also see from this curve in patients with residual disease that in spite of neoadjuvant and adjuvant pembrolizumab about 38% of patients with residual disease have an EFS event at 5 years, so clearly it tells us that even with the addition of pembrolizumab we have room for improvement, and we need better therapies for these patients.

These are further subgroup analyses of KEYNOTE-522 that were presented at San Antonio Breast Cancer Symposium just a few months ago, and you can see the benefit in patients with Stage II and III disease is similarly seen with similar hazard ratios. And when we look at patients with Stage III disease, again in spite of pembrolizumab introduction both neoadjuvantly and adjuvantly, about a third of patients have an EFS event. So, again, we need better therapies for these patients.

The benefit of pembrolizumab was also similarly seen in patients with node-negative disease and node-positive disease. But if you pay attention to the curve for patients with node-negative disease what we see is 77% of patients with node-negative disease who were treated with conventional chemotherapy did not have any EFS event. It also tells us that not all patients may need the treatment escalation beyond chemotherapy with pembrolizumab, so the scientific community still has a lot of work to do in terms of figuring out which patients may not need immunotherapy and can be spared the toxicity associated with it.

So in terms of summary for the interim analysis 6 from KEYNOTE-522, after a median follow up of greater than 5 years neoadjuvant pembro plus chemo followed by adjuvant pembrolizumab continues to show clinically meaningful improvement in EFS compared with neoadjuvant chemo alone. The EFS benefit was consistent across subgroups based on TNM stage and nodal status. The follow up for OS is ongoing. EFS improvement in pembrolizumab group was observed regardless of the pCR outcomes.

One point I would like to make here is because of the design of the KEYNOTE-522 is not possible to discern if this benefit is driven by the neoadjuvant component of pembrolizumab and how much the adjuvant component is contributing to improvement in outcomes.

Another point to make is the pCR and residual disease assessment for this international trial were done locally, and one can kind of think about the reliability of local assessment of pCR in this international trial. So I would take the results of the pCR and RD with this grain of salt.

The next update that we saw at ESMO was from the NeoTRIP trial. Again, this was a neoadjuvant trial that looked at combining atezolizumab with carboplatin/docetaxel neoadjuvant chemotherapy. This trial did not have an adjuvant immune checkpoint inhibitor component. This trial really focused and enrolled patients with very high-risk disease. Almost 90% of the patients had node-positive disease in this study.

We’ve already seen the pCR results from this trial. They have been published, and there was a small numeric improvement in pCR, which was not statistically significant with the addition of atezo to chemotherapy. However, they did see an interaction with the PD-L status and pCR improvement with atezo with higher pCR improvements seen in patients who have higher PD-L1-positive tumors with atezo.

So at ESMO we saw the EFS results from this trial, which was actually the primary endpoint of this study. There was no improvement in EFS with the addition of atezolizumab to chemotherapy in this trial, and overall the pCR status, positive PD-L1 status, earlier stage, higher sTILs were all prognostically linked to better EFS but were not predictive of atezo benefit in this small trial.

So why was there no pCR or EFS benefit in the NeoTRIP trial? Could it be the PD-1 versus PD-L1 inhibitor, the difference between the pembro and atezo here? We do know from IMpassion031 neoadjuvant trial that the addition of atezolizumab to AC-T chemotherapy did improve pCR rate, and there was numeric improvement in EFS seen in that trial.

We also know from the metastatic TNBC studies that atezo plus taxane were not statistically superior to taxane in terms of overall outcomes, and that led to removal of approval of atezolizumab for metastatic triple-negative breast cancer. Could it be the anthracycline backbone? Could it be the chemotherapy backbone as NeoTRIP did not have an anthracycline in the backbone? Could it be the anatomical risk of the enrolled population?

As I said, 90% of patients that enrolled in the NeoTRIP trial had node-positive disease compared to 30% in the IMpassion031 trial and 50% in the KEYNOTE-522 study. With the higher anatomical risk of these patients would the efficacy be more in line with what we see in metastatic TNBC, where the atezo plus taxane didn’t meet the bar to retain approval of that combination? Could it just be the difference in tumor biology of the patients that enrolled in different studies? We do know that patients in the chemo-alone arm of NeoTRIP had higher sTILs given that control arm inherent better prognosis compared to the IO arm. Or could it be just chance in this small trial?

Now moving on to ALEXANDRA, or IMpassion030 Phase III trial. So this was an adjuvant alone IO trial. Patients that enrolled in this trial underwent surgery first and then received adjuvant chemotherapy either alone or with immunotherapy with atezolizumab. And as you can see on the right there was no improvement in invasive disease-free survival with the addition of immunotherapy to adjuvant chemotherapy in this trial. In fact, this trial was stopped early because futility boundary was crossed.

What this tells us is that the timing of IO matters, and neoadjuvant IO is more effective than adjuvant IO. This is something that we’ve actually seen in previous animal data, and these clinical studies are mimicking what we’ve kind of learned from animal studies. And similar observations have been seen in other cancers, like melanoma, where the neoadjuvant IO is more effective than the adjuvant IO.

DR LOVE: Along those same lines I should ask you, how often do patients end up getting radiation therapy as part of their initial treatment when they get neoadjuvant therapy, and is there any correlation between whether they get radiation therapy and benefit?

DR SHARMA: So most patients that have breast conservation, meaning lumpectomy, will have radiation after surgery, and patients with locally-advanced disease, Stage III disease, will have radiation even if they have a mastectomy.

The abscopal effect in breast cancer has been evaluated in some studies and has really not panned out like it might in some other tumor types, but I think the immunotherapy aspect is if you have the intact tumor the immunotherapy has a higher chance of enhancing the innate immune response with the intact tumor already being there as opposed to removing it and now focusing on only just micrometastatic disease for these patients.

DR LOVE: The reason I brought it up is because there’s been out there for a while the big study in non-small cell lung giving IO after chemoradiation, durvalumab, really great effect. More recently there was a study in cancer of the cervix, same thing, after chemoradiation. So this hypothesis was maybe radiation is in some way doing something to neoantigens, potentiating it. I mean in breast though you’re not really radiating tumor exactly, so I don’t know. I was just curious if they’ve looked at it.

DR SHARMA: And both of those studies were also with the tumor intact, right? So the chemoradiation followed by immunotherapy —

DR LOVE: Absolutely. Absolutely.

DR SHARMA: — was given the tumor was still intact.

DR LOVE: Right. Right.

DR SHARMA: And in breast most of these cancers, at least early on, are quite chemosensitive, so we don’t have to have radiation to get the response rates to that level.

DR LOVE: Right. Right.

DR SHARMA: But that’s an intriguing thought. For tumors that already have a very high immune infiltrate and have the best immune microenvironment, for those tumors could you just do away with chemo completely? Do a little course of radiation, get your microenvironment built up to where it needs to be, your sTILs increase, and then just to immunotherapy. So from a research perspective I think that’s a pretty intriguing concept, we just have to find the right low-risk population to be brave enough to take away chemotherapy, at least in breast cancer.

DR LOVE: Okay. Please continue.

DR SHARMA: So the next trial I’m going to talk about is the NeoPACT trial that was published in JAMA Oncology recently that again looked at an anthracycline-free chemotherapy backbone of carbo, docetaxel, and pembrolizumab in early-stage triple-negative breast cancer. This trial also did not have an adjuvant immunotherapy part since it was primarily conducted before the approval of KEYNOTE-522 regimen.

And in this trial the pathological response rate — complete pathological response rate was noted in 58% of patients, which is quite in line with what we see with our more complicated chemotherapy regimens. And immune enrichment was identified in almost 50% of the patients using different measures, sTILs, PD-L1, or signatures, and patients who had tumor immune enrichment had a pCR rate exceeding 70% in this trial.

The 3-year EFS was 85%, pretty consistent with what we have seen in KEYNOTE-522. So this trial provides data on alternative anthracycline-free chemoimmunotherapy regimen for patients who are not eligible or wish to avoid anthracycline-based regimens and support further evaluation of this regimen as the chemotherapy de-escalation strategy in randomized studies. In fact, an ongoing NCI sponsored study, SWOG 2212 or SCARLET, is evaluating this question in a large noninferiority trial.

So let’s talk a little bit about immune toxicity of the KEYNOTE-522 regimen. So at San Antonio we saw this very intriguing data that was compiled from 17 sites looking at the real-world toxicity of the KEYNOTE-522 regimen. And what we saw here is 71% of patients reported immune-related AEs of all grades. AE led to early discontinuation of therapy in almost 40% of the patients. And again, almost 40% of the patients — about a third of the patients reported Grade 3 or more immune-related AEs. Immune-related AEs led to hospitalization in almost 40% of the patients, and intriguingly there were differences in the rates of immune-related AEs between blacks and whites, with higher proportion of whites noting immune-related AEs, so this is something I think we should investigate across tumor types.

The pCR reported in this retrospective analysis across 17 sites with this regimen was actually in the mid 50s, so a little bit lower than what we’ve seen in the KEYNOTE-522 study, which is probably a real-world reflection of the tolerability of this regimen. If about 40% of the patients are unable to take the entire regimen it clearly will have an impact on the efficacy of the regimen.

At San Antonio we also saw data from UCSF regarding delayed and late toxicity of immunotherapy in breast cancer patients. So about 35% of the patients reported delayed toxicities which occurred 3 months after starting of checkpoint inhibitor therapy, and about 8% of patients reported post-treatment toxicity which occurred actually 2 months or more after stopping all immunotherapy.

Another aspect of immunotherapy that’s pertinent to our patients is potential impact on fertility. This paper from 2022 suggested that at least in mice immunotherapy treatment can have an impact on the ovarian reserve. We really have no good data on the effect of immunotherapy on fertility or ovarian reserve or fertility markers in breast cancer patients, especially as we are treating now all of these patients with early-stage breast cancer. These are young women with triple-negative breast cancers. It’s an important clinical question. So in the ongoing trials they’re collecting biospecimens to address this question, so hopefully in the coming years we will learn more about it.

So in terms of the side effects and toxicity of immune checkpoint inhibitor plus chemo combinations. Especially in breast cancer I think we still have a lot more to learn in terms of the extent of side effects and our management skills to be able to counsel our patients appropriately.

Now I’ll move on to metastatic triple-negative breast cancer. And the abstracts and papers I will review are divided in these following sections. We will review first-line chemo plus IO data. This is the additional analysis from KEYNOTE-355 study. We will review the newly presented TORCHLIGHT study. I will review updates from DESTINY-Breast04 and ASCENT quality of life, and these are both approved ADCs in breast cancer. We will review some data of ADC plus IO combination from the BEGONIA trial. I will review some early data on novel ADCs. We will look at optimal capecitabine dosing in metastatic breast cancer and then some updates on OlympiAD study regarding PARP inhibitor and BRCA1 and 2 mutation-associated breast cancer.

So first an update from KEYNOTE-355. As a reminder, KEYNOTE-355 established the role of pembrolizumab in combination with chemotherapy for patients with metastatic breast cancer that expresses PD-L1 tested by CPS score. In this trial there was clinically meaningful and statistically significant improvement in both progression-free survival and overall survival with the addition of pembrolizumab to chemotherapy.

This updated analysis that we saw was an exploratory analysis to evaluate the efficacy in patients who had disease control and then discontinued chemotherapy but stayed on pembrolizumab and also look at the efficacy based on immune-mediated AEs or not. So this table on the right shows PFS and OS in patients treated with pembro plus chemo who either had an AE or not, and we can see about 30% of the patients within the PD-L1 CPS 10 group — I’m going to focus on that because that’s where we clinically use pembrolizumab in metastatic TNBC right now. So about 30% of the patients did experience an immune-related AE, and the median PFS and OS in patients either experienced an immune-related AE or not appeared quite similar.

Now looking at patients who derived benefit from the combination, meaning that they had either a complete response, partial response, or stable disease for over 24 weeks, again, looking at patients with PD-L1 score of greater than 10, about 24% of the patients with CR, PR, or stable disease actually discontinued pembrolizumab before — discontinued chemotherapy before they discontinued pembrolizumab, probably from the side effects — probably related to the side effects from chemotherapy. And what we see is the PFS and OS of patients who discontinued the chemo before pembrolizumab is quite similar to the overall PD-L1 CPS greater than 10 population. In fact, numerically it looks better than the overall PD-L1 CPS greater than 10 population. This is quite exploratory, so no statistical testing was done.

So in summary, what this update tells us is about a quarter of patients with a clinical benefit from the combination of pembro plus chemo stopped chemo prior to stopping pembrolizumab in KEYNOTE-355. And there was no evidence of detrimental effect on PFS or OS among patients with clinical benefit who discontinued chemo and continued pembrolizumab monotherapy. This is reassuring data for our patients who get initial benefit from chemo/pembro combination and then have to stop chemotherapy for side effects or quality of life, and I would say chemotherapy can safely be stopped after 6 to 8 cycles for side effects. In this subgroup analysis the median number of chemotherapy cycles for patients who ended up stopping it and continuing pembro was around 6. And also in patients who experienced an immune-related AE the benefit from pembro was similar compared to those who did not experience an immune-related AE, so that shouldn’t deter us from continuing the immunotherapy.

TORCHLIGHT was a randomized, double-blind, Phase II trial of another PD-1 antibody, toripalimab, that was presented at ASCO this year. This trial compared toripalimab versus placebo in combination with nab-paclitaxel for patients with metastatic or recurrent triple-negative breast cancer and enrolled 531 patients. The trial used a new PD-L1 assay to assign PD-L1 positive status, and a CPS of greater than 1 was considered PD-L1 positive, and by this criteria 56% of the patients had PD-L1-positive disease.

The data presented at ASCO was prespecified interim analysis 1 of PFS in PD-L1-positive population and the intention-to-treat population. Even though the trial allowed both first and second-line patients the majority of the patients enrolled in this trial were receiving this treatment in first-line setting, and about a quarter to a third had de novo metastatic disease. One thing to point out, that only 15% of patients had received a prior taxane for early-stage disease. This is a little bit different than what we would see in our contemporary population here in the United States, where most patients with early-stage disease would have received some form of taxane as part of their curative therapy.

So there was a numerical and statistically significant improvement in PFS with the addition of toripalimab to nab paclitaxel with a p-value that met the prespecified boundary for statistical significance.

So this trial had a hierarchal testing statistical plan. If the PFS in the PD-L1-positive population was positive they would test the PFS in intention-to-treat population, followed by OS in PD-L1 positive, followed by OS in the entire trial population. So since the PFS and PD-L1-positive population was positive they tested PFS in the ITT population, which numerically looks better but did not meet their prespecified boundary to meet statistical significance. So the further overall survival at this point is exploratory since the hierarchal testing wasn’t met. And we can see that the overall survival looks numerically better, but again, this is too early to make final conclusions.

Very limited toxicity that was presented. Per authors, there were new safety signals with this combination. Details of percentages of immune-related AEs or the types of immune-related AEs were not provided, and we hope to see that data in future presentations with longer follow-up.

So in summary, at the interim analysis A, first interim analysis, PFS was significantly prolonged with the addition of toripalimab to nab-paclitaxel in PD-L1-positive population. In fact, the hazard ratio and median PFS we see in this Phase III study are very similar to what we noted in KEYNOTE-355. The PFS and ITT did not meet the prespecified hierarchal testing boundary. OS data is also immature so further follow up is ongoing. Again, PD-L1 testing was done but yet another assay and threshold. This is becoming a common problem for us in clinic. Each checkpoint inhibitor is tied to a different PD-L1 assay and the threshold for different tumor types.

This trial enrolled patients only in China, so if overall after further follow up the trial is positive I’m uncertain of what the approval path for this drug would be in other countries. Limited toxicity data was provided at this initial presentation, and we look forward to hearing more on toxicity at future presentations.

ASCENT had an update that was published last year. ASCENT was a trial that led to the approval of sacituzumab govitecan in patients with metastatic triple-negative breast cancer and compared SG to treatment of physician’s choice in this patient population. So in this trial health-related quality of life was assessed on day 1 of each treatment cycle using the EORTC QLQ-C30 score. Data was collected up to 6 cycles, and the analysis population for this health-related quality of life study included a substantial portion of the main trial population.

So this publication showed that SG was superior to treatment of physician’s choice for the global health status, physical functioning, fatigue, and pain, but chemotherapy treatment of physician’s choice was superior to SG in terms of nausea, vomiting, and diarrhea. We know that these are all known side effects of SG. SG arm had a significantly better QLQ-30 summary score, median change from baseline, than TPC arm. Median time to worsening of quality of life and global health was similar for both treatment arms, but looking at individual elements of it the median time to first clinically meaningful worsening was longer for SG than for TPC for physical functioning, role functioning, fatigue, and pain.

So we know that SG prolongs progression-free survival and overall survival compared to chemotherapy treatment of physician’s choice in patients with metastatic TNBC/breast cancer. And what this data shows is SG is still really associated with greater improvements in delayed worsening of quality of life compared with chemotherapy of physician’s choice, supporting its continued use in patients with metastatic triple-negative breast cancer in preference of using this over treatment of physician’s choice.

We saw an update from DESTINY-Breast04 at ESMO with an updated OS and safety data. At the initial publication the median follow up of this study was 18 months, and here we see a much higher median follow up. So with the longer follow up there are actually no new cases of ILD/pneumonitis since the primary analysis cutoff in 2022. With this longer follow up in terms of overall survival in the hormone-positive cohort and in all patients median OS was consistent with the results from the primary analysis, showing a 31% reduction in risk of death for patients receiving T-DXd compared to those receiving chemotherapy treatment of physician’s choice.

Exploratory subgroup analysis for efficacy in the hormone-negative cohort shows that numerically the median PFS at this updated analysis is a little bit lower than what we saw at initial analysis. However, hazard ratio still continues to support T-DXd. Now the confidence interval around the hazard ratio is crossing unity. But again, it’s an exploratory analysis. We see similar findings in terms of overall survival, where the overall median OS is a little bit lower than the initial analysis. The hazard ratios continue to be the same, but the confidence interval is now crossing unity.

So results from this 32-month median follow up for the DESTINY-Breast04 confirm the sustained improvement in PFS and OS for T-DXd over chemotherapy treatment of physician’s choice in HER2-low metastatic breast cancer. We do see a little bit of a drop in median PFS and OS in the T-DXd arm in the hormone-negative cohort, but again, this is an exploratory analysis. And the hazard ratio continued to be similar to the initial analysis even for this hormone-negative cohort.

With longer treatment duration the overall safety profile of this agent is consistent with the primary analysis, and no new cases of ILD or pneumonitis were reported at this longer follow-up.

Since we’re talking about T-DXd, where we need to identify the HER2-low disease to find eligible patients, this abstract I have was presented at ASCO is a right fit looking at the impact of repeat biopsy in HER2-low status among patients with triple-negative breast cancer. So this is a retrospective study from a single institution that included 512 patients with triple-negative breast cancer who had at least 1 biopsy with a known HER2 IHC status. Many patients had more than 1 tissue sample available, but if you look at the box on the right, box in the green on the right we can see the majority of patients had 1 or 2 tissue samples with a few that had more than 2 tissue samples.

And what we see is HER2 status changes for half of the patients throughout the disease course. It actually changes in both directions, from HER2 low to 0 and from 0 to HER2 low. And we see this change both between the core biopsy and the surgical sample for patients who’ve had preoperative therapy in between the surgical sample and the metastatic biopsy for patients who developed metastatic disease.

So is this really a true change or known technical problems with the HER2 IHC? We know that the HER2 IHC testing was developed and put into practice really to identify HER2 overexpressed and amplified breast cancer tumors. It really isn’t calibrated and designed to accurately assess lower levels of HER2 expression. And if you look at the HER2 0 and concordance of HER2 0 among pathologists to call HER2 0 it’s really poor, under 20%. So HER2 sample — HER2 0 call in one pathologist’s hand will turn into 1 and vice versa. The same pathologist; you have the same looked at twice you can have a change. So we really need more reproducible and quantitative assays if we’re going to have drugs that are tied to these lower levels of expression of HER2.

In the meantime, we know in DESTINY-Breast04 HER2-low status, which was centrally confirmed, on either primary or metastatic tissue was allowed for eligibility. So in current clinical practice we use the same guidance, HER2-low status, on any sample that’s used for T-DXd recommendations, and at this point I wouldn’t change that current clinical practice.

We saw data from BEGONIA study looking at the combination of durva plus dato-DX. Dato-DX is another TROP2-directed ADC with a TOPO 2 payload.

BEGONIA is a large trial with different combinations and it was Arm 7 that was looking at the combination of durva plus dato-DX in first-line setting for patients with metastatic triple-negative breast cancer. You can see that only about 50% of the patients had had prior cytotoxic chemotherapy for early-stage disease in this study, and again that would be a little bit different than what we would see in our contemporary use population, where most patients with early-stage disease would have received some sort of chemotherapy.

In this trial we saw very impressive objective response rates of over 70% with the combination of dato-DX plus durvalumab. Median PFS was also very robust, and median duration of response was quite durable.

In terms of safety the most common adverse events with this combination were GI. Stomatitis is a known side effect of dato-DX and was the AE that led to most dose reductions. There were 5% adjudicated ILDs with this combination, and the side effects led to dose reduction/delays for both durva and dato-DX in a substantial number of patients, and this is something that is being looked at in the early-stage trials with supportive measures to mitigate some of the toxicities.

So in terms of summary for ADC plus IO combination in triple-negative breast cancer, this combination yielded robust activity in first-line setting. Again to point out that only half of the patients had had prior chemotherapy for early-stage disease and would that have impacted the robust effectiveness we see here? This was a single-arm design so it prohibits assessment of combination versus dato-DX monotherapy efficacy. We know from the TROPION-PanTumor01 study, where patients had prior 3 lines of chemotherapy, that the dato-DX monotherapy objective response rate is about 33% in that patient combination.

Ocular toxicity is unique to dato-DX. In fact, all trials require pretreatment ophthalmologic evaluation for patients before they receive dato-DX. This is something that is going to have clinical implication if the drug gets approved in terms of patients and practices having access to eye doctors to get this exam done before the patient can start on therapy.

Stomatitis has become to be a known side effect of this agent, and in the upcoming trials we’re using prophylactic steroid mouthwash to mitigate this toxicity.

There are several Phase III trials of ADC plus IO in TNBC that are ongoing. TROPION-Breast02 is actually not a combination. It’s dato-DX monotherapy. It’s looking at the efficacy of this ADC in first-line setting in patients who have PD-L1-negative disease or are not a candidate for checkpoint inhibitor therapy. TROPION-Breast03, which is also recruiting, is looking at dato-DX with or without durva versus treatment of physician’s choice in TNBC with residual disease. And TROPION-Breast04 is a neoadjuvant trial introducing dato-DX plus durva in a neoadjuvant setting followed by adjuvant durva and comparing that to KEYNOTE-522 regimen.

ASCENT 04 is combining sacituzumab with pembrolizumab versus treatment of physician’s choice in the first-line setting in patients with metastatic TNBC that is PD-L1 positive. And ASCENT 05 is assessing saci plus pembro in patients with residual disease and comparing that to standard treatment of physician’s choice.

So we await results of all of these trials to inform how the combinations of ADC with IO will fit into our clinical practice both for early-stage and advanced stage triple-negative breast cancer.

HER3-DX, or patritumab deruxtecan, is another new ADC that’s showing good clinical efficacy in Phase I/II trial. This data was published last year in JCO. So HER3-DX is a HER3-directed ADC with a TOPO 2 payload, deruxtecan derivative, and a cleavable linker.

The data that was saw published comes from this Phase I/II study, and it had 3 elements, a dose-escalation, a dose-finding, and dose-expansion cohorts. And all of these cohorts were combined for reporting, and we see efficacy reported by the 3 breast cancer subtypes.

The HER3 in this trial was determined by IHC in archival tumor tissue, and HER3 positivity was defined as either IHC of 2+ or 3+ for the dose-expansion and dose-finding cohorts, and greater than 25% membrane positivity at the 10X for the dose-expansion cohort. And between the HER3 positive they divided the HER3 positives into HER3 high, which was greater than 75% membrane positivity, versus HER3 low, which was over 25% but under 75% membrane positivity.

In terms of toxicity with this drug, GI and hematological toxicity were the most common toxicities. ILD was noted in 7% of cases, and most cases were Grade 1 and 2.

Now looking at efficacy, it demonstrated antitumor efficacy across all 3 breast cancer subtypes, which we can see the objective response rates of 30% in hormone positive, 22% in triple negative, and 43% in HER2 positive. The median duration of response appeared to be durable, and antitumor activity was also demonstrated across the range of HER3 expressions.

The reported safety profile was similar between the 4.8 and 6.4 mg doses that were tested in this trial, and an in-between dose of 5.6 mg is being evaluated in breast cancer in many ongoing trials with this agent.

HS-20089 is another ADC. We saw preliminary data for this ADC at ESMO last year. And this was studied in the advanced solid tumors. So briefly to describe this ADC, B7 homolog 4 protein is a transmembrane protein which is highly expressed in various solid tumors with low expression in normal tissue. This ADC is a B7-H4 targeted ADC, has a TOPO1 inhibitor payload with a drug to antibody ratio of 6 and a protease-cleavable linker.

This trial has various dose escalation cohorts and included various tumor types, enrolled 52 patients. A majority of the patients that had actually enrolled had breast cancer; 48 out of 52 had breast cancer, and 32 out of these 48 had triple-negative breast cancer. The patients had had prior 3.7 lines of therapies for metastatic disease. The MTD was determined at 5.8 mg/kg. Most common Grade 3 adverse events were hematological toxicity. They did note nausea, vomiting, and liver function test increase in about 20% of patients as well. No interstitial pneumonitis or infusion reactions were reported in this initial report.

In terms of efficacy we see an objective response rate in triple-negative breast cancer that is over 20%, and the responses were seen both at the 4.8 and 5.8 mg doses. They did have a small number of patients, 7 patients, who had received prior IO with an objective response rate of 42%. They didn’t quite report objective response rate in patients who had received a prior ADC. I think that’s going to be an important clinical question. PFS and median duration of response was also not noted. The authors reported that the expansion cohort in triple-negative breast cancer is ongoing, so we await that data.

In terms of ADCs in triple-negative breast cancer, kind of a brief summary of present and future, we have TROP2 ADCs. We have TROP2 ADCS with TOPO1 payload, which is sacituzumab, which is approved for metastatic triple-negative breast cancer regardless of the HER2 status. And then 2 other TROP2 ADCs, dato-DX and SKB264 are in clinical trials. We have a HER2 ADC, T-DXd, which again also has a TOPO1 payload, which is approved for HER2-low metastatic triple-negative breast cancer. HER3 ADC and B7-H4 ADC, again, both of which have TOPO1 payloads, are in clinical trials right now, so are bispecific antibodies.

All of these ADCs we see have a TOPO1 payload, and I think in terms of cross resistance we need ADCs that have different payloads. As we start to sequence them the payload might matter. Very little is known about mechanisms of resistance, whether it’s target, payload, both, or some other factor. We need more data on efficacy of ADC after ADC. Is there an optimal sequence? Should the level of target expression guide decisions about how we sequence these drugs? Clearly the current immunohistochemistry-based assays are not optimal to make those decisions, and many of these questions will become even more important as these ADCs move in the curative setting. And if many patients are already exposed to these drugs in curative setting how will we sequence them in metastatic setting and what the efficacy might be of these agents. So a lot of work ahead for the investigators and pharma to answer all of these important questions.

So moving on from novel drugs to an old drug, capecitabine, can we teach this old drug some new tricks in terms of metastatic breast cancer? So capecitabine is a commonly used oral chemotherapy drug for many cancers, including breast cancer. X-7/7 trial compared the standard BSA-based dose of 1250 mg/m2 given 14 days with a 7-day break to a fixed dose, non-weight-based dose of 1500 mg twice a day 7 days on, 7 days off. It enrolled 153 patients with metastatic breast cancer at any line of therapy. The majority of patients that enrolled actually had hormone-positive disease, and more than half were receiving capecitabine as first-line therapy.

What we saw was progression-free survival was similar with fixed dose versus the BSA-based dose in this randomized study, and the overall survival showed similar findings with similar overall survival.

The toxicity was different between the 2 schedules, and that’s expected. The BSA-based dosing had higher Grade 3 or 4 toxicity, more treatment discontinuation because of toxicity, and more dose modifications because of toxicities. So what this data tells us is the fixed-dose regimen has similar efficacy, at least for metastatic hormone-positive breast cancer, with lower toxicity and better tolerance, and provides an alternative schedule for patients with metastatic breast cancer.

A caveat is very few patients in this trial had triple-negative breast cancer or HER2-positive breast cancer, so I’d be a little cautious about applying this data to those patients. And again, patients enrolled in this trial had metastatic breast cancer, so I don’t think we’re ready to apply this data to patients who are receiving adjuvant capecitabine for their triple-negative breast cancer and residual disease. There I would still favor using the BSA-based dosing.

OlympiAD, which led to the approval of olaparib for germline BRCA1 or 2 mutation-associated metastatic breast cancer, published extended follow up for overall survival and safety. As a reminder, this trial demonstrated statistically significant clinically meaningful improvement in progression-free survival with olaparib compared to chemotherapy of — treatment of physician’s choice in patients with BRCA1 or 2 mutation-associated metastatic breast cancer. The median overall survival with olaparib versus TPC was similar and was not statistically superior for olaparib.

With the updated follow up, which is over 2 years longer than the initially reported OS and safety data, the median overall survival continues to be the same with olaparib and chemotherapy treatment of physician’s choice. As we look at the subgroups for overall survival between the hormone-negative population and triple-negative breast cancer, again the overall survival with olaparib versus chemotherapy is the same. But when we look at patients who had received prior chemotherapy for metastatic disease versus who had not olaparib performs much better than treatment of physician’s choice in patients who had not received prior chemotherapy in terms of overall survival.

No new safety signals were reported with this longer follow up. Importantly there were no cases of MDS, AML, or pneumonitis throughout the study.

So these findings from — updated findings from OlympiAD really kind of fall in line with what we have in NCCN Guidelines, which recommend a PARP inhibitor as first-line therapy for patients with germline BRCA1 or 2 mutation-associated, hormone-positive, HER2-negative breast cancer that is endocrine refractory. So once you’re ready to move on beyond all endocrine treatment it recommends a PARP inhibitor over chemotherapy in the presence of germline BRCA1 or 2 mutation in the patient. And for triple-negative breast cancer that is PD-L1 positive clearly the first-line options are chemo plus pembro, but for triple-negative breast cancer that is PD-L1 negative and in the presence of germline BRCA1 or 2 mutation olaparib or platinum is preferred first-line treatment option.