Dr. Adam Brufsky:
Welcome to The ASCO Post Roundtable Series on PI3K Inhibition Strategies for Hormone Receptor–Positive, HER2-Negative Metastatic Breast Cancer. My name is Dr. Adam Brufsky. I'm a Professor of Medicine at the University of Pittsburgh and a breast cancer medical oncologist. Joining me today is Dr. Heather McArthur. Heather, you want to introduce yourself?
Dr. Heather McArthur:
Certainly. Thank you for including me in this terrific program. My name is Heather McArthur, I'm the Clinical Director of the Breast Cancer Program at UT Southwestern in Dallas, Texas. Thank you.
Dr. Adam Brufsky:
And then we have Dr. Seth Wander. Seth?
Dr. Seth Wander:
Great to be with you both. Thanks for the invitation. My name is Seth Wander. I'm a medical oncologist at Massachusetts General Hospital in Boston, and an Assistant Professor of Medicine at Harvard Medical School.
Dr. Adam Brufsky:
Great. So we have a lot of really cool options now, and kind of interesting options. And believe me, after ASCO, all three of us know they're getting a lot more confusing. But I think that for now we're going to talk about the treatment and management of metastatic hormone receptor–positive, HER2-negative breast cancer in three patient studies. And we're going to go through a lot of questions about things that I think people have on their minds, but we'll focus it kind of how and when we need to use PI3K inhibitors.
And so we're going to start with the first-line treatment of a hormone receptor–positive metastatic breast cancer. And the first case, it's a 64-year-old woman, she's postmenopausal, has a history of asymptomatic left bundle branch block on her EKG as well as a longstanding history of irritable bowel syndrome. About a decade ago, she was diagnosed with a 3-cm ER 80%, PR 30% infiltrating ductal carcinoma of the left breast. She underwent a left mastectomy and left axillary dissection.
The pathology showed an infiltrating ductal carcinoma, 3.5 cm with 3 of 14 axillary lymph nodes positive. So again, I guess Heather, let's start with you. I mean, what would you do now in this case? This is actually 10 years ago, but let's say it's now, around now. I mean, what would you do?
Dr. Heather McArthur:
So this is a postmenopausal woman with ER-positive disease with one to three positive lymph nodes. This is someone who would have qualified for the RxPONDER trial. And so I would do genomic assay here to understand her risk of recurrence and to understand the predictive impact of chemotherapy, potentially. And she would be someone who would be considered for an aromatase inhibitor. And now with the recent approvals of CDK4/6 inhibitor therapy, we would also consider that in this patient as well.
So a lot of options here, we can talk about what CDK4/6 inhibitor we might consider for someone with a history of left bundle branch block. But it's a testament to the unprecedented number of successful drug approvals in this space that we have really improved outcomes for this patient population.
Dr. Adam Brufsky:
No, I would agree. And let's just say, for the sake of argument, she had a 21-gene assay, she had an Oncotype, and it was 27. So Seth, what would you do in this case?
Dr. Seth Wander:
I completely agree with all the points Heather just made, and I think we're entering a world where we have so many more choices. It makes these discussions a little bit more complicated. For a patient like this who meets criteria for RxPONDER—postoperative, fewer than four positive lymph nodes, with a score above 25—I would typically treat with combination chemotherapy. I would typically give something like docetaxel and cyclophosphamide here. We would reserve more aggressive therapy with an anthracycline for patients with more extensive regional lymph node disease, probably N2 disease or more, or patients who had lower ER expression, somebody who might be on the borderline of triple-negative tumor, somebody with maybe low ER and PR negativity.
And then I agree with all the great points that Heather just brought up about the new adjuvant CDK agents. The rubber will meet the road here a little bit when we have some more of the pathologic details. What's the grade, et cetera, would she have actually explicitly met criteria for monarchE vs NATALEE? And then to the point you just alluded to with her cardiac issues, which agent might we want to pick, thinking about potential toxicities?
Dr. Adam Brufsky:
So we'll get to an adjuvant CDK case in a few minutes. So let's hold on that thought. So it's still 2015, and her Oncotype DX score 27, and I think we'd all agree that we would give her chemo. The question I really was trying to get at here is hormonal therapy. I mean, this is right around the time, I don't even think it came out yet, of NSABP B42. So obviously this woman—she was postmenopausal 10 years ago when she was still 54. So let's say high Oncotype 54 hormonal therapy. How many years of hormonal therapy are you going to give her? She's obviously going to probably get an aromatase inhibitor, but how long do you think she's going to get? You want to start, Heather?
Dr. Heather McArthur:
It's sort of an antiquated question to be honest.
Dr. Adam Brufsky:
It is.
Dr. Heather McArthur:
In a modern era, so she probably would've received, and there's B42, MA-17, are all those studies that informed our thinking about extended therapy. I think we probably in someone with node-positive disease with three nodes involved would have erred on the side of overtreatment rather than undertreatment. I think she probably would've received, at that time in 2015, I think she would've received probably a dose-dense AC-T, and then a recommendation for extended therapy with 10 years of endocrine therapy.
And I'm not sure that that recommendation has changed much. We saw data from San Antonio showing that patients with high-risk node-positive ER-positive disease, some of them still derive benefit from anthracyclines. So I think that those options are still relevant. Of course, now we have genomic assays like Breast Cancer Index that we can apply at 5 years to refine our thinking about extended therapy. But in 2015, I think she would've received the full prescription for chemotherapy and extended therapy.
Dr. Adam Brufsky:
Got it. So Seth, do you use assays? Do you use BCI in your decision making?
Dr. Seth Wander:
I agree with all these points. I think everybody would have given her chemo, and I think in the modern era, whether it's a two-drug or three-drug combination would depend a little bit on how you interpret some of those prior studies in patient populations. I think it's also a little hard to know the exact benefit of the anthracycline given some of the better non–chemo-related adjuvant therapies we have today, including extended duration therapy and adjuvant CDK blockade.
But regardless of whether she got the AC-T or something like a dual regimen, I also would probably optimize her endocrine therapy by extending it past 5 years. It's interesting, Adam, your question about using some of these assays. I think if you go to different institutions, and I'm curious to hear from both of you. There are some places culturally where they like to use these additional assays beyond Oncotype to help further risk stratify. There are other institutions where culturally we don't use them quite as much.
We have not adopted it, for example, as much as MGH. The concordance between your clinical judgment and clinical acumen here vs the outcome on Breast Cancer Index—it’s I think, about an 80% to 90% concordance between factors like tumor grade, nodal status, size, et cetera, vs what you would predict you would see on the Breast Cancer Index. But regardless, I think most of us would err on the side as long as she was tolerating it well of extending duration past 5 years for some additional benefit.
Dr. Adam Brufsky:
Got it. Heather, anything to add to that before we move on to the real meat of this case in a minute?
Dr. Heather McArthur:
No, I agree with that. And I think that we are increasingly using tools like Breast Cancer Index, but there are so many online tools too, that are adding on to these genomic assays and helping us to refine our understanding about the integration of clinical plus genomic risk. And one additional issue, which I don't mean to throw a curveball, but that we've been struggling with is what's the role of adjuvant bisphosphonates in this setting where we now have CDK4/6 inhibitors. ASCO guidelines recommend that we consider adjuvant bisphosphonates in almost everyone. How much further risk reduction can we expect in a modern treatment area? And we've been really struggling with that question a lot on our end.
Dr. Adam Brufsky:
Yeah, I mean, that's a whole other discussion that we'll get us off the PI3K issue, but I'll throw my 2 cents in here as obviously a very bone-centric person who developed a lot of this stuff. And I will say that the meta-analysis showed a 3% overall survival benefit, breast cancer–specific survival benefit at 10 years in postmenopausal women for adjuvant bisphosphonates. So that's a survival advantage. It's not a DFS advantage, it's a survival advantage. But I agree with you, Heather. We just don't have any data.
And I think that this is going to be one where eventually with AI and probably mining some of the big databases, will probably answer some of those questions now that we have a couple of years of adjuvant CDK4/6 data. I think it's a great question. It's not going to be answered in a clinical trial, but probably with AI.
All right, so let's move on. So she does get dose-dense AC-T. She gets postmastectomy radiation, she's placed on an astrozole but only for 5 years. They only give her 5 years. Again, this is kind of before B42, and MA-17 was out there, but they figured they didn't do it for whatever reason. So she completes it in 2021, and it's now 3 years later, it's about a 1.5 years ago, she complains of back pain, which has been worsening for 3 months. You do a PET-CT, and she's got multiple lytic lesions to thoracic and lumbar spine, all of which are PET-avid as well as a 3-cm PET-avid lesion of the liver.
So now what do you do? Okay, so that's the next question. So obviously, do you rebiopsy her? I mean, that's a pretty standard question. Guys, would you rebiopsy her, and where?
Dr. Heather McArthur:
Always. We always rebiopsy to confirm a diagnosis of metastatic disease, you never make an assumption. So I would always biopsy, and in this case, if the liver lesion was accessible and it was safe and feasible, that's the lesion that I would go after. We worry about decalcification with bone biopsies and you can lose some of the integrity of the biomarker data, which we'll launch into. But that would be my preference to go after the liver lesion if it were safe and feasible.
Dr. Adam Brufsky:
Seth?
Dr. Seth Wander:
I completely agree for all the reasons Heather just alluded to. I've had this issue with some of my own patients with the decalcification and you're trying to kind of wonder how much true hormone receptor positivity is left. I would also think about targeted sequencing both on the solid assay as well as potentially concurrent blood-based liquid assay, which I know we'll get into, Adam, in terms of some of the newer targeted therapies.
Dr. Adam Brufsky:
Right. Well, let's ask. I mean, so would you do biomarker? Would you do next-generation sequencing on her liver? Let's say we do a liver met.
Dr. Seth Wander:
I personally would, and I think part of it is to kind of put that information in the bank. We have truncal mutations that are present at the time of metastatic diagnosis that can alter therapeutic choices in the future. We also have acquired mutations that are going to arise under selective pressure on various medications and we can get into what some of those alterations are. Having a baseline solid and liquid assay tells you sort of where you're starting, and then what I would typically do is repeat the liquid circulating tumor DNA assay at progression to look for dynamic changes, acquisition or loss of specific variants.
I think the field is moving more and more in that direction. Obviously, there are research questions to be asked there, but there are now actually concrete clinical actionable alterations that may appear and disappear over time that I think could change the therapeutic approach.
Dr. Adam Brufsky:
Fair enough. Let me move on to the next slide because I'm going to ask Heather this question because that's a great question. So she gets an ultrasound biopsy of liver, it's ER 50%, PR 20%, HER2 1+ adenocarcinoma, consistent with breast. You do next-generation sequencing on the tumor and she's got a PI3K mutation, the usual H1047R. So again, Seth, you would do liquid at this point on top of this or not?
Dr. Seth Wander:
I typically do concurrent solid and liquid at the time of metastatic diagnosis. There's about an 80% to 90% concordance between the two, but you can have some alterations that are missed on one or the other. The reason in my mind to have the liquid here is not only to pick up things you could have missed on the solid, but also because we're going to be comparing the progression samples, liquid to liquid, right?
We're going to use the same assay in the future. I think there's variability here, Adam. I don't think it's wrong to do one or the other. Our practice, and again, we do a lot of translational research and genomic type medicine has been to acquire this data across both the solid and liquid sites at tumor initiate, at treatment initiation in metastatic setting. I think you'll get pretty much the same information from both for the most part.
Dr. Adam Brufsky:
Got it. Heather, would you do a solid on it to begin with? You would wait?
Dr. Heather McArthur:
I mean, I understand where Seth's coming from with the EMERALD data for elacestrant and ESR1 mutations. We move from doing next-generation sequencing on diagnostic biopsy to waiting to first progression and doing liquid biopsies. And then if we didn't get the information that we needed for PIK3CA, we would go back and look at the tumor tissue. That was the recommendation at the time. However, with the INAVO120 data—and this patient doesn't meet criteria for INAVO120, that was a very high-risk rapid progressor population—but we now have that FDA-approved triplet combination in the first-line setting for those with PIK3CA-mutant,endocrine-resistant tumors.
So there's been a little bit of a shift. So I now have gone back to doing next-generation sequencing on tissue at baseline just for information gathering. But honestly, I'm not doing anything about it. It doesn't directly inform first-line therapy decisions because we're still giving an AI and CDK4/6 inhibitor or fulvestrant and CDK4/6 inhibitor in the first-line setting regardless of any of that information.
And I'm very allocation––minded. And so I'm now doing, honestly, NGS on tumor tissue at baseline to try and understand the biologic landscape and then going on to liquid biopsy at first progression on first-line therapy.
Dr. Adam Brufsky:
Got it. So she obviously gets that. And so, you're right. I mean, the big question here is, what do you do? And so, you're thinking kind of a PI3Kinhibitor, you're not quite sure of the INAVO120 data that we'll talk about in a minute. And so you do a fasting glucose, and it's a 120, her hemoglobin A1C is 6.2%. So what do you do now? What do you give this lady? Again, a history of left bundle branch block irritable bowel syndrome, and she now has bone metastases and then again does not meet the high-risk criteria, which are less than 6 months on an AI, or less than 6 months after completion of an AI in the adjuvant setting. So what would you guys give her? Heather, you want to start?
Dr. Heather McArthur:
I would probably give her fulvestrant and CDK4/6 inhibitor as first-line therapy given the interval since last exposure to an aromatase inhibitor.
Dr. Adam Brufsky:
And would you care about which one you give?
Dr. Heather McArthur:
Not really, no. In terms of CDK4/6?
Dr. Adam Brufsky:
Yeah.
Dr. Heather McArthur:
I probably give abemaciclib here because she didn't it adjuvantly in 2015, right?
Dr. Adam Brufsky:
Yes.
Dr. Heather McArthur:
So she didn't get adjuvant CDK4/6 inhibitors. So because of her history of left bundle branch block, I would probably, all things being equal, favor abemaciclib for her.
Dr. Adam Brufsky:
Seth, would you add a PI3K now or not? Give her a doublet?
Dr. Seth Wander:
I would do a doublet for the same reason. She doesn't meet the criteria for INAVO120 based on the timing of her relapse. I think as we're sitting here, three of us, if you ask five of us, you might get a couple of different answers as to which doublet you would give. I would probably stay away from ribociclib unless I had a good talk with her cardiologist and we were looking at the EKG and there weren't any concerns as I think Heather's alluding to.
I think the question of which endocrine backbone here is also a little bit debatable. She doesn't have the ESR1 mutation, and she's, I think, Adam, 3 plus years off of the AI when she recurs. So I don't think it's wrong to have that discussion. Whether you give AI plus CDK or fulvestrant plus CDK,, I think either would be reasonable. I'd be thinking about integrating either a PI3K or AKT inhibitor in the future, as I'm sure we're going to talk about with these upcoming cases.
Dr. Adam Brufsky:
Yeah. One last comment. Assume, say, it's 3 months. Now she's 3 months after completion of her AI. So someone had decided to keep her on the AI, and it had just stopped within a couple months. Would that change your guys' thinking at all? Would you think about the triplet?
Dr. Seth Wander:
It would, yes. In that case, she would meet INAVO120. I definitely would not put her back on the AI. I would go with the fulvestrant-based therapy and probably do the triplet.
Dr. Adam Brufsky:
Heather?
Dr. Heather McArthur:
I agree with that, yep.
Dr. Adam Brufsky:
Perfect. All right, so let's just talk about the key clinical takeaways, I think, that hopefully we've got here. The choice of 5 vs 10 years of endocrine therapy in the adjuvant setting is evolving. NSABP B42 did show an absolute 3.2% disease-free survival advantage in an unselected population. I think we've discussed genomic assays may help contribute to deciding who will benefit most.
The choice of a next-generation sequencing at diagnosis of an ER-positive metastatic breast cancer—instead of at first progression like we've been used to—is now really driven by the possible inclusion of inavolisib, which is PI3K inhibitor and palbociclib and fulvestrant in a triplet, in the INAVO120 trial, which did demonstrate a significant progression-free and overall survival advantage.
Again, we'll talk a little bit more about this in a bit, but hypoglycemia with PI3K inhibitors is a concern. I think baseline metabolic status should be assessed. The choice of CDK4/6 inhibitor, again, is usually driven by toxicity and patient comorbidities, in this case a bundle branch block.
And the choice of a doublet or triplet I think really should be determined with multiple factors, including pace of disease, bulk of disease, and whether they've relapsed on an adjuvant aromatase inhibitor. So I just want to thank both of you very much. And this brings us to an end of this case. Please see the other segments for further discussion about the latest research in metastatic breast cancer—that’s hormone receptor–positive, HER2-negative—or visit ascopost.com.
