We are going to start with a patient case.

[00:34:19]

Patient Case: A 67-Yr-Old Woman With de Novo MBC

This is a 67-year-old lady who was diagnosed with de novo metastatic breast cancer to her bones 3 years ago. At that time, her bone and breast biopsy revealed adenocarcinoma, which was ER-positive, PR-positive and HER2-negative. She had tumor genomics at the time that did not reveal any actual mutations. She started treatment with ribociclib plus letrozole and now 3 years later, her scans show disease progression in her bones. At this time, ctDNA analysis of the patient's blood sample is performed.

[00:34:55]

Pretest 1

Pretest 1, which of the following is the most likely molecular alterations that will be seen on the ctDNA analysis?

1. ESR1 mutation only;
2. ESR1 and P53 mutations;
3. PIK3CA mutation only; or
4. PIK3CA and ESR1 mutations.

Okay. We will talk about this during the talk.

[00:35:42]

Changing Landscape of Targeted Agents in ER-Positive Metastatic Breast Cancer

This is kind of our, I would say, ever-changing landscape of targeted agents in ER-positive metastatic breast cancer. Tamoxifen has been around for a long time. Then we had the aromatase inhibitors, then fulvestrant, then everolimus. But as you can see on this slide, in the last I guess 8 years or so, we've had multiple agents approved including of course the CDK 4/6 inhibitors agents that target the PI3 kinase pathway like alpelisib, elacestrant that targets ESR1, capivasertib that targets the PI3 kinase, AKT, altered pathways. Then most recently inavolisib. Erica is going to cover a lot of the pivotal trials as is Joyce later on in the event.

[00:36:27]

Significant Heterogeneity in Outcomes With Dual CDK4/6 Inhibitor + Endocrine Therapy

What we are looking at here is on the left-hand part of the slide is the MONALEESA-3 study. This was patients either treated at first line or second line who received fulvestrant without ribociclib for hormone receptor–positive HER2-negative advanced breast cancers. The progression-free survival you can see here was about 13 months in the placebo group, was increased to 20.5 months in the ribociclib group. This is very consistent with all of the studies that we've seen previously. But what was kind of interesting, and I think we've probably all seen this in our practice, is when the INAVO 123 study came out. Just to remind you, these were patients who had hormone receptor–positive, HER2 negative metastatic breast cancer that harbored API3 kinase mutation, but very importantly had either developed metastatic disease on adjuvant endocrine therapy or within 12 months of receiving adjuvant endocrine therapy, so you would imagine endocrine resistant disease. They were randomized to receive fulvestrant plus palbociclib with placebo or with inavolisib, which is one of the PI3 kinase inhibitors. When you look at this study here, what you can see is that the inavolisib was associated with a progression-free survival of 15 months vs 7 months in the placebo arm. But what is very notable here is how different the placebo arms did on both these studies. This really made us think that not all patients that present with hormone receptor–positive metastatic breast cancer do the same in the first line setting.

[00:38:07]

Mechanisms of Resistance to First-line CDK4/6 Inhibitor + ET

There has been a lot of work trying to work out what are mechanisms of resistance or is there a biomarker that could tell us which patients are going to benefit from a CDK4/6 inhibitor. In truth, the only one that we really have had is estrogen receptor and there has been a lot of work done on this.

[00:38:22]

Emerging Taxonomy of Mechanisms of Resistance to CDK4/6 Inhibitors

This slide kind of gives an overview of what are some of the causes of resistance to the CDK4/6 inhibitors. I think as you all know, CD4/6 or G1 checkpoint kinases coupled to D cyclins, they promote cell progression from G1 to S phase through phosphorylation of protein Rb. This then triggers release of E2F, which then supports CCNE1 expression by increasing Rb phosphorylation and resulting in accelerated progression. The CDK4/6 inhibitors, obviously they stop, essentially, cell progression. Some of the resistance mechanisms that have been described are firstly hyperactivation of the target. Hyperactivation of CDK6 for example, loss of PTEN MAP kinase for example. The second one is loss of target, so loss of Rb for example is another mechanism. Then just like we've seen in endocrine resistance bypass mechanisms for example, increase in cyclin E, loss of P53 for example are all associated with resistance to these agents. But none of this has really panned out clinically as yet.

[00:39:31]

Predicting PFS on First-line CDK4/6 Inhibitor + ET: Machine Learning Model

But there was a very interesting presentation at the San Antonio meeting from the Memorial Sloan Kettering group where they used a machine learning approach to develop a model to basically predict how patients will do on first-line endocrine therapy with CDK4/6 inhibition. What they did was they took patients who had hormone receptor–positive, HER2-negative metastatic breast cancer who'd all received a CDK4/6 inhibitor with endocrine therapy in the first-line setting. They also had to have tumor sequencing before or within 3 months of initiating first-line treatment. They had about 1100 patients, about 760 in the training cohort, and 326 in the validation cohort. Using clinical factors and also genomic factors which were related to some of the resistance mechanisms I showed you on the prior slide, they came up with this mixed clinical genomic model that very nicely predicted how patients would do on first-line therapy with the CDK4/6 inhibitor and endocrine therapy. What you can see here is that in blue are patients who did really well. Patients with lower risk cancer I guess you would say, where their median progression-free survival was as you can see, 29 months. Conversely, this group down here in the purple, they only had a 5 month median progression-free survival, which is a little lower than we saw in the INAVO120 study, but definitely suggests there are a significant number of patients don't benefit from this approach. Then the intermediate group, as you can see shown here was split into patients that did a little bit better at 20 months vs those that had about 11 month progression-free survival, so really showing the heterogeneity of responses to this first-line CDK4/6 inhibition with endocrine therapy.

[00:41:22]

Role of the Multimodal Risk Model

Then they took it to see if it would actually improve on what we've traditionally used. Using conventional clinical risk criteria, which was treatment free interval less than 12 months and measurable disease, what they found was that there was patients who did well at 19 month progression-free survival, and then patients that did less well with 10 months progression-free survival. When they applied their clinical genomic machine learning model, what they found was in patients who were predicted to do well on first-line CDK4/6 inhibitor endocrine therapy, there was a proportion of patients that actually did quite poorly as you can see shown here in the purple and the green. This made up about 20% of the total population and 27% of the clinical low risk as you can see. You can see that there were patients that did very well in this group, but also patients that did poorly. Then conversely, when you look at the patients using the conventional risk who did the clinical genomic machine learning model, what you found was there were patients that actually did very well there with a median progression-free survival of 28 months that you can see shown here and 20 months in the intermediate better risk group. That made up about a third of the clinical high risk group as you can see shown here. Basically taking this together, their model was able to predict better than what we've done before in predicting how patients were going to do on first-line treatment.

[00:42:49]

Genomic Landscape of MBC

Let us talk now about the genomic landscape of metastatic breast cancer. This is somewhat of an old study where what they did was they had tissue from primary breast cancers, then also from metastatic tissue, and they went ahead and basically looked at gene expression in the metastatic cancers shown here. You can see TP53 is one of the most common mutations that is seen, ESR1, PI3 kinase as you can see shown here. But what is very important here is that TP53 and PI3 kinase, these are present at diagnosis. They're not acquired mutations. In contrast and really what is so important today, I think if any of you were at the session today, this was a lot discussion about this, are these acquired mutations. We will talk about ESR1 in a second, but that is actually a pretty common mutation for patients who have received endocrine therapy. Also, ERBB2, there is some data using neratinib in patients whose tumors acquire ERBB2 mutations, and also AKT1, which is a target for capivasertib, which I think Erica is going to talk about.

[00:43:57]

Prevalence of Actionable Alterations in Higher-Risk HR+/HER2- MBC

Just going back to the Memorial Sloan Kettering group. Remember I said there was a low, intermediate low, intermediate high, and high group. What was really interesting here is what they showed is that as patients had a higher risk cancer, either intermediate, intermediate high, or high, you can see that they appear to acquire more of these targeted mutations. PTEN, AKT1, obviously ESR1 as you can see shown here. You can see here that those mutations were much more likely to be present in the intermediate high and the high risk group compared to the lower risk group. That is important because this is what we are going to hear about. We have agents that target those particular mutations.

[00:44:35]

TP53 Mutation Rate With Progression on CDK4/6 Inhibitor + ET in HR+/HER2- MBC

Then TP53 as you can see, as you became more higher risk, in other words, we are going to do worse with the CDK4/6 inhibitor and endocrine therapy, patients were more likely to have P53 mutations as well.

[00:44:49]

Role of Biomarker Testing in ER+/HER2- MBC

In terms of biomarking testing, obviously we have 2 options. We can either do it off tissue, which we've done historically, and then more recently doing it through liquid biopsies for ctDNA. What we are looking at here is that if you look at ctDNA overall, it appears to pick up at least more ESR1 mutations compared to biopsies. Just to go through, this a little bit complicated, is that the orange is concordant negative; this is both the tissue and the blood are both negative. The yellow, the tissue and the blood are both positive. Then the purple and the green are discordant, where the tissue is positive or the ctDNA is negative or vice versa. When you look at AKT1, you can see most of them are concordant. However up here you can see there are definitely some samples where the ctDNA picked up an AKT1 mutation while the tissue wasn't present in the tissue. It is more notable for ESR1, as you can see shown here, fairly significant number here, and were picked up in the blood but not on the tissue. Even for PIK3CA, you can see there are some that are picked up in the blood and not in the tissue. The reason or maybe the rationale for this is that we've known for a long time that in patients with metastatic disease, not all the metastatic areas are the same. For example, some of them express more ER than others, but we know that some of them have mutations and others don't. But the thought is that the most dominant mutation you have will be detected in the bloodstream.

[00:46:26]

ESR1 Mutations in Advanced Breast Cancer

Just to stop at ESR1 mutations a little bit more. As I mentioned, they are acquired to, it says aromatase inhibitor therapy, but really I think it is probably endocrine therapy. They function in a ligand independent manner, so they're all in the ligand binding domain of estrogen receptor. They're very rare in primary cancers, but very highly enriched in metastatic breast cancer. This table looks at a number of different trials that have been done either in the first-line or pretreated setting, looking at the rate of ESR1 mutations. You can see in the MONALEESA-2 study at baseline, only 4% of patients had ESR1 mutations. When you start looking at these pretreated trials here, you can see much higher rate, 29%, 37%, 25%, 39% as you can see shown here. They become really quite common as patients develop resistance to endocrine therapy and probably CDK4/6 inhibitor therapy as well.

[00:47:22]

Utility of Tumor Genomic Tests in MBC

The question is when and how should we perform these tests? There's no, I think, right or wrong here, but I think we have to check them for sure. You have the option of the metastatic tissue or via the ctDNA liquid biopsies. I think a lot of times we use biopsies at the diagnosis of metastatic disease because we almost always do a biopsy to confirm metastasis in patients. But I think more and more what we are doing is rechecking genomics on ctDNA at disease progression, but even sometimes before there is disease progression. You can kind of predict what you're going to be given the patient when they develop disease progression. Of course, as I mentioned, this is all looking for these acquired alterations in ESR1, ERBB2 and PTEN and AKT1. Because PI3 kinase mutations are present from the beginning, you could check them in primary tissue if you want to, but there is probably no need to do that because we have ctDNA. Of course as we are going to hear about from Erica and Joyce, this is to determine what is the best treatment for our patients once they have had disease progression on a CDK4/6 inhibitor and endocrine therapy.

[00:48:31]

Biomarker Testing Methods

In terms of methods, so RNA sequencing provides a dynamic view of gene expression splicing and DNA sequencing captures a snapshot of the genome including substitutions, insertions and deletions. Most of the sequencing assays, of which we have quite a lot, actually test for both RNA and DNA sequencing using tumorous tissue or increasing more blood. Our guidelines from ASCO, ESMO, NCCN recommend genomic testing in blood or tissue at disease progression to look for these acquired mutations. CtDNA I think is probably optimal because it is not invasive, but also as I showed you, perhaps it picks up mutations that you may not see in tissue.

[00:49:13]

Key Ongoing Studies Using ctDNA Monitoring for ESR1m to Guide the Switch to SERDs

Now, we are probably going to talk quite a bit about this today, but can we use these liquid biopsies or the ctDNA monitoring to guide when to switch treatment for patients with metastatic disease?

[00:49:25]

PADA-1: Palbociclib + Fulvestrant vs Palbociclib + AI in Patients With Rising ESR1m After Palbociclib + AI

The PADA-1 study shown here was a study basically designed to see if you saw rising levels of ESR1m in the blood before there was disease progression on scans. If you switched treatment at that point, would that be more effective than just waiting to see if there is progression on the scans? What they did here was these are all patients with endocrine-resistant hormone receptor–positive metastatic breast cancer untreated with no evidence of visceral crisis. You can see with over 1000 patients and everybody got palbociclib with an aromatase inhibitor. Then they checked liquid biopsies or ctDNA for ESR1 mutations I believe it was every 2 months. Then if the ESR1 was rising and they did not have progression on their scans, they were basically randomized to stay on the aromatase inhibitor with palbociclib, which is a standard of care, or switch from the aromatase inhibitor to fulvestrant and continue the palbociclib. They were stratified based on visceral involvement and time from inclusion to the ESR1 mutation detection. The primary endpoint was PFS and the intent to treat population by investigator and also grade 3 AEs, and then there was a bunch of secondary endpoints as well.

[00:50:51]

 PADA-1: ESR1 Mutation Status During CDK4/6i + ET

So, the objective of this was to assess if you change your endocrine therapy during ESR1 mutation emergence if this will benefit patients. If you look here, this is some of the numbers here, the number of patients who had wild type ESR1 cancers at disease progression during the first 6 months of an aromatase inhibitor and palbociclib was 81% with 19% having a mutation. If they looked after 6 months of an AI and palbociclib, there was a much higher rate of ESR1 mutations at 53% with only 47% being a wild type.

[00:51:36]

PADA-1: PFS

Here is what they found. As I mentioned, these patients had rising levels of ESR1 in the blood, no evidence of disease progression on their scans. They either stayed on an aromatase inhibitor or switched to fulvestrant. You can see there was a doubling of progression-free survival from 6 months in the patients who stayed on the aromatase inhibitor, actually it was 7 months, up to 13 months if they switched to fulvestrant. That is for PFS. Looking at PFS2, you can see here again, quite a marked benefit from 14 months if they stayed on an aromatase inhibitor vs 29 months if they switched to fulvestrant. But I think as we are going to hear about in the plenary session tomorrow, we will see whether this is true or not in the SERENA study, which Erica is going to introduce you to in a few minutes.

[00:52:24]

Conclusions: Utility of Genomic Testing Following First-line CDK4/6i + ET

In summary, I think genomic testing on tissue or blood is recommended at diagnosis of metastatic disease, especially for those patients that would've met criteria for the INAVO120 study, so patients who had either disease relapse on endocrine therapy or soon after completing endocrine therapy. I think the standard of care now is to recheck genomic testing at disease progression looking for these acquired mutations such as ESR1 mutations, AKT1 mutations. The ctDNA may be optimal because they're noninvasive, but also perhaps because they pick up mutations you don't see in tissue. There are many assays available as you all know and most of them do RNA and DNA sequencing. Emerging data, actually, we will talk about this later, but at least the data from PADA-1 and what we will see tomorrow suggest that changing treatment based on ctDNA may prolong progression-free survival. But of course the long-term question is does it really improve overall survival?

[00:53:25]

Patient Case: A 67-Yr-Old Woman With de Novo MBC

Let us revisit our patient case. Again, this was a lady, she has been on ribociclib and letrozole for 3 years. Her genomics were normal or did not reveal actionable mutations at the time she was diagnosed with metastatic disease. She now has disease progression in her bones and we sent ctDNA for analysis.

[00:53:46]

Posttest 1

Just to revisit this question, would you expect her to have an ESR1 mutation only because we are thinking about acquired mutations? Would she have ESR1 and P53? Would she have PIK3CA mutation alone keeping in mind that she didn't have that at baseline or could she have both of those mutations, PIK3CA and ESR1?

I didn't convince everybody, but I convinced some people I think. The correct answer is ESR1 mutation only because remember I mentioned that P53 and PIK3CA are present from the primary diagnosis, they're not acquired. Because she didn't have those mutations at the baseline, her cancer couldn't have acquired them basically, if that makes sense. I think that is my thing. I'm going to hand it over to Erica.

[00:54:55]

Clinical Evidence for Next-Generation Approaches Targeting the Estrogen Receptor

Dr Erica Mayer (Harvard Medical School): Okay. Thank you so much, Ruth. That was a fantastic overview and we are going to move on and we are going to do a few more cases and then we are going to take a deep dive into the oral SERD category, which is a rapidly enlarging exploding category of exciting medicine. We will take that journey together.

[00:55:21]

Patient Case: A 59-Yr-Old Woman With de Novo MBC

First, we are going to just run a few cases. Here's another case, it is a little similar, perhaps. 59-year-old postmenopausal woman was diagnosed with de novo metastatic hormone receptor–positive, HER2-negative breast cancer in 2020. At that time, she initiated first-line therapy with the CDK4/6 inhibitor, palbociclib and letrozole. At the time genomic testing showed no specific actionable mutations. She has done well for the past 5 years, but now she has been restaged and she has asymptomatic progression in the bone. You have repeated her ctDNA and now she has an ESR1 mutation, but there is no PIK3CA pathway mutation seen. There is no germline BRCA1/2 mutation.

[00:56:05]

Pretest 2

At this time, which of the following would you recommend as the best next step for this patient using the currently available choices?

1. Abemaciclib and fulvestrant;
2. Everolimus and fulvestrant;
3. Elacestrant monotherapy;
4. Imlunestrant and abemaciclib; or
5. Trastuzumab deruxtecan monotherapy.

Please vote. Okay. Great. Thank you. Oh, people are still voting. Super.

[00:56:48]

Pretest 3

Next case or next question. We are going to get specific now. In the phase III EMBER-3 trial for patients with hormone receptor–positive HER2-negative advanced breast cancer, which of the following was reported with imlunestrant plus/minus abemaciclib vs standard of care endocrine therapy?

1. Improvement in both PFS and OS with imlunestrant monotherapy vs standard of care endocrine therapy in all patients regardless of ESR1 mutation status;
2. Improved PFS only with imlunestrant monotherapy vs standard of care endocrine therapy in all patients regardless of ESR1 mutation status;
3. Improved PFS with imlunestrant and abemaciclib vs imlunestrant alone in all patients regardless of ESR1 mutation status; or
4. Equivalent PFS with imlunestrant and abemaciclib vs imlunestrant alone in patients with previous treatment with a CDK4/6 inhibitor.

Please vote. Great. Still voting.

[00:58:09]

Patient Case: A 62-Yr-Old Woman With MBC

We will move to our third case. This is a 62-year-old postmenopausal woman who has stable disease on first-line letrozole plus ribociclib for hormone receptor–positive, HER2-negative advanced breast cancer for the past year. She has enrolled in the phase III SERENA-6 trial, which as we are going to be talking about, involves ctDNA surveillance every 2-3 cycles for the emergence of an ESR1 mutation. At her most recent ctDNA check, she now has the presence of an ESR1 mutation in the absence of clinical progression. She moves on in the study and is randomized to receive camizestrant and abemaciclib. Now, you are in a morning huddle with your clinical team and you're reviewing patient care for this patient.

[00:58:56]

Pretest 4

At this time, what toxicities are you going to discuss in your huddle specific to camizestrant? What are you going to talk about with your clinical team? Will you discuss:

1. Alopecia and hypertension;
2. Fatigue and pruritus;
3. Hearing loss and arthralgia;
4. Photopsia and bradycardia; or
5. QT prolongation and bleeding?

Please vote. Very good. We will come back to these questions in just a moment.

[00:59:41]

Treatment Algorithm for HR+/HER2- MBC

Moving on, let us talk about how we take care of advanced hormone receptor–positive HER2-negative breast cancer. For the majority of patients, first-line therapy consists of an aromatase inhibitor or fulvestrant with a CDK4/6 inhibitor and a subset of patients are also eligible to receive the PI3 kinase inhibitor inavolisib. What happens after that, however, becomes very heterogeneous. Much of that is determined by the presence or absence of actionable genomic mutations. For the purposes of our talk, I really want to focus on this step here when patients develop an ESR1 mutation, as we just heard from Ruth, because this is one that develops over time. I want to talk about the medicines that are currently available in this category and what is coming in this category and why that is important.

[01:00:24]

Novel Endocrine Therapies May Address Endocrine Resistance in MBC

Let us talk a little bit about what we have available. ESR1 mutations indicate an element of endocrine resistance and it has been demonstrated that oral SERDs may be particularly helpful in this setting. Oral SERDs’ mechanism of action is twofold. By binding the estrogen receptor, they prevent expression of estrogen responsive genes, but they also lead to degradation of the estrogen receptor. It is profound inhibition of ER. There are actually multiple novel agents available targeting ER including the SERDs. We are also going to touch on the PROTACs and CERANs, which are other exciting areas. These are all oral agents. They have better bioavailability than the currently available SERD fulvestrant and they're particularly well suited for cancers with ESR1 mutations.

[01:01:11]

Select Ongoing Trials of Novel Oral Endocrine Therapies: Oral SERDs

Currently, there are 4 oral SERDs that are in active development: elacestrant, giredestrant, camizestrant, and imlunestrant. You can see the variety of trials here in the pretreated metastatic setting, the first-line metastatic setting, and I just want to point out the trials listed in green are the trials that are in the early stage setting, either neoadjuvant or adjuvant. Some of these are some very large adjuvant trials that are going on right now. We are not going to touch on these today, but if you are a provider and you have access to these trials, I strongly encourage you to please enroll your patients because these are very important trials for us to learn from.

[01:01:46]

Select Ongoing Trials of Novel Oral Endocrine Therapies: Oral PROTAC and CERAN

We also have the oral PROTAC vepdegestrant. We just saw some data on that agent today from VERITAC-2. And the oral CERAN palazestrant being studied in the OPERA trials and we will mention these in just a moment.

[01:01:59]

EMERALD: Elacestrant vs Investigator's Choice SoC ET in ER+/HER2- MBC

Let us jump in and we are first going to talk about oral SERD monotherapy and our currently approved agent is elacestrant as approved in the EMERALD study. This was a phase III study for patients with pretreated metastatic breast cancer. All patients in this study had prior exposure to a CDK4/6 inhibitor and patients could have had up to 1 line of chemotherapy for advanced disease. These are important details because it helps us interpret the data and put it in the context of other trials. Patients are randomized to receive elacestrant monotherapy or provider-choice endocrine treatment. Coprimary endpoints included PFS in the ITT population and in patients with ESR1-mutant disease.

[01:02:39]

EMERALD: PFS With Elacestrant vs SoC in ITT and Population With ESR1mut+ Disease

This was a positive trial. Patients receiving elacestrant had a prolonged PFS compared to those who received standard of care endocrine therapy. But it is important to try to piece together who is getting the greatest benefits here. If we look between the ITT population on the left and the ESR1-mutant disease population on the right, you can see a more profound benefit with an improvement in the PFS and a hazard ratio of 0.55. It is also important to note whenever looking at the EMERALD figures that you see this very steep drop off in the very beginning. About half the patients are lost at time of first restaging. That is a good reminder to us that patients who have progressed on first-line endocrine therapy, some of them have profoundly resistant disease, they're not endocrine sensitive and they may not be best served by further endocrine therapy and that is something for us to continue to try to identify upfront.

[01:03:31]

EMERALD: PFS by Duration of CDK4/6i in ESR1m Population

Further analysis of EMERALD has looked at outcomes based on prior duration of CDK4/6 inhibitor therapy. This does not have to do with CDK4/6 inhibitors, it has to do with endocrine resistance and in patients who had had at least a year of disease stability on their prior regimen, had a more substantial benefit. We can see here an improvement from about 2 months to almost 9 months with the use of elacestrant. It helps us begin to put it together an idea of who are best candidates to receive elacestrant as monotherapy.

[01:04:01]

EMERALD Subgroup Analysis: PFS by Subgroup in ESR1m Patients With ≥12 Mo of CDK4/6i

A subsequent subgroup analysis has been quite helpful to demonstrate that many molecularly defined or clinically defined subgroups are getting benefit from elacestrant. We point out here that the dual mutation subgroup with both ESR1 mutation and a PIK3CA mutation derives benefit improvement from 2 months to about 5 and a half. However, if you look kind of up and down the line here, the 5 and a half here does indicate to us that even with the oral SERD therapy, this is more resistant disease, as Ruth has alluded to, and we are going to see shorter PFS than we would compared to less resistant disease.

[01:04:38]

SERENA-2: Camizestrant vs Fulvestrant in Postmenospausal Patients With ER+/HER2- Advanced BC

Let us move on now to SERENA-2. This is a study of camizestrant monotherapy. This was a trial for patients who had prior endocrine therapy and were randomized to what is essentially 2 dose levels of camizestrant. The top level was stopped early, so camizestrant 75 milligrams, camizestrant 150 milligrams, or fulvestrant monotherapy with a primary endpoint of progression-free survival.

[01:05:05]

SERENA-2: PFS by Investigator Assessment

This is the PFS in the ITT population and we can see that there was an improvement here from 3.7 months to about 7 months. This is a positive study.

[01:05:15]

SERENA-2: PFS in Patients by Detectable ERS1m

But interestingly, when you look at this based on ESR1 mutation vs not, you can actually see quite a difference. In the patients whose disease has an ESR1 mutation, there's a substantial difference to whether patients get camizestrant or if they're getting fulvestrant monotherapy. Whereas if the ESR1 is wild type, there is not much difference among the arms. Again, a signal that the oral SERDs may achieve the greatest benefit in patients whose disease has ESR1 mutation.

[01:05:47]

EMBER-3: Imlunestrant ± Abemaciclib vs SoC for ER+/HER- ABC

Most recently we've seen EMBER-3. This was presented at San Antonio this year and this is one of our first studies looking at oral SERD combination therapy. This is a trial for patients with metastatic hormone receptor–positive HER2-negative breast cancer. Now, importantly, prior treatment with CDK4/6 inhibitor was not mandated. About half the patients in the study had exposure but it wasn't everybody and there was no prior chemotherapy in the study. There was a randomization to one of 3 arms: imlunestrant monotherapy, imlunestrant with abemaciclib, so a combination arm, or standard of care endocrine therapy. There were 3 primary endpoints of the study. The first 2 were for the monotherapy arms looking at progression-free survival in the imlunestrant vs standard of care ITT, and in the imlunestrant vs standard of care in the ESR1-mutant population. Then the third primary endpoint was for the combination comparing imlunestrant alone to imlunestrant with abemaciclib. It is a little complicated there, but 3 primary endpoints.

[01:06:51]

EMBER-3: PFS and OS With Imlunestrant Alone vs SoC

Let us first consider the monotherapy arms. For the patients who did not have an ESR1 mutation, which is the figure shown on the right, there was not a significant difference in progression-free survival whether they received imlunestrant or standard of care. However, in the patients who had the ESR1 mutation, here is again where we see the benefit from the SERD monotherapy and improvement from 3.8 months to 5.5 months with a hazard ratio of 0.62. This was positive in the ESR1 mutant population. Overall survival data is still maturing, but for patients with the ESR1 mutation, this is trending favorably for overall survival with a hazard ratio 0.55.

[01:07:36]

EMBER-3: PFS With Imlunestrant ± Abemaciclib

Now, the combination arms we see a significant difference. Patients who received imlunestrant and abemaciclib had a prolonged progression-free survival compared to those who received imlunestrant alone, 9.4 months vs 5.5 months; 9.4 months is one of the longest progression-free survivals we've seen in the post CDK studies. Although again, I would point out that this population is a little different because it is not as heavily CDK4/6 pretreated and the patients did not receive prior chemotherapy. That being said, this is a quite impressive number and this benefit was seen whether or not an ESR1 mutation was present. It is almost as if when you give the doublet it may cancel out the need to require the ESR1 mutation.

[01:08:18]

EMBER-3: INV-Assessed PFS by Subgroup

Subgroup analysis from the first presentation of this data demonstrated that all of the subgroups that were being evaluated had a consistent PFS benefit. I would point out one that is quite of interest is the previous CDK4/6 inhibitor because this is looking at abemaciclib vs not. It is quite interesting that the perhaps stronger benefit was seen in patients who had received prior CDK4/6 inhibitor and less benefit seen in those who had not, even though this was apparently their first exposure to CDK4/6 inhibitor. But it does suggest to us that the results are consistent whether or not a patient has had prior CDK4/6 inhibitor.

[01:08:57]

EMBER-3: PFS by ESR1 and P13K-Pathway Comutation Status Patients Pretreated With a CDK4/6i

We've also seen updated data just recently at ESMO Breast, specifically looking at outcomes in EMBER-3 in that population with prior CDK4/6 inhibitor. We can see in all patients shown in the left Kaplan-Meier, we can see the significant benefit with the combination imlunestrant, abemaciclib and improvement 5.5 months to 11.1 months. I'm sorry, that is for all patients, and the CDK4/6 pretreated patients on the right. This is actually the select population of patients who have the dual mutations ESR1 and PIK3CA. Similar to what we saw in EMERALD, where the use of elacestrant was active even in the setting of the presence of a PI3 kinase mutation, we see that also in EMBER-3; when dual mutations are present, the combination of imlunestrant and abemaciclib is superior to imlunestrant alone. Using an oral SERD in the presence of the dual mutations was highly helpful.

[00:09:51]

EMBER-3: INV-Assessed PFS by Subgroup

This is the subgroup analysis in the CDK4/6 pretreated patients. This is the 50% of patients who had prior CDK4/6 inhibitor. Again, you can see that there is maintained benefit in all of the subgroups. I'll point out that it doesn't matter which CDK4/6 inhibitor patients had, the majority with palbociclib, ribociclib, very few had prior abemaciclib. It also did not matter how long the prior exposure to CDK4/6 inhibitor was, whether it was a little shorter or a little bit longer, so strong signals throughout.

[01:10:22]

ELEVATE: Elacestrant Combinations in ER+/HER2- ABC

Now, in terms of other combination experiences, the ongoing elevate study is a multi-cohort study looking at combinations of elacestrant and targeted partners. So far we have 5 cohorts that are either enrolled or are enrolling combining elacestrant with these agents. Data so far has shown favorable PFS, elacestrant with everolimus showing a PFS of 8.5 months, elacestrant and ribociclib with a PFS of 7.2 months. There has also been data presented on side effects and toxicity with these combinations. As we will get to in a moment, oral SERDs are very well tolerated, and the toxicities that are seen here are really primarily from the partner agent. For example, in the everolimus arm, the most common grade 3 or higher treatment emergent events included things like stomatitis and diarrhea, which are classic for everolimus, neutropenia with ribociclib. We are seeing the side effects from the targeted partner but not from the oral SERD itself.

[01:11:21]

SERENA-6: CDK4/6i + Camizestrant or AI in ESR1m HR+/HER2- MBC in Patients Without PD on 1L CDK4/6i + AI

Finally, just another shout out for the very exciting data that we are actually going to see tomorrow at the plenary session from SERENA-6. SERENA-6 is a study that was built on PADA-1 and this idea of evaluating in real time for the emergence of ESR1 mutations for the emergence of resistance and then acting on it and making a change in treatment before the emergence of clinical progression. SERENA-6 has 2 steps, step 1 and step 2. Step 1 enrolled patients with metastatic hormone receptor–positive HER2-negative breast cancer who were receiving aromatase inhibitor and a CDK4/6 inhibitor in the first-line setting and were stable for at least 6 months. They could have been on therapy much longer than that, but they needed at least 6 months of stability. In step 1 patients had ctDNA testing every 2 to 3 months ideally to be timed with their routine restaging. If the patient had ESR1 negative, then they just continued screening. If they were found to be ESR1 positive and they had progression at the same time, they were off of the study. The key population were the patients who developed an ESR1 mutation but did not have evidence of clinical progression. This idea of trying to catch this change early before the tumor has demonstrated measurable growth. Patients then went on to step 2, which is the treatment step. Those patients then were randomized. They either stayed on their current aromatase inhibitor and CDK4/6 inhibitor with the addition of a placebo or they were switched to the oral SERD camizestrant with their CDK4/6 inhibitor and with a placebo.

[01:12:56]

SERENA-6: Key Points

We know from a press release that this trial has met its primary endpoint of progression-free survival and we will look forward to seeing this trial presented tomorrow.

[01:13:05]

Select Ongoing Phase III Trials of Oral SERDs in ER+/HER2- MBC

There are many ongoing studies right now with Oral SERDs for advanced hormone receptor–positive disease. I'll point out these are trials in the first-line settings such as SERENA-4 trials that are for first line but more resistant disease like persevERA or pionERA and trials in the pretreated setting such as evERA trial, so many more trials particularly of combinations, which I think is a topic of great interest to all of us.

[01:13:32]

Safety of Oral SERDs

Now, briefly, let's talk about safety. These are some of the side effects we might see with oral SERDs. GI toxicity can occur. Usually this could be mild diarrhea or nausea vomiting. Bradycardia can occur particularly with camizestrant, as we will take a look at, and visual disturbances, which is called photopsia, which is a patient seeing halos or tracking lights in low light conditions. Hot flashes and arthralgias are listed here, although honestly we tend to see less arthralgias with oral SERDs compared to drugs like aromatase inhibitor. I would say in general these are very well tolerated drugs.

[01:14:09]

SERENA-2: TRAEs

This is data from SERENA-2 looking at the side effects, and again, I point out photopsia and bradycardia as 2 of the more common treatment emergent side effects. But in general this tends to be almost all grade 1 and very mild. It is also interesting to look in SERENA-2 that this is dose dependent and you can see the rate of photopsia increases with dose, the rate of bradycardia increases with dose. This dose, 75 milligrams is what has moved forward in subsequent camizestrant trials. Overall these are very low rates of these low grade toxicities. And we do know that the photopsia associated with camizestrant does not affect visual function. There is no change in visual acuity. It is a reversible side effect. It doesn't appear to be dangerous in any way. With the bradycardia this is asymptomatic. Patients, if they exercise, their heart rate comes up. It is something which one might encounter in clinic, in triage when they're getting vital signs, but it is not a dangerous side effect.

[01:15:07]

VERITAC-2: Phase III Study of Vepdegestrant vs Fulvestrant in ER+/HER2- Advanced BC

Briefly, the PROTAC is another exciting area. These are very interesting drugs that target protein degradation. The lead agent here is vepdegestrant or vepdeg, which is meant to help target the estrogen receptor for degradation. We just saw data earlier today for the VERITAC-2 study. This is a trial for patients who had prior exposure to CDK4/6 inhibitor and endocrine therapy and then were randomized in the second-line setting to get vepdegestrant monotherapy or fulvestrant monotherapy. Unfortunately we weren't able in the hour between the presentation and now to pop in the slide, but we did see that this was a positive study. The fulvestrant-alone arm was 2.5 months. The vepdegestrant arm was 5.5 months and this benefit was seen in the ESR1-mutant population. Again, building on this idea that monotherapy with these estrogen specific agents may be best served in the ESR1-mutant population.

[01:16:05]

VERITAC-3:Vedegestrant + Palbociclib vs Letrozole + Palbociclib in ER+/HER2- Advanced BC

We also are awaiting results from VERITAC-2 and this is now combination. This is a first-line trial of vepdegestrant and palbociclib vs letrozole and palbociclib. We will await these data.

[01:16:18]

ELAINE 2 and ELAINE 3 Trials of a CDK4/6 Inhibitor, Abemaciclib ± a SERM, Lasofoxifene for ESR1m+ ABC

Lasofoxifene is an oral serum, so another method of targeting the estrogen receptor. This agent has shown high activity in the ELAINE 2 study and is currently being studied in ELAINE 3, which is a study for patients post CDK4/6 inhibitor with ESR1 mutations randomized to lasofoxifene and abemaciclib vs fulvestrant and abemaciclib.

[01:16:38]

OPERA-01: Palazestrant vs SoC in ER+/HER2- Advanced BC

Then finally OPERA-01 is looking at the oral CERAN, which is a complete estrogen receptor blocker, palazestrant. This is again a post CD case study of palazestrant monotherapy vs investigator choice endocrine therapy and we await results from OPERA-01.

[01:16:57]

Treatment Algorithm for HR+/HER2- MBC

In conclusion, coming back to our figure here, as we said at the outset, we do our genomic testing in the second-line setting and then if we see an ESR1 mutation, elacestrant monotherapy has been our option. But I think as you see there are some really exciting other agents that are in development that have shown a lot of activity for ESR1-mutant disease. Perhaps the most exciting is that when we combine these agents with targeted partners, we can see substantially more activity. Because oral SERDs are so well tolerated, they do seem to match well with targeted partners. I think in the next year or 2, we are going to have a lot more options in this setting and we will greatly expand and complicate my figure here.

[01:17:41]

Let's Retake Our Quick Survey

Let us retake our questions.

[01:17:44]

Patient Case: A 59-Yr-Old Woman With de Novo MBC

First we are going to talk about our patient with metastatic disease who has been stable for 5 years on her first-line AI and CDK regimen. She now has asymptomatic bone progression and she has a new ESR1 mutation. You've not been checking all along, like in SERENA-6, you checked at the time of tumor progression.

[01:18:02]

Posttest 2

What would be the best next step for the patient? I'll let you vote again. Well, you did very well before, selecting elacestrant monotherapy, and even better now. Thank you. Getting back to the answers, so we would want to pick elacestrant monotherapy because this patient is a really good fit for the EMERALD study. She has had a prolonged period of time on her prior CDK4/6 inhibitor. It again speaks to this idea of endocrine sensitivity and patients like this with asymptomatic progression are ideal to try elacestrant and especially as it is so well tolerated. This would absolutely be my choice in this setting.

[01:19:02]

Posttest 3

Next question. This was the really kind of complicated question about EMBER-3. I'm going to read these through just so we can remember them. This is in the phase III EMBER-3 trial for patients with ER-positive, HER2-negative advanced breast cancer, which of the following was reported for the imlunestrant plus/minus abemaciclib vs standard of care endocrine therapy? Here we go.

1. Improvement in both PFS and OS with imlunestrant monotherapy vs standard of care in all patients regardless of ESR1 mutation status;
2. Improve PFS only with imlunestrant monotherapy vs standard of care in all patients regardless of ESR1 mutation status;
3. Improve PFS with imlunestrant and abemaciclib vs imlunestrant alone in all patients regardless of ESR1 mutation status; or
4. Equivalent PFS with imlunestrant and abemaciclib vs imlunestrant alone in patients with prior CDK4/6 inhibitor.

Please vote. Great. Definitely learning here. I'm sorry this is such a complicated question, but you got it right that the correct answer was C because remember when we use the combination of the SERD and the CDK4/6 inhibitor, this provided benefit regardless of the ESR1 mutation status. Again, this builds a nice argument that the combinations may be a great way forward for this class of drugs.

[01:20:42]

Patient Case: A 62-Yr-Old Woman With MBC

Then finally we are going to get to our toxicity question. This is your patient who is on SERENA-6. In step 1, she has her ESR1 mutation detected in the absence of clinical progression so she moves to step 2. The secret here is that you actually know she is on camizestrant, but in SERENA-6 it was actually placebo controlled. But we are going to pretend that we know that.

[01:21:07]

Posttest 4

You're in your morning huddle and you're talking to your team about treating patients with camizestrant. Which camizestrant toxicities are you reviewing with your team? Please vote on what is related to camizestrant. Perfect. 100%. Wow. Yeah. Again, photopsia and bradycardia are the toxicities that are linked to camizestrant, although these are grade 1 and they are relatively infrequent, but they are a little unique in the oncology world, and so as we are becoming more familiar with this agent, we are also becoming more familiar with these toxicities. All of the other choices, while they happen with many of our drugs, but they are not associated with camizestrant.

[01:22:04]

Moderated Roundtable

Well, I'm going to pass this over to Joyce. Thank you.

Dr Joyce O'Shaughnessy (Sarah Cannon Research Institute): Very nice talk and such good learning we are having this evening. We just have a few minutes. I just wanted to ask Ruth and Erica the question of now that we've heard all about these targeted agents, super important, probably the agents are most appropriate at different times in the metastatic course. How do you use the molecular testing in your practice, tissue and blood? Let us start with Ruth.

Dr O'Regan: I mean, I've been leaning more and more to using blood, but I will sometimes do tissue genomics at baseline, but I think the liquid biopsies are fantastic because you don't put patients through biopsies. Also from the data that I showed the ctDNA may be more accurate anyway. I would do ctDNA at disease progression or like I was saying sometimes even before if the patient is really asking what other options they have.

Dr O'Shaughnessy: But after each progression, generally looking for the next target basically?

Dr O'Regan: Yeah.

Dr O'Shaughnessy: Cool. Cool. How about you Erica? What are you doing?

Dr Mayer: I do a similar pattern. I try to get tissue NGS at time of diagnosis of metastatic disease, ideally from a metastatic biopsy. But from then on forward I'm doing ctDNA. But I think what has become more important is the timing of these tests, and as I think you're going to talk about with inavolisib, we need to know the PIK3CA status at the time of diagnosis of metastatic disease in order to pick our first-line therapy. When I'm suspecting a patient has metastatic disease, I'm sending my ctDNA, I'm sending my tumor biopsy for NGS, I need all that data immediately and before I might've waited until I knew they had metastatic disease and then I would start sending it, so it has moved that up. Similarly, Ruth, I really appreciate what you say about sending the ctDNA early. When I'm seeing a patient and I'm a little worried that things are brewing, that is when I may send the ctDNA. When I see them back a few weeks later when their scans are done, I have the data and I say, ‘Oh, you have an ESR1 mutation. Let us look at what is available for you.’

Dr O'Shaughnessy: Trying to time with the scan results so you can have a comprehensive conversation about what is next. Basically very, very similar. I'll make a couple of nuanced comments and as I talk about PIK3CA, AKT and PTEN, but I think that is a very good representation of where we are. I'm amazed at how much ctDNA testing we are doing, but I think that really is reflective across what I've seen in the community as well with each progression looking for those targets we don't want to miss.

[01:24:38]

Targeting PIK3CA/AKT/PTEN Alterations: Key Clinical Trials

I'm going to now turn us to PIK3CA/AKT/PTEN alterations and targeting this very, very important resistance pathway that can lead to endocrine therapy resistance or really often does lead to endocrine therapy resistance in our patients.

[01:24:53]

Faculty disclosures

Some disclosures.

[01:24:54]

Pretest 5

A quick survey here. What would you tell your patient with progression on first-line CDK4/6 inhibitor plus endocrine therapy about the rationale for use of an oral SERD or PI3 kinase AKT pathway inhibitor?

1. Alpelisib and capivasertib inhibit the PI3 kinase pathway in PIK3CA-mutant metastatic breast cancer, but capivasertib is better tolerated with fewer patients requiring discontinuation vs alpelisib;
2. ESR1, PIK3CA and AKT mutations are commonly acquired when HR-positive HER2-negative metastatic breast cancer develops resistance to CDK4/6 inhibitor plus AI therapy;
3. Elacestrant has better efficacy compared with fulvestrant following disease progression on a CDK4/6 inhibitor regardless of ESR1 mutation status;
4. It is important to biopsy metastatic breast cancer tumor tissue after progression on a CDK4/6 inhibitor to optimize the detection of the presence of an ESR1 mutation or any mutation in the PIK3C, AKT, or PTEN pathway alterations.

Please vote. We will come back here. There are a couple of split votes there, so we will come back here at the end.

[01:26:29]

Rationale for Targeting PIK3CA/AKT/PTEN Alterations

Let us look a little bit at the rationale for why we really are so happy that we finally have some detection methods for these very important driver mutations.

[01:26:40]

Resistance Drivers Define New Therapeutic Targets

As Ruth was saying about the resistance to CDK4/6 inhibitors, is a variety of mechanisms here that have to do with cell cycle regulators, increased expression of cyclin E, loss of Rb, increased Aurora kinase A, increased cyclin 6, CDK6 among others. There is a lot of adaptive changes and expression of cell cycle regulators, but also you get upregulation of oncogenic growth-signaling mediators, and very important PI3 kinase, AKT, mTOR pathway. These cancers, for example, that have a PIK3CA mutation, they're under selective pressure stress, they don't have any estrogen, CDK4/6 is being shut down. These cells elaborate neuregulin, for example, they can autocrine signal down through HER2 and HER3, just overcoming that blockade of CDK4/6. This pathway is very, very important. When the PIK3CA AKT already exists, those can really be upregulated to try to overcome the blockade of CDK4 and 6. Very, very important resistance mechanism in general for all of our cancer therapies, but particularly in the context of endocrine therapy resistance.

[01:28:03]

Biomarkers Associated With FDA-Approved Therapies

The NCCN, these are the guidelines from NCCN, the most recent ones really showing us our approved agents here that we've heard about and we will hear about now, and really endorsing looking for activating mutations in PIK3CA, AKT, or loss of function, PTEN alterations as well as ESR1 mutations. Then as you know, there are others as well that are less common: germline BRCA1 and 2, germline PALB2, etc. But these are very, very definitely endorsed by NCCN as standard of care that we look for these mutations.

[01:28:36]

Prevalence of PIK3CA, AKT and PTEN Alterations in HR+ MBC

How common are they? This is a large series from Memorial Sloan Kettering, looking at how often these were found in metastatic cancers and then kind of comparing that to primary cancers from TCGA 41% of metastatic HR-positive, HER2-negative samples with PIK3C mutations, 9% with PTEN loss and about 7% with AKT gain of function mutation. Now, in this particular series it was similar mutation frequencies between primary and metastatic tissue. But if we look over here at another publication, and this was from 49 centers in the Netherlands and they weren't looking as well to where should we find these mutations, the blue is primary breast cancer and then this is their metastatic prospectively collected. These are their own data and these were from the TCGA and they found that in the metastatic lesions that there was a little bit of an enrichment for PIK3CA.

The way I think about it rightly or wrongly is these are really mainly in the primary breast cancer, but we've seen some data, for example in PALOMA-3, that they may be acquired perhaps in up to 10% of patients. If you don't find them, they could still be acquired in breast cancers. PTENs more really commonly enriched in the metastatic setting. Historically, our ctDNA assays haven't always across the board been as good one to another at calling copy number alterations. I think they're getting better, but not all of them can pick up PTEN loss of function. That is why I will get those biopsies upfront metastatic. But also if I've got a new side of disease like the bone goes to liver for example, I will biopsy that and see maybe what has changed because I think the copy number alterations, particularly loss of function mutations are not as easy to find in all the blood tests as they are in the tissue and you can find them in the primary breast cancer, but they're enriched in the metastatic breast cancer. I think occasionally biopsying is not a bad idea. Then AKT, as we heard, is one of those truncal mutations that initiates the breast cancer, but it too can be acquired but it is a point mutation, easier to find in the blood. Blood mainly, and I think increasingly as our assays get better, blood, but right now I think there is still a role for tissue in the metastatic setting.

[01:31:13]

PIK3CA/AKT/PTEN Alterations and Survival Outcomes in HR+/HER2- ABC

Now, are these poor prognostic factors? This series published recently looks at any of the 3 alterations in HR-positive HER2-negative breast cancer, looking at overall survival in patients. Of course treated in different ways, but pretty much per the standard of care from the start of first-line therapy. You can see median survival of 31 months in those altered patients and 44 months in those who are negative who are wild type. This really has played out in many studies. This is a meta-analysis now looking specifically at PIK3CA alterations in advanced breast cancer HR-positive HER2-negative breast cancer. You can see that this is the actual percentage here. It was about a 13% decrement in median overall survival. You can see the range here sort of in the same range. There's a lot of consensus that these PIK3CA mutations, AKT mutations, PTEN mutations in the metastatic setting really do cause resistance and shorten survival. These are poor prognostic factors.

[01:32:23]

BOLERO-2: Exemestane ± Everolimus in HR+/HER2- ABC

Finally we do have some now multiple agents to target this pathway. We will just kind of review some of the data here. Everolimus, the mTOR inhibitor, the BOLERO-2 trial done before we had CDK4/6 inhibitors, they were AI-resistant patients randomized to placebo exemestane vs with everolimus.

[01:32:45]

BOLERO-2: PFS With mTOR Inhibition

We know that both by local and central assessment, those a really pretty big improvement. When those curves split this early, you know that the agent that you've added is hitting the mark for resistance. They really split early. In the context of that AI resistance, that PI3 kinase AKT pathway was really upregulated and the everolimus did a good job in inhibiting that and it was really regardless of mutation status.

[01:33:16]

BOLERO-2: PFS by PIK3CA Mutation Status

If we look here for example at contemporaneous PIK3CA mutations in the blood right now, so the patient's disease was progressing, they were just entering the BOLERO-2 trial, they get baseline ctDNA and indeed they got a PIK3CA mutation, so suggesting it really is operational in the resistance of their disease. You can see that whether they were wild type or whether they were mutant for PIK3CA, the everolimus worked equally well. That is really quite interesting. Here in the dark blue, these were the tumor tissue patients and these were wild type patients, the dark blue and the dark orange, and you see obviously the improvement with everolimus. Then the lighter colors were patients who were mutant, and you can see that the mutant, with the mutant PIK3CA here, these patients didn't do quite as well with everolimus, but those were historical and this was more contemporaneous in the blood right now. I think the thinking is that regardless of PIK3CA mutation status, we can get equal benefit from everolimus when we need to think about that agent for patients. Now we have of course inhibitors specifically of PIK3CA, but still good to know.

[01:34:38]

PIK3CA/AKT1/PTEN Alterations and Survival Outcomes in Hormone Receptor+/HER2- ABC

Not only was BOLERO-2 done with everolimus, but we had the TAMRAD trial with tamoxifen, we had the PrECOG study with fulvestrant and you can see this was a different mTORC1, TORC2 inhibitor here. But if we look at these 3 here in the middle, we see very nice improvements in progression-free survival as well as sort of a trend in survival here as well as in the TAMRAD. Look at the hazard ratio for survival here 0.45. These are some very nice effective options for patients. Everolimus is a drug that we still have the option to utilize, particularly I think in patients that we don't find one of the mutations in PIK3CA, AKT, or PTEN today.

[01:35:22]

PFS and OS With Fulvestrant and Everolimus After CDK4/6 Inhibitor

Now, what about everolimus after progression on a CDK4/6 inhibitor? This was a prospective trial done and published recently in Oncogene looking at 57 patients here whose disease had progressed on a CDK4/6 inhibitor. They were treated with fulvestrant and everolimus and we can see the median PFS here of 6.8 months.

[01:35:46]

PIK3CA/AKT1/PTEN Alterations and Survival Outcomes in Hormone Receptor+/HER2- ABC

We go back here and say it was a little bit higher prior to the introduction of CDK4/6 inhibitors, more like 8 to 10 months, but post CDK4/6, about 7 months here and survival of about 38 months.

[01:36:02]

PFS and OS With Fulvestrant and Everolimus After CDK4/6 Inhibitors

There is evidence of activity of the everolimus in patients whose breast cancer has progressed on a CDK4/6 inhibitor. This was just to show that it really didn't make any difference how long the patients had been on the prior CDK4/6 inhibitor, which was mainly palbociclib. They had the same median PFS and OS whether they had had below the median time on the palbociclib or greater than the median time. That didn't seem to make a difference. Everolimus does have activity post CDK4/6 inhibitor therapy.

[01:36:35]

 evERA Breast Cancer: Giredestrant + Everolimus vs Exemestane + Everolimus in ER+/HER2 ABC

Of course we'd like to make an advance because we know that patients who are progressing on CDK4/6 inhibitors, we can find ESR1 mutations in up to 50% of them. Using exemestane is not the optimal endocrine therapy for those patients. This evERA trial randomizes patients, their prior CDK4/6 inhibitor therapy randomized to everolimus exemestane vs the oral SERD giredestrant everolimus. Hopefully this will be an advance for patients because there are quite a few of ESR1 mutations which is a key secondary endpoint. The intent to treat population is the primary endpoint.

[01:37:14]

Regulation of mTORC1

Hopefully, we will still see more advances with everolimus for our patients. But when patients are on the everolimus long enough and we are inhibiting TORC1, usually TORC1 phosphorylates S6 kinase which feeds back and turns off IRS. It actually says we are fine over here, we are signaling down through a TORC1, we don't really need the PI3 kinase upstream activation, but when you inhibit TORC1, this negative regulation goes away and now you have activation of IRS and PIK3CA and now you signal down through PIK3CA, AKT. This is a definitely very, very important mechanism of resistance. Perhaps we want to go upstream and particularly when there is PI3 kinase mutation.

[01:38:10]

Phase III SOLAR-1: Alpelisib + Fulvestrant vs Placebo + Fulvestrant in HR+/HER2- Advanced Breast Cancer

Of course alpelisib did that and showed us the benefit in the PIK3CA mutant cohort, only 6% of patients in the SOLAR-1 had had prior CDK4/6 inhibitor. They had all had a prior AI, and as you know, fulvestrant placebo vs fulvestrant plus the oral inhibitor of PIK3CA but it inhibits both wild type and mutant PIK3CA.

[01:38:35]

SOLAR-1: Survival Outcomes

Again, see these curves splitting very early this PIK3CA mutant cohort. Indeed the cancer was really being driven by PIK3CA and we see a very big improvement from 5.7 to 11 months in median PFS. Now, most of the patients had both tissue and also ctDNA and they did this analysis where they looked at patient's plasma ctDNA and they did not find a PIK3CA mutation in 188 patients. What they did is they went back to the tissue of these patients and they found that in these negative ctDNAs they were found to have a PIK3CA alteration in 38% of their tumor tissues. How did they do with the alpelisib? You can see those that were truly not altered, both in the blood and the tissue, that is the darker colors, the navy blue and the orange, the alpelisib really didn't do anything. But in those that were positive in their tumor tissue but negative in their blood, you can see that alpelisib really did have a positive benefit in those mutant patients compared to the orange where they just had the placebo. This is a reason that if we don't find the tumor, the PIK3CA mutation in the blood or the AKT mutation in the blood or PTEN, we go back to the primary cancer and sequence it or metastatic tissue if we have it. But going back to the primary may be easier. We just don't go back to the primary of course for ESR1 mutations. That really is a blood-based assay there.

[01:40:14]

BYLieve (Cohort A): Alpelisib + Fulvestrant after CDK4/6i in PIK3CA-mutated HR+/HER ABC

Of course we want to know how did alpelisib do post-progression on CDK4/6 and this is the BYLieve trial. We are looking here at fulvestrant plus the alpelisib in patients who had progressed on AI plus CDK4/6 and we see a median PFS quite respectable here of 8 months. Yes, alpelisib and fulvestrant in the PIK3CA mutant population is effective in patients whose breast cancer is progressing on AI plus CDK4/6. The issue with alpelisib of course is very difficult to give enough of it and sustained long enough because of the toxicity. It is difficult for patients to actually benefit from alpelisib because of hyperglycemia, rash, and overall fatigue.

[01:40:59]

Double PIK3CA Mutations

Now, it is interesting that we will find as we get more and more ctDNA results, we will find multiple PIK3CA mutations. And this analysis, and there is multiple of them, PIK3CA mutations in cysts means that there is 2 mutations in the same gene on the same chromosome. In the same cell that PIK3CA gene has 2 mutations. That is really, really an active cancer driven by PIK3CA. You can see that here with these dual mutations, the brown staining is phospho-AKT and when you have activation of PIK3CA, one of the key PD readouts is phospho-AKT. You can see that you can find these multiple mutations both in the primary and in the metastasis and you can see it is in HR-positive HER2-negative, really not much in HER2-positive or triple negative breast cancer. Does this indeed predict for benefit from PI3 kinase inhibitors?

[01:41:59]

Tumors With Double PIK3CA Mutations and PI3Ki Sensitivity

The SANDPIPER trial was looking at taselisib, which was another PIK3CA, PI3 kinase inhibitor. In the ctDNA population with single PIK3CA mutation, we can see the objective response rate here is higher with taselisib but it wasn't significant. Whereas in patients with multiple PIK3CA mutations in their ctDNA, there was a very large improvement that was significant in favor of adding the PI3 kinase. Then in the BYLieve trial with the alpelisib and endocrine therapy patients who had more than 1 PIK3CA mutation detected did better with regard to PFS with alpelisib compared to only 1 mutation. It is not a bad thing when we find this in our ctDNA, we are seeing multiple PIK3CA mutations, that might be a patient whose cancer is really, really dependent on the PIK3CA.

[01:42:51]

BOLERO-2: Exemestane ± Everolimus for MBXC With Prior Exposure to an AI

Now, what about ESR1 mutations? When we want to give a patient everolimus endocrine therapy, does the presence of an ESR1 mutation decrease her outcome? This is actually all comers in the BOLERO-2 trial looking at survival wild type vs mutant for ESR1. You see the mutation really does portend a worse survival for these patients treated. Now, this is with exemestane with or without the everolimus. Then if you look at overall survival in the wild type for ESR1 single mutations or double mutations, we see survival getting increasingly worse. Now, not shown here, but with the D538G, the everolimus had the same improvement in outcome in the PFS as it did in the wild-type population, but at least in this trial it was ineffective in those with the Y537S, it really did not help those patients. That is a retrospective analysis, I would not use that in the clinic, but I did want to mention that. These ESR1 mutations, again, we are seeing a more poor prognosis with these patients here in the context now of exemestane plus everolimus.

[01:44:05]

CAPItello-291: Capivasertib + Fulv vs Placebo + Fulv in AI-Resistant, HR+/HER2- Advanced Breast Cancer

In addition to targeting PIK3CA, we of course would like to target AKT because heretofore we didn't have any way to directly target AKT nor PTEN loss. When you lose PTEN, the PD marker that really goes up is phospho-AKT. That really increases AKT signaling. Capivasertib is an oral AKT1, 2, and 3 inhibitor, a very welcome addition to the armamentarium because especially we didn't have anything for our PTEN-loss and AKT patients. Plus we need some additional effectiveness against PIK3CA because the alpelisib is very difficult for patients to really take. Of course the CAPItello-291 trial, where patients had had an AI therapy and 70% of them had had progression on a CDK4/6 inhibitor, 700 patients randomized to fulvestrant and placebo or with capivasertib, which is 400 milligrams PO BID 4 days on and 3 days off with a primary endpoint of investigator assessed PFS.

[01:45:16]

CAPItello-291: PFS in Overall Population and Patients With AKT-Pathway Alteration

We are familiar with the data and the strongest statistically significant improvement in progression-free survival here from 3.1-7.3 months was in the AKT pathway altered population, which was about 40% of the patients in this trial. About 30% were PIK3CA, 5% were PTEN, and 5% were AKT mutations. In this trial there was a trend in the nonaltered population. It was also a trend towards benefit but it wasn't statistically significant when you just looked at the nonaltered population. Of course, capivasertib was approved in this AKT altered population.

[01:45: 57]

CAPItello-291: PFS by Subgroup

If we look again here at that altered population and we look at a subset analysis of prior CDK4/6 prior chemo, yes, no, liver metastases, yes, no, which are particularly difficult to treat. We see that the hazard ratios here are basically on top of each other right along that 0.5, so regardless of prior CDK, regardless of liver metastases, etc, patients still benefit with that same proportional reduction in the risk of progression with capivasertib.

[01:46:26]

CAPItello-291: Exploratory Analysis of PFS by AKT Pathway Gene Alteration

Likewise, whether it is a PIK3CA, an AKT, or PTEN alteration, patients benefit no matter which of these alterations they have when they get the capivasertib.

[01:46:36]

FINER MA.40 TRIAL

We just saw the results of the FINER trial, the MA 40 trial, which again was in patients progressing on AI, all of whom were required to have a prior CDK4/6 inhibitor randomized to fulvestrant with or without a different AKT inhibitor, ipatasertib. This schedule is once a day, 21 days on and 7 days off on a 28-day cycle. It was placebo controlled and the primary endpoint, which was PFS by the investigator, in the intent to treat population 1.9 months with fulvestrant going up to 5.3 months with the addition of the ipatasertib, and that was highly significant. The AKT pathway altered population, so again PIK3CA/AKT/PTEN loss, we see even a little bit larger benefit, 1.9 to 5.45 and again the hazard a little bit better at 0.47. Steve Chia said that in the AKT pathway nonaltered, you see that trend towards improvement here with the hazard of 0.75 but it was not significant. This benefit from ipatasertib was seen regardless of ESR1 mutation status. Survival is immature but about 2 years in both arms right now. Diarrhea a high percentage of patients. I think it was about 80% of patients had diarrhea, some nausea, vomiting, fatigue but not much in the way of hyperglycemia. Not very different from the capivasertib results as well as the toxicity profile, at least to my estimation there.

[01:48:11]

Preclinical Studies in PIK3CA-Mutant Breast Cancer Models

The way it works when we got PIK3CA that is being effectively inhibited here, you get downregulation of AKT, mTOR and you get less production of cyclin D1, so less there to dimerize with CDK4/6 and the whole thing basically turns off. But if mTOR, for example, and this could happen as well with AKT, is activated through bypass pathways, then they become active and then you get upregulation of cyclin D1 expression, increased phospho-Rb and driving transcription through E2F proliferation. It turns out that preclinical data support the use of combinations of PI3 kinase inhibitors and CDK4/6 inhibitors as well as endocrine therapy blockade by basically inhibition at vertical nodes within the PI3 kinase pathway includes efficacy, so going after both the CDK4/6 and PI3 kinase inhibition.

[01:49:22]

Palbociclib + Fulvestrant ± Inavolisib for PIK3CA-Mutated ABC

That of course was what was done in INAVO120. We've seen already the schema here and these patients were recurring on their adjuvant endocrine therapy or within 1 year of finishing their adjuvant endocrine, there'd been no therapy for metastatic breast cancer. They were found to have a PIK3CA mutation, so a very poor prognostic group. And they received palbociclib fulvestrant placebo with or without inavolisib 9 milligrams daily. This is a PI3 kinase inhibitor.

[01:49:53]

INAVO120: PFS and OS

We had seen this was their initial PFS results from the 7.3 to the 15 months. We saw that updated today with longer follow-up of 34 months. We saw that the PFS got a little larger with the inavolisib 17 months vs 7 months. That is a big improvement in PFS. OS: there is now a stat SIG improvement in OS from 27 months without the inavolisib up to 34 months with it. Here is the hazard ratio, here's the P value, and median time to chemotherapy was really 2 years longer with the inavolisib, this triplet in this really poor prognostic group of patients, 35 vs 12 months, again, statistically significant. As we've heard, we need to check that PIK3CA mutation status in our first line. Patients who recur on or within a year finishing their adjuvant endocrine therapy, we got to see if they have a PIK3CA mutation. This is an excellent triplet for these patients.

[01:50:49]

Summary of Toxicity of PI3K/AKT/mTOR Inhibitors

Now, there are toxicities as we've heard. We know with alpelisib, hyperglycemia, diarrhea, and rash, 25% discontinuation rate in SOLAR-1. Inavolisib only 6.8%. There is hyperglycemia. You can see here not too much in the way of grade 3, much less than alpelisib. You see that here, there is stomatitis, you do have to use the prophylactic mouth rinse, no question, with dexamethasone. Capivasertib, it is mainly some diarrhea and rash. We prophylax with a nondrowsy antihistamine to prevent the rash and diarrhea, dietary modification, loperamide. But that is why the 3 days off helps so much. Four days on, 3 days off the diarrhea goes away. Everolimus we know we just have to prophylax with the dexamethasone mouth rinse.

[01:51:36]

INAVO121: Fulvestrant + Inavolisib or Alpelisib for PIK3CA-Mutated HR+/HER2- ABC

The next steps for INAVO121 is to go in patients who are previously treated with a CDK4/6 inhibitor and randomize to fulvestrant alpelisib vs fulvestrant inavolisib in those patients who have a PIK3CA mutation.

[01:51:52]

CAPItello-292: Palbociclib + Fulvestrant ± Capivasertib for HR+/HER2- ABC

This trial is ongoing as is the CAPItello-292 to build a triplet, now with capivasertib. We saw the triplet with inavolisib. These are also patients recurring on adjuvant endocrine therapy or within a year of finishing their adjuvant endocrine therapy or they progressed within 6 months of starting endocrine therapy for advanced breast cancer and they're randomized to palbociclib and fulvestrant vs with the addition of capivasertib. But this is actually for all comers. This is regardless of mutation status in PIK3CA or AKT. There is a secondary endpoint that will look at that altered subset of population. This would be great to post. These patients had not had a CDK4/6 inhibitor, they had not had one, but it would be great for patients who have this poor prognosis who have endocrine therapy resistance recurring on adjuvant endocrine therapy or within a year to have a triplet regardless of PIK3CA mutation status. Very, very interesting trial.

[01:53:00]

Next Frontier: Tumor/Mutant Selective PI3Kα Inhibitors

Now, the next frontier is to develop PI3 kinase inhibitors that do not inhibit wild-type PIK3CA but just inhibit mutant PIK3CA, and there are 2 of them. Lilly has one and Relay has one, and they're in the advanced trials now heading towards phase III. I think these are going to make the combination strategies more tolerable and very likely more effective. You can give higher doses when you're just inhibiting the mutant PIK3CA.

[01:53:33]

Conclusions

In conclusion, this pathway is very frequently altered in breast cancer, drives cell signaling growth and metabolism. PIK3CA is mutant in about 40% of patients. It is an independent poor prognostic marker. Everolimus/exemestane, alpelisib/fulvestrant for PIK3CA, capivasertib/fulvestrant for PIK3CA/AKT/PTEN-altered are approved for patients with HR-positive metastatic breast cancer. The next generation mutant selected PI3 kinase are in early phase. They're actually very close now to launching phase III studies and the goal was to improve the therapeutic index of the currently approved PI3 kinase inhibitors with really proactive management of toxicity, stomatitis, hyperglycemia, rash being as very proactive. This triplet now improves overall survival, the inavolisib, palbociclib, and fulvestrant. We have to know the PIK3CA mutation status for those patients who recur on their adjuvant endocrine therapy. Then the CAPItello-292 phase III fulvestrant and CDK4/6 inhibitor—they're going to probably enlarge it beyond palbociclib with or without capivasertib in the same population now regardless of mutation status—is indeed underway.

[01:54:45]

Let's Retake Our Survey

With that, let us retake our survey here.

[01:54:48]

Posttest 5

What would you tell your patient with progression on first-line CDK and endocrine therapy about the rationale for the use of an oral SERD or PI3 kinase AKT pathway inhibitor?

1. Alpelisib and capivasertib inhibit the PI3 kinase pathway in the PIK3CA mutant patients, but capivasertib is better tolerated and fewer patients require discontinuation vs alpelisib;
2. ESR1, PIK3CA, AKT mutations are commonly acquired when HR-positive HER2-negative metastatic disease develops resistance to CDK4/6 inhibitor therapy;
3. Elacestrant has better efficacy compared with fulvestrant following progression on a CDK4/6 inhibitor regardless of ESR1 mutation status;
4. It is important to biopsy metastatic breast cancer tissue after progression on CDK4/6 to optimize the detection of the presence of an ESR1 mutation or a PIK3CA/AKT/PTEN pathway alteration.

Please vote. That is good. The top one is correct actually. The metastatic biopsy down the bottom there, we won't find the ESR1 mutations reliably. Remember they're acquired and if somebody has multiple mutation metastatic sites and one is biopsied, you could certainly miss it because it is acquired. It is not uniform across all of the metastatic sites necessarily. The blood is definitely the best way to integrate what is going on in all of those metastases post progression on CDK4/6. That is why D is not the right answer there. It is blood for ESR1 mutation. That is the right answer there on the top. Capivasertib, much better tolerated than the alpelisib. Thank you very much.

[01:56:57]

Discussion

Dr O'Regan: We've got a few minutes left and we are going to go through these very quickly just in the interest of time.

Erica, what are the key treatment decision making factors: NGS results, duration on CDK4/6 inhibitor, safety profiles? Do you repeat NGS testing? I think we've already discussed that actually. What are the major adverse events you see in your patients and how do you manage these? I think we've kind of covered a lot of that.

What about this one? How do you decide between an oral SERD or a PI3 kinase inhibitor in a patient who is eligible for both? Actually that was one of the questions we got as well.

Dr Mayer: Yeah. This is one of the key questions right now because with all of our NGS we are detecting dual mutations and the question is, which agent do you use? On the one hand, as I showed you the analysis from EMERALD and from EMBER-3 show us that in ESR1-mutant and PIK3CA mutant, the oral SERD has activity vs the comparator. We know that oral SERDs work, but we've also seen that if you have a PIK3CA mutation, this impends on poor prognosis. That might make you lean more towards offering a PI3 kinase pathway inhibitor. But when we look at the types of agents available to us, elacestrant monotherapy, this is available as of today, elacestrant monotherapy vs alpelisib or capivasertib combinations, these are different experiences for patients. Additionally, if we are picking patients for oral SERDs, there are some key characteristics that we want to be looking at, like things like duration on prior line of endocrine therapy. I think that it really becomes a discussion with the patient looking at the patient's profile, their disease history, the side effect profile, their robustness. In general, I would be tending to lean towards the PI3 kinase inhibitor just because again, I worry about resistance, but in patients who have more indolent disease, an oral SERD would be a fine choice. I think ultimately what we'd love is the combination of the SERD and the PI3 kinase pathway-targeting therapy together because then we'd hit both.