Year in Review: On Demand — Significance and Relevance of Recent Data Sets and Publications in the Management of Breast Cancer
Year in Review: On Demand — Significance and Relevance of Recent Data Sets and Publications
in the Management of Breast Cancer
Featuring key slides from the most important recent breast cancer presentations and publications and a video interview with Dr Hope S Rugo.
3 AMA PRA Category 1 Credits™
To complete the Post-test and Educational Assessment and Credit Form, click here. A statement of CME credit will be issued for scores of 80% or better. Your statement of credit will be mailed to you within 3 weeks or may be printed online.
Efficacy and safety of CDK4/6 inhibitors in premenopausal women with ER-positive, HER2-negative advanced BC DR LOVE: So let’s start out with these 2 papers on CDK inhibitors in premenopausal women. First a paper that you had at ASCO on abema and then the other paper that had been previously presented with ribo. Can you go through those 2 papers? DR RUGO: So cyclin-dependent kinase 4/6 inhibitors now are a huge change in the way we treat hormone receptor-positive breast cancer, but all of the trials included women who were postmenopausal, either because they had an oophorectomy or they met standard criteria for being postmenopausal, and they specifically did not allow patients who were on chemical ovarian suppression. So there were 2 trials that actually included a fair number of patients. The abemaciclib MONARCH trial that allowed patients who were premenopausal and were on ovarian suppression and then MONALEESA-7 that actually conducted in a single — I think the first of its kind — prospective trial testing the addition of the CDK4/6 inhibitor ribociclib in premenopausal women who were on chemical ovarian suppression and then took an aromatase inhibitor. Both trials showed that the impact of the CDK4/6 inhibitor was identical in premenopausal women on ovarian suppression and taking an AI compared to the postmenopausal women who’d been included in the other randomized trials. There’s also a subset analysis of PALOMA-3 that looked at the premenopausal women who were allowed on study who were on ovarian suppression and showed similar results. Very similar hazard ratios, almost a doubling in progression-free survival. And, interestingly, reasonably similar all other outcomes compared to the postmenopausal women. In talking to the sponsors of these various trials over time, when we’ve really wanted to include premenopausal women on ovarian suppression, there’s been concern that the ovarian suppression might not be complete enough. But I think that now investigators and physicians who are using these agents are really monitoring estradiol and making sure that the suppression of the ovaries is adequate for use of aromatase inhibitors, and that’s really the key. Then when you give the hormone drug — whatever it is — and you add in the CDK4/6 inhibitor, as long as the ovaries are suppressed by whatever means you’re suppressing them, you’re going to get a similar improvement in progression-free survival. DR LOVE: So it’s a theme in general with CDK inhibitors in terms of comparative toxicity between trials. But just for the sake of looking at it, when you look at these 2 studies, anything that you could tease out in terms of toxicity profiles indirectly, abema versus ribo? DR RUGO: These trials — I think you really have to take the entire spectrum of trials. Interestingly, again, the primary toxicities of these agents really didn’t differ in terms of pre- and postmenopausal women. But if you take a premenopausal woman and you suppress her ovaries — depending on her age and how much ovarian function she has — she will have, of course, estrogen deprivation toxicities more than a postmenopausal woman. But in terms of the known toxicities, for ribociclib being more neutropenia, Grade III and IV neutropenia without an increase in febrile neutropenia and for abemaciclib more diarrhea, which can be controlled with antidiarrheal agents, dose reduction and dose delay. Very, very similar toxicities seen in the pre- and postmenopausal women. And then for ribociclib, the very small increase in QT prolongation — again, without serious toxicity — was very similar in the- pre and postmenopausal women. DR LOVE: So what about premenopausal women in terms of CDK inhibitors combining it with tamoxifen? And what about with fulvestrant? DR RUGO: Those are very good questions. Now, we have great data with fulvestrant. We have the randomized trials in patients whose cancers progressed on nonsteroidal aromatase inhibitors that led to approval of palbociclib initially and then abemaciclib. And then we have the most recent MONALEESA trial, where fulvestrant was used in both the first-line setting in patients who were faster relapsers and then in the second-line setting as well, after progression. And in all cases, the addition of a CDK4/6 inhibitor to fulvestrant has resulted in similar outcome and no differential toxicity. DR LOVE: But did that include premenopausal women? DR RUGO: Those trials — so there is data only in fulvestrant in the second-line setting in the trials where premenopausal women were allowed on study if they had ovarian suppression. So there is data now, and it actually is a really good point because we didn’t — people have always said, “If you’re using fulvestrant, it has the selective estrogen receptor degradation as opposed to modulation, do you have to be in menopause?” And all of the studies we have with fulvestrant have treated women who are either in natural or chemical or treatment-induced menopause, so no premenopausal women. The tamoxifen data only comes from MONALEESA-7, which is the trial in premenopausal women. And in that trial, women could take tamoxifen and then the CDK4/6 inhibitor, ribociclib, or placebo. But there is concern about an interaction of tamoxifen with ribociclib in terms of increasing the QT effect, a QT prolongation. So it’s not part of the label, and one shouldn’t combine tamoxifen with ribociclib based on MONALEESA-7 because of that concern. DR LOVE: So I don’t know how often this might occur, but if you have a patient who’s premenopausal, young, who gets ovarian suppression and an AI adjuvantly and then recurs while they’re on the AI, how would you think through their next therapy? DR RUGO: I think that we really — in our ASCO guidelines, although now in need of updating from just a few years ago because we have all these CDK4/6 inhibitors, that they reflect the same thinking, which is that in that setting, you’re thinking of the premenopausal woman as being postmenopausal. So I would do the same thing as I would do in a postmenopausal woman, which is continue the ovarian suppression and use fulvestrant and a CDK4/6 inhibitor as first-line therapy. DR LOVE: Just out of curiosity — because I don’t know — is there data with palbo in premenopausal women? DR RUGO: The data with palbo comes from PALOMA-3 in the subset of patients published by Sibylle Loibl looking at patients who were premenopausal on ovarian suppression. Because in the second-line setting, for whatever reason, which is not entirely clear to me, mostly premenopausal women on ovarian suppression are allowed in those trials. Whereas in the first-line studies, they weren’t allowed. DR LOVE: So would you have equipoise about all 3 of the CDK inhibitors in that situation of a premenopausal woman? DR RUGO: Absolutely. And I think that the most important thing is making sure you have adequate ovarian suppression. Clinical implications of the results from the Phase III MONALEESA-3 trial of ribociclib/fulvestrant DR LOVE: So let’s talk about the MONALEESA-3 trial. You referred to it. Maybe you could just briefly talk about what they looked at and what they saw and what you think it means. DR RUGO: It’s actually a really interesting trial design and kind of nice to see something a little bit different in the third trial looking at fulvestrant and a CDK4/6 inhibitor. So not just looking at the patients whose cancers progressed on a nonsteroidal aromatase inhibitor but actually looking at the patients who progressed with a relatively short period of time after their adjuvant nonsteroidal aromatase inhibitor. Those patients also could be treated with fulvestrant in the first-line setting. And patients who really just appeared with metastatic breast cancer, ER-positive. So they actually had 2 groups of patients, about half and half — those who were in the first-line setting in metastatic disease and those in the second-line setting. So patients who relapsed on their adjuvant nonsteroidal aromatase inhibitor were classified as being essentially second line, so they’d already had a nonsteroidal aromatase inhibitor. And, actually, what was really interesting was that the hazard ratios in these 2 populations really reflected the other trials. So you’re really seeing a significant benefit in adding the CDK4/6 inhibitor, whether you’re using fulvestrant in the first-line setting or whether you’re using fulvestrant in the second or greater — really a second-line setting after failure of a nonsteroidal aromatase inhibitor. So I think it’s very encouraging, because it suggests that we can now use fulvestrant in that first-line setting with the CDK4/6 inhibitor. I don’t think we know whether it’s better to use a nonsteroidal aromatase inhibitor or fulvestrant in these patients, and there weren’t enough numbers to be able to separate out patients who’d had a prior endocrine therapy more than a year before or those who hadn’t who were on that first-line fulvestrant arm. And there is actually a trial, which has completed accrual, called the PARSIFAL trial that looked at an AI, nonsteroidal AI, versus fulvestrant with a CDK4/6 inhibitor in the first-line setting. It’s kind of similar to taking the FALCON population, although not the untreated patient population, and then adding a CDK4/6 inhibitor to try and get to the question of whether or not there’s a best first-line hormone approach for metastatic hormone receptor-positive breast cancer. DR LOVE: So a practical clinical question — how do you approach the choice of endocrine therapy — presumably with a CDK inhibitor — first-line endocrine therapy in a patient who had prior adjuvant endocrine therapy and it was stopped? And there — it’s 1 year later, or it’s 5 years. DR RUGO: Right. DR LOVE: Does the amount of time since they’ve stopped matter? How do you select endocrine therapy? Let’s say it’s an AI. DR RUGO: This is all such new data, it’s actually really a nice situation to be able to choose now based on the patient’s individual situation. I think if a patient is within a year, certainly I would use fulvestrant if they’d previously had an exposure to a nonsteroidal aromatase inhibitor. In a patient who’s in that 1- to 2-year range, I think we’re in uncharted territory. It seems as though they might have some degree of greater resistance to a nonsteroidal aromatase inhibitor, and that’s a group of patients where I certainly would consider fulvestrant. Once they get farther out, until we have data to show that fulvestrant is better in that setting, I really have to talk to the patient and individualize their approach because for many of these patients, an all-oral regimen is preferable than getting the injection. And given the fact that we have equipoise right now, I think that we can still go with an all-oral regimen in a patient who’s more than a year, year and a half, 2 years out from their adjuvant nonsteroidal aromatase inhibitor. The other consideration is that there are many clinical trials ongoing now looking at newer targeted agent approaches, and so if you have a trial that’s available in your area, those trials are always including fulvestrant. So if you gave them fulvestrant already they won’t be eligible, and so that’s a little bit of a consideration. Data supporting the ongoing investigation of other novel agents and strategies for patients with ER-positive metastatic disease, including combinations of endocrine therapy (ET), CDK4/6 inhibitors and PI3K or mTOR inhibitors DR LOVE: I was just kind of curious whether you’re participating in any trials of what I call triplet therapy, which would be adding mTOR, CDK and endocrine therapy, and what you think about that strategy and where it might be heading. DR RUGO: The preclinical data is so supportive of that triplet therapy that you could really both avoid and overcome resistance to CDK4/6 inhibitor and to endocrine therapy. And so I’m intrigued by this, and I think it’s a really great clinical idea if we could make it tolerable. It’s actually quite similar to the issues we’ve had with PI3 kinase inhibitors, which are an enormously effective class of agents and where we have been blocked because of toxicity. So you just can’t deliver the PI3 kinase inhibitor when it’s blocking a number of different PI3 kinases and get a benefit, because people stop because of toxicity or dose reduce. So these triplet therapies have also been facing some toxicities. For example, alpelisib, an alpha-specific PI3 kinase inhibitor, was combined with ribociclib, the CDK4/6 inhibitor, and endocrine therapy, but they really couldn’t reach a tolerable combination in the Phase I study, so they didn’t continue to pursue that triplet. The triplet that’s been explored in the TRINITY study is better tolerated, and that’s everolimus at a marked dose reduction and then ribociclib — I think the — yes, the ribociclib and then endocrine therapy. And, actually, I have a patient who’s been on that whose cancer progressed on letrozole and a CDK4/6 inhibitor who’s now been on that for almost 2 years. So clearly this can overcome some degree of resistance, and we’ve found that that triplet combination is pretty well tolerated. That, I think, is a very interesting combination. We’ve also been working with a drug called gedatolisib that’s a PI3 kinase/mTOR inhibitor and combining that with a CDK4/6 inhibitor in endocrine therapy and also have had significant responses in patients whose cancers progressed on CDK4/6 inhibition. Again, I think the issue is figuring out tolerable doses in the right population to make this a reasonable approach moving forward. And I hope that we see more of the TRINITY-like combination as well as gedatolisib in future trials. DR LOVE: So I want to try to get back to the papers, but I have to ask you about something you just said that surprised me, which is, I think you said that there’s a lot of clinical activity that’s been observed with PI3 kinase inhibitors, but it’s toxic. Clinical activity like response, or more laboratory stuff? DR RUGO: No, clinical activity in terms of response. And we saw that in all the smaller trials, these Phase II trials, neoadjuvant metastatic, where patients really responded to a very resistant disease to the combination of a PI3 kinase inhibitor and endocrine therapy but — DR LOVE: No, I’m talk — oh, yes, really? DR RUGO: With buparlisib, there was a lot of clinical activity that led to 2 large Phase III trials, and the drug was just not able to be delivered, and potentially the right population wasn’t selected. There are 2 main issues — so toxicity, which really couldn’t be overcome, and then selecting the one-size-fits-all ER-positive disease population. Because when they did retrospective analyses of subsets, it looked like the PI3 kinase-mutant population was benefiting a lot. But because of toxicity, you really couldn’t pursue further development of buparlisib. It has a long half-life and a lot of various toxicities being a pan-PI3 kinase inhibitor. Pictilisib, another pan-PI3 kinase inhibitor, didn’t have as good efficacy. But taselisib, which was most recently studied and presented in the SANDPIPER trial, this was a very active drug in Phase II trials and then was not so active even in the PI3 kinase-mutant population, which was their primary population they studied in the SANDPIPER trial. But if you looked at the number of patients who dose reduced and discontinued, I mean, a lot of the patients never even really got much drug. They stopped. They got a lot of colitis. Colitis in these patients, second-/greater line, just not tolerable. And if you are going to manage those toxicities, you need to have an up-front management strategy. If you send the patient home with the drug, they get bad colitis, they’re not going to take the drug anymore or you’re going to hold and dose reduce, and then you really impact the efficacy of the drug. Our clinical experience using alpelisib, the alpha-specific PI3 kinase inhibitor that we’re now waiting for results from the SOLAR-1 Phase III trial, is patients who progressed after 18 months on everolimus in Phase Ib trials and then stayed on that drug for 4 years with a CR in liver. DR LOVE: Wow. DR RUGO: Visceral CR. I still have from our Phase Ib trial, which closed ages ago, I still have 3 patients on study who are more than 4 years on treatment. DR LOVE: Wow. DR RUGO: So these are very effective drugs. We’re really needing to have a better up-front strategy for managing toxicity from these agents. And, of course, maybe with new drug design we’ll find better agents. Updated PFS analysis of the Phase III PALOMA-2 study of first-line palbociclib/letrozole in patients with ER-positive, HER2-negative advanced BC DR LOVE: Let’s talk about a few more endocrine papers. First, your presentation at San Antonio, looking at data with palbo, can you talk about that? DR RUGO: We looked at the longer-term outcome with palbociclib, so just — the primary endpoint, of course, of PFS had already been met. But what we found in PALOMA-2 in the first-line setting, that the PFS in the treatment arm was out to 27 months in long-term follow-up now, and that is the final endpoint. Similar hazard ratio, but I think 27 months is really very impressive — the longest PFS we’ve ever seen in this population. So I think that was encouraging. But we also looked at a subset analysis trying to really evaluate all the clinically relevant subsets and basically saw that all patients benefited, and the hazard ratios were quite similar across the different prognostic groups, whether or not you’d had long or short disease-free interval. And we looked at it in — treatment-free interval really was what we were looking at, which I think is, in many ways, more important to us clinically rather than disease-free interval. And we looked at all sorts of different timeframes, different durations, and there really wasn’t any difference. All of these populations seemed to benefit. And it’s actually interesting, because it was brought up in a MONARCH nonfinal analysis — so it was a very early analysis where they hadn’t reached the PFS, the median PFS in many of the groups — that maybe there were groups who didn’t need a CDK4/6 inhibitor. But, actually, if you look at the patients who have luminal A and B disease, they get the biggest benefit from these drugs. Mechanisms of resistance to CDK4/6 inhibitors DR LOVE: So I want to ask you about the paper by Turner et al that was just presented at ASCO looking at the issue of resistance to CDK4/6 inhibitors. There’s been a lot of work done. I’m curious what you think about this paper and the whole issue of what we know about resistance. What do we know about, first of all, how CDK inhibitors actually work, and then why you see resistance. DR RUGO: I have to say, this whole area of resistance is just so cool right now that we’re able to make more progress now than ever before, so that has changed a little bit. But CDK4/6 inhibitors, the mechanism of action is very, very interesting. The whole idea that, in some ways, estrogen signaling is dependent on this proliferation pathway and that by reducing the proliferation pathway that you can, therefore, restore some sensitivity to endocrine signaling and then block it with the endocrine therapy. So it’s actually quite intriguing. And some of this ability of CDK4/6 inhibitors to work is dependent on having an intact tumor suppressor gene that you — basically Rb, retinoblastoma, that you release to allow it to suppress cell cycling. And so if Rb is mutated and can’t be restored to its functional state — like the phosphorylation can’t be reversed or blocked — then it may be that you have up-front resistance to CDK4/6 inhibition. So one of the reasons why it can work so well in hormone receptor-positive breast cancer is that Rb mutations or loss are quite rare up front. But we’re now finding out with resistance that you can acquire this, but it’s still a fairly minor pathway for resistance to CDK4/6 inhibitors. But there are many other pathways that aren’t so minor. And so the ability to learn more — I think back to when we were looking at whether or not we could predict which patients benefited from everolimus in a paper published by Gabe Hortobagyi. But I was involved in BOLERO-2 in that analysis. And, of course, looking at a tumor in patients enrolled in a registrational-style Phase III trial is limited by the fact that a lot of the tumor’s archival. And at the time, I remember bringing up in BOLERO-2, maybe we’re not getting to this because the tumor’s archival, and they said, “Nah, we had the same numbers, so archival versus new, so it won’t make a difference.” Now it turns out it makes a huge difference. The ability to look at cell-free DNA in blood, and Nick Turner’s really done a beautiful job with this, also in explaining it in a way that clinicians can really understand is, I think, changing the whole way we’re thinking about breast cancer, and it fits perfectly with what you see in the clinic. So you watch that patient in the clinic, and what you see is that under the pressure of treatment, cancer grows in different ways, right? So you suppressed the bone disease, but the liver is growing, and you biopsy it and the ER is no longer present, right? This is a mechanism of treatment, in some ways, treatment-induced resistance. And we learned this, actually, in the CDK4/6 inhibitor trials because of different eligibility requirements — that treatment, for example, with chemotherapy will induce some resistance to endocrine therapy. That, in other words, any treatment induces resistance. So what we’ve learned about that is — and that doesn’t mean you shouldn’t treat patients, obviously, but some cancers are going to be more susceptible to induction of resistance faster than others, and that’s what creates all the difference in the heterogeneity in these cancers who are responsive up front. So what we saw is that PI3 kinase mutations are, of course, acquired over time and that they tend to stay. So once you have a PI3 kinase mutation, you get a PI3 kinase mutation. And that even under fulvestrant treatment, you can gain mutations in ESR1, which is fascinating, because we’ve thought of that as something that’s acquired under treatment with nonsteroidal aromatase inhibitors first-line setting, any aromatase inhibitor, and that that may predict relative benefit to fulvestrant versus, for example, exemestane. And there’s also very elegant work that was done in blood testing in patients who were on the trials that were randomized to exemestane versus fulvestrant. But now we’re finding that even with fulvestrant, you can induce ESR1 mutations. But the fascinating thing about those mutations — you lose them. So then you give a treatment, and over time you might lose that ESR1 mutation and have some restored sensitivity to endocrine therapy of the same sort — an AI over time — so very interesting. And then there are other mutations that are induced over time like mutations in the fibroblast growth factor receptor. Very nice work by Carlos Arteaga has shown that that’s part of the mechanism of resistance to CDK4/6 inhibitors, changes in cyclin E or mutations in CDK2. And so now there’s agents that are blocking CDK2, 4 and 6 that are being studied in Phase I trials to try and get over that. And also more specific inhibitors of FGFR are being studied to try and see if we can overcome some of that degree of resistance and, of course, the PI3 kinase inhibitor combinations. DR LOVE: So I can just imagine our general medical oncology audience listening to you and thinking, “I wonder if breast cancer is going to become like lung cancer where now we’re doing liquid biopsies or repeat biopsies in people, let’s say, with EGFR tumor mutations with ALK and almost getting to a point clinically where you’re starting to maybe even be able to think about interventions.” How close are we to that with breast cancer? DR RUGO: I think that we’re getting closer to it asymptotically. It’s like a slow progress. But mainly because I think that people are still very interested in whether or not cell-free DNA is going to be good enough. I think it will be, because people will continue to improve the testing and it will be what we can do. So maybe you’ll miss some mutations, but you’re going to need a test, and then we’re going to have to be able to look at the percentages of cells that are mutated to try and understand responses over time. So yes, I think we are heading in that direction where we’re going to be able to look at blood over time. And now there are also some studies where you’re able to give a treatment to a patient, and maybe we’re going to be able to assess some of the efficacy by looking at the target going away in blood, which is another fascinating area. Plus, there’s great interest in looking at minimal residual disease in the adjuvant setting with cell-free DNA. So it’s kind of a field that is slowly exploding and I think is a hugely important thing as we poke these patients over time. Doing liver biopsies and lung biopsies put them at risk. But I’ve been struck by how heterogeneous disease is. I biopsy the liver, it’s ER-negative, but the bone is growing on whatever therapy I’m giving — that’s not endocrine therapy — because the bone is still very ER-positive. These are heterogeneous mutations over time in cancer, and cell-free DNA may give us a way of trying to look at this all together. DR LOVE: What happens to estrogen receptor levels on a patient on fulvestrant? DR RUGO: You downregulate ER on fulvestrant, but you can’t measure that easily unless you did biopsies and looked at time — DR LOVE: But I mean if you do a biopsy — does it, actually, on a biopsy, you do an ER level, does it drop? DR RUGO: You’d have to look at that by immunohistochemistry. And I think some of the very early studies, although these are much older studies, showed that you could downregulate the estrogen receptor on a biopsy by giving enough fulvestrant. But that when you stopped the drug, ER comes right back again. Efficacy and tolerability of mTOR inhibitors in combination with ET DR LOVE: So let’s talk a little bit about mTOR inhibition. But before you get into that, just kind of curious, is it your take in metastatic disease that by adding everolimus to a hormonal therapy you get a similar benefit to CDK but it’s more toxic? DR RUGO: That is, generally, I think, what we’re seeing if you look at hazard ratios across trials and similar trial populations. It’s not perfect. Everything is a little bit different in trial populations across trials. They’re not meant to be the same. There’s a lot of heterogeneity. But I have to say, the hazard ratios and the PFS is pretty similar. So I think that we are benefiting patients, and the noncross resistance, of course, makes these appealing in sequence, but there is clearly less toxicity of the CDK4/6 inhibitors. DR LOVE: So let’s talk about these papers. Any comments, everolimus plus fulvestrant and also this BOLERO-6 trial? DR RUGO: I think that both trials are really important to add to our body of data about everolimus. And, interestingly, I think that we had thought that the PI3 kinase inhibitors would rapidly supplant everolimus, but 2 things happened. One is the PI3 kinase inhibitors were not tolerable, and the second is that we made everolimus a lot more tolerable by using this preventive steroid mouthwash so that we could eliminate Grade III stomatitis and markedly reduce Grade II stomatitis, and that’s become a standard of care and really made it much more tolerable. And to the point, when you’re looking at toxicity from these 2 different trials with fulvestrant and then BOLERO-6, it’s important to keep in mind that steroid prophylaxis was not used. So there was more toxicity than we see now with the combinations. But fulvestrant is clearly a reasonable partner for everolimus, and that trial showed that again, the hazard ratios are very similar, it’s just more toxicity than we see now. And I actually did a Podcast on that, because it’s nice to have another option for an endocrine agent to combine with everolimus and especially as we learn more and more about the usefulness and efficacy of fulvestrant in these patients. And then I thought that BOLERO-6 rapidly published, really impressive. It was quite interesting, because it was a small study trying to answer the first question, and it did answer that question. So everolimus/exemestane versus everolimus alone — did you really need the hormone agent? And it looks as though their primary endpoint demonstrated that the combination was superior to everolimus alone. There is magic, indeed, in combination. We’ve seen it with the CDK4/6 inhibitors too, and that’s how these agents should be used. In combination is the preferable way to use these targeted agents. The secondary endpoint, which was comparing everolimus and exemestane to capecitabine, was more problematic. Because although it looked like capecitabine was better, the problem is that the arms were imbalanced, and they favored a better prognostic group in the capecitabine arm. So we didn’t really answer that question. It was just numerically better in capecitabine but, unfortunately, biased and needs to be interpreted with great caution. I think that we have benefits and risks to both types of treatment. And indeed, the trial, BOLERO-6, did not use the steroid prophylactic mouthwash, and so there was a little more toxicity from the combination than capecitabine. And it’s the immediate toxicity, and you get the stomatitis in the first 8 weeks, so you’re much more likely to pick that up. But we don’t see that now with the steroid mouthwash. Whereas capecitabine over time, the hand-foot syndrome can be quite problematic for active people because of just limiting activity. DR LOVE: Getting back to everolimus with hormonal therapy, nowadays using, as you mentioned, the mouthwash — particularly people like yourself who are so experienced with it — what do you see quality of life-wise when you use everolimus compared with a CDK inhibitor globally? DR RUGO: It’s, of course, much easier to use everolimus now than in the past. And I do tend to, if I’m worried about somebody, start at a lower dose and go up if they tolerate it well. But I still see more issues that are potential depending on the patient. Some people do great, no problems. But you do have to watch for hyperglycemia, and the fatigue may be a little bit more. It’s always a bit of a bias because we tend to use it in a later-line setting where patients are already sicker and have had more treatment. And patients who are using CDK4/6 inhibitors in later-line setting also get a lot of fatigue. But I just get the sense that there’s a little bit more in just toxicity management. And, of course, you have to watch out for interstitial pneumonitis, although I have to say I really haven’t seen much of it in the last bit of time. It may be that as we’re more experienced, and I don’t know, maybe those patients fall off, because we did see it when we first started using everolimus a lot more than I’m seeing it now. But I am very cautious in patients who have poorly controlled hyperglycemia. Those are not good candidates for use of everolimus, whereas CDK4/6 inhibitors aren’t going to make any impact in those patients. Phase II BOLERO-4 trial of everolimus in combination with ET for postmenopausal women with advanced BC DR LOVE: So any comments on the BOLERO-4 study? Basically looked at everolimus with endocrine therapy in the first-line setting. I’m not really sure what a clinical situation is where you would do that, but what do you think about this? DR RUGO: We may end up with one someday with CDK4/6 inhibitors being studied in close to — I don’t know, I think I figured out it’s around 13,000 women in the adjuvant setting with the 4 different trials that will eventually be done using CDK4/6 inhibitors in patients with early-stage breast cancer. Maybe we will have a setting like that. I think what was most striking to me about BOLERO-4 — I mean, it’s a small trial, so it doesn’t compare, power wise, with the registrational CDK4/6 inhibitor studies, but the hazard ratios are so similar. It’s really interesting — and it gets to your question earlier — do these agents all help endocrine therapy to work better? But really what we’re talking about is questions in tolerability and ease of use, and I think that that’s really what the question is. We’re not smart enough, and we haven’t done the right studies yet to be able to figure out if there’s one group of patients who benefits more from one type of therapy or another, one target versus another. And I doubt that we ever really will figure that out, in fact. But it’s comforting, I think, to know that you can combine these agents with hormone therapy — depending on the individual patient situation — and expect to see benefit. The problem with BOLERO-4 is, those patients hadn’t had a CDK4/6 inhibitor, and we just don’t have a lot of data in that group of patients yet from any study. My experience clinically is that patients respond well. And in an individual patient — because the resistance pathways may be different — you can actually get a better response to the mTOR inhibitor than a CDK4/6 inhibitor in a patient who has up-front resistance. And I’ve seen that in clinical practice. So I don’t think you should throw out the baby with the bathwater. You want to keep in mind that a patient who has ER-positive disease could actually take a break from chemo and go on everolimus and an endocrine agent. DR LOVE: I always got the feeling that people underestimated the efficacy of everolimus because they were so focused on the toxicity. But you look at those curves and hazard rate and look pretty impressive. DR RUGO: Yes, I think they are. And it really has to do with better management of toxicity and education. And that, I think it’s something we could get over with everolimus, and now we just have to learn to get over it with the PI3 kinase inhibitors. Role of antiandrogens in ER-positive BC DR LOVE: So 1 final endocrine question. I’m curious about a presentation at San Antonio in December by Denise Yardley on enzalutamide, and this was enzalutamide and exemestane. We’ve kind of been hearing that story creeping along the last few years. Haven’t heard a whole lot of excitement about it recently. What do you think about those data? What do you think about antiandrogens in breast cancer? DR RUGO: It’s an interesting area that’s I think generated a lot of excitement in triple-negative breast cancer and then kind of got muted because it’s such a tiny population of patients who benefit and very hard to identify. But in ER-positive disease, it’s interesting because it’s an old treatment, right? These drugs are just more specific and have less off-target toxicity, like enzalutamide. But we were giving fluoxymesterone, for example, 2 decades ago, but people responded to it. DR LOVE: Yes, but fluoxymesterone is an androgen, not an antiandrogen. DR RUGO: But, I mean, this is the whole idea: It’s like giving estrogen versus giving an NSAID or tamoxifen, that you’re basically targeting that pathway by that agent. DR LOVE: Okay. DR RUGO: And, obviously fluoxymesterone — because it was an androgen — had unacceptable toxicity, and then we got better drugs that didn’t, so we abandoned fluoxymesterone. But we all had patients who responded to it, those of us who are old enough to have used it. So we knew that pathway was something that we could target, and that hasn’t changed. It is a pathway we can target. But even in this enzalutamide study, where they had patients in the first- and second-line setting, they really saw a very modest differential in terms of efficacy, and they were looking at progression-free survival. I mean, we’re just — compared to our targeted agents, both in efficacy and toxicity, it’s just not something that’s going to move forward. And if you look at patients who have very resistant ER-positive disease, it doesn’t really have enough efficacy to make it worthwhile to pursue. So I think that although this is a drug which has some degree of efficacy, it can’t compete against where we are right now in the treatment of hormone receptor-positive disease. TAILORx: A Phase III study of chemoendocrine therapy versus ET alone for hormone receptor-positive, node-negative BC and an intermediate 21-gene Recurrence Score® DR LOVE: So let’s move on now and talk a little bit about genomic predictors of benefit of therapy and, of course, particularly the TAILORx study that was finally presented at ASCO and then published. What an exciting moment. Can you — I mean, I don’t know if we can do it briefly or not — provide an overview of the study? What they found and what you think it means. DR RUGO: TAILORx, I think, these genomic tests now have completely changed the way we think of treating patients with ER-positive early-stage breast cancer. TAILORx, of course, focused on patients with node-negative/ER-positive breast cancer, obtained a Recurrence Score and then basically put patients into 3 baskets with a revised scoring system. So a score less than 11 was low risk treated with endocrine therapy alone. A score of 20 — greater than 25 was high risk — chemotherapy, and then this group of 11 to 25 was intermediate risk. And expanding the intermediate-risk group so we could really look at a group of patients where we have uncertainty about the benefit of adding chemo to endocrine therapy and ask the question of whether or not chemotherapy benefits those patients. Really important question. Patients were randomized, of course, to receive endocrine therapy or endocrine therapy with chemotherapy of choice. And it’s a really big study, huge number of patients were screened in this trial. We already know that patients who have a recurrence score less than 11 — ER-positive disease, node-negative breast cancer — had an excellent outcome with hormone therapy alone. And we learned a number of different things from this intermediate-risk group presentation. One thing, I think, very interesting, is that patients with a score higher than 25, who all got chemotherapy, had a worse outcome even though they got chemotherapy. So we know they need chemotherapy, but they clearly have bad disease. And we need to be really thinking about what the right treatment for those patients is. And maybe those are a group of patients who will benefit from add-ons to endocrine therapy as well. But clearly, this predicts a worse outcome. DR LOVE: Before you go on, can I just clarify, though, this issue of 25 to 31 and greater than 31, is 25 to 31 the same thing as greater than 31 in your own mind from the point of view of risk, or is 25 to 31 a little bit farther down the line? DR RUGO: We just don’t have the — I don’t think we have the focus to really be able to answer that question. The ability to look at it wasn’t part of the prospective look, and they grouped all the patients together. And patients who have a higher score do have a tendency to have more disease and more aggressive disease if you look at clinical pathologic characteristics. So I think that we’re not in a situation where we can really separate out groups to that degree. Very difficult to do. So I think we really have to look at greater than 25 as a risk group. The original studies that looked at large numbers of patients enrolled in trials looked at the score that was over 31, right, so 31 and over. And now we’re looking at over 25. Can you parse out that group between 26 and 31 and say something about them? And I don’t think we can, as yet. We really don’t know. But we do know that a higher score correlates with a worse outcome. And there’s some continuity, it’s just that if you want to take the patient and all of the clinical pathologic risks and the score and try and divide it all up into its minutiae, it’s actually very, very difficult even with those large numbers. DR LOVE: So let’s focus on the real key issue, which is the 11 to 25 population, and what they actually saw there. And particularly this issue of the analysis they did of premenopausal versus postmenopausal women. And incidentally, was that ahead of time that they determined they were going to look at that, or a post hoc analysis? DR RUGO: There was a prospective plan to look at women who were younger versus older and to try and understand whether or not there was a difference. And there also was a prospective plan to look at differences within the intermediate-risk group. So that was a planned analysis, although still it’s a subset analysis. And it’s really interesting. So in the group of patients who were randomized in the intermediate-risk group, the median tumor size was small, 1.5 centimeters, and there were very few patients who had tumors that were greater than 2 centimeters. So it’s important to keep in mind that this is in a group of patients who have a 7-cm high-grade tumor with lymphovascular invasion and are 29 years old. I mean, we just don’t have enough information on that group of patients. Node-negative, relatively smaller tumors. And what they saw was there was absolutely no benefit in chemotherapy in this group of patients, of course, really leading to all these headlines — most patients with breast cancer don’t need chemotherapy — which, of course, was a gross misrepresentation of the data. But it’s true that there are more women who don’t need chemotherapy following TAILORx than there were before, and this is very, very helpful information. In looking at the subset analyses, it kind of goes along with what we understand already: There’s heterogeneity in breast cancer. Women who are younger, women who are under the age of 50 who had scores particularly between 21 and 25 seemed to obtain some benefit in terms of distant recurrence in getting chemotherapy versus not. And so I think in that group of patients the lesson to me is that I need to be looking at those patients based on their clinical pathologic characteristics and adding that into the mix when I’m making decisions about recommending use of chemotherapy or not. But by far the majority of women who have an intermediate score as defined by TAILORx won’t need chemotherapy if they have cancers that fall into that smaller node-negative category. DR LOVE: So the obvious thought is that maybe the chemo was affecting the ovaries. I think most of these women did not have ovarian ablation. Is that correct? DR RUGO: Right. TAILORx really predated any results from SOFT and TEXT, and most women did not. And I think there is a question about whether or not the chemotherapy was beneficial due solely to ovarian suppression. And I wish we could get at that answer. And it may be that future analyses over time will give us some more clarity, but it’s pretty tough to know. If you look at SOFT and TEXT, the women who benefited the most from ovarian suppression were the women who got chemotherapy because they had higher-risk disease. So we can’t even really separate — those patients in SOFT had to recover ovarian function, but they still benefited more from ovarian suppression. So I think if you look at that trial population and you think about the fact that most of those women got chemo too, it suggests that there may be some benefit from chemotherapy in these patients. DR LOVE: So just out of curiosity, in your own practice outside a trial setting, what fraction of patients who are premenopausal who are ER-positive are getting as part of their adjuvant therapy ovarian suppression or ablation? Do you do this in most patients? Do you do it in high risk? When are you doing it? DR RUGO: So I’ve used ovarian ablation — more in suppression, really, and not so much ablating the ovaries — for a long time and predating the results of TEXT and SOFT because I’m interpreting the data that we had so far. I had believed that patients who particularly are young and have highly functional ovaries, lots of estrogen around, that doesn’t make any sense to leave all that estrogen around. And data from preclinical and other subset analyses suggested benefit. Now, with SOFT and TEXT, I think, particularly in women who are younger — in their thirties or twenties — who have high-risk disease, I routinely use ovarian suppression. I think the combination with aromatase inhibitors can be quite challenging, and so I’m happy to use ovarian suppression and tamoxifen in women who have too much toxicity to really be able to maintain a quality of life with aromatase inhibitors. And I just try and get that aromatase inhibitor in for some period of time. But adherence is so much more important than sticking with a particular regimen that no one’s going to take because they’re so miserable and destroying their life, so we have to balance this. Once women get into their forties and have lower-risk disease, I’m less excited about using ovarian suppression in everybody. But certainly any premenopausal woman with very high-risk disease, residual disease after neoadjuvant chemotherapy, node-positive breast cancer, I use ovarian suppression routinely. Analysis of long-term risk of BC recurrence after discontinuing adjuvant ET in patients with ER-positive early BC DR LOVE: So let’s continue on then, and there are a couple of other papers related to genomic factors. First, long-term risk of recurrence and extended adjuvant therapy. There was a paper in the New England Journal by the Oxford group from the Early Breast Cancer Trialists, 20-year risk of breast cancer recurrence. And also a paper that came out of the ABCSG-16 trial that was presented at San Antonio looking at an additional 2 versus 5 years of anastrozole after 5 years of therapy. So both of these papers really relate to long-term endocrine therapy, long-term recurrence. Can you talk about what they looked at and what you think it means? DR RUGO: Yes, I mean, I think that the Pan paper, the Early Breast Cancer Trialists’ Group paper, is incredibly important. And I think it really validated a lot of our understanding already about the importance of looking at 5-plus-year recurrence. These were patients who had not recurred at 5 years. So you’ve already excluded the patients who have early recurrence. And they were supposed to have all taken 5 years of, generally, tamoxifen, some aromatase inhibitor, but a lot of tamoxifen. And at 20 years, there was this marked additional risk of recurrence that was very much dependent on the amount of cancer you had at diagnosis and the clinical pathologic characteristics of that cancer. So it was impacted by the amount of disease you had, the size of the tumor, the nodes and Ki-67, and the risk ranged from 14% to the high forties, close to 50% at 20 years. And that’s year 5 to 20, so you’re excluding all the recurrences that occurred in the first 5 years. Now, of course, this is a historical data set. So maybe we’re better at a number of different things now. We treat for HER2, but we tend not to think about even HER2-positive recurrences occurring way after 10 years. The ER-positive/HER2-positive — there are recurrences that occur between year 5 and 10. But it’s hard to know how much impact better assessment of receptors, assessment of HER2 and treatment with HER2-targeted agents impact that later recurrence. We do believe that use of aromatase inhibitors impacts that later recurrence. But clearly we need to understand this better and to try and better tackle dormancy and late recurrence risk, and there are big task force groups working on this now from the laboratory and translational science side. Of course, one area we’ve thought about for a longer time is just giving more endocrine therapy, and that is based on the fact that we have seen a carryover effect that was longer when you gave 5 versus 2 years of tamoxifen. And at 15 years, you saw less cancer recurrence if you got 5 years of tamoxifen versus 2 years — and less cardiovascular disease, which was interesting. So the idea was that maybe if you just gave more therapy. So we knew that giving an AI for 5 years after 5 years of tamoxifen had a big impact. And, interestingly, in MA.17, that impact was greater if you were — at least it was suggested in an unplanned subset analysis, if you were premenopausal and if you had positive nodes, you seemed to have more benefit from extending the endocrine therapy with an AI after tamoxifen. So then, of course, the next idea was, maybe you could just continue tamoxifen for 10 years. And that seemed to be better in 2 different large randomized trials, but lots of problems. ER-negative patients enrolled, people didn’t take their study therapy, so they excluded a bunch of people. And then there was more endometrial cancer and thromboembolic disease, so that wasn’t so great, but still it was better. So then maybe it just made sense to continue the AI. But from the MA.17R trial, it looked like all you were doing was preventing new cancers, and why would you take a drug that made you feel old and stiff and gave you vaginal dryness and sexual dysfunction if that was what your goal was? So these trials that looked at 5 versus 10 years of an AI or 7 versus 10 years, which was the Austrian trial, were really important. And, disappointingly and somewhat surprisingly, they didn’t show any benefit in longer-duration endocrine therapy. And I think the Austrian study is helpful for us psychologically, because they looked at 7 years. So now we could tell people, “You know, you have a high risk, you can continue to 7 years and then stop.” If you continued from 7 to 10 years, you didn’t improve your disease-free survival, but you increased the rate of fractures, so that wasn’t so hot. I think that we’re going to need genomic tests in order to decide who needs and who will benefit from longer-duration therapy. And that work, hopefully, is ongoing based on tissues collected in all of these trials, including the NSABP trial. DR LOVE: So I just had a thought. What would your best guess be on when we’re going to see the first adjuvant CDK data? How long you think it’ll be? DR RUGO: I don’t know. The PALLAS trial, which started enrollment first and has already enrolled about 4,500 patients, increased their enrollment to 5,600 patients recently because they had included Stage IIa. And when you go out and predict when you’re going to see a difference, it’s going to take until the next 2 generations or something, I mean, at least the next generation. And so the idea then was in MONARCH-E, looking at abemaciclib, was to enroll a very, very high-risk group of patients, but then it takes a little longer to accrue those patients, although that trial is open. And then the ribociclib trials are just opening, planning their first patients enrolled by the end of the year. So I think the NATALEE trial, looking at a higher-risk population as well. We have a fourth trial, which is PENELOPE-B, joined as a collaboration between the German breast group and the NSABP. And that trial, we enrolled patients who had residual disease and additional high-risk features after neoadjuvant therapy, but they only got a year of CDK4/6 inhibitor treatment. PALLAS and MONARCH trial are 2 years, and then NATALEE is going to be 3 years. So it’s going to be a long time. We figure that PALLAS, if you can look at the high-risk group, they’re going to have been accrued more recently, so — because IIa closed a while ago, Stage IIa. So if you figure you’re going to need 5 to 8 years of median follow-up, it depends, to some degree, on the power of improvement in disease-free survival, obviously. If we see a huge impact, we’re going to see it maybe in the next 3 to 5 years. If the impact is smaller, it’s going to be longer. DR LOVE: So when you were talking about these papers on delayed recurrence, I had a flash of what’s going to happen when those data start coming out. Because I’m going to go on record here and predict that there’s going to be — I’m going to say 0.5 hazard rate for disease-free survival. I’m always optimistic. DR RUGO: Cup half full. DR LOVE: I’m going to say 0.75 for survival and not statistically significant, but whatever. DR RUGO: That’s still huge. DR LOVE: But the real thing I was thinking about is if — I was thinking about the adjuvant trastuzumab studies where when they came out, there was this scramble and even trials looking at delayed. So if those trials are really positive, are you going to start giving it to all these people who never got it 5 years down the line because, as you just said, I mean, we don’t know that it’s going to prevent delayed recurrence, but clinically I wonder if that’s going to happen. DR RUGO: Yes, I wonder, too. I think there are people, particularly who have Stage III disease and multiple positive nodes, potentially people will use drug, but there will be no data to support it. I think that what’s going to be interesting is people — the first data to come out is going to be using 2 years of a CDK4/6 inhibitor versus not. We know that compliance will be very variable, right, so at least a year in most patients, right? There’s going to be problems with adherence after 1 year that are going to affect the number of patients who go out to 2 years. So then you think okay, you’re a patient who had high-risk disease, and you’re now 3 years out, and this data comes out, are you going to take the drug? And the likelihood is that that will have more uptake. So if you have somebody who is within their 5 years and has high risk, they’re going to be more likely to take a CDK4/6 inhibitor than somebody who’s way far out. SOFT and TEXT trials: Importance of tailoring adjuvant ET for premenopausal women with BC DR LOVE: So you mentioned already the SOFT and TEXT trials. I’m going to ask you about the New England Journal paper that came out recently with more data, and there was an editorial there and some other data presented. Can you talk about — and you’ve already referred to it, but what the bottom line is about what we’ve learned from these data sets. DR RUGO: SOFT and TEXT are incredibly important trials, and I think that not only were they important trials, but they also were able to collaborate, play in the sandbox together pretty well, which is really, really good. And, actually, I think that ability to work together collaboratively has had the biggest impact on young women’s lives, because we’d probably still be waiting for some of the data now. So I think what’s been nice is that over time, we’ve continued to see a significant benefit particularly in the young women and in women with higher-risk disease for whom the decision was made to give chemotherapy where ovarian suppression is better than not giving ovarian suppression. And ovarian suppression and an aromatase inhibitor is better than use of tamoxifen. I think that what we also have come to understand is that toxicity from treating a very young woman with ovarian suppression and an AI up front can be quite significant. And that in many situations I think we are starting with ovarian suppression and tamoxifen and then switching over to an AI at some point, where we can allow a patient to tolerate the treatment better. Certainly in the extremely high-risk populations, and in a woman who’s already had ovarian suppression from chemotherapy, the tolerance is generally a little bit better for using an AI up front because they’ve already had, like, 6 months or longer to adjust to ovarian suppression because of chemotherapy that they have received. I think 1 question that we still have, which is incredibly important, and one of the ABCSG trials tried to look at this question but didn’t randomize is whether or not there’s a patient population who you would have given chemo to but you could treat now with ovarian suppression and an AI. That question wasn’t asked because, basically, you spoke with your feet, as a clinician — you took the high-risk patients, you gave chemo, and the lower-risk patients didn’t receive chemo. The older patients, patients who were older and premenopausal, were less likely to get chemo. And then the patients who benefited were the patients who had higher-risk disease who got chemo primarily. But we just don’t know how much benefit the chemotherapy had in that group of patients and whether or not you could find a group who could just get hormone therapy. And I think that that’s where the genomic tests really come in to try and help us make those decisions. Clinical practice guidelines on the management of early BC DR LOVE: Let’s talk about a couple of practice guidelines papers that came out. One was the ASCO guideline paper on early breast cancer, and the other was what happens every year or two, the St Gallen meeting that also had a statement that came out. It was interesting, I actually referenced the St Gallen paper in a lymphoma conference recently because Hodgkin’s, there’s this big thing about deescalating treatment, giving less therapy in follicular lymphoma, et cetera. And the title of the St Gallen paper is “De-escalating and Escalating Treatments.” But anyhow, both papers have a lot of stuff in them. But if you could pull out 1 or 2 things from each paper that came out that you want to comment on. DR RUGO: I think the St Gallen paper is interesting. And looking at that paper, I reflected back to say 10 years ago where people were saying — maybe 8 years ago — “Does anybody in the US really look at the St Gallen guidelines, and does it matter, et cetera?” And the answer then was, “Oh, we don’t really use it.” But now, of course, completely different. And I think we’re much more of an international cooperative and collegial group in terms of our approach to breast cancer. And the St Gallen guidelines as well as the Advanced Breast Cancer Consensus really have had an impact I think internationally on treatment. This most recent paper is really great, and it reflects our current thinking, which is that we can use both genomic testing and maybe response to neoadjuvant therapy as a way to both deescalate and escalate therapy. And then the paper really talks about the fact that you don’t have to give — there’s not a one-size-fits-all, and really we need to be looking at the biologic characteristics of a tumor in order to better understand who needs more versus less therapy. And then we need to study that prospectively, and those studies are being planned or are ongoing now. So I think that that’s really the take-home message from the St Gallen paper. It’s worthwhile reading it, because I think it gives some interesting thoughts on this idea of using biology to try and determine treatment for an individual patient and using the available data we have on less aggressive versus more aggressive treatment regimens. And I will say that even in looking at some of the genomic tests and trying to find out whether or not we should add paclitaxel to an anthracycline-based regimen, we’ve never seen that they tell us which chemo to use or what sequence. And so we’re really having to use the biologic characteristics more to make those decisions in terms of risk of recurrence and response rather than the genomic test to think about specific differences in chemo. But we can decide about chemo or not. And maybe in some of the studies that are going on with HER2-positive disease to evaluate extent of treatment. The ASCO guidelines update is interesting because it talks about a few different things in terms of which patients should receive different therapies, specifically who you should consider for neratinib, who you should consider for a year of pertuzumab and then endocrine therapy and duration, et cetera. But mainly it’s talking about chemotherapy and what kind of chemotherapy regimens to use and how to take our newest data and apply it to clinical practice. And I think that all that kind of a guideline approach is limited by the data we have in the clinical trials. And basically I think that we’re thinking about more therapy for higher-risk disease and more aggressive cancers and using subset analyses from these trials to better direct how best to treat our patients. And so I think that it’s a timely update, so it gives some kind of — putting all the data in one place to help us, and that’s very useful. ExteNET: 5-year follow-up analysis of neratinib after trastuzumab-based adjuvant therapy DR LOVE: One of the interesting things that’s happened over the last couple of years has been the evolution of neratinib as post-adjuvant therapy on HER2-positive disease. And we saw this paper in The Lancet Oncology in December that updated the 5-year analysis of the ExteNET study. Can you go back through what that paper looked at, what this update said and where we are right now with using this strategy in practice? DR RUGO: Neratinib is a highly potent tyrosine kinase inhibitor that blocks the HER family, HER2 and to some degree HER1/EGFR, and had had difficulty in terms of its development in the metastatic setting largely because of its primary toxicity, diarrhea, and the comparison studies that were done. And that led to the issues with the ExteNET trial where 3 different companies, basically, owned the drug during the conduct of the trial. And during that time, the number of patients decreased and the follow-up duration also decreased to be 2 years. And then the final data using 1 year of neratinib after patients completed a year of trastuzumab and where the trial had been changed a little bit during the course of the study to focus on high-risk disease rather than lower risk, similar to what’s being done in the endocrine therapy trials. And also to try and focus on patients who were within 1 year of stopping trastuzumab as opposed to many years out where their risk is going to be reduced simply because they’re many years out. When they finally got the results, surprise, surprise, it showed that a year of neratinib was better than not — than taking a placebo after completing trastuzumab. And then, of course, having such great results, reconsented all the patients they could to try and follow these patients out to 5 years to have a longer endpoint. And they were able to reconsent a majority of patients on the trial. And the group that was reconsented has been analyzed very, very carefully, and it’s very similar to the entire trial population. So I think that we can feel fairly secure on this. And it showed that there is continued benefit and that the degree of benefit in terms of disease-free survival is very similar and extends out to 5 years. The other thing that was shown in the analysis of the ExteNET trial both at 2 and 5 years is that the benefit appears to be greater in a subset analysis, prospectively planned, in patients who have hormone receptor-positive disease. And there’s actually very nice preclinical data that suggested that this would be the case many years ago — from Kent Osborne’s lab and others — where it appears that relative endocrine resistance and your recurrence risk goes out much longer with ER-positive/HER2-positive disease. And that these oral tyrosine kinase inhibitors may be particularly effective at improving response to endocrine therapy, and we’ve seen that in the metastatic trial published by Stephen Johnston with lapatinib and endocrine therapy in ER-positive/HER2-positive disease. So it makes sense that you might see a greater benefit in hormone receptor-positive breast cancer, and that absolute difference in disease-free survival at 5 years was 4.4%, which is quite impressive and better than what we’ve seen with many other treatments that we’ve adopted. The biggest issue is diarrhea and Grade III diarrhea, which occurs early. So it’s an early event, and it’s difficult for patients, obviously. So, actually, there’s been a prospective, multicohort, sequential trial called the CONTROL study that’s been looking at ways to prophylax against the diarrhea. The label includes the standard antidiarrheals. But only some patients get diarrhea. The other patients get really constipated, and they hate that, too. So that’s a problem, a big problem, in patients with cancer, anyway. So the CONTROL trial actually looked at loperamide, but then it went on to look at an oral nonabsorbable steroid called budesonide and then colestipol. And with these last 2 cohorts, we really have shown a marked reduction in significant diarrhea, number of days of duration of diarrhea. So the duration total over the duration of time you’re taking the neratinib where you have significant diarrhea, Grade II and Grade III, and that then correlates with better adherence and tolerability of the agent. And now what they’re studying, actually, is using a dose escalation in the first month. Because since you see the diarrhea right away when you start neratinib, it may be that starting low and going up allows the bowel to adjust so you don’t get so much of this secretory diarrhea. And there are a number of other approaches being studied elsewhere as well. So I think neratinib with this approach — I like colestipol, but certainly loperamide is the least you should do — is much more tolerable for patients. And in our patients on the CONTROL trial, where we’ve enrolled quite a number of patients, it’s been quite tolerable. DR LOVE: So I’m curious, because I’ve heard this thing about diarrhea and colestipol a couple of — like lenalidomide diarrhea, other places. What’s your vision about the pathophysiology, at a basic level, why, let’s say, neratinib causes diarrhea and why colestipol would improve it? I know it’s a simple question, but I’m curious. DR RUGO: No, it’s a really good question, and it really has to do with the hypothesis in animal models that have looked at why these agents cause diarrhea, which is secretory. So it’s a secretory diarrhea. And if you sequester bile acids, you reduce the secretory diarrhea. DR LOVE: What do you mean by secretory? Like, what’s going on? When you say secretory, like, what’s happening? DR RUGO: So you stimulate the channels, which I’m going to get into chemistry I can’t explain — DR LOVE: But this is in the bowel. It’s something about a direct toxicity of the bowel — DR RUGO: Right. DR LOVE: — and then what, somehow the bile salts make it worse? DR RUGO: But the bile salts make it worse, yes, by increasing the — I think maybe osmotic load, and so you get more fluid rushing in. And you can really reverse that by binding up the bile acids. It’s actually quite intriguing how that works. And you think about it, colestipol might even help infectious secretory diarrhea as well. There are a lot of things if you just reduce the secretory nature, say, of cholera-associated diarrhea. People don’t die of the diarrhea, right? So it’s a similar idea where you just reduce the amount of fluid that goes in, you don’t get the diarrhea anymore, and then you avoid the constipation, which patients hate with the loperamide. DR LOVE: That’s fascinating. You mentioned the difference between ER-positive and ER-negative. And I’ve heard people say, “There’s no benefit in ER-negative, very substantial benefit, maybe 40% relative reduction in ER-positive. Therefore, I will only offer it to a patient who’s ER-positive.” Do you buy into that approach? DR RUGO: The study looked at all comers, and the FDA’s approach is to take the way the study looked in all comers. They did the same with pertuzumab in the APHINITY trial. But if you look at the subsets, it’s very convincing. And I think we do tend to use drugs based on subset analyses — however flawed they are — particularly when they’re prospectively designed subset analyses. For example, we tend to — I mean, if you even think about the SOFT and TEXT trials, we are going to use more ovarian suppression in younger women and women who needed chemo and of higher-risk disease — that’s all subset — to try and really define who benefits more from adding on toxicity and cost. So I think that in this situation, the patients who have ER-positive disease, it makes sense to me biologically that neratinib works in that group of patients. So that is the group of patients I would tend to use neratinib for. If I had a very high-risk patient with ER-negative disease, I wouldn’t steer away from using neratinib, it’s just going to be the very high-risk resistant group of patients. And the reason why I say that for ER-negative disease is, I know that some of those patients respond very well to tyrosine kinase inhibitors. So you have a patient with ER-negative/HER2-positive disease that’s growing on neoadjuvant trastuzumab/pertuzumab-based therapy. I have had patients then respond for a very durable response to an oral TKI. So I know it can overcome some degree of resistance, and I would use it in that setting. DR LOVE: Oral TKI — neratinib or lapatinib or what? DR RUGO: We have more experience with lapatinib, because neratinib has only been recently available, but both drugs can do that. DR LOVE: Interesting. One final question about this, which is that a lot of patients are getting pertuzumab nowadays in the neoadjuvant/adjuvant setting. Were there patients in the ExteNET study who got pertuzumab? Does it change the way you think it through if a patient has had pertuzumab? DR RUGO: Pertuzumab wasn’t around at all during the ExteNET trial, so, of course, no one received pertuzumab, and we don’t know how that affects the efficacy of neratinib. However, in the APHINITY trial, of course, the benefit from pertuzumab and in the neoadjuvant trials that led to the accelerated initial approval of pertuzumab, the benefit was greater in patients who had ER-negative disease. And we’ve seen this consistently, that the antibodies benefit all comers but that trastuzumab’s benefit has been greater in ER-negative disease. And we saw that in APHINITY, where in ER-positive disease it appeared the benefit was less or nonexistent for adding pertuzumab. And then in, of course, patients who had node-positive disease, there was even more benefit from pertuzumab. We don’t have the breakdown of if you had node-positive, high-risk disease and ER-negative versus ER-positive because there just wasn’t enough follow-up when the first APHINITY data was presented. So I’m actually very interested in that subset analysis when we see the next updated presentation of APHINITY, where there are more patients who have node-positive disease that will be represented. DR LOVE: So just to clarify, though, when you’re deciding whether or not to use pertuzumab as adjuvant therapy, do you consider the ER? DR RUGO: I do consider ER, and I consider the stage of disease. So I would use pertuzumab in node-positive breast cancer or higher-risk disease. I might use pertuzumab in a patient who had a T2 node-negative/ER-negative tumor, whereas I wouldn’t use it in an ER-positive tumor. Efficacy and safety of PARP inhibitors for germline BRCA1/2 mutation-positive advanced BC DR LOVE: So let’s do a little PARP inhibitor update. And, of course, we spend a lot of time nowadays talking about PARP inhibitors in ovarian cancer, 3 approved, amazingly enough. These seminars are, like, really interesting to get into. But we have one now approved in breast cancer, olaparib. And we saw the paper in the New England Journal this past year, after the plenary presentation of the data. We also saw at San Antonio your data looking at another PARP inhibitor that’s not currently approved, talazoparib. So can you kind of provide an update of what we know about PARP inhibition in breast cancer, particularly as it relates to these 2 PARP inhibitors? DR RUGO: The second PARP inhibitor, talazoparib, has been submitted — that data has been submitted to the FDA, so we’ll probably see it sometime in the future. And I think that it’s really amazing how long it took us to get to where we are now with approved drugs because they’re clearly very effective in patients who have cancer associated with BRCA1 or BRCA2 mutations. We know that by inhibiting PARP that we’re causing the cancer to accumulate DNA damage, and that results in cell death because we’re blocking the main alternative pathway for DNA repair in tumors that are BRCA deficient and can’t repair through that pathway. We also know that resistance develops to these agents, and resistance can develop through a number of mechanisms, but one intriguing mechanism is restoration of function of the BRCA gene — actually DNA repair, which is fascinating. So that leads to that data set, which has shown that in patients who have BRCA mutations in metastatic platinum-sensitive disease, that we can more than double the response rates by using a PARP inhibitor compared to chemotherapy of physician choice, which is really fascinating. I mean, the responses are rapid and significantly higher when you use a PARP inhibitor versus treatment of physician choice. And then the duration, the durability of the response is also much longer, so progression-free survival was also doubled by using the PARP inhibitor. And in clinical practice, you actually see this, where patients are better very, very rapidly when they start treatment with a PARP inhibitor. We had thought that maybe we would see in these trials differential benefit in cross-trial comparisons, which, of course, we’re not supposed to be doing. But what’s striking in looking at the trials with olaparib — the trial with olaparib and talazoparib — is that the hazard ratios and also the improvement in response and progression-free survival is almost spot on between the 2 trials. We’re seeing very, very similar results. Disappointingly, the OlympiAD trial did not show a benefit in overall survival. Wasn’t really powered to look at overall survival, but, of course, we go back and think about trastuzumab in a marker-driven trial. And I’m hopeful that by moving these agents earlier into the course of therapy that we will see a survival benefit because we’ll either delay or avoid this rapid development of resistance that is corresponding to patients then progressing and potentially not having this survival benefit despite the improvement in overall response and progression-free survival. Both agents seemed to be reasonably well tolerated. We have some bone marrow suppression and nausea, but generally you can overcome those toxicities pretty well. I think it really leads us to a few different areas. One, I think the most critical for patients with BRCA mutations is trying to move it earlier in the course of therapy. The second is whether or not patients who have other inherited mutations that result in defects in DNA repair will benefit from those drugs, and there’s a couple of trials looking at that. And then the third is whether or not we could identify a group of patients who have sporadic cancers by looking at a homologous recombination deficiency or defect where they might benefit from PARP inhibitors. That is a very elusive goal but is still being pursued. And the last one is an interesting — that’s 4 areas — the last one is really interesting, and it has to do with preclinical data that suggests that potentially PARP inhibitors, by causing an accumulation of DNA damage, might sensitize tumor cells to the effects of checkpoint inhibition through something called the STING pathway and people are very interested in. And that’s actually a fascinating area and being studied in another set of trials both in the neoadjuvant and metastatic setting. DR LOVE: And, of course, it’s being studied in ovary also. You mentioned early-stage disease. I kind of want to dissect out where you are right now about how you actually are integrating this into practice, specifically olaparib. But just to start out with earlier-stage disease. Are there situations, if you have a BRCA mutation carrier, for example, and maybe they have triple-negative disease, you’ve given them neoadjuvant therapy, they got massive — or let’s say they have significant residual disease. Any situations — where multiple positive nodes — where you could access it where you would use a PARP inhibitor early, olaparib? DR RUGO: It’s an interesting question. I’ve sat on a number of panels where I’m the only person who’s used off-label drug. So I think that it’s not common. And we would always want to enroll patients in the OlympiA trial, because these trials take forever to enroll, and it’s the patients who suffer when we have delayed trial enrollment and don’t get an answer. So I would say the first step is to enroll in the OlympiA trial, which is the adjuvant olaparib versus placebo trial. But I think that if you have a patient who’s not eligible for OlympiA or can’t access it and has really a lot of cancer — skin involvement, positive margins and has a BRCA mutation — those patients who have an extraordinarily high risk of recurrence may benefit from use of a PARP inhibitor, and I would consider it there. But it would really be a very high-risk patient. I would not routinely use these drugs. I think we don’t know what the long-term toxicity is, and there is certainly toxicity in the immediate time taking the medication, particularly for patients who have early-stage disease. DR LOVE: I mean — yes, I mean, that is a point. So certainly some of these people, and a lot of them are going to be cured. What about HER2-positive disease? Again, what do we know about it, and what do you do in your practice? DR RUGO: What about HER2-positive disease? We don’t have any data on using PARP inhibitors in HER2-positive disease. HER2-positive disease that is associated with a BRCA mutation, generally we would, I think, as a community, focus on using HER2-targeted therapy. I think a patient who has HER2-positive disease and a BRCA mutation who has disease resistant to HER2-targeted therapy, a PARP inhibitor should be considered. We don’t know that there’s any added toxicity from combining a PARP inhibitor with HER2-targeted therapy, but this has not been studied, and I would be very hesitant to use that combination without at least some data on tolerability. DR LOVE: So I want to drill down of course the 2 common situations, which is metastatic ER-positive/HER2-negative disease and triple-negative disease. And before you get into where and how you integrate it, I’m curious how you approach using a PARP inhibitor, specifically olaparib. And particularly in a patient who’s not had a platinum agent, whether you consider the ovarian model of chemotherapy, particularly with platinums, followed by PARP maintenance, or do you kind of stick with the way the trial was done in breast cancer, which was olaparib monotherapy? DR RUGO: Actually, I think that the ovarian model is a great one, and I wish we’d studied it in breast cancer because these studies take so long to do. I think in a patient who needs chemotherapy up front because they didn’t study PARP versus chemo followed by PARP in ovarian cancer, so we don’t know. I think in a patient who needs chemotherapy because of bulk of disease, this maintenance idea is very, very appealing. And I would use it without qualms. In a patient who doesn’t need chemotherapy up front, I’d feel comfortable using a PARP inhibitor to start with. And I would try and put the PARP inhibitor as early as possible into the course of treatment. How that’s going to be impacted by immunotherapy with a recent press release from IMpassion, stating that they meet the endpoint by adding atezolizumab to nab paclitaxel in first-line therapy for metastatic triple-negative breast cancer remains to be seen, because in that situation, of course, you’re going to be using the checkpoint inhibitor as maintenance. So I think that a PARP inhibitor’s going to depend on what happens in the adjuvant setting in that situation. DR LOVE: And we’re going to get into checkpoint inhibitors as our final topic. But just to get into the real question, which is when to bring in a PARP inhibitor. So first of all, a patient with ER-positive/HER2-negative disease — when do you start thinking about a PARP inhibitor? I’ve heard people say after CDK/hormones. DR RUGO: That’s what I’ve heard a lot, too. I have used a PARP inhibitor instead of a CDK4/6 inhibitor as first-line therapy combined with endocrine treatment in a patient simply because that’s what I know is driving her cancer. And this was a very unusual case where she didn’t tolerate her adjuvant endocrine therapy and refused to take it and then developed widespread bone metastases and needed, actually, immediate surgery and very complicated. Where neutropenia would have been a big issue for her, because she had big surgery in her hip and then she had to get radiation to her hip. And I’ve found it’s very hard to administer the PARP inhibitor within a fair bit of time after you’ve radiated a big chunk of marrow. Whereas I didn’t have a lot of problem giving olaparib and hormone therapy. And so I have used that. But I think when I sit on panels, most of my colleagues would tend to use the CDK4/6 inhibitor first. It’s just that that’s more of a global attack, right, on hormone therapy, whereas the BRCA, you know already the target and you also know that resistance can develop more rapidly the farther along a patient is with treatment. So it makes sense to me to try it earlier, because some people can stay on drug for a very long period of time. DR LOVE: So triple-negative metastatic disease, again, when are you thinking about a PARP inhibitor, olaparib? DR RUGO: It’s a tough thing with triple-negative metastatic disease, because a lot of those patients have fairly bulky disease and visceral involvement, so you’re needing to give chemotherapy up front. And where we’re standing today, where checkpoint inhibitor therapy is not our standard of care in the first-line setting, if I don’t have a clinical trial available for a patient with a targeted agent, I would give chemotherapy first and then use olaparib as maintenance in that situation. Emerging role of immune checkpoint inhibitors as monotherapy or in combination with novel agents in BC DR LOVE: So let’s finish out talking about checkpoint inhibitors. And we pulled a bunch of papers looking at that, starting out with a paper that you had at ASCO recently, updated efficacy and safety as well as PD-L1 status with ER-positive/HER2-negative metastatic disease, abemaciclib plus pembrolizumab. Maybe you can kind of provide the background first about where we are in general with checkpoint inhibitors in breast cancer and then some of these new strategies, such as combining a CDK inhibitor. DR RUGO: Checkpoint inhibitors, which have been so successful in many cancers, and there are tons of drugs now approved, the idea of targeting triple-negative breast cancer first was there’s no targeted agent. So okay, let’s throw out HER2-positive disease because we’ve got all these targeted agents. And it is one of the 2 subtypes of cancer, triple-negative and HER2-positive, where we see a lot of immune activity and where tumor-infiltrating lymphocytes have correlated with both response to therapy and long-term outcome in the early-stage setting. Whereas in hormone receptor-positive disease, we don’t see the immune filtration as much, except for in specific subsets. And tumor-infiltrating lymphocytes may indicate a worse outcome, actually, with the standard therapy that we’re using at the present time. So that’s the rationale behind looking at triple-negative breast cancer — it’s an unmet need, and we see this immune activation that correlates with outcome. We saw single-agent activity, which we were really excited about, in small Phase I trials with pembrolizumab and atezolizumab, and they panned out to kind of poor — in the 5% range — responses in patients who had received prior chemotherapy for metastatic triple-negative disease. But then in the first-line setting, for both pembrolizumab and atezolizumab, looked like it was in the 23% to 25% range, kind of exciting, a quarter of the patients. And then work by Sherene Loi and others have suggested that if you look at patients who have high TILs, or tumor-infiltrating lymphocytes, that response rate goes up even further because that’s a group of patients you can select out who benefit. And then, of course, then the idea is how can you even further improve it in giving chemotherapy, radiation, immune agonists, PARP inhibitors all together with the checkpoint inhibitors in triple-negative disease may enhance the host immune response and then make the checkpoint inhibitor work better. So that’s a strategy that was taken for registration purposes where there was a press release for the IMpassion study that stated that they improved. They reached their progression-free survival endpoint and had encouraging overall survival data as well, and that data is scheduled to be presented this fall. And shortly thereafter, likely we’ll see the data from a KEYNOTE trial that looked at pembrolizumab in a similar setting with 2 different chemotherapy options. Ongoing trials evaluating immune checkpoint inhibitors combined with endocrine interventions for patients with ER-positive metastatic BC DR RUGO: I think that the use of checkpoint inhibitors in triple-negative breast cancer is evolving very rapidly. And on top of that, we have both postneoadjuvant, adjuvant trials as well as neoadjuvant trials that are looking at checkpoint inhibitors in triple-negative breast cancer, and those are going to give us a readout, I suspect, in the not too distant future, and so very, very important. Then it leaves us with ER-positive disease, where checkpoint inhibitors in a paper we recently published demonstrated a 12% response rate, representing 3 patients out of the 25 or so patients who were enrolled in a Phase Ib basket trial. So not a great response rate in ER-positive disease. And at the same time, Shom Goel in Boston, at Harvard, actually published intriguing data that showed that CDK4/6 inhibitors actually enhance infiltration of tumor infiltrating lymphocytes and upregulated immune signature and improve response to checkpoint inhibition. So that led to the study with abemaciclib and pembrolizumab, where we studied it in hormone receptor-positive breast cancer naïve to CDK4/6 inhibitors. And also, there was a cohort with non-small cell lung cancer. And we now showed a 28% response rate with combination therapy, no endocrine therapy and no real relationship with PD-L1 positivity in the TIL analysis that’s still ongoing. So a new cohort has been added to that trial that’s looking at that combination as first-line therapy with endocrine therapy as well, so a triplet-type therapy. And then there’s a randomized trial going on led by Erica Mayer at Dana-Farber that’s called the PACE trial that’s looking at randomizing to checkpoint inhibitor with a CDK4/6 inhibitor and endocrine therapy versus endocrine therapy and endocrine therapy and CDK4/6 — has 3 arms to it. So I think that we’re in an area where we’re really exploring the ER-positive patients in terms of how checkpoint inhibitors work. There’s also some immune agonist combination studies under the MORPHEUS trial designs and, of course, other studies as well looking at other combinations. So that’s a big area of exploration. But for triple-negative breast cancer, we may be in a situation where we’re incorporating these drugs into our clinical practice in the not too distant future. ENHANCE 1 trial: Efficacy and safety of eribulin in combination with pembrolizumab for metastatic triple-negative BC DR LOVE: So speaking of triple-negative disease, Dr Tolaney had a Phase Ib/II study with eribulin in combination with pembrolizumab. Any comments on that? She presented that at San Antonio. And do you think that chemo plus checkpoint inhibitors is where we’re going to eventually land, at least in the next few years? DR RUGO: To answer the second question first, I think definitely we’re going to land with chemo/checkpoint inhibitors. There are, clearly, a group of patients who do very well and live a very long period of time with checkpoint inhibitor therapy alone. But until we have really good tests that can differentiate out that group of patients, likely we’ll give everybody chemo and a checkpoint inhibitor together followed by checkpoint inhibition as maintenance. And that will help us differentiate out the patients who respond to checkpoint inhibitors primarily or not. And it’s quite likely that treatment after a checkpoint inhibitor will also be improved — in other words, treatment response. And this has been seen in other malignancies, very, very interesting concept that will need to be looked at in the existing randomized Phase III checkpoint inhibitor trials. In terms of the eribulin combination, eribulin’s a great drug. It’s caused a lot of bone marrow suppression but is a microtubule inhibitor. And so not unexpectedly, eribulin and checkpoint inhibition is effective and tolerable. We have no idea whether adding a checkpoint inhibitor to eribulin improved the response that we would have seen with eribulin alone, because this was a nonrandomized trial. So I think what it does is it adds to our safety data, but it doesn’t tell us a lot else yet. DR LOVE: So when you talk about response rates in the twenties, of course, the issue comes up of the quality of response, and that’s what has people excited. I mean, the response rates even in lung cancer, at least to checkpoint alone, are not that high. But what has everybody excited is the possibility of prolonged response, even cures being discussed in some situations. What about that in breast cancer? In general, are the responses very prolonged? Is it qualitatively different than what you see with chemotherapy, for example? DR RUGO: It really is. And I think that still, we have these data from single-arm trials, Phase II trials. But with both pembrolizumab and atezolizumab, the patients who achieved a true CR/PR had this unbelievable survival. I mean, the curves are really quite dramatic. The CR/PR patients are sitting at the top, living out way past median duration of survival for triple-negative breast cancer, whereas the patients who aren’t responding are falling off and having more of what we think of as the natural history of triple-negative disease. One of the questions is whether or not if we go back and look at patients who have metastatic breast cancer who have high TILs, are they a group of patients who already were living longer with triple-negative disease on treatment? But I have a big triple-negative population in my clinic, and I have not seen this before where you can have isolated patients just hanging around on checkpoint inhibitor therapy for years without any evaluable disease. So I do think that we’re qualitatively changing what happens to patients by inducing a response — by seeing a response to checkpoint inhibition, and that’s why the response is so exciting. Unlike, for example, PARP inhibitors, where the response is great, but the survival may not be any different. TOPACIO/KEYNOTE-162 trial: Response to niraparib and pembrolizumab in metastatic triple-negative BC DR LOVE: So a couple of final quick questions. We talked before about PARP inhibitors and checkpoint inhibitors. There’s the TOPACIO study. It was a Phase II trial presented at ASCO of niraparib, approved in ovarian cancer, with pembrolizumab. Any comments about those data? DR RUGO: Yes, I think we saw some really interesting responses from combining the PARP inhibitor with the checkpoint inhibitor. The responses were, as expected, much higher in patients who had germline BRCA mutations than those who didn’t. But there were responses anyway. The trouble is that there was no randomization, so the next step is to understand, what is the PARP inhibitor adding to the response you would have expected to have seen from the checkpoint inhibitor alone? And that we don’t know. But we do know that the combination seems to be tolerable, and it’s intriguing. We actually have an arm in our I-SPY neoadjuvant trial now that includes low-dose olaparib and durvalumab along with paclitaxel. DR LOVE: Low durvalumab, paclitaxel and olaparib, wow. What have you seen so far? DR RUGO: So it’s very early, so I don’t really have any data. But there is data from the pilot trial that led to this that was run by Lajos Pusztai at Yale. And he found it very tolerable with high response rates. But we see high response rates in this group of patients — DR LOVE: Of course, yes. DR RUGO: — so we need the comparable data from I-SPY. |