Current and Emerging Strategies for Early-Stage HR-Positive Breast Cancer (BC) — Stephanie L Graff, MD, FACP

DR GRAFF: Hi, my name is Dr Stephanie Graff and I’m the Director of Breast Oncology at the Legoretta Cancer Center at Brown University and Lifespan Cancer Institute in Providence, Rhode Island. And today, I’ll be discussing current and emerging strategies for early-stage hormone receptor-positive breast cancer.

I’m going to go through some of the key breakthroughs over the last year, starting with CheckMate 7FL which was a randomized double-blind trial of nivolumab versus placebo with neoadjuvant chemotherapy followed by adjuvant endocrine therapy plus or minus nivolumab in patients with high-risk estrogen receptor-positive, HER2-negative, early-stage breast cancer. I think it’s important to start by calling out some of the key inclusion criteria. This is a very high-risk patient population. Patients either had Grade 3 tumors, which, frankly, we don’t see very often in hormone receptor-positive breast cancer, or were Grade 2 tumors and ER-low defined as having an estrogen receptor of 1 to 10%. They were also healthy with good performance status and available biomarkers for testing. And then patients were randomized to this neoadjuvant approach adding nivolumab to sort of our standard chemotherapy backbone followed by surgery with continuation of adjuvant nivolumab for those patients that received neoadjuvant nivolumab.

And the primary endpoint in the intention to treat population was the likelihood of achieving a pathologic complete response. And what we saw was that pathologic complete response was significantly improved for the whole study population from 13.5% to 24.5%. And then when you start to break that down by key subgroups including patients that were PD-L1 positive using the SP142 assay, we saw that that pCR rate gets even more pronounced with an odds ratio of 3.11 and a delta of 24.1% as compared to the population that was PD-L1 nonamplified who had a much smaller 4%, 3.6% difference in their benefit. And I think that this starts to give us some helpful information about how we might start selecting hormone receptor-positive patients for possible addition of neoadjuvant pembrolizumab, nivolumab, other checkpoint inhibitors just from this early data.

This was, of course, then presented around the same time that we saw the press release and ultimately, the results of KEYNOTE-756. KEYNOTE-756 is similar. It’s a Phase III study of neoadjuvant pembrolizumab versus placebo plus standard chemotherapy followed by adjuvant pembrolizumab or placebo plus endocrine therapy again in this high-risk hormone receptor-positive, HER2-negative breast cancer population. Again, when we look at those inclusion criteria, these are patients that have, on clinical staging, evidence of a large tumor, more than 2 cm, or bulky adenopathy, a clinical T3/T4 or an N1/N2 tumor. Again, all Grade 3 breast cancers. And so patients are randomized to pembrolizumab plus chemo versus chemotherapy alone, go to surgery, with the primary endpoint of reduction in the pathologic complete response. And from KEYNOTE-756, we haven’t seen biomarker breakdown. We’ve just seen this first round of data looking at the intention to treat analysis. But we did see an improvement in pathologic complete response of 8.5% which was statistically significant, increasing the pathologic complete response rate from 15.6 to 24.3 in this hormone receptor-positive population. Again, very high-risk. Most of these patients were young, premenopausal with a median age of 49, 100% of them had Grade 3 tumors, 90% of them had lymph node-positive tumors. So maybe not the average patient that’s coming through our clinics with hormone receptor-positive breast cancer. But nevertheless, an important population that there’s an opportunity to really improve outcomes which hopefully, we’ll see as a long-term follow-up on this study as a result of improving pathologic complete response.

Another study that was published this year by Taylor in Breast Cancer Research and Treatment looked at the role of Oncotype DX Recurrence Scores^® in defining which patients may benefit from neoadjuvant endocrine therapy. And I think you all remember from your practices during the height of the COVID-19 pandemic that sometimes just because of delays in the operating room, we all had to be a little adaptive in the way that we approached our cases. And so many of us in clinical practice were ordering Oncotype DX Recurrence Scores on biopsy specimens and then using that to decide whether or not a patient should have up-front chemotherapy while waiting for an operating room opportunity or start neoadjuvant endocrine therapy again while waiting for the operating room. And this kind of takes that a step farther in a randomized approach where patients underwent core needle biopsy, had it sent for the 21-gene Recurrence Score. And if their Recurrence Score was less than 11, they were randomized to receive hormonal therapy alone per clinical practice. If they had a Recurrence Score of over 25, they were randomized to receive neoadjuvant chemo, again, per clinical practice. And then in sort of a TAILORx/RxPONDER approach for patients that had Recurrence Scores of 11 to 25, that key TAILORx group B population. They were randomized to either receive neoadjuvant endocrine therapy alone or neoadjuvant chemotherapy followed by surgery. And the endpoints were the rates of breast conservation, the residual cancer burden, and the percentage of pathologic complete response. And what we see is that patients that were in that core randomization group with Recurrence Scores of 11 to 25 really don’t see a big benefit of the addition of chemotherapy across these different populations which further supports how we think about TAILORx and RxPONDER in our clinical application of hormone receptor-positive genomic predictors like the 21-gene Recurrence Score. And I think that this is a notable effort that we can use to inform our clinical decision making.

Next, I want to look at this trial looking at identifying patterns and barriers in Oncotype DX Recurrence Score testing in older patients with early-stage hormone receptor-positive breast cancer that was published in Journal of Oncology Practice. And this cohort utilized a group of patients at a large academic medical center, Dana-Farber, and evaluated how they were using their own reflex criteria. So they had a reflex protocol to send 21-gene Recurrence Scores on patients who were under the age of 65. And that was very clearly defined. They only sent them on T1c tumors that were Grade 2 or Grade 3. They sent them on T2 or larger tumors. They sent them on N1 tumors of any grade. And they would send them on T3N0 or N1 tumor. And they did that for both pre- and postmenopausal patients. They saw that evolve where they included a little bit more Grade 1 tumors as their data matured. But for patients over the age of 65, it was really let up to the decision of the investigator, or of the clinician in this case, on whether or not to do 21-gene Recurrence Score testing. And so Dr Trapani and colleagues looked at what happened for patients that were over the age of 65. And in this population, what we saw was that when Recurrence Scores were sent as opposed to when Recurrence Scores weren’t sent, the patients were a little bit likely to be younger.

The patients that they omitted Recurrence Scores were older with a median age of 73. Recurrence Scores were more likely to be sent in patients with Grade 2 and Grade 3 tumors. Patients with Grade 1s, they omitted more often. And patients were more likely to not have any data about their lymph nodes when they omitted the 21-gene Recurrence Score. I think that this aligns with the thinking that if you have a patient who maybe isn’t the best candidate for a standard axillary staging with a sentinel node, they’re also probably not a patient that you’re going to be considering for chemotherapy and sending a 21-gene Recurrence Score. But what we saw in the population of patients that did have Recurrence Scores sent over the age of 65, really their Recurrence Scores broke down around classic breakdowns that we expect. Around 20% had low-risk scores, around 60, 65% had intermediate scores defined by the classic definition, and about 15 to 18% of patients had high-risk scores, scores over 31. And then about 20% of them go on to receive chemo, which means there’s a lot of nuance in the decision making for patients in those intermediate scores, but most of the patients with high scores are probably going on to receive chemotherapy. So for me, this says that I do think that we should include patients early in the decision who are over the age of 65 about whether or not Recurrence Score would change the way that they made decisions in their own cancer and partner with geriatric oncologists or our own knowledge base of how our elderly patients do with chemotherapy. And still consider scores like the Oncotype DX Recurrence Score to help select patients for chemo. Dr Love?

DR LOVE: Yeah. Actually, I was just having a flashback to ASH where we had this 87-year-old woman with lymphoma who had a complete response to CAR T therapy, no problems, in terms of chemo for older women. But anyhow. I wanted to ask you about neoadjuvant use. You were talking before about that. Because we actually did a survey of investigators and found a fair number of them were actually using Oncotype in the neoadjuvant setting. If you think about that paper you were discussing, it kind of implies that if they have a low Recurrence Score, the standard would be hormonal therapy or, I don’t know, maybe send them to surgery or something. But do you think that is actually being done? Do you use Oncotype or any genomic assay in the neoadjuvant setting? And do you think investigators in general are or not?

DR GRAFF: I think that most of us practice in very multidisciplinary environments. And I would say that that’s a broad statement. I came from a community practice at Sarah Cannon prior to being here at Brown, and it was the same thing. I practiced in a big multidisciplinary group. I had great working relationships with my breast surgeons and my radiation oncologists. And what I would say is that there are lots of patients who can very quickly have surgery and sentinel node that will help inform, especially in a premenopausal patient if you even want to get a 21-gene Recurrence Score. In older patients, will help have sort of that comprehensive look. Your radiologists and your pathologists can be your allies in saying this is a representative sample of the tumor. If you have a 5 cm breast mass and it’s Grade 1. If the needle was a little bit to the left or a little bit to the right, might there be more than meets the eye there? Maybe. Maybe not. But that’s also the patient that you potentially have the opportunity by giving them neoadjuvant therapy, whether neoadjuvant endocrine or neoadjuvant chemo, to reduce the size of their tumor and change their potential surgical outcomes. So I think just engaging not only the patient, but your colleagues across the aisle, whether they’re your surgeons or your radiation oncologists, pathologists, on the quality of your specimen and if we believe that it’s representative of that patient’s breast cancer and appropriate to send for testing. And then how early therapy, regardless of which of the 2 forms we use, chemo or endocrine therapy, might change that patient’s surgical approach. Keeping in mind that the likelihood of a pathologic complete response, the likelihood of changing your surgical approach is still probably low, 20 to 30%, in a hormone receptor-positive breast cancer.

The other scenario that I encounter in my own clinical practice that perhaps is the reason I’m most likely to send Oncotype on a biopsy has a lot to do with the way that we consider breast reconstruction. Some of my younger patients, my patients with BRCA2 mutations, for example, that are also candidates for things like PARP inhibitors, but may be considering bilateral mastectomies even for smaller primary tumors, trying to coordinate something like a deep flap or a breast reconstruction just takes time. And if you know that you’re going to be waiting weeks to go to the operating room and you could be using that time to maximize their systemic therapy and push off a surgical plan for a later date when everyone is more coordinated. Again, this is just an opportunity for team-based decision making.

DR LOVE: All right. Please continue.

DR GRAFF: Thank you. So this is going to bring us from the endocrine therapy space into the CDK4/6 inhibitor space. And of course, over the last year, we’ve seen the results of both NATALEE using ribociclib and the monarchE study using abemaciclib. And so this year, we really sort of started to dive into the NATALEE ribociclib data. I do want to just pause here and say that of course the data I presented first looking at CheckMate 7FL and KEYNOTE didn’t include CDK4/6 inhibitors because those trials were accruing synchronously and was before that was really emerging as a standard. So how these different treatments are changing the landscape of how we treat early-stage hormone receptor-positive breast cancer is still a question. And we’ll have to see how those pieces work together to ultimately improve our patients’ outcomes. NATALEE, of course, used ribociclib plus the non-steroidal aromatase inhibitors as adjuvant treatment in patients with hormone receptor-positive HER2-negative early breast cancer. This was a scheduled analysis of survival, on invasive free survival analysis from NATALEE. NATALEE, as a reminder, did expand that monarchE population to include some patients with lymph node-negative breast cancer.

That included tumors that were Stage II anatomically, so larger than 2 cm, and either Grade 2 with a high 21-gene Recurrence Score, Grade 3, high Ki-67, you can see it there in the first panel, or that were lymph node-positive. We did, NATALEE also did include patients that were lymph node-negative and T3 or T4 although, again, those are smaller populations. And randomized patients to either receive letrozole or anastrozole plus or minus goserelin if they were premenopausal or male plus or minus ribociclib. And so ribociclib was really our randomization there. And the primary endpoint was invasive disease free survival. This initial analysis was presented in January and, at that time, had 426 invasive disease free survival events with about half of the patients having discontinued ribociclib including only 20% that had completed their full 3 years. This update which is just 6 months later, which if you think about it is really just a tiny march forward, included an additional 70, 80 invasive disease free survival events for a total of 509 invasive disease free survival events with now 78% of the patients on this study having stopped ribociclib with about the same amount that discontinued therapy early, 35%, and now 42% having completed the full 3 years of ribociclib. So we’re starting to get more of a snapshot of what this therapy means once you’re off it which, remember from the earliest analysis of monarchE, was one of the big concerns was if we were just really delaying time to recurrence. Median follow-up of the study overall is 33 months with a maximum of about 51 months of therapy.

And so the results here show that compared to just that 6-month advance in time and that slight increase in the number of invasive disease-free survival events, we’ve seen the benefit of ribociclib improve from 1.5% to 3.1% where now at this 36-month analysis, we see the benefit of ribociclib improving invasive disease free survival from 87.6% to 90.7%. That’s a hazard ratio of 0.749 which is an absolute benefit of ribociclib of 3.1%, 25% relative reduction. And I think that this compares very similarly to what we had seen with monarchE. Of course, it’s never a great idea to do cross-trial comparison, but we do it all the time. And with monarchE which was, again, a higher-risk population including patients that were all lymph node-positive, the difference at 3 years was about 4.8% and the hazard ratios actually looked really, really similar. And so I think that this is a favorable sign for the NATALEE study but, of course, we’ll continue to wait to see for this data to mature.

Adjuvant abemaciclib on the monarchE trial also, of course, got an update for us. And so this is the overall survival interim analysis including the 5-year efficacy outcomes. And what we see with monarchE, again just as a reminder, all hormone receptor-positive early-stage breast cancer, all lymph node-positive, patients that had 1 to 3 positive lymph nodes also had 1 other high-risk feature including either a high Ki-67, a large tumor size or a Grade 3 tumor, randomized to abemaciclib plus endocrine therapy versus endocrine therapy alone. And at this 60-month time interval, we see that the hazard ratio for favoring abemaciclib moves to 0.68 with a benefit changing from 76.0% for the endocrine therapy alone to 83.6% for the abemaciclib arm, a delta of 7.5% — sorry, 7.6% absolute difference which is a 32% relative risk reduction with the addition of abemaciclib. This is, of course, published in Lancet and Dr Nadia Harbeck had presented this at ESMO. So, again, continued benefit of abemaciclib for our patients with high-risk early-stage breast cancer. This is certainly a part of my clinical practice for high-risk patients. And I think we just all have to continue to consider and select our patients for these opportunities to ultimately improve long-term outcomes.

Adjuvant exemestane and ovarian suppression in premenopausal breast cancer which was the TEXT and SOFT trials also got an update in 2023. I always have to go back to the beginning and remind myself about the TEXT trial versus the SOFT trial. The TEXT trial included premenopausal women who had been planned to receive ovarian function suppression and either had no chemotherapy or chemotherapy with 1,053 patients getting no chemo, 1,600 patients getting chemotherapy. And they were randomized to tamoxifen plus ovarian function suppression or exemestane plus ovarian function suppression. The SOFT trial included patients that were premenopausal, had no chemo, again 1,400 patients there, or who remained premenopausal after chemo, another 1,600 patients in that cohort. And they were randomized to tamoxifen alone, tamoxifen plus ovarian function suppression, or exemestane plus ovarian function suppression. So as we look across the TEXT and SOFT trials, the group of patients that did not receive ovarian function is just from SOFT and is a pretty small population. But on this longer follow-up that was published in JCO in 2023, we see that the benefit of exemestane, the blue line, plus ovarian suppression versus tamoxifen, the red line, plus ovarian function suppression is an absolute difference of 4.6% for the aromatase inhibitor with an improvement from 75.9% at 12 years of follow-up versus 80.5% at 12 years of follow-up for disease free survival.

And I think that, again, this is really important. I’ve previously pinned an editorial talking about premenopausal ovarian function suppression. And I think that for our patients, my approach in clinic is to start patients on aromatase inhibitors plus ovarian function suppression and then to really meet the patient where they are in terms of managing toxicity. A 5% improvement in overall survival, disease free survival at 12 years is really powerful. But not if it absolutely destroys that patient’s quality-of-life to get to that 12 years. And so I think continuing to see your patients early and often with premenopausal women and if they’re having a lot of problems with the aromatase inhibitor, know that coming down to tamoxifen plus ovarian function suppression can still be really beneficial, keeping an eye on bone health. And then obviously, dropping the ovarian function suppression for patients that are having a difficult time tolerating that. And really just continuing to personalize that care for the best outcome that allows adherence to therapy for our premenopausal patients.

Of course, I’m talking today on hormone receptor-positive breast cancer, but one of the questions I hear from patient advocates all the time is, what about co-positive or triple-positive patients? And so I really wanted to highlight that in this update in JCO, they do give us the breakdown by HER2-positive and HER2-negative. And interestingly, again, I’m going to remind the audience that the red lines are the tamoxifen and the blue lines are the exemestane. In the HER2-positive hormone receptor-positive patients, it does seem that the benefit is greater for tamoxifen than for exemestane in combination with ovarian function suppression. And so I think that there’s lots we still don’t understand about the intersection between the estrogen receptor and the HER2 receptor and how there’s crosstalk and signaling between those pathways. But there may be some benefit in our co-positive patient populations to choose tamoxifen based on this subset population. Again, small numbers of patients included here. Not necessarily practice changing or the highest level of evidence. I’d love to see a follow-up trial design looking at this population in particular. But certainly, food for thought as we’re counseling and choosing agents for our patients.

Another update on the role of ovarian function suppression came from the Early Breast Cancer Collaborative Trialists’ Group. A meta-analysis Dr Gray presented at ASCO this year, published, hopefully, we will continue to see data emerge, looking at the role of ovarian function suppression. Now this is a patient-level meta-analysis, so the rawest data you can possibly pool on nearly 15,000 patients. And looked at the benefit for patients who either did not receive chemotherapy or remained premenopausal after chemotherapy. And we see a substantial benefit in the 15-year reduction in the risk of recurrence for the addition of ovarian suppression with an absolute 15-year gain of 9.8% for the addition of ovarian function suppression. When we look at the patients who were premenopausal before chemo and who were uncertain on what their hormonal status is after chemo, we see that that really levels out and narrows down to a 1.3% benefit. I do think it’s often hard for us to clinically identify who is who in clinical practice. And so being really thoughtful about our clinical approach here will help us determine the benefit of ovarian suppression for our patients. For patients that did not receive chemotherapy specifically and remained premenopausal, there was a further analysis breaking it down by age. And again we see that the very young women under the age of 45 seem to be the patients that get the largest benefit with an absolute gain of 10.9% compared to patients age 45 to 54 where that benefit narrows to 7.5%. That’s not to negate a 7.5% absolute benefit for a woman aged 47, for example, particularly who has extensive lymph node involvement or higher-stage disease because obviously, there’s shades of gray in that absolute number. But, again, can be a distinguishing factor as we’re considering who to offer ovarian function suppression to.

Looking deeper across that data, one of the important aspects was an analysis on mortality. When we look at breast cancer specific mortality, we do see a decrease with the addition of ovarian function suppression with a 20-year reduction in mortality of 10.9% which is huge. But then when we look at death without recurrence and all-cause mortality, importantly and perhaps flying in the face of what we’ve traditionally been told, we see no difference in the patients that were offered ovarian function suppression. And I think that we perhaps were taught that patients that get ovarian function suppression at an early age are going to have a higher risk of cardiovascular events or other causes of mortality. And here, the death without recurrence is no different between the control and the ovarian suppression arm. Dr Love?

DR LOVE: So I’ve got your really great comprehensive paper from JCO Oncology Practice on treatment of premenopausal women. And I’m curious, who are the women that you don’t give adjuvant ovarian suppression or ablation to and, for example, give just tamoxifen?

DR GRAFF: Yeah, I think the things that go into my decision are their stage of disease. Obviously, I’m not giving ovarian function suppression to a T1a or a T1b tumor. Their age is certainly a factor as well. Patients that are 49 and whose mother and 3 sisters all went through menopause at 49, I think that they’re naturally going to transition into menopause without it. Interestingly, in the SOFT and TEXT trial data when you parse down the granularity of the data, the patients that got the biggest benefit in SOFT from ovarian function suppression were actually the patients who also received chemotherapy. And so that’s actually another factor in my decision making. I think that it’s easy to assume that our patient’s chemotherapy-induced ovarian function suppression is going to be durable. And we know it’s not always, especially for patients that just get TC. And so I am more likely to offer ovarian function suppression to patient that were also candidates for chemotherapy. I think that’s one of our unanswered questions from RxPONDER and TAILORx is whether the impact of chemotherapy on ovarian function suppression is driving the differential benefit that we see in premenopausal women. And at least that little nugget of detail that comes out of SOFT suggests that maybe there is addition there. And so, again, I try to be very adaptive in meeting patients where they are. I bring back my premenopausal patients for more frequent initial follow-up when they’re starting ovarian function suppression to try to troubleshoot the side effects that arise and to deescalate, if need be, right-size their treatment.

DR LOVE: So let me tell you what happens when we ask investigators that question I just asked you about tamoxifen because it brings up something that doesn’t usually get into papers and I’m not sure if you got into it which is what we hear when we say who do you give tamoxifen to is, A, age is an issue, like over 40, for example, or whatever number you want to come up with. But the big thing that we saw that really surprised me, it didn’t surprise me but it surprised me it’s not in the guidelines, is node-negative low Oncotype.

DR GRAFF: Yeah.

DR LOVE: Simple as that.

DR GRAFF: I think you’re exactly right. There is obviously, the foundational data that the 21-gene Recurrence Score is built on was a group of patients that pretty exclusively got tamoxifen. And so I think that if somebody’s Oncotype Score is 2, their endocrine sensitivity is so powerful that they are probably going to get a lot of bang for their buck, if you will, from tamoxifen without ovarian function suppression and aromatase inhibitors.

DR LOVE: So that also makes me wonder whether ovarian suppression is being underutilized in clinical practice. I’m curious if you have any thoughts about that.

DR GRAFF: I 100% think it’s being underutilized in clinical practice. We see that when we do large analysis of available data. Actually, at ASCO in 2023, we presented a poster using the concert AI, the concert data set, and looked at the rates of ovarian function suppression, basically replicating this data that you see here, that ovarian function suppression in addition to either aromatase inhibitors or tamoxifen improves outcomes for premenopausal women. We see that the number of patients being offered ovarian function suppression in those electronic medical record datasets is lower than perhaps guidelines would select patients for ovarian function suppression.

DR LOVE: All right. Yeah, that’s one of my things, so to speak, because you’re talking cure.

DR GRAFF: Yeah.

DR LOVE: Right? Potential to — I’m not saying everyone, I’m just bringing up the issue of whether it should be offered. People can make their own decision whether they want to do it. They could try it, whatever, but I don’t know.

DR GRAFF: Yeah. When I speak to practitioners, they often tell me that patients just feel so beat up at that point. Their tenacity has started to wane as they reach the decision for ovarian function suppression. And often, I think clinicians approach the decision as an escalation strategy, that they start on tamoxifen and if the patient feels good, they change them to ovarian function suppression plus an aromatase inhibitor. But in my experience, it’s really hard to convince somebody to stop doing something that they’re comfortable with and that is working. It's easier to convince somebody who is miserable to stop doing something. It’s just important that you’re making yourself or your APPs or your team that supports you in clinical practice available to your patients and that you’re using that all important sentence, if this is hard, if you’re having problems, you have to let us know because there’s always a next step, so that patients feel empowered to reach out and be like, man, this aromatase inhibitor stinks so bad there’s no way I’m going to take it, and don’t just stop coming to your office altogether.

DR LOVE: Absolutely. Okay, please continue.

DR GRAFF: And what a great transition, right into when we need to stop endocrine therapy. So Ann Partridge led what I consider just pioneering work on the POSITIVE trial which is pregnancy outcome and safety of interrupting therapy for women with endocrine responsive breast cancer. This was published in The New England Journal of Medicine this year. And this trial took patients that were interested in achieving pregnancy after a diagnosis of hormone receptor-positive breast cancer. Patients on average were 37 years old with about 34% of the patients under the age of 35, 23% of the patients age 40 to 42, so a fairly young group of people, 75% of patients had not had a prior pregnancy. Only 6% had Stage III disease which I do think is a caveat for me in my clinical practice. Although those patients were eligible for participation, it’s such a low number it still makes me anxious to offer cessation of therapy to the highest-stage disease. Again, here we see that kind of breakdown of what people are really doing, right, 36% of patients got tamoxifen plus ovarian function suppression, 42% of patients got tamoxifen alone, 16% of patients got aromatase inhibitor plus ovarian function suppression. And then 62% of these patients had had prior chemotherapy which aligns with that kind of higher-risk hormone receptor-positive disease that we tend to expect in patients this age. Patients were eligible if they had completed up to 2 years of endocrine therapy, and then were enrolled on the study.

Once they were enrolled on the study, they had a washout period and then were allowed to interrupt their endocrine therapy for up to 2 years to allow for pregnancy, conception, delivery, breastfeeding and importantly, fertility treatments if needed to achieve their pregnancy. If no pregnancy had been achieved by a year, patients were encouraged to go to fertility assessment to try to achieve that pregnancy during the time that was allotted. And then after 2 years, patients were strongly encouraged to resume their endocrine therapy. And so the primary outcome of the study was breast cancer events. And they compared to a historical control group using the SOFT and TEXT data and saw that there was really no increased risk in having a breast cancer recurrence despite this 2-year interruption of endocrine therapy and attempts and successes at pregnancy with a hazard ratio of 0.81. That confidence interval is pretty wide, 0.57 to 1.15, telling us that this is an appropriate option for our patients. Out of the patients that were enrolled in the study, 497 women, 368 or 74% of them did have at least one pregnancy. And out of those 368 women that had at least 1 pregnancy, 317 or 86% of the ones who became pregnant, ended up having a live birth. I think that particularly when you consider that these are women in their late 30’s, early 40’s, 75% of whom had never had a prior pregnancy, so you may be selecting a group of patients with a higher likelihood of infertility at baseline seeing 64% of all enrollees achieving pregnancy. Especially after 62% had been treated with chemotherapy is really important and also can give hope to our patients that are worried about their probability of going on to become pregnant after breast cancer and initial breast cancer therapy. 11% of the pregnancies did have complications. That’s actually pretty in line with age-matched populations. And birth defects were low at 2% of the live births. And so this is hopeful data. We do want to see continued long-term follow-up. This is just 48 months of follow-up which for a hormone receptor patient population, of course, we know is a short snapshot in time, but does provide some initial hope for our patient population. And I do think can inform clinical practice today as long as you’re counseling your patients on all of these caveats.

Moving to maybe what’s up and coming in the hormone receptor-positive breast cancer space, we look at SERENA-3. SERENA-3 was a randomized presurgical window of opportunity study assessing camizestrant which is one of the oral SERDs that in development for postmenopausal women with estrogen receptor-positive HER2-negative primary breast cancer. And so this study took patients really with a palpable tumor, any size, hormone receptor-positive, HER2-negative, biopsy confirmed, and randomized them to camizestrant for a relatively short timeframe, in just that time between diagnosis and when surgery was ordinarily planned. Many of these patients received camizestrant for a very short timeframe, 15 days is sort of the longest window of time, some as short as 5 to 7 days of camizestrant. And looked at whether or not camizestrant could change their estrogen receptor immunohistochemistry H score. I’m sorry, that’s very science-wordy. I’m going to say it again in non-science talk. As well as things like proliferation index like the Ki-67. So this is really, to say it back using different language, is really a proof of concept study that oral SERDs are effective reducing estrogen consumption by cancers and antiproliferative in that they stop markers of growth in that cancer. This lays the backbone that this could be a strategy, like Dr Love and I talked about earlier, of selecting patients with something like a 21-gene Recurrence Score for neoadjuvant endocrine therapy and really maximizing endocrine therapy for patients to try to get their best reduction of their tumor or their best optimally timed surgical plan for that patient’s life or lifestyle. And what we saw was that most of the patients enrolled did have ductal carcinomas. Most were early-stage. Most were Grade 2, so these weren’t even necessarily high proliferative tumors. And most were lymph node-negative at 69% of the population.

And then I think I have this breakdown. And, again, to save everybody the super science talk here. In that first slide, we look at the estrogen receptor H score. And what you’ll see is that camizestrant at all doses and at all lengths of therapy, whether 5 to 7 days, 75, 150 or 300 mg, or 12 to 15 days, 75 mg or 150 mg, significantly reduced the rates of estrogen receptor expression. It effectively degraded the estrogen receptor for these tumors despite really short durations of exposure which not only means that this drug works, achieves its mechanism of action, but it does so quickly. For me, as we start to look at these drugs in the metastatic setting, that also translates to mean that we can expect pretty brisk responses in our patients that we’re offering oral SERD therapy to. Moving into this Ki-67 degradation, again we see that after 5 to 7 days of exposure, Ki-67 is reduced across these studies. This is, the dot diagrams here are showing the reduction of Ki-67 where the bottom of the line is 100% reduction and the kind of midpoint of the low section there is a 50% reduction in Ki-67. So we’re seeing substantial reductions in Ki-67 across doses and across time intervals of exposures to this medicine.

And I think that moves me to one of my final presentations which is an update on KATHERINE. Again, I was assigned to talk about the estrogen receptor-positive cohort here, so I want to look at patients that co-expressed hormone receptor-positive and HER2-positive breast cancer. I think just to go back over the KATHERINE study, KATHERINE takes patients with neoadjuvant HER2-positive breast cancer treated with chemotherapy who did not achieve a pathologic complete response, randomized those patients to complete 1 year of adjuvant trastuzumab versus 1 year of adjuvant T-DM1. We see at 7 years of follow-up, the absolute benefit in the intention to treat population of switching to T-DM1 is an improvement in invasive disease free survival of 13.7% at 7 years. Obviously, practice changing. I know many or all of you have already incorporated this into your practice. And so this is just that longer-term follow-up really continuing to show the power of doing that. We had seen in an earlier analysis that although in order to be eligible for the KATHERINE study you only had to prove that you were HER2-positive on your initial biopsy, not after surgery.

Even patients that lost HER2 expression after neoadjuvant therapy, so whose residual cancer volume still had lost HER2 expression and had become HER2-negative, still benefited from this change. So we’re really making our decisions based on that initial biopsy on whether or not to switch patients to T-DM1. Importantly, this is a co-positive trial, 72% of patients enrolled on this study were positive for both the hormone receptor and HER2. Which in my clinical practice, I see patients that are co-positive, HER2-positive, hormone receptor-positive, are less likely to have pathologic complete response, so it makes sense that we’re sort of selecting them into this KATHERINE design. And when you specifically look at the hormone receptor-positive cohort, the invasive disease-free survival hazard ratio is 0.52. We see that same breakdown in benefit of changing to T-DM1 from trastuzumab for patients that express both hormone receptor and HER2. Also, one of the updates that we saw in this most recent analysis was site of recurrence. They did look at CNS specifically, and we see a small but I think significant reduction in the likelihood of having a CNS recurrence. To quote Nancy Lin who is one of our CNS experts in breast oncology, a drug that’s more effective for everyone is more effective for reducing brain metastasis. And I think that that really bears out here where we see the CNS recurrence drops from 7% with trastuzumab to 5.1% with T-DM1.

Final, Dr Matteo Lambertini at San Antonio and synchronously in JAMA published this international multicenter retrospective cohort study looking at pregnancy after breast cancer in women with germline BRCA1 and BRCA2 pathogenic variants. I am proud to have participated in this cohort. And what Dr Lambertini was able to showcase was that across patients with Stage I, Stage II, Stage III breast cancer who were under the age of 40 at diagnosis, so again, slightly younger than Dr Partridge’s POSITIVE trial, who had a known germline BRCA1 or BRCA2 mutation then were matched for whether or not they had achieved pregnancy. The cumulative incidence of pregnancy at 10 years was 22%, so these women are able to go on and successfully have pregnancies. That doesn’t even factor in their intention, if they wanted to go on to receive pregnancy. So maybe if we’re selecting more of a population like POSITIVE where they wanted to go on to get pregnant, we would see higher numbers. The likelihood of having a pregnancy was slightly lower in the hormone receptor-positive cohort which was probably influenced by our understanding and patient counseling at the time. Median time from breast cancer to conception was also slightly longer in that hormone receptor-positive cohort, 4.3 years. But no significant difference in pregnancy outcome compared to the general population. And I think that that again is really valuable information that we can use to counsel and support our patients with germline BRCA1 and BRCA2 populations. And I think between this analysis and the POSITIVE study, I’d be comfortable considering patients with a germline BRCA1 or BRCA2 for that POSITIVE cohort as well.

Additionally, Dr Lambertini does give us a look at the results of their risk of recurrence which is disease free survival here as the curve. And we see that patients that had no pregnancy in the blue line versus patients who became pregnant. It does seem that pregnancy may have a slight protective effect in hormone receptor-negative breast cancer with a hazard ratio of 0.76. But no effect, good or bad, neutral in hormone receptor-positive breast cancer with a hazard ratio confidence interval from 0.95 to 1.76. Again, I think sending the message that this is a population that perhaps could benefit or be considered candidates for achieving pregnancy in the future.

And that is my final slide. Thank you for your time today.

DR LOVE: So that was really awesome. I just wanted to ask you one — do you know what a colada is? That’s, we’re in Miami, that’s Cuban coffee.

DR GRAFF: Yeah, yeah, yeah.

DR LOVE: I feel like your talk is like a colada for me this morning, just, like, okay.

DR GRAFF: That’s good.

DR LOVE: I’m trying to grasp all this stuff.

DR GRAFF: I might need one.

DR LOVE: I’m telling you, there were a couple papers I’d never even heard of that were so great. But one of the things that I wanted, the KATHERINE one is one. I did not know that. But in any event, what I want to ask you, just one quick question is you kind of alluded to the fact that in HER2-positive disease, hormone receptor-positive tumors tend to have lower path CR. But in this study, the hazard rate looks very similar for ER-positive and ER-negative, right? It’s 0.54 versus 0.52.

DR GRAFF: Yeah. And I think that that is, again, you’re looking at the hazard ratio for the benefit of the addition of T-DM1 over trastuzumab which doesn’t consider the likelihood of achieving a pathologic complete response. So I think one of the — we already know that we’ve selected the patients that did not achieve a pathologic complete response for KATHERINE. And so when you —

DR LOVE: I was just alluding to general antitumor efficacy in general.

DR GRAFF: Absolutely.

DR LOVE: That lower path CR. And I think some people feel like hormone receptor-positive, HER2-positive patients tend to not respond as well. But at least in terms of this now, granted it’s T-DM1, but it looks like it’s about the same.

DR GRAFF: Yeah. I think that we still need to do much better work as a community of researchers and scientists trying to answer questions about the intersection of the HER2 receptor and the hormone receptor, especially in the premenopausal population which tends to be the patients that are presenting with that co-positive disease expression about how much additive benefit you’re getting in that population for ovarian function suppression versus tamoxifen versus T-DM1 versus, you know, the whole kitchen sink. Again, I showed in the SOFT and TEXT long-term follow-up that that co-positive group seemed to get a bigger benefit from tamoxifen plus ovarian function suppression than aromatase inhibitors plus ovarian function suppression. And we don’t have a great answer for why that might be.

DR LOVE: Right. Yeah, that’s interesting. Did they ever look at the CLEOPATRA data to see length of benefit in metastatic disease, ER-positive or ER-negative? And do you —

Do you in your own practice, do you think that you see less, you know, not as long responses in ER-positive, HER2-positive?

DR GRAFF: I do not necessarily think that in the metastatic setting that I see a difference in length of response between my co-positive and my hormone receptor-negative, HER2-driven breast cancers. I know that in CLEOPATRA in the subgroup analysis in those forest plots, they do give a breakdown by hormone expression positive and negative, and we see similar benefits. In my clinical practice when I’m treating somebody, for example, with first-line metastatic disease with a taxane plus trastuzumab plus pertuzumab, once I, pardon me, once I drop the taxane, likely due to peripheral neuropathy, and I’m just continuing the trastuzumab and pertuzumab alone, the doublet HER2 blockade, that’s when I always make sure that I’ve added back in endocrine blockade. And there’s been a wealth of studies looking at whether CDK4/6 inhibitors or SERDs or next generation estrogen blockade would be more beneficial there, but they’re always pretty small numbers. So trying to really counsel a patient on the best endocrine partner has been a challenge.

DR LOVE: Do you think in your own practice you see lesser path CR if it’s ER-positive?

DR GRAFF: Yes.

Relevant Advances in the Care of Patients with HR-Positive Metastatic BC (mBC) — Erica Mayer, MD, MPH, FASCO

DR MAYER: Hi, everybody. I’m Erica Mayer from Dana-Farber Cancer Institute in Boston. I’m going to be speaking today about advances in the care of patients with hormone receptor-positive, HER2-negative metastatic breast cancer and an update for 2023.

So just to get started, what a year it has been. Mirroring what we’ve seen in the rest of medical oncology, breast cancer has been an incredibly fast-paced field. We have many new agents, new drug approvals, lots of evolving data to stay on top of. And the new drug approvals and these new results means that our treatment paradigms are continuously in evolution and under modification. I do think that toxicity and tolerability have become really key and maybe differentiators among all of the agents that we are now looking at. And the new findings that we’re going to talk about hopefully result in better and longer outcomes for our patients.

So let’s jump in. So first we’re going to be looking at hormone receptor-positive disease, and we saw updates from the MONARCH 3 study. MONARCH 3 is one of the first-line studies looking at endocrine therapy with or without a CDK4/6 inhibitor. In this study the CDK4/6 inhibitor is abemaciclib. This is one of the studies that’s been previously presented, and what was shown at San Antonio from Matt Goetz was an overall survival update.

Notably in this study this was a 2:1 randomized study, so two-thirds of the patients received abemaciclib and aromatase inhibitor, one-third placebo and aromatase inhibitor. And the study enrolled a total of about 500 patients. This is first-line setting.

What was seen in the final overall survival results was that the patients who received the abemaciclib had a longer overall survival, 66.8 months, at what is now about 8-year follow-up on this study, and that’s in comparison to the control arm of 54 months. So numerically this is longer. It’s more than a year longer. However, this did not meet statistical criteria for the study, which was fairly rigid and stringent due to the specific design of the trial. The prolonged overall survival was also seen in the subgroup with visceral disease, which tends to be a group that we think of in particular for use of abemaciclib.

I’ll also point out that they presented updated progression-free survival, and this continues to be dramatically longer with the use of abemaciclib, about 30 months compared to about 15 months, so a doubling. And if you look at the curve on the right, the red line, I think a very interesting observation is that at this 8-year point about 23%, so about 1 in 5 patients, are without evidence of disease progression. And these are patients who started therapy an average of 8 years before. I think we all have these people in our practices who started CDK4/6 inhibitors around the time of initial approval and continue to do extremely well on the drugs, so I really think that speaks to the power of CDK4/6 inhibitors when used in the first-line setting.

If we put this into context these are all of the randomized trials using CDK4/6 inhibitors in the first-line and pretreated setting looking at overall survival outcomes. We can see that the trials with palbociclib, PALOMA-2, PALOMA-3, unfortunately did not have improvement in overall survival. The trials with ribociclib, the MONALEESA series, all did have a significant improvement. And we can put MONARCH-3 into context here that this did not meet statistical significance.

But I think it’s really interesting, again, to look at the numbers, with the improvement of 13 months, from 53.7 to 66.8 months, in MONARCH-3. If you look 1 line down to MONALEESA-2, a study with a very similar design, just using ribociclib, we can see that the numbers are almost identical, about a 13-month improvement. And so it’s interesting to think about we see very similar activity, very similar improvement, and yet the significance is a little bit different. And this ultimately may reflect trial design and not so much the specific agent.

I’ll also point out that we did see at San Antonio this year a rapid update of a study called PARSIFAL-LONG. This was a study that was with palbociclib in the first-line setting using palbo with fulvestrant or palbo with AI. And there was no difference in the arms from the original PARSIFAL study, but they combined them to do long-term follow up, and they presented that data at San Antonio. The median overall survival in that study was 65 months. Gosh, that’s pretty similar to what we see in MONARCH-3 and MONALEESA-2, quite different than what was seen in PALOMA-2, which was the official first-line study.

And from my perspective I think that just speaks to a lot of the variability and statistical design and power that exists among all of these trials, making it really hard for us to know exactly with precision what are the relative benefits. That being said, I think we still continue to see a lot of activity from abemaciclib from the MONARCH-3 update.

And so in summary for MONARCH-3, again this was a smaller trial than the other first-line trials, it had a different statistical design, including the 2:1 randomization, and this presents some statistical challenges versus the other larger trials, and some power issues, particularly with long-term follow up. We also know whenever we look at overall survival benefits can be washed out with subsequent therapies, and that again weakens power. But I do think it’s really neat to look at this 1 in 5 patients without progression at 8 years. I think that really speaks to the power of the drug. And the PARSIFAL-LONG data makes me think about palbociclib as another option, which some of us may not be using that much now after that PALOMA update from last year.

For the time being I think many of us are selecting ribociclib as our CDK4/6 inhibitor of choice in the first-line setting given the positive overall survival benefits with that drug, but I find that the MONARCH-3 update is not making me think any less of abemaciclib. I think this is a reasonable alternative, particularly if ribociclib is not feasible in this setting.

So let’s move on now, and we’re going to talk about other updates from another exciting study, CAPItello-291. So CAPItello-291 is looking at a drug called capivasertib, which is an oral AKT inhibitor. This was a study in pretreated patients with metastatic hormone receptor-positive, HER2-negative disease who had had up to 2 prior lines of endocrine therapy, and they wanted at least half the patients with prior CDK4/6 inhibitor therapy.

I’ll point out just a few important details. So patients in the study were randomized to capi and fulvestrant or placebo and fulvestrant. The capi dosing is 400 mg twice a day. Now this has an interesting dosing pattern, 4 days on and 3 days off, and this is intentional to help avoid buildup of side effects and toxicity.

I’ll also point out that in the eligibility patients were allowed to enter with diabetes not requiring insulin and with a hemoglobin A1c less than 8%. And this is different than the trial SOLAR-1, which is for the other agent that targets the PI3 kinase/AKT pathway, which is alpelisib, which did not allow patients with diabetes into the study. So this is more flexible in terms of the eligibility.

All right. So let’s look at what we found in this study. So the primary endpoint was progression-free survival in the overall and in the AKT pathway-altered population, which was about 40% of the patients who entered the study. In the overall population there was an improvement in progression-free survival, 3.6 months to 7.2 months, hazard ratio of 0.6. And in the AKT-altered population, meaning an alteration in PI3 kinase, AKT, PTEN, there was a similar improvement, 3.1 to 7.3 months, and an even more favorable hazard ratio of 0.5.

And based on the strength of these results this very recently led to the approval of capivasertib in combination with fulvestrant in patients with an AKT pathway mutation. This approval just happened around November of 2023, so that was fairly recent at the time of this recording, so we are just beginning to bring capivasertib into our clinics.

Regarding the side effect profile, notable side effects included diarrhea and rash compared to the control arm. And I’ll point out that a vast majority of patients experienced some-grade diarrhea; about 10% experienced Grade 3 diarrhea. The rash numbers here are maybe a little bit of an underestimate because there are other categories for skin alterations that could be included, and so the rates of rash may be a little bit higher. We do know that a number of patients had to take either topical or even oral steroids to manage rash. And so that may be more of a toxicity issue.

But I’ll also point out hyperglycemia rates were low. And remember, this was a trial that allowed patients who had diabetes or an elevated hemoglobin A1c to enter, and even with that more broad population rates of hyperglycemia were 16% all grade and only 2% with Grade 3; very different than what we see with alpelisib.

And so we know that there was more dose holds and dose reductions in the treatment arm, but only 13% of patients had to discontinue for adverse effects. I’ll also add that unlike some of our efforts now using prophylactic medicines there were no mandated prophylactic medicines as part of the CAPItello study. And if one perhaps used prophylactic, for example prophylactic antihistamine, maybe the rates of rash might be lower.

So we did see an update at San Antonio looking at outcomes based on the specific AKT/PI3 kinase/PTEN mutations, and in that subgroup of patients who had a mutation this was looked at based on whether the mutation was PI3 kinase, AKT, PTEN, or if they had different combinations. And very reassuringly the benefits of capivasertib were maintained across all of these groups. So if you have a patient who has let’s say an AKT mutation but not PI3 kinase, and you’re wondering will I get the same benefit, the answer is yes. Or if they have PTEN the answer is yes. So I find this very reassuring that as we look at our genomic testing, and if we see these mutations we know that no matter what the mutation we can still be offering capivasertib.

So the summary for capivasertib is that this combination is now an approved therapy for pretreated hormone receptor-positive, HER2-negative disease with a mutation in the PI3 kinase/AKT/PTEN pathway.

Now, you need to know your patient has this mutation, so you need to get tumor genomic profiling. Many of us are getting this at baseline using a tissue assay, but we can also be getting it with ctDNA. This performs extremely well to provide this information for us. And I think as we consider the agents that target this mutational pathway the toxicity profile can be a differentiator. Capivasertib can cause diarrhea and rash, to a lesser extent nausea and fatigue, but importantly less hyperglycemia than what we see with alpelisib. And as I’m offering this to patients I’m thinking about giving prophylactic antihistamine to help prevent that rash and telling patients to use antidiarrheals aggressively if they begin to have diarrhea. So I am now sequencing this before I would be offering a patient alpelisib.

DR LOVE: So I’m just kind of curious whether or not this kind of intermittent, 4 days on 3 days off, schedule has been looked at with alpelisib.

DR MAYER: I’m not aware of efforts to look at that intermittent schedule with alpelisib. I think most of the efforts to prevent and manage toxicity with the drug have been built around appropriate patient selection and use of prophylaxis. For example, there was as study called METALLICA that looked at using prophylactic metformin for a week before beginning alpelisib, and that study showed a lot of success in reducing rates of Grade 3/4 hyperglycemia. Now we do know that metformin can cause some diarrhea, and so it’s a little bit of a tradeoff there. But that dosing schedule has not been looked at.

I will say that it’s one thing to be in a trial and have patients being watched very closely to use that dosing schedule. It’s another thing to have real-life patients, some of whom struggle taking a daily medicine once a day, in what we think of as the most easy way to take a drug, learning how to take it intermittently. So we’re going to have to use all of our resources in clinic, all of our patient education resources to make sure that patients take the drug in the correct schedule, otherwise they may increase their risk of toxicity.

DR LOVE: Yeah. I heard Adam Brufsky, I worked with him a few weeks ago, he said he tells his patients take the weekend off. But I think 3 days off, that’s not going to be — I think that could be a problem. Have you seen that as an issue so far, or is it too early to say?

DR MAYER: Yeah. I think it’s a little early to say because we’ve only had the drug available for like 6 weeks at the time right today. But what we’re doing at Dana-Farber is using our pharmacists who do drug education, our APPs who do drug education, and we’re probably going to do a lot of direct outreach to patients to check in with them and make sure are you taking it properly. I also hope that there can be a way for the drug to be dosed in blister packs, which I think is a much better way for people to keep track of what they’re taking than just in a bottle.

DR LOVE: Yeah. Right. Right. Yeah. That’s a great idea. Yeah because it’s hard to think of another situation like that. I’m trying to think of another drug in oncology. I’m sure there are. I just can’t think of one right now. But anyhow, yeah. That’s very interesting.

DR MAYER: Yeah.

DR LOVE: All right. Whenever you’re ready please continue.

DR MAYER: All right. Let’s move on.

So one of the real hot topics at San Antonio this year was a late-breaking presentation of the INAVO120 study. INAVO120 studies inavolisib (GDC-0077), which is an alpha-specific PI3 kinase inhibitor. This was a study looking at a triplet. The study enrolled what I would consider very high-risk, first-line, hormone receptor-positive, HER2-negative patients whose tumors have a PIK3CA mutation. Now I call this high risk because these were patients who progressed during or within 12 months of completion of adjuvant endocrine therapy. So these are the patients recurring on their aromatase inhibitor where you are really quite worried about endocrine resistance, plus they have a PI3 kinase mutation.

So these patients were randomized to receive the treatment arm of inavolisib plus palbociclib plus fulvestrant versus placebo plus palbociclib plus fulvestrant, so palbociclib/fulvestrant, a very reasonable first-line option for these patients compared to a triplet, which is a little unusual. We don’t use too many triplets. And what was shown in this study was that for the primary endpoint of progression-free survival there was a significant improvement, from 7 months to 15 months, with a hazard ratio of 0.43, so close to a 60% improvement in progression-free survival. So this is a positive study.

I will point out that when we look back on the first-line studies of CDK4/6 inhibitors our improvement from the placebo to the treatment arm tends to be more in the range of like 12 months to 24 months, and here we see the lower numbers, and I think that speaks to the high-risk nature of this population and that when there is a PI3 kinase mutation these cancers can be more resistant at baseline. But even in this more resistant group we are seeing a benefit. I’ll also point out that if you look at the very beginning of the Kaplan-Meier curve for the control arm there’s a drop off that happens pretty quickly. We avoid that with the use of the triplet, so a lot of activity seen there.

Regarding side effects, the inavolisib does cause several side effects, including stomatitis. We do see some rates of hyperglycemia, although perhaps not as much as we see with alpelisib, some diarrhea, nausea, rash. This is typical for an alpha-specific PI3 kinase inhibitor. But in general, I would say that this was a relatively manageable toxicity profile, and when we think about toxicity when we use triplets, this is where some of the triplet studies have not been able to move forward because of excessive toxicity. This study was manageable and able to complete, and I think that speaks to palbociclib as an ideal candidate for combinations since of the CDK4/6 inhibitors it tends to be a well-tolerated drug. I’d also point out that there was no toxicity prophylaxis used in this study, so rates of these toxicities might be better if some prophylaxis was used.

So very reassuring to see outcomes improve in a very high-risk cohort of patients. This, again, shows us that we need to know the PIK3CA mutation status from the time of diagnosis, so when we are getting that baseline biopsy or blood work to confirm the patient has metastatic disease we need to immediately get tumor genomic profiling to know if PI3 kinase mutation is present. And palbociclib is a CDK4/6 option for metastatic disease, and I think it’s really an ideal partner when we’re thinking about triplets.

There was a very nice discussion of this data from Hope Rugo at San Antonio, and she pointed out some features of the patient population that are unique, including that this was a very tightly regulated patient group. They had a low BMI, there wasn’t a lot of diversity in the group, and considering the hyperglycemia and mucositis that was seen in the toxicity profile this might be different in a more real-world group of patients, particularly where there’s a greater range of BMI. Additionally, we do have a whole new generation of mutant-specific PI3 kinase inhibitors that are really designed to target the mutant PIK3CA and not wild type and dramatically reduce rates of hyperglycemia, and that’s really going to be the next generation of where we’re going with these drugs.

DR LOVE: So a couple quick questions. How much second-line and beyond data do we have on inavolisib?

DR MAYER: At this time, not a great amount. There are other studies that are looking at this drug not only in hormone receptor-positive disease but also in HER2-positive breast cancer where patients who have PIK3CA-mutated tumors also have somewhat inferior outcomes. But we’re just starting to learn more about how this drug performs.

DR LOVE: The other thing that’s almost shocking is it looks like there’s like the control group had 80% relapse rate in 18 months.

DR MAYER: I think that that red curve, the control arm here, really speaks to how this is a high-risk group. A little earlier when we were talking about MONARCH-3 and how we can have 1 in 5 patients being disease free at 8 years, that’s one of our more favorable hormone receptor-positive groups of patients. This is the flipside of that. This is really our most bad actors, and we can see that these patients are really not doing very well, and so the idea of escalation to a triplet makes sense.

DR LOVE: No. I mean it’s kind of amazing, that kind of prognosis. 80% relapse in 18 months.

DR MAYER: Yeah. I think when we consider hormone receptor-positive disease as a group it’s our largest subset of metastatic patients, and it’s really heterogeneous. And anyone who’s in clinic knows this because you have patients where the path report looks somewhat similar and yet dramatically different outcomes. But I really think tumor genomics here is beginning to unlock this for us and help us from the jump identify the bad actors and escalate when we need to, and then on the flipside identify people whose disease is more latent and more indolent, where something like this would not be appropriate. So I think this really will help us much more with treatment tailoring.

DR LOVE: I’m just kind of curious. Right now if you could, would you use this strategy?

DR MAYER: I find this data very compelling, and I think exactly as you point out, the inferior outcomes with the control arm, we see this in clinic. We see these patients where we are really hoping they’re going to get years of disease stability on their first-line CDK4/6 inhibitor, and they don’t, and it’s very discouraging for us as providers and for our patients.

And so if I know that my patient would be on that red line, and I know I have a way to help them do better and have over 50% improvement in progression-free survival, keep them going twice as long on their agent, I would absolutely do this, but I think that the toxicity profile is significant. And so again, it was a very tightly regulated population. I think we need more information about how to prevent and manage the side effects so that patients can comfortably stay on this and have a good quality of life.

DR LOVE: I wonder what you’d do with your patients who are not PIK3 positive in that situation.

DR MAYER: Who have resistance but no PI3 kinase mutation. Yeah. I mean I think right now our plan is to give first-line endocrine and CDK4/6 inhibitor, but as we’ll talk about in a moment we’re going to be looking for other resistance mutations and other strategies that we can use.

And speaking of that, let’s move along to talk about a whole new category of drugs, the oral SERDs. And we saw some updates from the EMERALD study, which is a Phase III study that’s looked at the oral SERD elacestrant. This was a trial for pretreated patients, all of whom had had prior CDK4/6 inhibitor, hormone receptor-positive, HER2-negative disease, randomized to elacestrant monotherapy or endocrine therapy of provider choice.

Patients who were enrolled in this study were very, I would say, real world. All of them had had CDK4/6 inhibitor, about 70% with visceral metastases, about 20% with a line of chemotherapy, so it really does resemble what we see in clinic.

The overall study for intention to treat was a positive study, but I think the better way to look at the data was an updated analysis in patients who had an ESR1 mutation looking at the duration of prior CDK4/6 inhibitor. So outcomes were more significantly positive in patients with an ESR1 mutation, which we’ll talk about in a moment, is a preferred mutation for oral SERDs. And as we can see from these figures patients who had longer duration of time on their prior line of therapy, which had CDK4/6 inhibitor, had greater benefits from the use of elacestrant. In particular patients who were on for at least a year beforehand had an improvement in their progression-free survival from about 2 months on the standard of care endocrine to close to 9 months with elacestrant, hazard ratio 0.4.

And so I think as I consider use of elacestrant in my clinic I find this data very helpful. And it’s not really about the CDK4/6 inhibitor here. It’s really about endocrine sensitivity. If a patient was on their prior line of therapy for at least a year that’s indicating that their disease is endocrine sensitive, and they are more likely to respond to another line of endocrine therapy. So I think this is a very helpful way for us to identify best candidates. The drug is approved for pretreated patients whose cancers have an ESR1 mutation.

And so we saw updated data at San Antonio this year looking at biomarkers and subgroups. So they looked at patients who had ESR1 mutations as well as an additional feature. And I highlight here 2 of the categories that I think are particularly helpful. First of all, I’ll just say that all of the patients demonstrated benefit from the use of elacestrant over standard of care. These are all the patients with ESR1 mutations who had had at least a year of prior CDK4/6 inhibitor in their prior line of therapy.

Now patients who have both ESR1 mutation and PIK3CA mutation did have a benefit from about 2 months to about 5 1/2 months, so elacestrant was active in this group and provided benefit.

You’ll notice that that 5 1/2 months seems to be somewhat less than the other categories that were examined here, and that again speaks to the underlying resistance of patients who have cancers with PIK3CA mutations. But I find this very helpful because if I have a patient who has both mutations, and I’m wondering should I offer that patient elacestrant or should I offer that patient something that targets the PIK3CA, be it capivasertib or alpelisib, I feel comfortable offering elacestrant in that situation, particularly if the patient has perhaps lower-volume disease, no visceral disease, has somewhat more latent disease, then I think it’s very reasonable to offer elacestrant. If the patient seems to be in more of an impending crisis situation then endocrine with targeted agent may be a preferred choice. But I think this is very helpful to manage those patients.

Additionally, for patients with ESR1 mutation and HER2 low we see a real nice improvement, 2 months to about 9 months, with elacestrant and again suggesting here there’s no reason if they’re HER2 low to start thinking about T-DXd in this setting; we can stay with endocrine therapy. So I think this is very helpful data.

So the EMERALD update is that elacestrant monotherapy is an option for patients with hormone receptor-positive metastatic disease and an ESR1 mutation and who have retained endocrine sensitivity, which we can identify based on how long they were on their prior line of therapy. In that situation the ER pathway is still the driver of disease even if there is coexistence of another mutation or HER2 low, and so we can use this to prioritize elacestrant before the other options. I think that’s really nice because elacestrant’s a very well-tolerated agent.

Now I would love to use elacestrant in combination with a targeted therapy. There are studies. For example the ELEVATE study is looking at these combinations; so we await this data.

And a very important thing to note is that ESR1 is a kinetic mutation. This develops over time, and this means that we need to check the ESR1 status at the time of progression in patients with metastatic hormone receptor-positive breast cancer. And we know this from a variety of experiences looking at the rates of ESR1 mutations at different timepoints in treatment. We know that patients in the first-line setting, for example the MONALEESA-2 patients, have a very low rate of ESR1 mutation, just 4%, but in studies looking at pretreated patients that rate is much higher, somewhere between 25-40%. So this is a mutation that develops over time. It’s kinetic, and that means that we can’t just check at baseline and not check again. We have to check continuously to know if that mutation is developing.

Additionally, how do we check? Well, we really do think that liquid biopsy or ctDNA would be a preferred way. If you look on the left you can see that if you compare looking for ESR1 based on tissue or based on liquid the purple bar in the ESR1 column indicates where there’s discordance between tissue and liquid, where the tissue is negative, and the liquid is positive, so ctDNA is a better way to look for ESR1 than doing a biopsy.

And another reason why it’s good is that as we know, cancers are not homogeneous. They’re not the same throughout the body. There’s heterogeneity, and this can include the different parts of the cancer can have different resistance mutations. CtDNA is looking at the whole cancer. It’s an overview as opposed to like a liver biopsy, which is only looking at the liver. So it really is the best way to provide the overview.

And this year we did have an update, a rapid ASCO Guideline update that Hal Burstein and his team published looking at how best to treat hormone receptor-positive, HER2-negative disease recommending routine testing for ESR1 mutations at time of recurrence or progression on endocrine therapy. We should do tissue testing or blood testing at the time of progression because the ESR1 mutations develop over time. And importantly that if you test again at time of progression, and the ESR1 mutation is not there, you’re going to check again in the future. We just have to continue doing it. And I think a paradigm shift for all of us is building this into our framework of how we take care of patients and knowing that we have to do this routine testing.

So why is this important? Well, we have this category of therapy called oral SERDs, and we know that these are agents that are particularly helpful in the setting of an ESR1 mutation. They can bind to the mutated estrogen receptor, not only blocking estrogen ligand stimulation, but they also can lead to degradation of the estrogen receptor, so can be more potent than other therapies. And we don’t just have the oral SERDs, but we have oral SERMs, such as lasofoxifene, and the exciting agents in the PROTAC and CERAN categories, so we have a whole collection of new endocrine agents which are currently in development.

Let’s just take a look at some of the ones that are in development. So we have seen recent updates of the SERENA-2 study. This is a trial that’s looking at the oral SERD camizestrant. This is a Phase II study that enrolled pretreated patients to be randomized to get fulvestrant monotherapy or different dose levels of camizestrant, 75 mg or 150 mg. There was a higher dose, 300, and that closed early.

The study population again I would describe as real world in this study. About half of them had prior CDK4/6 inhibitor, about 60% had lung or liver metastases. And if we look at the intention-to-treat results for progression-free survival patients who received camizestrant had an improvement in PFS compared to those who received fulvestrant monotherapy, from about 3.7 months with fulvestrant to about 7 months with the camizestrant arm, with a hazard ratio around 0.6, so camizestrant outperformed fulvestrant in the pretreated setting. And specifically if we break this down by whether the patient had an ESR1 mutation detectable at baseline or not we can see that the benefit here is really driven by the cancers with the ESR1 mutation, where we see an improvement from 2.2 months to as high as 9 months with camizestrant. In the ESR1 wild-type population we really don’t see a significant difference amongst the arms.

And so we saw updated data at San Antonio this year looking at outcomes in patients who had 1 ESR1 mutation, more than 1 mutation, or specific mutations, including the D538G or Y537C mutations. And if you can see these Kaplan-Meier plots you’ll see that there is a preserved benefit of camizestrant in all of these populations over fulvestrant. So camizestrant monotherapy has shown promising data over fulvestrant, particularly in those with an ESR1 mutation, and it doesn’t matter which mutation or how many mutations are present.

So there are many ongoing studies with camizestrant, including looking at combinations. There’s other studies looking at other oral SERDs, including giredestrant and imlunestrant, which we’re going to take a look at. And then we also have the other agents, the PROTAC ARV-471 and the oral CERAN OP-1250, which have also shown some really exciting new data. So overall I think this is going to be a very rapidly expanding space in the next 1 to 2 years. We’re going to see a lot of data coming out for these novel endocrine agents.

And just to take a bird’s eye view you can see that there are many trials going on with all of these agents, and I’ll actually point out in red are the very large, randomized adjuvant trials which have launched. With giredestrant we have lidERA, with camizestrant we have CAMBRIA-1 and CAMBRIA-2, and with imlunestrant we have EMBER-4. And these are trials that are designed to capture tens of thousands of patients, so we have a lot more data coming out about these agents in the next few years.

So now let’s turn to imlunestrant. And we had updates of the EMBER study from Komal Jhaveri. This is a multi-arm study, and we saw updates looking at imlunestrant monotherapy, as well as combinations, imlunestrant and abemaciclib, and even a triplet of imlunestrant, abemaciclib, and aromatase inhibitor. And the randomized portion was in patients who were less heavily pretreated, no more than 1 prior line of therapy for advanced disease, no prior CDK4/6 inhibitor.

And here are some waterfall plots looking at these groups. First on the left is the imlunestrant monotherapy at the recommended Phase II dose. This is a population of patients who was more heavily pretreated, prior endocrine therapy for advanced disease, prior CDK4/6 inhibitor, and the response rate in this group was 12%, clinical benefit rate 55%, PFS 7.2 months. 7.2 is very respectable. We’ve seen PFS post CDK with fulvestrant in the 2-3-month range. We just looked at that. And here 7.2 months is really quite robust, so that’s a very nice efficacy signal.

In the combinations you can see very nice-looking waterfall plots where almost every patient is experiencing some element of response. Response rates here in the 30-60% range, clinical benefit rate in the 70s, and the median progression-free survival of at least 20 months, and we would expect that with the inclusion of the CDK4/6 inhibitor in this setting.

In terms of side effects imlunestrant is associated with some GI side effects, including diarrhea, and if we combine this with abemaciclib, which also has risks of diarrhea, that’s something for us to be mindful of. In the combinations with abemaciclib there was a rate of all-grade diarrhea that was between 70-90%, so almost every patient had some degree of diarrhea. The rate of Grade 3 diarrhea was 10% or less, and that would be the toxicity to pay attention to in these types of combinations.

So the summary for EMBER is that imlunestrant monotherapy in general is well tolerated, with nice post-CDK activity, the drug can be combined with abemaciclib and even with aromatase inhibitor with no new safety signals, and that there are large Phase III trials that are now ongoing looking at imlunestrant. This includes EMBER-3, which randomizes to imlunestrant, provider-choice endocrine therapy, and a combination arm of imlunestrant and abemaciclib post CDK, so that’s actually a CDK after CDK study. And EMBER-4, as we just alluded to, is a very large adjuvant trial with imlunestrant.

So if we think about how to put this all into context for hormone receptor-positive disease, in the first-line setting many of us are giving endocrine therapy with a CDK4/6 inhibitor, but in the second- and third-line setting here’s where we have a lot of choices. We can give endocrine therapy with CDK4/6 if we didn’t do it already. If there’s a BRCA mutation we can give a PARP inhibitor. We can give an everolimus combination, and that is not mutation selective. But everything else we need to know the mutation status. If there’s a PIK3CA/PTEN/AKT mutation we have capivasertib, we have alpelisib. If we have an ESR1 mutation we have elacestrant. And so I think this really, again, supports the need to know tumor genomic profiling as we move down the pathway.

So now let’s cross over and talk a little bit about antibody-drug conjugates, which is a hot area throughout all of oncology and certainly in breast cancer. We have a whole variety of agents that are either approved or under study. First we have the HER2-targeted agents, the original OG ADC in breast cancer, T-DM1, and then the next-generation agent T-DXd, which targets HER2 with a DXd payload. We also have the TROP2-targeting antibodies. This includes sacituzumab, which has the SN-38 payload, which is like irinotecan, and a new drug in development dato-DXd, which like sacituzumab targets TROP2 but has the T-DXd payload of DXd. And we’re briefly going to touch on patritumab, which is a HER3-targeting antibody that also has the DXd payload. So there’s a lot going on in the ADC space.

So first let’s review overall survival updates from the TROPiCS-02 study. So we’re talking first about hormone receptor-positive, HER2-negative disease. This was pretreated patients that had at least 1 line of endocrine therapy, they’d had CDK4/6 inhibitor, and they’d had 2-4 lines of chemotherapy in the metastatic setting or if they rapidly progressed in the adjuvant setting randomized to sacituzumab or treatment provider choice, and this was cape, vinorelbine, gemcitabine, or eribulin. And updated data on progression-free survival continues to show an improvement in PFS from 4 months to about 5 and a half months, hazard ratio 0.65, and this is what supported the approval of sacituzumab in this space.

Overall survival is positive as well. Patients who received sacituzumab have a median overall survival of 14.5 months, improved over treatment provider choice, which was 11.2 months, hazard ratio of 0.79, so about a 20% improvement in overall survival, and this was statistically significant.

So extended follow up of TROPiCS-02 showed that improvements in PFS and OS are durable and that this — there was also some subgroup analysis showing that sacituzumab improved outcomes in subgroups, including depending on TROP2 level, HER2 low, HER2 0, so all of these patients benefitted from sacituzumab. And so this reinforces that this is an agent for us to consider in endocrine-resistant, pretreated hormone receptor-positive disease.

Now, we’ve also seen updates from DESTINY-Breast04. This is a study using T-DXd, so targeting HER2, and this was a study that enrolled patients who had HER2-low disease, which means IHC 1+ or IHC 2+ and FISH negative. This was a study for patients who had up to 1-2 prior lines of chemotherapy in the metastatic setting and enrolled mostly hormone receptor-positive patients, although there was a small population of triple-negative patients who enrolled, as well; triple negative with HER2-low.

And then overall survival updates in both the hormone receptor-positive cohort, as well as all patients, continued to show substantial improvements, from about 17 months to about 23-24 months, hazard ratio around 0.7, so significant improvement in overall survival and continued dramatic improvements in progression-free survival from about 4 months to about 9 months. So a highly active agent for these patients. Follow up on the hormone receptor-negative small cohort showed results that really mirror what we see in the hormone receptor positive in terms of the benefits gained.

Now we did see updates looking at subgroups, including those with PIK3CA, ESR1, intrinsic subgroup, CDK4/6 resistance marker status, and across all of these subgroups we continue to see benefits from the use of T-DXd over treatment provider choice. I think it’s important to show this data because we do know these things about our patients, although as we’ve talked about knowing PIK3CA and ESR1 is a little more important when we’re picking endocrine therapies. By the time we’re into chemotherapy these are of lesser importance, but it’s very reassuring to see how broadly across the board T-DXd is improving outcomes.

And so results of this long-term follow up of DESTINY-Breast04 show the benefit of T-DXd over TPC in HER2-low populations, benefit is regardless of PIK3CA or ESR1 mutation status, and this has really become our preferred agent in the second-line and beyond setting. So first line we still don’t have data from DESTINY-Breast09, which is looking at activity in the first-line setting, but for the time being T-DXd lives in the second-line setting for our patients with HER2-low disease. If they are not a candidate for T-DXd then we might be considering sacituzumab based on the TROPiCS-02 study.

We also saw data from TROPION-Breast01. So this is looking at that new antibody-drug conjugate dato-DXd, which targets TROP2 but has the DXd payload. This was a study that was in patients with hormone receptor-positive disease, pretreated, 1-2 prior lines of chemotherapy, but this was not for HER2 low the way we think of for T-DXd. This was just HER2 negative. Patients were randomized to dato versus investigator choice chemotherapy, eribulin, vinorelbine, gemcitabine, or capecitabine, and results of the study showed a significant improvement in progression-free survival, 4.5 months to 6.9 months, with a hazard of 0.64, so definitely a positive study.

In terms of adverse events, neutropenia was an adverse event of clinical interest. Not that dato causes neutropenia but to show it actually doesn’t really cause neutropenia, in contrast to chemotherapy, which certainly does. So we see less neutropenia with the dato than with chemo. We do know though that dato can cause stomatitis. In fact, there was about half the patients had any-grade stomatitis, about 6% with Grade 3, and this was a relatively frequent cause for dose reduction. So that is an adverse event that is quite notable for the agent.

So dato improved PFS compared to treatment provider choice in the second- and third-line setting. We await the more mature overall survival data. There’s less neutropenia than chemo but more stomatitis. There’s also some dry eye which you see with dato. So this could become our fourth ADC approved for breast cancer.

And the tricky thing with all of the ADCs is to understand how to sequence them. So for example with sacituzumab they both target the same receptor, so if we have both available which one are we going to pick as the preferred agent? And there are trials that are launching now that are looking at sequencing strategies trying to figure out if we can identify best candidates for one strategy versus another. But if dato’s approved it’s going to be ending up in a very similar place to sacituzumab, so we’ll have to figure out how to select one agent versus another.

And then we’ll move on to the trial of patritumab, which is a HER3-targeting ADC with the DXd payload. This is a study with several parts to it, an initial part of monotherapy in HER2-negative breast cancer, and then Part B and Part Z, which are currently enrolling, is looking at expansion in groups based on HER3 expression, based on ER expression, and post T-DXd.

Now HER3 is expressed in about 30-75% of breast cancers. It’s quite common. Overexpression is associated with a poor prognosis, however. And we know that the ADC can target HER3 and brings the DXd payload.

So here’s the outcome from Part A of the study that Erika Hamilton presented this past year. First if we look on the left, these are swimmers’ plots looking at outcomes based on HER3 membrane expression. And we can see that there’s some patients who are having extremely nice responses to this agent, but the responses are actually seen in all of the categories of HER3 expression. Even patients who have lower expression there’s still those who have over 6 months of disease stability with the drug.

And if we look at the waterfall plot on the right we can also see that outcomes do not appear to be significantly linked to having higher levels of HER3. So it may not be that knowledge of HER3 itself is necessary, but I think we’re going to learn more from the expansion cohorts.

And these are the Kaplan-Meier plots for progression-free and overall survival for the patients in the study, again showing that all of the subgroups for HER3 expression are having improved outcomes or having nice outcomes with this agent.

The safety profile is notable for nausea, fatigue, diarrhea, many of the things that we tend to see with the DXd-targeting ADCs. But patritumab is active regardless of level of HER3 expression. These responses appear to be durable. And I’ll point out that there’s very little ILD to date, which is also very favorable given what we’ve seen with T-DXd. But again, it’s unclear how many patients would respond to this drug after prior treatment on T-DXd or even dato-DXd, and figuring out how to sequence the ADCs and how this drug in particular would fit into the sequence is tricky as well. So that’s going to unlock a whole new generation of clinical trials for us looking at these sequencing strategies.

So this brings us to the end of our updates from 2023. It’s quite a whirlwind, and I think, as you see, there’s a lot of new data, a lot of new agents, some great new agents that are helping our patients today, but a lot of questions that come up about how to insert these into our treatment paradigms and how to pick best candidates. So I very much look forward to 2024 and hope that we’re going to see a lot more discovery and development that will continue to refine our portfolios and paradigms for our patients.

DR LOVE: So that was awesome. Just 2 quick follow up questions. I’m curious to what extent you’ve had any experience with dato and how you find mouthwash prevention.

DR MAYER: So thank you. We do have dato available in our clinical trials, and the mucositis is real. And we are instructing patients to use the prophylactic steroid mouthwash, the same one that we use with everolimus. If we think back to the old SWISH study that Hope Rugo led, that was so successful in showing us that prophylactic steroid mouthwash prevents mucositis. It’s a wonderful tool that we can really expand to use for other drugs that cause mucositis. So it is something that is included in ongoing dato protocols.

DR LOVE: But have you yourself seen that it works?

DR MAYER: It does work. And so patients do use it, and they have less mucositis. I will say that some toxicities we see with cancer therapies occur in the beginning, and once you get over the hump then you can kind of taper off your prophylaxis. It’s not clear what the timing and kinetics of the mucositis toxicity is, whether it’s just in the beginning and then you could taper off, or whether it’s something where a patient would need to continuously be using the mouthwash. We offer the mouthwash. We tell people to use it 4 times a day. That’s a lot, and so it’s nice when you can get over the hump and then taper down. And we don’t know yet what the rhythm will be with dato.

DR LOVE: Yeah. We get a lot of questions from the lung people about this. So do you find that the mouthwash is similarly effective as it is with, say, everolimus?

DR MAYER: I definitely think that it helps treat and manage mucositis. It’s not going to make Grade 3 mucositis go away. That may require a dose reduction. But it definitely can delay the appearance of mucositis and diminish the seriousness of the mucositis, so it’s a very helpful tool to have available.