Case: A woman in her mid-30s with de novo ER-positive, HER2-negative metastatic breast cancer (mBC) achieves a complete clinical response to palbociclib, letrozole and ovarian suppression

DR LOVE: For this issue, which focuses on estrogen receptor-positive breast cancer, I met with Dr Matthew Goetz, and to begin, he presented a patient from his practice.

DR GOETZ: She presented with a palpable mass and had appropriate workup at that time. Apart from the palpable breast mass she was asymptomatic, and she was found to have a locally advanced breast cancer. Estrogen receptor was strongly positive. HER2 was negative. And, of course, given the locally advanced nature of the tumor, she did have staging workup that showed evidence for multiple skeletal metastases.

DR LOVE: Did you have any sense for whether this is just the biology of disease, or had she delayed seeking care?

DR GOETZ: Actually, this particular individual was someone who was very much attuned into her health. She was actually — I would call her perhaps even a health nut. She exercised regularly. She was somebody clearly who did not delay seeking attention. This was something that — what we would call truly de novo Stage IV, somebody presenting at the time with disease both simultaneously in the breast and systemically as well.

Importance of providing support to patients and their families coping with a diagnosis of mBC

DR LOVE: Before we start to kind of get into the clinical research that relates to this kind of case, any thoughts about what it was like kind of dealing with this woman who’s so health avid, now has metastatic disease?

DR GOETZ: I think this is, of course, a huge issue, because first of all, being a young woman who has so many different things that she’s dealing with — her career, her young family, young children. Knowing that she’s done everything possible up until this point to take care of herself. And by the way, I didn’t mention this, but there was no family history of breast cancer. And the genetic workup that we did was negative. There’s clearly a lot of unanswered questions for this individual. And so moving through the diagnosis and going from “I have breast cancer” to “It’s Stage IV” was obviously, in light of her family and everything else, was obviously a very difficult situation.

DR LOVE: Our group has had an interest for a long time in the issue of the minor children and grandchildren of patients with cancer, particularly those with incurable cancer. And I’ve been kind of disappointed about the lack of national support. A lot of research centers will have counseling, et cetera. Any thoughts about how you approach this and how your team approaches it?

DR GOETZ: I think one of things that’s important up front, of course, is to really — when we see a patient with Stage IV disease — is to provide as much information as we can about the biology, about what the prognosis is. And there needs to be some realism so that they can plan.

I think probably the worst-case scenario is a situation where the patient doesn’t completely understand the prognosis and is not able to prepare. And I think especially for individuals with young children, you want to be able to — if you really have an opportunity, whether that survival is 3 years or 2 years or 5 or 10 years, you have an opportunity to leave a legacy and to impart that upon your children, upon your family. I think it’s critically important that clinicians, social workers, et cetera, provide that support that allows those patients and their families to know realistically what’s going on. But, of course, we have to provide hope, and the good news, of course, is, there has been a lot of advances that really have improved the outcomes of these patients.

Therapeutic options for patients with de novo ER-positive, HER2-negative mBC

DR LOVE: Maybe you can talk a little bit about the options that a woman in this situation might be considering or you might be considering. And also, you mentioned she had locally advanced disease. Were you concerned about losing local control?

DR GOETZ: I think that’s a really good point. And certainly it’s one of the things we think about in these patients, where we are thinking about systemic therapy. And this really gets into the issue of how confident are we our therapy is going to work? In this situation, from the start it appeared this woman had a truly estrogen-dependent tumor, at least on the face, from what we could see. And that if we were able to provide adequate systemic therapy that we should be able to achieve both local as well as distant systemic control.

DR LOVE: Maybe you can talk about from your point of view what were the options that were considered and what you ended up treating her with.

DR GOETZ: I think things certainly have evolved especially over the past — even the last 5 years for the treatment of metastatic ER-positive breast cancer. And certainly now with the data from MONALEESA-7 with regard to the treatment of premenopausal patients.

But I think in general the point has been, of course, that in the metastatic setting we’ve treated women with ER-positive breast cancer much like we have in the metastatic setting, and that is to adequately suppress estrogen. And that really means, in the premenopausal setting, both a drug like goserelin or oophorectomy along with an aromatase inhibitor really gives the best outcomes. And I think in this situation, in 2016, the question that we were faced was, should we additionally give a CDK4/6 inhibitor given the emerging data in the postmenopausal setting?

Activity and tolerability of gonadotropin-releasing hormone agonists

DR LOVE: One of the issues I’m curious about, you mentioned the goserelin, and I hear about this fairly frequently, is concerns that physicians in practice have about whether it’s adequate ovarian suppression and how to determine that. And how often it can actually be a problem.

DR GOETZ: This is a really important point. I don’t think we completely know the answer. If we were to look at data in the postmenopausal setting, where we give aromatase inhibitors routinely, there are emerging data and data that will be presented at some upcoming meetings looking at this question of the levels of estradiol and how well one suppresses that and whether that actually is associated with outcomes.

In premenopausal women we have very few data, and I think clinicians, rightfully in this situation, aren’t thinking about this in the correct manner. What we do need to know is, are we adequately suppressing the ligand? And if that is important, we think it is important, should we be measuring it? Right now we don’t measure it on a routine basis, but certainly we’ve all experienced patients in our practice that for whatever reason did not have adequate ovarian function suppression, and we’ve perhaps gone on to do different things, such as oophorectomy or other measures to try to deal with that.

DR LOVE: What about with this woman? Did you bring up surgical oophorectomy with her?

DR GOETZ: Actually, I did. And what we decided at the time of the consultation was that we would get going with the goserelin. That’s an easy thing to start with. I like to start with goserelin for my patients because it allows them, if you will, to be able to get used to the effects of ovarian function suppression and to know perhaps after a couple of months that it is tolerable, which it is for most patients in this situation. And then they’re quite willing to move right into oophorectomy.

Occasionally I’ll have patients who say, “Hey, I just want to move right into oophorectomy.” And of course, as we know, that can take anywhere from a few days to weeks to get that scheduled surgically.

DR LOVE: She had goserelin, and I see she also had letrozole and palbo. And certainly a common choice. How did she physically feel once she got on the therapy?

DR GOETZ: This is a case of a patient who actually tolerated the therapy beautifully. She was able to take the medication and really have very minimal, if any, side effects. Now, this particular patient, as I mentioned, was someone who was working out regularly. She was running anywhere from 1 to 2 miles a day. And she did mention to me that she had some fatigue, but she described it as being very mild and really not interfering with her activities of daily living.

DR LOVE: That’s interesting. Do you find that patients who exercise tolerate these agents better?

DR GOETZ: I think that there clearly is — in general, we see that patients who are metabolically fit can oftentimes do better with the therapies which we know have a lot of effects on a variety of different organ function. Certainly in this situation, and I know clinicians would feel the same way, is that if you have a patient with an ECOG performance status of 0 and is doing all their activities of daily living that they’re going to do much better than those patients that have multiple comorbidities and are starting out with difficulties as it relates to fatigue or other side effects from their chronic medical illnesses.

Selection of patients for treatment with CDK4/6 inhibitors in combination with endocrine therapy

DR LOVE: It’s interesting, you may know that we are always asking people, both investigators as well as docs in practice, what they do in various situations. We did a big survey where we presented all kinds of scenarios — ER-positive, HER2 negative, metastatic disease. And I was kind of surprised, particularly in patients who present with metastatic disease, that a substantial minority do not use a CDK inhibitor. And if we throw in 1 or 2 liver mets, you’ll start seeing a lot of chemotherapy also. Can you kind of talk from a broad point of view, before we get into the specific agents, about these kinds of questions? Are there patients where you don’t need a CDK inhibitor? And what would it take to use chemotherapy?

DR GOETZ: These are obviously important questions, and I think as we’re beginning to see more and more data about the CDK4/6 inhibitors and their efficacy, we’re beginning to see survival data. I think it’s beginning to be harder and harder to find patients, let’s say, that we should be omitting CDK4/6 inhibitors.

Now, that being said, we’ve all seen patients in our clinic who are known to have hormonally driven cancers. A patient, perhaps, has a long disease-free interval from the time that she was originally diagnosed, and let’s say completed her adjuvant hormonal therapy, who presents perhaps with minimally symptomatic disease. We can think about these patients and know it’s probably not unreasonable to treat these patients with hormonal, let’s say, monotherapy, like an AI.

But in general, as these data emerge showing clear evidence for now survival advantage, we see, for example, data that we can prolong the time to chemotherapy. And these are important endpoints, clearly, for patients. And I think as we begin to see these data more and more, it’s going to be more and more difficult to find these patients, let’s say, that should be treated with just endocrine monotherapy.

DR LOVE: Another thing that I recall from that survey that was interesting is, again we started to see a tail-off in use of CDK inhibitors in older patients, particularly 80-plus. Any justification for that?

DR GOETZ: This is a thing where a lot of clinicians really do think about tolerability. We know that as a person ages, their organ function changes. Their ability to metabolize drugs changes. And so rightfully, I think clinicians are cautious in this situation. I think what we can say in this situation is that when people have looked very carefully thus far at elderly patients, they seem to tolerate the CDK4/6 inhibitors well. And specifically I’m thinking about palbociclib.

And despite the fact that we do see neutropenia, and of course all of us as clinicians when we first saw the data with CDK4/6 inhibitors, when we saw neutropenia, we thought about, oh my gosh, this is going to be chemotherapy-like neutropenia. But really it is quite a bit different. And I think what we can say is that the elderly can do well. They can tolerate CDK4/6 inhibitors, and they clearly can gain benefit from them.

Mechanism of action of CDK4/6 inhibitors

DR LOVE: Just to start with the basics, can you go back through your vision, how you explain to your fellows basically how these agents work?

DR GOETZ: When I think about these agents, and first of all, stepping back, when we think about ER-positive breast cancer and knowing a little bit about how breast cancer cells utilize the estrogen receptor to grow, we of course know that blocking the estrogen receptor or controlling the ligand with aromatase inhibitors is a quite effective treatment but that with long-term treatment there can be resistance that develops.

Now, the mechanisms of resistance that we know are quite diverse, both de novo as well as acquired resistance. And one of the things about CDK4/6 inhibitors is that CDK4/6 is a critical aspect that interacts with cyclin D. And cyclin D is a major transcriptional, if you will, gene that is regulated by estrogen receptor. But it’s also importantly regulated by a number of different pathways.

And so if we think about why, for example, people have observed the fact that CDK4/6 inhibitors have activity in HER2-positive breast cancer, again, it’s because some of those same pathways downstream actually also target cyclin D1. And so these pathways emerge on CDK4 and 6 and cyclin D. And because of that there is an ability to potentially deal with resistance that occurred, both de novo, by the way, as well as acquired.

One of the things, of course, that we’ve all been quite interested with CDK4/6 inhibitors is this question of ESR1 mutations, which, as we are hearing, is a major mechanism of resistance to, for example, aromatase inhibitors. And again, knowing that the cyclin D1 is downstream of the estrogen receptor, despite the fact that the receptor is mutated, we have the ability, if you will, to rescue that particular cancer cell by inhibiting this common pathway. That is CDK4/6 and cyclin D ultimately going through and affecting the cell cycle.

Role of everolimus and alpelisib in the management of ER-positive mBC

DR LOVE: How do you bring in mTOR inhibitors and PI3 kinase inhibitors into this kind of biologic model?

DR GOETZ: I think in this situation what we have is really a question of, with CDK4 and 6, two aspects. Number one is, what are the mechanisms of resistance to CDK4/6, and if one of those mechanisms, perhaps, is activation of the PI3 kinase/AKT pathway that perhaps it would make sense to start thinking about combination therapies. And, in fact, there are preclinical data. Those data do suggest that that pathway is important, and then actually doing studies where you combine with mTOR inhibitors or PI3 kinase inhibitors may make sense.

I think we get into that age-old question which we’ve always had, and that is, can you combine therapies, and will it be tolerable? Because now all the sudden you bring in different pathways, different side effects, and indeed when people have looked at the so-called triple therapy, and we did a study at the clinic where we used a TORC1/TORC2 inhibitor along with targeting the estrogen receptor along with CDK4/6. In general, you’re seeing more side effects, and for a lot of patients this is just not going to be tolerable.

Then we get back into this question of is it best to probably do sequential therapies, and is there a subset of patients where absolutely we should be thinking of triplet therapy? And I think for right now I see it more of sequential. I see it’s a great idea. I think scientifically it makes sense. Clinically it just hasn’t been — we haven’t been able to achieve bringing in multiple drugs and doing so-called triplet combinations without substantial toxicities.

Implications of ESR1 mutations for the management of ER-positive mBC; use of selective estrogen receptor degraders (SERDs)

DR LOVE: You mention ESR1 mutations. Can you talk a little bit about exactly how that works and in terms of what the result is? You hear you can use fulvestrant or agents like that, but not ligand-removing agents like AIs. What’s the biology of that?

DR GOETZ: Sure. This is, of course, an area that has really just, if you will, exploded in the last 5 years. Suzanne Fuqua did some very important research on this many, many years ago, actually more than 20 years ago, and at the time it was actually observed that these ESR1 mutations do confer resistance in the laboratory to endocrine therapy. But it felt like that these ESR1 mutations were not really observed in patients’ samples.

The field did not really move forward, and it was really because of the advent of next-generation sequencing that people really started seeing these mutations. And what they saw was that they were emerging in metastatic disease, and they were emerging really in patients who have so-called acquired resistance.

If you long-term estrogen deprive a cancer cell, in this case, using specifically aromatase inhibitors, the observations were that these clones could emerge, and these clones — what am I talking about? They are mutations in the estrogen receptor gene. They confer resistance, and these mutations essentially turn on the estrogen receptor. They cause it to signal despite the fact that the ligand perhaps is removed, in this case with an aromatase inhibitor.

These ESR1 mutations are a problem because now as a clinician, we use a medication, we’re effectively getting rid of the ligand, but despite the fact that we’ve removed the ligand, the cancer cell and the estrogen receptor is still signaling. It’s still on. And so if we think about strategies that would potentially target these mutations, 1 strategy, of course, is to simply get rid of the receptor completely. And that has been really shown already that fulvestrant seems to have activity in patients with ESR1 mutations.

But if you look at the data, one of the problems with fulvestrant is really more of a drug delivery issue. You can’t give enough of the drug in patients in an intramuscular buttock shot to actually completely and totally degrade the receptor. In fact, at the concentrations that are achieved in the laboratory, fulvestrant is one of the best drugs we’ve ever used. But in patients we just don’t get enough fulvestrant in to achieve optimal estrogen receptor degradation.

That really has prompted a lot of focus on developing new drugs that target the estrogen receptor. We call them next-generation SERDs, and what’s really exciting about this, Neil, is the fact that there are a number of companies that have these drugs. They effectively target the receptor for degradation, and we’re seeing data both preclinically, in ESR-mutant models where these drugs work, but also we’re seeing clinical data. And so I think we’re very excited to move these drugs into the clinic and to see if they can be a step forward for treating ESR1-mutant breast cancers.

Activity of the selective estrogen receptor modulators endoxifen and lasofoxifene for mBC with ESR1 mutation

DR GOETZ: Now, another approach is actually to use a SERM. And so our group has studied the drug endoxifen for a while. We have a randomized clinical trial that will read out later this year comparing endoxifen with tamoxifen. And when we published our data a few years ago in Journal of Clinical Oncology, we showed that endoxifen had antitumor activity in ESR1-mutant tumors.

Similarly, there is another drug called lasofoxifene, and if you’ve been in breast cancer for a while, you’ll think lasofoxifene, that’s an old drug, and indeed it was developed years ago. And lasofoxifene is a drug in the PEARL study, which was a study that was evaluating women at risk for osteoporosis or perhaps who had vulvovaginal issues. Lasofoxifene in that study as a secondary endpoint reduced breast cancer by over 80%.

It was not taken forward but recently has been picked up by another company, and there’s renewed interest, and specifically because there’s data that lasofoxifene fits into the estrogen receptor and specifically causes a distortion in such a way that in these ESR1-mutant models can actually inhibit ER transcription and proliferation and perhaps even better than the drug fulvestrant. There’s a study ongoing with lasofoxifene specifically in these ESR1 mutants. I would say there’s an explosion of, if you will, of studies specifically looking at this very critical mechanism of resistance, which is ESR1 mutations.

Tolerability profiles of the CDK4/6 inhibitors palbociclib, ribociclib and abemaciclib; dose reduction to mitigate side effects

DR LOVE: Getting back to this patient, you chose to give her, in addition to ovarian suppression, letrozole and palbo. And I think that would really give us an introduction into the issue of what we know about (A) the choice of up-front endocrine therapy to start with, because you mentioned fulvestrant. There have been trials, really, in the pre-CDK era, where fulvestrant and an AI was used, or fulvestrant. And then, of course, the issue of choosing a CDK inhibitor, specifically in a premenopausal woman. Maybe you can kind of go through an overview of what we know about really the 3 CDK inhibitors that are approved right now.

DR GOETZ: It’s a remarkable time, Neil, because we have 3 CDK4/6 inhibitors, as you said, that are approved, and they’ve all shown efficacy. And remarkably they actually have different tolerability profiles. Now palbociclib and ribociclib appear to have quite similar tolerability profiles, and perhaps ribociclib has a few additional differences than palbociclib. Abemaciclib stands out as being quite a different drug. Different not only in terms of its effect on CDK4 and 6, more potent toward CDK4, the fact that abemaciclib is dosed continuously as opposed to discontinuously, but also abemaciclib different in, obviously, the tolerability profile or the side-effect profile.

The side effect we see, of course, with these drugs across the board is neutropenia. And, in fact, this was really the defining side effect that had concerned most clinicians when these drugs first came on board. And as we’ve all gotten used to taking them, we realize that the neutropenia that we see is something that can be dealt with. And it’s really rarely associated with substantial problems related to infections.

If we think about another class effect, I think what we’re seeing is fatigue is clearly a class effect. We might see a little bit more alopecia with these drugs, although that is not a predominant side effect, for the most part. When we think about abemaciclib we see less cytopenias, so we see fewer problems with neutropenia, and diarrhea appears to be the predominant side effect of abemaciclib. And most of the studies that have been done more recently have really focused on utilizing a more aggressive antidiarrheal regimen to deal with the side effects or the diarrheal side effects, excuse me, of abemaciclib.

I think if we look across these drugs, clearly they are different and different side effects. But also I think one of the things we can say is that we can use information about the side effects to help individualize therapy. So, for example, I’ve started some patients on CDK4/6 inhibitors, and they’ve had profound neutropenia despite multiple dose reductions, and we’re actually unable to keep them on the drug. Now, that’s unusual, but certainly we can all think of patients in our clinic where we’ve gone down to the 75-mg dose days 1 through 21 and still had problems.

And I think where clinicians begin to be concerned about this, Neil, is where we have prolonged time off the CDK4/6 because we’re worried about this so-called rebound G1/S, where we’re going to see rebound proliferation. And I think that concern is probably most acute early on. Once patients are on the drugs for a prolonged period of time, one of the observations that has been made, both preclinically as well as clinically, is that you might get senescence. And so, therefore, there’s less critical problems, if you will, if a patient is off the drugs, let’s say, for 2 weeks as opposed to 1 week. Nevertheless, I think a theoretical advantage of abemaciclib is that we don’t have to worry about the cytopenias as much, and we can dose the drug continuously, and there’s perhaps theoretically less worry about this rebound effect with growth.

CNS penetration of abemaciclib; management of abemaciclib-associated diarrhea

DR LOVE: One of the things I’ve heard mentioned about abemaciclib, I’m curious what you think about it, is the CNS penetration and whether that actually is a clinical consideration.

DR GOETZ: I think it’s an interesting observation. Certainly when we use these drugs in the first-line setting, most of our patients, of course, really don’t present with CNS metastases. That’s not a defining early event. I do think that as we get better about identifying who is at risk for CNS events, then utilizing a drug that has CNS penetration is going to be critical. Right now, the frequency of this is so low and I just don’t think we can necessarily use this, especially in the earlier stage or in the first-line setting.

DR LOVE: You mention also the issue of diarrhea with abemaciclib. I’m curious what you actually do practically. Do you start patients on preemptive medication? Do you just tell them to contact you if they start to have a problem? What do you actually do?

DR GOETZ: What I do in my clinic is to make sure that patients are informed about the risk. I give them a prescription for loperamide. I make sure that they’re ready to take it. I have not preemptively given it in all my patients, because I frankly had patients who’ve done that and developed constipation problems. The patients that I think about as being where this could be problematic are those perhaps that had prior problems with diarrhea for whatever reason. Obviously we’re pretty careful not to put patients with inflammatory bowel disease or other chronic conditions on these drugs but in general approach to make it available, and so they can use it at the first time of diarrhea seems to work.

Effects of site and extent of metastases on outcomes with CDK4/6 inhibitors

DR LOVE: Another issue I want to ask you about is the issue of liver metastasis, which this patient didn’t have. But I was mentioning in this survey, when we put in even a couple asymptomatic liver mets, all of a sudden we start to see a lot of chemotherapy being used. What do we know about hormonal therapy with CDK in terms of visceral mets and liver mets?

DR GOETZ: This has really been a subject of some secondary analyses, and we reported first at San Antonio a couple of years ago and then published data with regard to what were the critical, if you will, important prognostic factors associated with outcomes in the control arms. And control arms here in both MONARCH 2 and MONARCH 3, speaking about either fulvestrant alone or the AI alone.

And what really came out of that was that really data that we’ve already known for some time but really provided with much more granularity the fact that patients with liver metastases do very poorly with endocrine monotherapy. And this is where the data with regard to abemaciclib showed substantial benefit, both in terms of improving response rates, response rates that approached actually up to 60%, as well as a more than doubling of progression-free survival.

And by the way, I think both the studies with regard to palbociclib as well as ribociclib again have shown benefit in patients with visceral metastases. What we can say, Neil, in this situation, is that perhaps in the past where we would say liver metastases, oh my gosh, I need to go directly to chemotherapy. I think that what we can say is that in those patients that are relatively asymptomatic, have disease with little or no synthetic problem as it relates to their liver function, that a CDK4/6 such as abemaciclib really leads to response rates that probably are comparable to single-agent chemotherapy or even some of the doublet chemotherapies that we used years ago.

And I think that’s obviously exciting for patients because, just like I said with this particular patient that I mentioned, you really don’t want to be coming back and forth, if you can, with a new diagnosis and a young family, back and forth to the chemotherapy unit if you don’t need to. And this has been in my mind one of the great advances of the CDK4/6 inhibitors.

Efficacy of CDK4/6 inhibitors in combination with endocrine therapy for ER-positive mBC

DR LOVE: You mention the response rate with abema being close to 60 but the control, I guess hormonal therapy alone, being only 20%.

DR GOETZ: Right, and I think in the past, Neil, as we have given hormonal therapy in the metastatic setting, we certainly have known that there might be certain groups of patients that have, let’s say, better or worse outcomes. But these large randomized Phase III trials have given us the opportunity to really examine more closely, what are those factors that are associated with endocrine insensitivity or worse outcomes? And again, with the idea to say that as we get smarter and we can identify these patients up front, we can do hopefully an intervention that’s going to lead to better outcomes earlier on.

DR LOVE: You were mentioning the fact that there are 3 CDK4/6 inhibitors approved. Kind of reminds me a little bit — there’s 6 checkpoint inhibitors approved right now in oncology, and it kind of seems in general, when you have these situations, I’ve seen people putting up slides, doing indirect comparisons and saying “I’m not supposed to do this, but they all look the same.”

But then you find people, even if they believe the effects are the same, follow the data. First of all, do you in your heart feel there’s any way to distinguish efficacy? You mention toxicity, but efficacy of the three. And when you start to look at the data that we have to work with, you mentioned survival data that’s starting to come out, also data specific to premenopausal, how do you differentiate the available data?

DR GOETZ: This is obviously difficult, and as you said, one has to be very cautious, because doing these cross-trial comparisons is fraught with issues. I think what we can say, Neil, is that thus far we have data that the CDK4/6 inhibitors in high-risk populations can improve not just progression-free survival but overall survival.

Survival outcomes with CDK4/6 inhibitors and endocrine therapy for ER-positive mBC

DR GOETZ: Let me give you an example. I think what I was most impressed with from the MONALEESA-7 data was that this is a group of patients that is premenopausal patients, where in which really the event rate, and the event rate here are deaths, is actually occurring at a relatively early time point. For example, we don’t have data as of yet in the first-line palbociclib or the first-line ribociclib or first-line abemaciclib with regard to survival simply because in those patients we don’t have the events. Those patients are doing well.

And we’ve always said, “Is an ER-positive breast cancer that occurs in a premenopausal woman the same as a postmenopausal woman?” I think what we can say here is that the benefit of the CDK4/6 inhibitors is not just a progression-free survival benefit, especially in these premenopausal women, but it’s a survival benefit. And that that benefit is probably occurring because there are more events. This is a more aggressive, if you will, tumor. And again, there’s a variety of reasons for that. It could just be host factors. More estrogen around, so more possibilities of a need for additional ways to control proliferation. I think the MONALEESA-7 data to me are convincing that these drugs are leading to meaningful improvements in an endpoint that we all agree is survival.

We don’t have the data as of yet, but there was a press release with regard to abemaciclib showing that abemaciclib, in MONARCH 2, and of course MONARCH 2 as the second-line setting with fulvestrant has improved survival, and we’ll await those data to see that. And of course we have the data from the PALOMA-3, and of course that was the trial that looked at fulvestrant plus palbociclib published in The New England Journal of Medicine within the last year, showing a survival advantage, at least in the more hormonally sensitive group.

And so here’s the question, and that question is for clinicians and for me as well, is that which drug should I use? Is there a drug perhaps in a more hormonally sensitive setting that I should be thinking about more than in, let’s say, that patient that has 3 liver metastases and would otherwise be considered to have more hormonally insensitive disease?

Those are the issues that I think we’re going to need to sort out. It looks to me like these drugs are not the same. They all have efficacies, no doubt, clearly different tolerability profiles, but they may have differences in these different settings. That is more hormonally sensitive versus more hormonally insensitive tumors. And so we’ll just have to see the data, Neil, as time goes on.

Response to palbociclib and endocrine therapy in patients with ER-positive mBC

DR LOVE: I want to get your thoughts on beyond first-line therapy. Let’s come back to your patient. Can you talk about what happened to her tumors as she was treated?

DR GOETZ: I think it’s quite interesting to me when we see the, if you will, the pace of response to these medications. We can all think about patients in our clinic where we see just this slow response over time. And then I’m thinking about this particular patient, where she came back at her first scan, I think it was about 3 months after we first saw her, and her disease, basically — her PET scan showed a complete response to therapy.

And in that type of response, of course, those are the patients we love to give that news to. They are gratified. They are in tears. They are excited. Obviously this has implications for their planning. And this was a case exactly like that, Neil, where it was almost a complete response right away.

Interestingly, she had called our nurse probably about 2 weeks after she started the drug, and she said, “Within 2 weeks I’ve noticed already that the tumor within my breast has decreased.” And so there was almost an immediate, if you will, a sense by the patient that something good was happening. And obviously that was confirmed when we got her first scans.

DR LOVE: It is interesting she had this clinical complete response. You hear people say you don’t see path CRs when you give neoadjuvant hormonal treatment. But what do we know right now about neoadjuvant hormonal therapy with CDK4/6 inhibition?

DR GOETZ: What we know, Neil, is that when we treat, obviously, with a drug like tamoxifen or an aromatase inhibitor, we can inhibit proliferation. And how well you inhibit proliferation has implications long term. We know that in those neoadjuvant studies that have evaluated, for example, abemaciclib or palbociclib, that we inhibit proliferation much better. In fact, proliferation reductions in Ki-67 are substantial. That’s the only word I can think of.

And so with that in mind, I think the observation and the thought is that given that much inhibition of proliferation that this certainly should have implications as we think about the adjuvant treatment of ER-positive breast cancer and as we await the outcomes of these trials evaluating CDK4/6 inhibitors in the adjuvant space.

Selection of therapy for patients with locally advanced ER-positive, HER2-negative breast cancer

DR LOVE: It also makes me think a little bit about how you might have approached this patient if she did not have metastatic disease. Would you have used neoadjuvant chemotherapy? Would you have considered hormonal therapy? I don’t know if you ever would use a CDK inhibitor outside a trial in a neoadjuvant setting. Would you have gotten a genomic assay?

DR GOETZ: I think what I would think about in this situation is to say, let’s say this particular woman who presented with a locally advanced breast cancer, ER-positive, HER2-negative, without metastatic disease, I would really be prompted to think about the data from the SOFT and TEXT trials, knowing that (1) ovarian function suppression is critical, (2) knowing that an aromatase inhibitor gives you a little bit better outcome and (3) you mention the genomic assays. I think certainly what we’re hearing now with regard to the use of the genomic assays, especially in premenopausal women in this, if you will, secondary analysis from TAILORx where it appears that the cut points maybe different in premenopausal patients. And does that have implications for choosing chemotherapy?

In this particular patient, had she presented with a locally advanced disease, let’s say a 3-cm tumor with 1 positive lymph node, and she was 35 years old, I think in the past without a doubt we would say that that woman should get chemotherapy and should then be treated with the drug tamoxifen. But I think and what we’re certainly seeing is, there’s a group of these patients that probably have highly endocrine-sensitive tumors, perhaps that she had presented with a Grade 1 highly ER/PR-positive, HER2-negative, Oncotype Recurrence Score^® of, let’s say, 5, I think we would all question whether chemotherapy might just be a toxic way to suppress the ovaries.

Now, this is still an important question, but the TAILORx studies do suggest that there may be a group of premenopausal, or women less than the age of 50, that still don’t derive benefit of chemotherapy when their recurrence scores are low. And I think this is something that we’re all thinking about in our practice.

DR LOVE: In getting back to this lady. Now she’s had this great response. Hopefully she’ll tolerate therapy. How is she tolerating the therapy, incidentally? Any issues with neutropenia?

DR GOETZ: No. So, I mean, the neutropenia that she’s had is mild. It’s been well tolerated. We’re talking about neutrophil counts perhaps during that week off of therapy in that range of about 1,000. And then after a while, she’s been on it for so long we don’t necessarily check a neutrophil count with each cycle. We’ll check in about every 3 months with her CBC. But she really has continued to do well. Excellent quality of life.

Therapeutic approach for patients who experience disease progression on a CDK4/6 inhibitor

DR LOVE: If and when she were to develop progressive disease, first of all, how would you work her up? What kind of genomic assays would you do? And what would be treatment options, assuming she was clinically stable without visceral disease?

DR GOETZ: I think 2 things. One, we would certainly think about doing a biopsy, and people still ask me this question of is there value of doing a biopsy after you progress? And of course in this situation we know that there still are patients who progress where a HER2 clone can emerge where the tumor can be estrogen receptor negative. This is uncommon, but it’s common enough, and in an age of individualized medicine it’s certainly something that I think about. And especially at that time of first progression, where, let’s say, in this case the patient has been on therapy for quite some time.

I think the question we try to deal with is, we think about these patients, and what’s running through my mind is, what are those mechanisms of resistance? Certainly we’re beginning to understand that the pathway, and specifically RB, which is this gatekeeper, is critical for the activity of CDK4/6 inhibitors. And people have observed either RB loss or RB mutation as one mechanism for resistance.

But we also know that certainly ESR1 mutations can be present, and we also know that these PIK3CA alterations and clones can be present either from the start or they can emerge. And so I think as the field is moving forward what we’re seeing, Neil, is that we’re beginning to think about using the biopsies not just to assess whether ER and HER2 is present but also, given the data with regard to alpelisib and PIK3CA mutations, that we should be thinking about retesting the tumor and that it may not be adequate enough to use information from that tumor that was, let’s say, biopsied 3 years prior.

DR LOVE: And for practical purposes, does that mean NGS? Or do you get a specific PI3 kinase assay?

DR GOETZ: I think it is both. I think a lot of us in practice have been doing NGS. I think not everybody, but — and the advantage of an NGS assay, of course, is that it can provide information potentially on whether — not only on, in this case, using alpelisib, but also potentially prompting and getting us information on future therapies down the road.

I usually like to tell patients when I’m seeing them in the clinic it’s a little bit like a chess match. You want to be able to think 2 to 3 to 4 moves ahead. And so having that information can provide at least potentially that — some options, whether it be a clinical trial or whether it be another therapy down the road. And we can use that information to be thinking about subsequent therapies as well.

DR LOVE: How would you think through the approach to this situation of progressive disease after a good initial response? PI3 kinase-positive versus -negative? Why don’t we start with the negative first?

DR GOETZ: I think that one of the important questions that’s unanswered is, what is the activity of, if you will, second-line therapies after you progress on a CDK4/6 inhibitor? And we looked at this in our own Mayo patients recently. And what we found was actually quite sobering. And that is, patients that were treated with endocrine monotherapy after progression on a CDK4/6 inhibitor had actually very, if you will, short median progression-free survival of somewhere in that range of 2 to 4 months.

Interestingly, if you look at data from SOLAR-1, there was a small subgroup of patients in that trial who had progressed on CDK4/6 inhibitors, and it was a very small number, but again, those patients were randomized to fulvestrant or the combination with alpelisib. And again, in that small number, the median progression-free survival, if I recall, was something around 2 months.

And so one of the questions that we think about in this situation is, does endocrine monotherapy have activity in this post-CDK4/6 space? I think in the past we would say, “You progressed on an AI. Let’s give you fulvestrant.” We know it has activity, or we can sometimes use tamoxifen. Or we can use another endocrine agent. I think what we’re seeing, Neil, is that these tumors are probably rewired. And they’re probably rewired in ways that we don’t completely and totally understand.

And so as we think about next steps, certainly in that setting where the patient does not have a PIK3CA mutation, I’m thinking about a clinical trial. What about in practice? If the patient has minimal disease burden, let’s say bone-only disease, I don’t think it’s unreasonable to use endocrine monotherapy.

But one of the important questions that’s unanswered here, and that is, should we continue the CDK4/6 inhibitor past progression? I’m seeing some people do this in practice. And, quite honestly, I think there’s some emerging preclinical data — this data’s not published yet — that might suggest that that may have value. There are trials that are looking at addressing this issue. Of course this is going back in time and thinking about trastuzumab all over again when we were dealing with this years ago. Should we continue trastuzumab past first progression? And I think the question is, is this pathway so critical that if we target other pathways we should continue inhibition of CDK4 and 6? I think preclinically I’m thinking that that may be the case, and clinically we’ll just have to see the data.

DR LOVE: Prior to alpelisib coming on the scene, a lot of people talked about everolimus with exemestane, for example, after first-line endocrine therapy. Is that something you do or consider? What do we know about that after a CDK4/6?

DR GOETZ: It is, and actually in the data that we’ve presented at San Antonio last year when we looked at the outcomes of patients treated in our institution that had progressed on first-line CDK4/6 inhibitor, as I mentioned, we saw that the median progression-free survival with endocrine monotherapy — and this was, again, fulvestrant and tamoxifen — was really quite short. But, interestingly, we found that patients treated with exemestane and everolimus actually did quite well. And, in fact, we had in the relatively small group of patients that we were looking at, about 10 to 15 patients, the median progression-free survival was somewhere in that range of around 10 to 12 months.

One of the questions that we’ll need to understand is, is the mTOR pathway an important component of resistance to CDK4/6 inhibitors? And if so, if a patient progresses on a CDK4/6 inhibitor, should we be targeting that pathway? We have data, of course, coming up hopefully fairly soon from the BYLieve study, which is essentially looking at the activity of alpelisib along with fulvestrant in patients that have progressed on a CDK4/6 inhibitor. So we’ll be able to gain some data with some certainty on the activity of targeting this pathway in the post-CDK4/6 setting.

Activity of alpelisib for ER-positive advanced breast cancer with a PIK3CA mutation

DR LOVE: Can you talk a little bit about the data we have on alpelisib and what your experience is both in terms of efficacy and tolerability?

DR GOETZ: The data have been presented and recently published. And of course we know that this has just recently been FDA approved. What we know about alpelisib in the PIK3CA setting is that certainly it provided robust antitumor activity in terms of progression-free survival. We saw essentially a near doubling of progression-free survival in patients treated with alpelisib compared to fulvestrant alone. The overall survival data at this point are immature.

But I think what was quite convincing from the SOLAR-1 data is, this drug doesn’t have activity in patients that don’t have the PIK3CA mutation. And so this is truly individualized medicine, and this is really a scenario where we have to test in order to know whether the pathway is activated.

To me what’s really interesting, Neil, is the fact that we’ve known about the pathway. We’ve known about targeting the pathway for some time. It has taken a while for us to finally get to this point where we have a drug we can use in this setting. But, interestingly, the alpha-specific inhibitors, when they’ve been tested in the early-stage setting, meaning even in the neoadjuvant setting, really have not shown a lot of activity.

And of course we know that PIK3CA mutations, when present in early-stage breast cancer, are actually associated with a better prognosis. And so one of the things that’s fascinating to me, of course, is that at some point PIK3CA becomes a driver. And we know from data in the metastatic setting that when it’s present in a resistant setting this is associated with very poor outcomes.

For example, there was a secondary analysis of MONARCH 2 recently presented at ACR by Sara Tolaney where they looked at PIK3CA mutations in patients treated with fulvestrant monotherapy or the combination with abemaciclib. And those patients with the PIK3CA mutations really did quite poorly treated with endocrine monotherapy again.

Clearly in that setting, in an endocrine-refractory setting, having the mutation is associated with a poor prognosis. That does not necessarily mean that it’s the case in the early-stage setting. And so it will be interesting to see how this drug moves forward into the earlier-stage setting, and we’ll await data, obviously, from those trials.

Side-effect profile of alpelisib

DR LOVE: What about tolerability? Prior to alpelisib coming along, I was hearing a lot of negative things about PI3 kinase inhibitors, about toxicity. Then this came along and seemed like it was better. But it wasn’t really clear exactly what happens.

DR GOETZ: I think what we can say is that the attempts in the past to really focus on PI3 kinase have been fraught with the toxicity issues that you mention. And some of those toxicities actually included psychiatric issues. For example, there were some drugs that just couldn’t move forward because of some of those side effects, including depression and other psychiatric side effects that were substantial to patients.

What we can say is, as companies have done a better job of developing selective or alpha-selective inhibitors, that some of those issues have gone away. But we still certainly do see side effects that are unique to targeting this pathway, including — the big one is hyperglycemia. And with the alpelisib, the hyperglycemia is pretty substantial in a number of patients that require drugs like metformin.

And so I think as clinicians begin to use these medications, there’s going to be that familiarity issue. And that is, how do you deal with the hyperglycemia? When should you start metformin? Are those patients that should be started up front, or can we wait to see what type of response in terms of hyperglycemia that’s going to develop? And as I think with all these drugs — I can think of the mTOR inhibitors — there’s a certain comfort level that certainly comes just when you prescribe it enough and you know how to deal with these side effects. And I think clinicians will get to that point.

I think the other side effects we see with alpelisib are some of the other classic side effects associated with targeting the PI3 kinase mTOR pathway, including rash, including diarrhea. And certainly it’s unclear as of yet whether we’ll see any issues, for example, with pneumonitis, like we did with everolimus. These will be issues that — obviously once a drug gets on the market we know about the common ones, and we’ll have to work out even the uncommons as well.

I didn’t mention the issue of stomatitis. Obviously that was an issue with the mTOR inhibitors that we’ve begun to use the dexamethasone mouthwash, and I think certainly we will be thinking about using that as well, to help ameliorate or prevent some of those side effects as well.

DR LOVE: Do you have any general clinical experience? I don’t know how many patients you’ve actually treated alpelisib, but how do patients feel on the drug?

DR GOETZ: I have to say that our group at Mayo was not involved in the direct development of alpelisib. We certainly have treated patients already on this drug. And what we are seeing is the side effects that really have been reported, which is the diarrhea, the rash, certainly some fatigue as well. As all of us gain more confidence about using these drugs I think we’ll hopefully have a better sense, really, of who’s at risk for the side effects and how best to deal with them.

Case: A woman in her late 60s with high-grade, ER-positive, HER2-negative breast cancer and a 21-gene assay Recurrence Score (RS) of 52 receives adjuvant chemotherapy

DR LOVE: Let’s move down to the localized stage of ER-positive, HER2-negative disease. Particularly interested in your thoughts on the new data that’s come out now from the TAILORx study that you were so much involved with that was just presented at ASCO. And maybe to get into that we can hear about your 69-year-old lady, just to get started.

DR GOETZ: This was a patient that presented a couple years ago in my clinic. She presented with early-stage breast cancer. This was a T1c breast cancer, about 1.6 centimeters. It was high grade. Estrogen and progesterone receptor were strongly positive. HER2 was negative. And she had treatment with lumpectomy and radiation therapy. And, interestingly, postoperatively we obtained a Recurrence Score on her. And she had, actually, Neil, one of the highest recurrence scores that I’ve seen in my practice. She had a Recurrence Score of 52.

And so this was a setting, a situation where having this information about the Recurrence Score really obviously changed how we thought about this. And perhaps, for example, in a 69-year-old female in the past we might have tried to hem and haw and try to avoid chemotherapy. But this is one where we really felt like we were concerned that endocrine monotherapy alone is really not going to completely hold her disease back.

And so we did recommend a course of adjuvant docetaxel and cyclophosphamide. And this patient really had a lot of difficulties with the docetaxel. She actually had a hypersensitivity reaction. We’ve all seen these in our clinic, and despite rechallenging her — actually, we rechallenged her twice — we were unable to get the drug in her. And so we ended up moving on, in her situation, to CMF. The patient actually had a cardiac history, and she was quite adamant she did not want to be treated with an anthracycline. And so we were able to effectively deliver the oral CMF regimen, and actually she tolerated that quite well.

DR LOVE: That case really struck me, because whenever you talk about Recurrence Score, people always think about the women who are avoiding chemotherapy. And when I saw that 52 and age 69 that was exactly what I was thinking, probably you wouldn’t have thought about chemotherapy in a patient like that, and yet it seems like she would need it.

DR GOETZ: Exactly. And I think this patient was really struck as well. She was a very intelligent woman who really — I have to laugh. When I sometimes see my breast cancer patients and I walk in the room that they actually know more than the trainees know about their cancer. And she certainly was one of these, was very well read, and she was quite motivated to receive adjuvant chemotherapy.

DR LOVE: What happened next with her?

DR GOETZ: She actually finished the CMF chemotherapy, as I mentioned. She tolerated that quite well. And, interestingly, although we don’t necessarily do this in all our patients, this patient had requested a new baseline mammogram. And, in fact, it had been probably about 9 months from her prior mammogram, and she came in and requested that. She was traveling from quite a ways away, and so we felt like it was potentially reasonable.

And so the new mammogram, interestingly, demonstrated new calcifications. And so this prompted a biopsy, and the biopsy showed DCIS. And so she was taken to the operating room for a mastectomy, and, interestingly, at the time of mastectomy was found to have additional invasive disease that was present.

Obviously this is sobering because this is a, if you will, this is a scenario where we think about where we’re giving adjuvant chemotherapy to eradicate disease that we can’t see. In this situation, she had just finished 6 cycles of CMF chemotherapy and despite that had residual invasive disease in her breast.

DR LOVE: But she hadn’t started on adjuvant endocrine therapy at that point.

DR GOETZ: She had not started on adjuvant endocrine therapy. That’s correct.

DR LOVE: It is amazing, though, that it popped up so quickly.

DR GOETZ: Yes. She finished her treatment. She actually — at the time of surgery, she had a right skin-sparing mastectomy. Interestingly, she had some disease that also involved the muscle as well. And she had a single cluster of isolated tumor cells. And because of the nature of the disease, the involvement of the muscle, obviously we were very concerned.

We actually repeated the PET scan at that time. And, interestingly, the PET scan demonstrated an area of sclerosis in one of the ribs. Of course the question, of course, was, was this treated disease or not? We’ve gone on to treat her with an aromatase inhibitor. She is receiving zoledronic acid. And obviously we’re following her very closely, as she remains at high risk for developing a distant event.

Role of bone-modifying agents for patients with ER-positive breast cancer and bone metastases

DR LOVE: What was the rationale behind the zoledronic acid? Was she osteopenic? Or were you trying to prevent mets?

DR GOETZ: We really were doing it in this situation more in the adjuvant setting based on the data from the Oxford Overview that demonstrates that these patients have a benefit in terms of breast cancer survival, reduction in bone metastases. This is a patient, by the way, that previously, years ago, had a history of osteopenia and probably about 7 or 8 years ago had been treated with bisphosphonate. She had been off of that for many years. And so the intent here was to, if you will, to complete a course of treatment that would be curative.

DR LOVE: Is that something you routinely do? And what about denosumab? We saw some data on that at ASCO.

DR GOETZ: I think the best data here that we have about these bone-modifying agents is really with the bisphosphonates, and of course the ones used commonly are the aminobisphosphonates. The data with regard to denosumab, although certainly we know that that can prevent bone loss and it does it quite well, the data have yet to convincingly show that denosumab can effectively reduce breast cancer events, specifically in the bone, and also survival. I think from a standpoint of using these agents in the adjuvant setting with the intent to prolong survival, prevent breast cancer morbidity and mortality, the best data that we have are with the bisphosphonates.

Ongoing investigations of CDK4/6 inhibitors in the adjuvant setting

DR LOVE: I wonder, you said this patient was very proactive and involved, whether she brought up the idea of either getting a CDK inhibitor off trial or participating in a trial. Where are we right now with adjuvant CDK inhibitors, and when are we going to see some answers?

DR GOETZ: Some of the large adjuvant CDK4/6 inhibitor trials have already completed enrollment. For example, the PALLAS trial, which is a trial that examined the role of palbociclib in women with ER-positive breast cancer that had been treated with curative intent. And in that trial, patients are randomized to either palbociclib or just standard endocrine therapy. All women, of course, in this trial get standard endocrine approaches.

And in the monarchE study, this was a trial evaluating abemaciclib, a little higher-risk population, patients with locally advanced breast cancer, as well as a subset in the monarchE that have high levels if Ki-67. Again, looking to see whether abemaciclib when added to standard endocrine therapy can improve invasive disease-free survival. And of course there’s data in studies ongoing in the adjuvant space with ribociclib, and those trials have either completed accrual or will be close to completing accrual soon. To answer your question when are we going to see the data, I expect probably in that range of perhaps 2021 when we would begin to potentially begin to see the data mature.

Effects of CDK4/6 inhibitors on immunity; potential for combination with immune checkpoint inhibitors

DR LOVE: What do we know about the issue of immunity in breast cancer, but specifically with CDK4/6 inhibitors? And what about the strategy of combining them with checkpoint inhibitors?

DR GOETZ: This is one of the interesting things about CDK4/6 inhibitors that has really just emerged just recently. And, in fact, sometimes I find that when I look at a drug that works extremely well, and I’m thinking about the drug trastuzumab, when that drug was first developed, we all thought, this drug is simply inhibiting HER2 in the downstream pathways. But as it turns out, the drug has profound effects on the immune system, and specifically on ADCC.

It’s not unsurprising that when we look at CDK4/6 inhibitors and we see them working so well that we might find additional mechanisms by which these drugs work. And, in fact, in some ways it makes sense. I think the issue of the immune effects, what’s been observed, Neil, is that in preclinical models these drugs have effect on things such as Tregs. And Tregs we know are important as it relates to developing resistance to standard chemotherapy, associating with early recurrences, and there have been really not a lot of, if you will, dedicated strategies that we have at targeting Tregs. And the observations were that simply by inhibiting CDK4/6 that you could have some effects preclinically in a number of these different models, both in mouse models as well as seen in patients potentially on these immune subsets or these immune cells, the Tregs.

I think that we don’t completely know or understand exactly how these drugs effect the immune system to some extent, other than the observations that there do appear some effects on the immune system. There was a paper that was published by Shom Goel in Nature about 2 years ago that really outlined for the first time the potential immune effects of CDK4/6 inhibitors. And what this has prompted people to think about is to say, if that’s the case, would there be value of combining, in this case, a CDK4/6 inhibitor along with checkpoint therapy? And, in fact, in some of the preclinical models there was a suggestion indeed that there was synergy in an ER-positive breast cancer model combining a CDK4/6 inhibitor along with checkpoint therapy.

What we can say is that those trials are ongoing. We don’t have necessarily randomized data as of yet. What we can say is that we are obviously concerned about the possibility for the side effects which we know can occur when you use a checkpoint inhibitor and along with the side effects that occur with CDK4/6, and is this going to be a tolerable approach?

The data combining abemaciclib along with a checkpoint inhibitor, pembrolizumab, those are really single-arm data. But I would say they certainly do look interesting. But the real question is, will this be a tolerable approach?

The other thing I think is really important, and that is, as breast cancer oncologists, we certainly have not thought about estrogen receptor-positive breast cancers being immunogenic. We think about triple-negative breast cancer. We think about HER2-positive breast cancer. And in general, when we think about ER-positive breast cancer, I think of these cancers as flying under the radar of the immune system. Think about a lobular breast cancer. These are genomically quiet. They don’t tend to have a lot of mutations.

And so I do think that there’s going to be emerging work and attempts to really define a subset of ER-positive breast cancers that may be immunogenic. And I think we can only look at — I’m thinking out loud here, the data from I-SPY 2 that was recently presented within the last almost 2 years by Rita Nanda, where they looked at the role of pembrolizumab in the I-SPY 2 trial, where you saw that pembrolizumab improved path CR rates. But it was not just in triple-negative, but it was also in the ER-positive. And just to be clear, patients to get onto I-SPY 2 had to be high by MammaPrint^®. They had to have this high-risk assay suggesting a higher risk for recurrence.

And so there probably is a subset of ER-positive breast cancers that we might think about that not only may be immunogenic but may be responsive to estrogen-targeted therapies along with immune strategies. And certainly this is going to be an area of great research, and I look forward to seeing these data moving forward.

TAILORx trial: Adjuvant chemotherapy guided by a 21-gene expression assay for women with ER-positive, HER2-negative, node-negative breast cancer

DR LOVE: Speaking of waiting for big trials to accrue, actually, George Sledge was showing me a picture of a water bottle. We did this display at San Antonio I think in 2008 of ongoing trials. And we actually had a water bottle when the TAILORx study was just getting started. Can you talk about that study that you’ve been so involved with? And particularly the data that was just presented at ASCO.

DR GOETZ: The TAILORx data that we’re familiar with and that has been so widely disseminated and talked about really, as we know, took any woman who had ER-positive breast cancer. Just to clarify, there were both pre- and postmenopausal women in that study. And the study really examined this group of patients with a so-called intermediate Recurrence Score, a score between 11 to 25. And these are patients that we really didn’t completely know or understand, did chemotherapy add to the outcome or improve the outcomes for these patients?

And in the TAILORx study, patients with low Recurrence Scores were treated with endocrine monotherapy. Patients with a high Recurrence Score were treated with standard chemotherapy followed by endocrine therapy. And then this intermediate group was, of course, the subject of randomization. And as was presented originally over a year ago, we know that overall there was no benefit of chemotherapy in this intermediate group.

Impact of clinical risk on prognosis and prediction of chemotherapy benefit by age and RS in the TAILORx study for patients with early breast cancer

DR GOETZ: I think that the more recent data, which has been quite interesting, that was published by Joe Sparano in The New England Journal of Medicine and recently presented at ASCO, looked at this issue of clinical risk. And I think this is, of course, something that we’ve been waiting for for some time, because when we see a Recurrence Score on a given patient and we think about the benefit or lack of benefit, we see data without information about the clinical characteristics. And as we think about these genomic assays, it’s been pretty clear that the clinical, if you will, characteristics still matter.

As we look at something, for example, like the EP assay, or EndoPredict assay, one of the things that has been shown is that you get a much more robust sense of the value of that assay when you bring in the clinical factors. And indeed, the analysis from the TAILORx study is that first of all, the clinical risk does matter. And so, for example, if you have a small tumor, let’s say less than 1 centimeter, low grade, that the implications in terms of the risk for a late disease event, that clinical risk still matters despite your genomic data.

What was important to note, however, was, despite looking at these varying different clinical risks, that in postmenopausal women that there was still no value of giving chemotherapy, whether you had high clinical risk or low clinical risk. And that is so if you had a 65-year-old woman that had more, if you will, high clinical risk features, larger tumor, we could still say with confidence that chemotherapy was not having value.

What was interesting, however, was looking at the premenopausal women, Neil. And that is that there appears to be something different about premenopausal women. And I think one of the caveats about interpreting the TAILORx data in the premenopausal setting is that these patients were enrolling at a time when most of them were getting tamoxifen monotherapy. In fact, about 85% or so received tamoxifen monotherapy. We know that today that just giving proper endocrine therapy, that would be adding ovarian function suppression, and in some patients even adding in an AI, that actually gives you a large delta, a large benefit.

And so knowing the history of the fact that chemotherapy-induced amenorrhea is an important prognostic factor, what we don’t know in this situation is, how much did chemotherapy-associated amenorrhea impact the outcomes of these patients? That being said, it’s pretty clear that there is a subgroup of patients within the intermediate group, perhaps in that 16 to 20 and 20 to 25 range, that appear with high clinical risk to be benefiting from chemotherapy.

We can only surmise at this point, is it because of the chemotherapy effects on the ovaries or is it something unrelated? And one of the things that I’m kind of interested about, Neil, is the fact that we have patients who are premenopausal who are more likely to have pathogenic mutations in something like a BRCA1 or a BRCA2 gene. And we know already that deficiencies in DNA repair confer chemotherapy benefit in some of these patients.

I think a question is, are they benefiting because we’re ablating their ovaries? Or is it just the chemotherapy works better in these younger patients? And I think that’s an unanswered question, but certainly these are provocative data.

DR LOVE: That’s interesting. I haven’t really thought — when I think about DNA repair, I think about platinums. But you think that applies to chemotherapy in general?

DR GOETZ: I think it does, and I think the idea here is to say that, for example, when one utilizes a regimen such as docetaxel and cyclophosphamide, when you’re using a drug that targets DNA like cyclophosphamide, or obviously the classic one are the anthracyclines. We know that these drugs do benefit patients with DNA repair deficiencies. We think, obviously, that platinums might be the optimal way to go. But we actually don’t have, if you will, Level 1 data that we should choose a platinum over our standard therapies.

And so I think a question moving forward in the TAILORx study, especially in these young premenopausal women, is, is the benefit that we’re seeing perhaps restricted to certain types of patients? I don’t think we know the answer to that, but I think there’s an interest in looking at that moving forward.

DR LOVE: Getting back to these recent data looking at clinical factors. The one that seemed to strike me as being very common, as you mentioned, was size. And I think it was 3 centimeters that was the breaking point. But does that mean for you it’s biology not anatomy? I mean, 4 and 5 centimeters, are you still, in a postmenopausal woman, comfortable not using chemotherapy in an intermediate score?

DR GOETZ: I think I’m quite confident about not using chemotherapy in that postmenopausal woman. I think what we can say is that risk does not confer benefit from our therapies. What’s pretty clear is that as we look at TAILORx, but not only that, but the recent Oxford Overview with regard to the late recurrence outcomes that were published a few years ago, that these clinical characteristics have implications with regard to risk for recurrence.

I like to think of it this way, is that higher disease burden, larger tumor, nodal status certainly imparts a higher risk for distant metastases but does not necessarily give me input, necessarily, that those patients, at least based on TAILORx, I can confidently say that our best approach right now needs to be to deliver better targeted therapies focused on the estrogen receptor for those patients as opposed to chemotherapy. And the nice thing about this, and when I speak to patients in the clinic, I like to think of it this way, very simplistically. We have types of breast cancers where targeting the estrogen receptor leads to excellent long-term outcomes, and we’re very confident, for example, based on TAILORx, that we can focus on estrogen-targeting therapies. And then we have these patients with a higher Recurrence Score, the patient I just mentioned, where we’re actually quite concerned that endocrine therapy alone is really not going to have substantial effects on inhibiting the growth of presumed micrometastatic disease.

Perspective on the use of gene expression assays to determine risk of recurrence and guide adjuvant decision-making

DR LOVE: You were mentioning other genomic assays. You were talking about EndoPredict. Of course the 70-gene assay’s out there. It seemed like we were talking more about those a few years ago. Now maybe it’s because the TAILORx study came out. It seems like most of the discussion is more centered around the Recurrence Score.

But what do we know? We talked earlier about sometimes you have a bunch of alternatives but yet there’s more data with one. Do you think there’s a fundamental difference in what they’re looking at, what we know about these studies that have looked at the same assay or different assays in the same patient? And in terms of clinical utilization, is there a role for any of the others?

DR GOETZ: I think that what’s pretty clear is that all of these assays provide robust prognostic information. If you look at one of the common denominators across these assays, it’s that they measure proliferation. Without a doubt, if I wanted simply an assay that would give me prognostic information, I really could choose a number of these assays. But really, the question that a lot of us face is this issue of clinical utility, and that is, what should I do with the information? And I think that’s where data with regard to the Recurrence Score has been most helpful in terms of providing help with decision-making for clinicians.

And that is, I can really confidently say that, okay, based on this result and based on the prospective data from TAILORx, I don’t need to give chemotherapy in this situation. I think one of the other issues that’s really untapped, and untapped meaning we need more information, and that is the role of these assays in looking at the risk for long-term recurrence as well as the benefit of long-term endocrine strategies — that is, utilizing endocrine therapy in the years 5 through 10.

And again, we have some data with some suggesting that they may be prognostic for late recurrence. We also have some data from the so-called Breast Cancer Index^® that combines a 5-gene assay, which is more of the proliferation part, with the so-called HOXB13/IL17. And that assay really suggests that there may be a group of patients not only at higher risk for a late recurrence but perhaps who may benefit from extended therapy.

Cancer cell dormancy and risk of late recurrence

DR LOVE: One of the other talks that I saw that you gave that I was interested in actually focused on that and the issue of dormancy. I’m just kind of curious, what’s your hypothesis about the biology of late recurrence?

DR GOETZ: This is really a hot topic. And I think it’s a hot topic, Neil, because we’ve been successful with our early strategies, or we’ve been successful at reducing early recurrences. If we weren’t successful in reducing early recurrences, we wouldn’t be worried about late recurrences. But the reality is, we’ve gotten much better and we’ve started discharging these patients from our clinic low and behold, to see them come back in 10, 15, 20 years. And I always tell audiences that the latest recurrence I’ve ever seen was in a Mayo Clinic breast cancer patient who had a 40-year relapse —

DR LOVE: Wow.

DR GOETZ: — and it was an interesting scenario where she had had a bilateral mastectomy in 1970 and was being treated by her hematologist for her CLL when all of a sudden, her anemia really, if you will, her cytopenias really increased dramatically. Much more problems with anemia, requiring blood transfusions, prompting a bone marrow, and low and behold finding not necessarily the CLL had progressed but finding breast cancer cells that had taken over the bone marrow. We do see these late recurrences, and as I mentioned, it’s because we’ve been successful, I think, in part, by treating some of the early recurrences.

This idea of tumor dormancy is an interesting one, because the suggestion is that we know that there are breast cancer cells. These cells that are going out, sometimes — and I think about them as spreading long before they ever come to my office. And we know that these cells can be in an equilibrium, so-called, with their microenvironments, such that they’re not causing problems. We know that we can inhibit the growth of them with drugs like tamoxifen, and we can certainly eradicate some of these micrometastases with our chemotherapy agents.

But we also know that these micrometastases can be present for many, many years. And, in fact, there’s probably many patients that are dying of things other than breast cancer with micrometastases present in a dormant state, where they’re just not causing problems.

The interesting approach — or the interesting issue, of course — and that is, who is at risk for developing late recurrence, and which of our patients have these dormant cells (A) that would never cause problems and we should actually leave them alone, and (B) those that are actually going to cause problems down the road. What are the host factors, the genetic factors, the microenvironments, the immune factors that would all potentially be driving the reemergence and causing these dormant cells to no longer be dormant but to be actively growing?

This is certainly an area that is of great interest. One of the things that people have started thinking about, of course, is, can we utilize these very sensitive assays such as circulating tumor cells or circulating tumor DNA to pick up and identify patients early? And perhaps by doing that prevent some of the morbidity that occurs with breast cancer. And, of course, these assays are out there. We know that they’re reproducible, but we’re trying to find utility for them.

DR LOVE: So interesting. I was just flashing on your patient with a 40-year-later recurrence who had CLL. And I’m curious, did she get a BTK inhibitor, venetoclax, and how that would have affected the breast cancer? So fascinating.

DR GOETZ: Yes. Yes, this particular patient really had a quite virulent course after the diagnosis of breast cancer. She had really been observed with regard to her CLL. And it was the observations: her CLL was quite indolent, and she hadn’t reached the point where her hematologists were even treating her. But it was the change in her CBC that prompted the workup. And she was treated at that time, initially, with endocrine therapy. This was prior to CDK4/6 inhibitors, moving through to chemotherapy and then unfortunately had a fairly rapid course.

DR LOVE: Yes, there was a case sent into me from a doc in practice of a patient with marrow disease who was also ER-positive, HER2-negative. I was kind of curious which CDK inhibitor you would give. This patient started out on palbo. Neutropenia got even worse. Would you, in that situation, use abema?

DR GOETZ: I think I would, because in that situation, knowing that the patient, let’s say, already has problems with cytopenias, and knowing that we’re going to probably run into some problems, I think that would be a situation where I would think of abemaciclib definitely first.

Efficacy and safety of the novel SERD elacestrant for ER-positive breast cancer; Phase III EMERALD trial evaluating elacestrant versus endocrine therapy

DR LOVE: I met with Dr Virginia Kaklamani, and to begin, I asked her about emerging data on the use of novel selective estrogen receptor degraders, SERDs, in the management of ER-positive breast cancer.

DR KAKLAMANI: If you look at the trials in the metastatic setting, fulvestrant really seems to be the superior endocrine therapy. It’s beaten anastrozole in a head-to-head trial. The only limitation is that it’s an IM injection, so when you look at the adjuvant setting it’s kind of hard to be thinking of it as moving into the adjuvant setting. Also, the other limitation is the dose. We can’t give as high a dose as we want to because of the IM injection.

That’s why companies started developing oral SERDs: so that they can bypass that limitation. Several companies are out there. They’ve developed SERDS. They are in Phase I/maybe II trials. I think the furthest along is elacestrant.

DR KAKLAMANI: And we did the Phase I trial, and I can tell you, my 1 patient who is still on therapy — I put several patients on the trial, but one of them, I just saw her and she’s on her 34^th cycle of elacestrant. And so this is a drug that works well. It’s pretty well tolerated. A little bit of GI toxicity, especially with an older preparation, but now with the newer preparations, pretty easy to tolerate.

Some of the other ones have had some GI side effects as well. I think where the benefit is going to be is probably in the ESR1-mutated patients. And so the trials are now focusing on looking at CDK4/6 inhibitor post-therapy with or without ESR1 mutations. And in patients that have been on fulvestrant, it still seems to be an active drug. Several of the patients on the Phase I trial had been on previous fulvestrant and enjoyed a pretty good response to RAD1901.

Now, the Phase III trial has already started. There have been some patients that have been enrolled. We’ll see how it goes. But it’s a registrational trial, so if it’s positive, the drug will be approved.

Incidence of ESR1 mutations; activity of elacestrant after disease progression on fulvestrant and a CDK4/6 inhibitor

DR LOVE: How often do you see ESR1 mutations at initial diagnosis, localized disease, first diagnosis of mets and later on? And how much of a correlation is there with benefit? Do you see responses with AIs or tamoxifen?

DR KAKLAMANI: In the primary setting they’re very rare. Usually they develop over time, mostly with AI use. And so when you look at the metastatic setting, they’re a little bit more common, and that’s where the studies have shown that fulvestrant works better than anastrozole, which kind of makes sense, since a lot of these patients have already been exposed to anastrozole and have developed those mutations on anastrozole.

Now when you look at CDK4/6 inhibitors, the studies have shown that it’s pretty equivalent, the benefit, whether the patient has or doesn’t have an ESR1 mutation, but really the issue is going to be on what endocrine backbone to give to patients if they have an ESR1 mutation.

DR LOVE: And in terms of this particular agent, I know it’s actually a trial in progress, actually, but I guess — of course, the obvious question is, have you seen responses in people who progressed on fulvestrant?

DR KAKLAMANI: We have. There were some data that was presented at San Antonio and ASCO, and the full data set will be presented at San Antonio this year, and as you’ll see in that data set as well, there are several patients that have been on fulvestrant. Several patients that have been on CDK4/6 inhibitors progressed on both agents and then went on to RAD1901 and had a good response.

Results from the Phase II ANETT trial of the mTORC1/2 inhibitor TAK-228 with tamoxifen as neoadjuvant therapy for ER-positive breast cancer

DR LOVE: Another ASCO paper, a poster you had, and I see you’re working with Jenny Chang. She really has done some amazing work, and this seems like it’s kind of lined up with that — “Open label, phase II trial of neoadjuvant TAK-228 plus tamoxifen.”

First, neoadjuvant always gets my attention. And Jenny has done a lot of stuff with HER2-positive. But anyhow, what’s TAK-228, and what’s up with this trial?

DR KAKLAMANI: That’s a dual mTOR inhibitor, and we participated in the trial. There was some concern previously — there was an oral presentation at SABCS I want to say 2 years ago, looking at a dual mTOR inhibitor and, actually, the drug was not effective at all. And that was our concern initially with this trial. But this is a little bit of a different mTOR inhibitor.

The results seemed pretty promising. It seems like this might be an agent that we may be using. The benefit, as you know, in the neoadjuvant setting is that you can get results quicker. As long as you have a drug that is relatively well tolerated and we know the toxicity profile, it’s easier to try to do a neoadjuvant trial with it than to try to do a metastatic trial, especially in these 1- or 2- or 3-institution trials.

This might be an agent that we’ll hear a little bit more about, probably in the metastatic setting. The good news is that now we have, what, 3 different targeted agents, taking away the endocrine therapy. And so with this we might have 4 or 5, especially when you add AKT inhibitors to the mix.

DR LOVE: Actually, what really caught my attention was the neoadjuvant aspect of it. Because, I mean, I don’t know, kind of looking back, it always seemed like such a great idea. In endocrine therapy I guess you can look at the Ki-67, or I guess that’s been done. But I haven’t heard too much about using it to — you have the I-SPY out there. I mean, they’ve done some things. But do you see a resurgence in neoadjuvant therapy in the ER-positive/HER2-negative situation, both research as well as practice?

DR KAKLAMANI: I think in research more than practice. Now, maybe with CDK4/6 inhibitors, since they’re such active agents, we may see a little bit of a change. My issue with giving endocrine therapy in the neoadjuvant setting is that then you may not know whether you should be giving chemotherapy in the adjuvant setting or not. These tumors are slow growing, and they’re slowly responding to treatment as well, so you have to treat them for at least 6 months of endocrine therapy, which is a little hard to take. Some of the patients we’ve had on endocrine therapy for 9 months, and they continue to have slow responses.

It’s hard to think in practice of giving these patients neoadjuvant endocrine therapy, unless there’s limitations — for example, they’re older patients, you don’t want to take them to surgery or they’re patients that have other comorbidities. We have several patients that are on the list for liver transplant and all of a sudden they get diagnosed with breast cancer, but liver is not good shape.

And so in these patients it may make sense to give them the adjuvant endocrine therapy. But in most cases, I think, standard of care, we’re still going to be looking at adjuvant therapy. But for clinical trials, this is how we learn. This is how we look at these agents and see whether the Ki-67 responds, whether there’s other biomarkers of response that can help guide us in the adjuvant and metastatic setting.

Novel approaches under investigation in the neoadjuvant setting

DR LOVE: Any new strategies in terms of endocrine therapy out there, as long as we’re talking about new agents and strategies, maybe looking at the neoadjuvant setting that you’re excited about?

DR KAKLAMANI: The window trials are also pretty interesting, where you’ve giving a very short course, maybe 3 to 4 weeks of treatment and then you’re really assessing whatever your biomarker is, whether it’s Ki-67 — I think that, again, is going to help us design trials. It’s definitely not going to help us find a new standard of care, but it’ll help us design trials in the metastatic setting or maybe in the adjuvant setting. I think when you look at agents, the one class of agents that seems to be the furthest along are the AKT inhibitors, together with the oral SERDs, which again is another endocrine approach.

And then, obviously, how you combine these agents — does it makes sense to combine a CDK4/6 inhibitor with an mTOR inhibitor? Are you really going to achieve something by that or just sequencing those agents?

DR LOVE: What happens when you give endocrine therapy alone as opposed to endocrine therapy plus CDK neoadjuvantly to Ki-67?

DR KAKLAMANI: You’re going to have a little bit more drop by giving the CDK4/6 inhibitor with endocrine therapy instead of endocrine therapy by itself. The response is going to be faster as well. But whether you’re going to have any long-term benefits, that remains to be seen.

We have 2 large adjuvant trials with CDK4/6 inhibitors that are going to answer this question in our higher-risk patients. And then we’re going to have a third one, I think a smaller trial that’s going to also be looking at a third CDK4/6 inhibitor. But, obviously, these trials are going to report in several years from now, given how slowly this cancer progresses.

DR LOVE: Sometimes we get surprises. But whenever it does happen, it certainly wouldn’t be a shock to see them be positive. But then when you think about it, it’s going to change the whole platform of adjuvant therapy. Things that weren’t that big a deal now maybe are going to look bigger if you’re going to have to do it for a long time.

DR KAKLAMANI: And that’s the thing: You’re adding more agents. You’re adding more toxicity, but then you’re also trying to figure out, what are you going to do in the metastatic setting? If you have somebody that is on a CDK4/6 inhibitor for a year in the adjuvant setting and then recurs 3, 4 years later, do you rechallenge them with these agents or not? Those are going to be unanswered questions.

Perspective on the use of neoadjuvant therapy to optimize surgical outcomes for patients with ER-positive, HER2-negative breast cancer

DR LOVE: When I brought up the issue of neoadjuvant endocrine therapy clinically, you gave the answer — kind of an approach that characterizes I’d say about two thirds of breast cancer investigators, because there’s another third that use genomic assays and will use endocrine therapy. And one of the situations that I hear about — we actually had a case that a general medical oncologist presented at one of our meetings, and I don’t know how often it happens, is the patient with ER-positive/HER2-negative disease who really needs tumor shrinkage to have breast conservation.

And in that situation, I don’t know, what do you do? I mean, would you give chemotherapy? Would you look at Recurrence Score and then decide? What would you do?

DR KAKLAMANI: I think in that setting, looking at the Recurrence Score makes sense. Again, there’s not large trials, but it’s pretty obvious when you look at several neoadjuvant trials with the Recurrence Score that if you have a low Recurrence Score, you’re not going to get at least PCR with chemotherapy, because that’s how we’ve measured chemotherapy benefit. You may get some benefit.

But again, if I have a patient, as long as she’s not too young — and she has a Recurrence Score of 10, me trying to give her chemotherapy in the neoadjuvant setting is probably not going to achieve my goal. But if I feel I have enough time — and giving her endocrine therapy in that setting I think makes sense. Again, I’m going to have to treat her for at least 6 months, but that tumor is slowly going to respond, hopefully making her into a surgical candidate.

The other approach I’ve used recently, especially if you kind of think that this patient is locally advanced and maybe not metastatic but definitely micrometastatic, is whether you can add a CDK4/6 inhibitor to that mix, and then you’re going to get a better response based on our metastatic data. That might help, again, making her into a surgical candidate.

Activity of CDK4/6 inhibitors for locally advanced ER-positive breast cancer

DR LOVE: That’s interesting.

How do you approach the use of, outside a trial setting, CDK inhibitors in locally advanced breast cancer?

DR KAKLAMANI: Yes. The one patient that I’m doing this on is a locally recurrent breast cancer. I guess I’m kind of getting away with the nonstandard of care approach, which I just mentioned. That patient, she had her initial cancer 20 years ago. Now she has a local recurrence, but it’s inoperable and it’s strongly ER/PR-positive. She had been on tamoxifen previously.

And so the question for us was, she can’t have surgery. Do I really want to give her chemotherapy? Should she have radiation? Initially we talked about radiating her, and then I presented her at tumor board, and when we all kind of started talking to each other, we realized that radiating her and then subjecting her to surgery might be not the best approach. And since she was strongly ER/PR-positive, we decided to give her endocrine therapy with a CDK4/6 inhibitor. We’re going to watch her every month. Every 3 months we’re going to repeat the imaging. And then hopefully in 6 months or so she’ll be a surgical candidate and we can take her to surgery.

DR LOVE: Now you’ve got me interested in this case. Did she delay coming in, is that why it was locally advanced? Or was it aggressive? Or what did it look like?

DR KAKLAMANI: It actually did not look as bad. It was a chest wall recurrence that’s encasing some of the vessels. That’s really more of why it was inoperable. It wasn’t some exophytic mass that you could see. I think that’s what delayed her coming in.

DR LOVE: She had a recurrence 20 years later and the recurrence was inoperable? And it went into what vessels?

DR KAKLAMANI: It was on the pec. And so I’m told that it was the subclavian vessels that were kind of deeming her inoperable.

DR LOVE: Wow! I have to get the follow-up on what happens to her with treatment.

DR KAKLAMANI: I’m curious as well. This is really my first case of treating a local recurrent disease with CDK4/6 inhibitors. I’m interested as well.

DR LOVE: Wow, yes. And also, I mean, this is critical: this woman’s quality of life. I mean, it sounds like not a good situation.

Case: A woman in her late 60s with ER-positive, HER2-negative breast cancer and 1 of 3 positive axillary lymph nodes receives a 21-gene assay RS of 18

DR LOVE: Let’s get into the more common scenarios. I asked you to pull a few people from your practice. And, of course, I want to try to get into the clinical relevance of the big TAILORx presentation that just happened at ASCO, really an important presentation. And to get to that, I asked you to pull a couple of cases where genomic assays were used in your patients. And I thought we could start out with maybe the big issue, let’s say, prior to ASCO or prior to some of the recent data coming out, which is patients with node-positive disease.

You wrote up this case here, 68-year-old woman, 2.1-cm tumor, 1 of 3 positive axillary nodes after lumpectomy for ER-positive/HER2-negative disease. And I guess the issue there is, with 1 positive — these are sentinel nodes?

DR KAKLAMANI: Sentinel nodes.

DR LOVE: Then, the issue is, what about genomic assays in a patient like that? Can you talk about your view of genomic assays in general in the node-positive situation and what happened this lady, too?

DR KAKLAMANI: When you look at the data with genomic assays, again, we have MammaPrint. We have Oncotype. We have several other assays. The reason I look at Oncotype only is because that’s where we have the good predictive data. The other assays have exceptional prognostic data and even more solid than Oncotype does, but when you look at predictive data, that’s where I think Oncotype, it tends to kind of have more solid data.

You look at the node-positive patient population and kind of the more established trial is the 8814, Kathy Albain’s trial, where you’re looking at patients that were node-positive. Oncotype was done retrospectively. Patients were given either CAF, CAF and tamoxifen or CAF with tamoxifen combined. And the initial trial showed that giving tamoxifen concurrently with chemotherapy is probably not a good thing. But CAF followed by tamoxifen is a good option.

Role of the 21-gene expression assay in predicting chemotherapy benefit for patients with ER-positive breast cancer

DR KAKLAMANI: Now, on the trial when they looked at Oncotype, what they found was very similar findings to the adjuvant lymph node-negative trials where patients had a benefit from chemotherapy if they had a high Recurrence Score and did not have a benefit if they had a low Recurrence Score. And they didn’t seem to get a benefit from the intermediate Recurrence Score, even though numerically I’m always concerned because the numerical difference between the chemo versus no chemo was pretty striking, in my mind. And so that Recurrence Score of 18 to 31, I’d be very careful in the node-positive patient population as to how to treat them. But if you have a low Recurrence Score, based on that data set it doesn’t seem that chemotherapy is going to benefit these patients.

Having said that, I think it’s important to recognize that these patients are going to have a higher recurrence risk than if they were node-negative. Prognosis is worse, but chemotherapy prediction ends up being the same. And that’s where I kind of differentiate. I’m not a proponent of doing Oncotype on every single node-positive patient, but I think in postmenopausal women where the trial was based on, I think there we can consider doing it, especially in women that are willing to not get chemotherapy, given the limitation of data, because this is not a prospective trial.

Now we have the prospective trial, RxPONDER. It’s completed accrual years ago. I’m hoping that soon we will have the results from that trial.

DR LOVE: And of course the big issue here, as you just mentioned, is really the patient and what the patient’s view is. This lady, 68 years old, so she’s certainly not in the very elderly population. What were her thoughts about chemotherapy? And did you talk to her about even before you sent it, or did you send it and then talk to her?

DR KAKLAMANI: I always talk to them. I don’t want to order Oncotype if I don’t discuss it with them. Eric Winer years ago, I want to say back in 2005 when all these assays were new, came up with these very nice clear guidelines as to who we should be doing the assay on. And one of them was if the patient is biased or if we are biased, we shouldn’t be doing the assay.

If I already know what I want to do, then the assay is not going to help me. And if the patient already knows what they want to do, the assay is not going to help them, either.

In this patient, and what I usually also do is, I discuss the fact that the standard of care for her would be to receive chemotherapy. I think that the chemotherapy benefit would be minimal, maybe moderate at best. And I think that, based on some data, which have limitations because it’s not a clear prospective trial, that we can order Oncotype as long as we’re willing to look at the limitations of the data that we have and as long as we think that that will help us make decisions. And in this case, like most 68-year-olds, you kind of usually have to convince them to get chemotherapy.

And so she was very happy to know that we were going to look at Oncotype and decide in that sense whether to give her chemotherapy or not. And with the results, that definitely helped make that decision.

Clinical and genomic risk to guide the use of adjuvant therapy for breast cancer

DR LOVE: The result of 18, though, is kind of interesting to discuss. How did you discuss it with her?

DR KAKLAMANI: What I told her was, based on the TAILORx data on the node-negative patients that were postmenopausal, the result of 18 did not seem to give us a chemotherapy benefit. I discussed some of the node-positive data such as the SWOG trial, such as the data sets from SEERS, which is a large data set, that again pointed toward the direction that she would not get a benefit from chemotherapy. And so I think, based on all of that data, she was comfortable not receiving chemotherapy.

DR LOVE: And, of course, it’s, I guess, biology versus maybe anatomy kind of, though. And it’s interesting that you drew the analogy from the node-negative to apply to her. And also, if you think about it, these data that we’re going to talk about right now looking at clinical higher risk in node-negative, it kind of falls into that same biologic concept.

DR KAKLAMANI: It really does. And we’ve had data with genomic assays, such as Oncotype previously, on the clinical risk versus the genomic risk. And what that data shows is, again, the clinical risk gives us prognostic data, but it doesn’t give us predictive data. And with TAILORx the same thing was shown, that the prognosis can be tweaked by using the clinical risk data, but prediction ends up being the same.

And so when I look at all that data, I try to shield myself from that clinical risk and focus on the genomic risk and then focus on the chemotherapy toxicities. Because again, if you have a patient who you’re giving a 2% to 3% benefit from chemotherapy but based on their comorbidities you think that the risks are higher, it’s really not worth giving chemotherapy, regardless of the benefit in the breast cancer recurrence. Those are the things that I kind of think about when looking at clinical risk versus genomic risk.

DR LOVE: It’s interesting. When you began, you were talking about the fact that you used the Recurrence Score more than the others because of the predictive data. And I was mentioning that our primary target here is the general medical oncologist, and I think you’re seeing that throughout oncology. I don’t know, I was flashing on CLL for some reason, where there was, like, a million of these prognostic factors but yet really there’s only 3 that predict treatment — del17p, p53 and IGVH — and those are the only 3. And those are the ones right now that are being used in making clinical decisions.

It’s interesting. I think really, I guess breast cancer really, I think, started this whole targeted therapy thing, if you really think about it.

DR KAKLAMANI: I think so too. Now, there’s another school of thought that things like prognosis and prediction is similar. I don’t agree with that. I think prognosis and prediction is different. And you can again look at Oncotype in the colorectal cancer arena, where they have good prognostic data, but prediction did not pan out. Again showing that in my mind, at least, that I want to differentiate on what I want to achieve here. Do I want to achieve prognosis, or do I want to achieve prediction? And based on that, I decide on what assay to use.

DR LOVE: I guess if you think about it in terms of prediction, I guess the first predictor we ever had in oncology was ER. I mean, it is a predictor. The only time I’ve heard an argument about prognosis that kind of was a little persuading was if through some kind of genomic assay you can show that the risk for recurrence is, like, less than 5% or less than 3%. It doesn’t really show a predictive thing but, like, you can’t make it any better or risk of treatment is the same.

What do you think about that? I’ve heard that applied sometimes with genomic assays.

DR KAKLAMANI: And I think you’re right. And when you look at the MINDACT data set, that’s exactly the point, right, that the prognosis is so good that how much better can we get?

DR LOVE: That’s what I was thinking.

Case: A premenopausal woman in her late 40s with ER-positive, HER2-negative, node-negative breast cancer receives a RS of 19

DR LOVE: You’ve got another case here that looks like it’d be very interesting, 48-year-old woman with a 3.1-cm tumor, node-negative, ER-positive/HER2-negative. Gets a lumpectomy. You didn’t say whether she’s pre- or postmenopausal.

DR KAKLAMANI: Premenopausal.

DR LOVE: Premenopausal. Okay. So that’s even more interesting. And so she gets a Recurrence Score, and it’s 19. First of all, let’s talk about what was presented at ASCO and then track it back into this case.

DR KAKLAMANI: Sure. Again, at ASCO we looked at the prognosis versus prediction. We looked at the clinical risk score versus the genomic risk score and how can the clinical risk be incorporated into the genomic risk? And the clinical risk improves prognosis, but it doesn’t really change the prediction, even in the younger patient population.

The case here, in my mind what makes it interesting is that you have a 48-year-old with a 3.1-cm tumor, and a lot of people will say, “Oh, geez, I need to give her chemotherapy.” And I think that’s what helps with us with the ASCO presentation because you kind of have to take away that 3.1-cm tumor, because if she had had a mammogram, you don’t know when she had a mammogram. Maybe she hadn’t had a mammogram in 5 years. She had a mammogram 2 years before, that tumor would not have been 3.1 centimeters, but the biology of the tumor would have been the same.

That’s why to me, the size of the tumor doesn’t make that big of a difference. And again, we’ve seen that with all of the data.

Clinical implications of the TAILORx trial comparing chemoendocrine therapy to endocrine therapy for patients with ER-positive, node-negative breast cancer and an intermediate RS

DR KAKLAMANI: Then you look at the fact that she’s 48 and she’s premenopausal, and you look at the presentation from last year, where in this patient population the group that had a Recurrence Score of 16 to 20 had marginal benefit from chemotherapy, mostly in the local recurrence, not as much in the distant recurrence.

And that’s where I think I struggle, because the discussion ends up being very complicated. And if I showed the patient the curves are really very, very close together and you need a magnifying glass to see the difference — but you’ll see a difference of about 1, maybe 2 percentage points. And then the patient gets to decide if that’s good enough.

Now, what Joe Sparano looked at at ASCO, and we had kind of discussed this when we were looking at all the slides, was the last slide that he put together that incorporated endocrine therapy with Oncotype and with TAILORx, which only looks at chemotherapy benefit. And the thought is that these patients that have maybe a higher clinical risk, maybe not a genomic risk that will mean that they should get chemotherapy, but based on that intermediate-risk category, maybe we should optimize their endocrine therapy by offering them ovarian suppression. I think that’s where the gap in data is, where we don’t have clear understanding of whether we should be giving them OFS or not. But in my mind, that’s where the benefit in this woman would be, mostly in ovarian function suppression versus chemotherapy.

DR LOVE: And as you say, that debate really started 2018 ASCO. That was the first time I really saw menopausal status really being factored into interpretation of the Oncotype. And you kind of saw the same thing this year at ASCO, but now looking at something else, which is the issue of adverse clinical features like tumor size. And did you come away with the same message in general that for the postmenopausal women, pretty straightforward, they’re not going to benefit from chemo? Premenopausal women, not clear. Maybe they’re benefiting from chemo. Maybe they’re benefiting from chemo-induced ovarian suppression. Did that theme go through this year too, do you think?

DR KAKLAMANI: I think so. And again, if you’re a purist, you’re going to look at the data and see all of the gaps in data and say, “Geez, we can’t really conclude that way.” And you’re probably right. But in oncology we work with gaps all the time, and we have to make the best of it.

When you look at TAILORx, they looked at menopausal status in the beginning of the trial, but they did not measure estrogen levels. They did not look at menopause after chemotherapy. All of these are total unknowns.

Tailoring adjuvant endocrine therapy for premenopausal women with ER-positive breast cancer

DR KAKLAMANI: When you look at TEXT and SOFT, they looked at OFS, but again, there’s limitations in those trials as well. And the benefit from OFS is not huge, and it does depend on clinical risks, which is why that STEPP analysis is so important.

When you try to marry the 2 trials that are totally different, they have different patient populations. They look at different endpoints, but at the end of the day, we are faced with these patients everyday.

I, at least, get back to what the Europeans have been doing for a long time, that what we’re doing, what we’re achieving with chemotherapy, especially in the younger patients, is making them postmenopausal, and do I really need to subject them to chemotherapy to make them postmenopausal when I can give them a shot or perform an oophorectomy and then make them postmenopausal?

I think in that at least in that intermediate Oncotype risk score where you’re struggling with the decision on what to do, adding OFS is probably beneficial and probably helps you not have to subject the patient to chemotherapy.

DR LOVE: And everybody has different opinions on this, and, of course, everybody also individualizes it. It’s kind of interesting. This lady is 48 years old, a little bit older, premenopausal. She was having regular menstrual periods? Completely premenopausal?

DR KAKLAMANI: She was. She was.

DR LOVE: She has this Oncotype come back as 19. How did you discuss it with her, and what did you do?

DR KAKLAMANI: What I discussed with her was the fact that she was in this really marginal category of 16 to 20, where the benefit from chemotherapy, if any, would be small, that I personally would not recommend chemotherapy based on that, but many people around the country would definitely discuss it with her. If we were to give her chemotherapy, it would be 4 cycles of probably docetaxel and cyclophosphamide. And then I told her that she would need endocrine therapy, that I thought the best endocrine therapy was going to be an aromatase inhibitor and that I thought that giving her ovarian suppression would probably give her as much benefit as the chemotherapy would in her case. Given all of that, and the fact that she was close to menopause anyway, she agreed to the OFS and endocrine therapy.

DR LOVE: OFS or surgery?

DR KAKLAMANI: She ended up agreeing to OFS. What I usually do — I’m not a surgeon, so I try not to tell patients they need surgery unless they really do. I usually give them a month or two of OFS to make sure that they actually tolerate it well before we discuss ovarian suppression versus oophorectomy.

And many of these patients are perfectly happy coming once a month for their shot. Many of these patients just want to get the oophorectomy and not have to come back for their shots.

Benefit of endocrine therapy with or without chemotherapy for patients with ER-positive breast cancer and an intermediate RS; consideration of clinical and genomic risk in assessing the likelihood of disease recurrence

DR LOVE: I’m just kind of curious, I don’t know if you actually got into numbers with this patient or if she wanted, but if she wanted it or if a patient like this wanted numbers, so she’s got a 3.1-cm tumor, node-negative and a Recurrence Score of 19. If she were to say, “What’s the chance that the cancer is going to come back if I take ovarian suppression as well as an aromatase inhibitor?” — incidentally, how long would you continue that therapy?

DR KAKLAMANI: The therapy I’d continue for 5 years, at least with the ovarian suppression. The question would be then, do we want to give extended endocrine therapy? And that’s another big dilemma and question for us. The benefit from chemotherapy I would probably say would be in the one and a half percent points’ category, which most women do not want to be subjected to chemotherapy based on that benefit.

The benefit of ovarian suppression, that’s a question. And when you look at the TEXT and SOFT, they did the STEPP analysis. What I usually do, and I don’t remember if I actually did it in this case, but I look at the STEPP analysis and I’m able to calculate what sort of benefit she would get from ovarian suppression plus an AI versus tamoxifen. And then I can have a little bit more of a discussion with them, depending how much data they want. A lot of patients don’t want a lot of data.

A lot of patients want to be told what to do or want to be given kind of a direction. But some patients really are data oriented, and those patients I sit down with the software and I calculate exactly what the benefit from ovarian suppression would be for her.

DR LOVE: Yes, what I was wondering about is, because I was kind of flashing on a fantasy of a couple years from now, waking up, checking your phone and saying, “Oh, CDK is positive in adjuvant setting.” I was curious about the control arm and how much better we have to try to make it. This woman’s a little bit bigger than a typical node-negative patient, 3.1 centimeters. But what would you say if she were to ask you, maybe she did ask you, “Even in spite of all your therapy, or endocrine therapy you’re talking about, what do you think my long-term risk for recurrence is?” What do you think you would say for her?

DR KAKLAMANI: I would again look at Oncotype and the score of 19. I would take into account — because now it’s prognostic data — so you’d want to take into account the tumor size.

With a score of 19, the risk of recurrence probably is around somewhere between 10% to 15%. And because the size of her tumor is a little bit bigger, it probably would be closer to that 14%, 15%. But that really would be it. Again, with tamoxifen. That data is with tamoxifen. I don’t think it would be much higher than that.

If you give ovarian suppression and an AI, you're decreasing that by a few percentage points. You’re getting closer to that 10% long-term risk of recurrence.

DR LOVE: It’ll be interesting. I don’t even know whether a patient like this would fit into some of these adjuvant trials. But if we did move towards a situation of adjuvant CDK and you think that she’s got a residual risk of 10%, and if you could cut it in half, that’d be pretty substantial for a therapy that really doesn’t have any life-threatening toxicities.

DR KAKLAMANI: It would. If I recall, at least one of the adjuvant trials that we participated in, they either wanted positive lymph node patients or they wanted patients that had a high Oncotype score. I don’t think this sort of patient would be incorporated in many of these trials, probably for good reason. Because a lot of them, you consider their risk lower. And if the risk is lower, that means it probably will take a while for their tumor to recur, which also means that these trials are never going to be completed. But also, the benefit would be smaller than somebody who, let’s say, has a 30% risk of recurrence on endocrine therapy. And then if you decrease it by half, which is what we see in the metastatic setting, then you’re getting to 15%, which is a huge improvement.

DR LOVE: Of course we’re being optimistic by hoping it’s going to have that kind of risk reduction, but I do think it is going to be interesting for patients like this, though, because they maybe couldn’t have gotten on the trial, and if you do see a significant risk reduction, it might be tempting to do.

Anyhow, it looks like it’s something interesting to look forward to other than checkpoint inhibitors in the future.

Case: A woman in her early 40s with ER-positive, HER2-negative breast cancer and metastases in the liver and bones attains a partial response to ribociclib in combination with endocrine therapy

DR LOVE: Let’s talk a little bit about metastatic disease, and you had a couple of cases here. How about this 43-year-old lady? Originally 4-cm primary. She had a metastatic workup and found to have a liver met, which on biopsy was positive, ER-positive/HER2-negative, and bone mets. She presented with what appeared to be localized disease but actually was found to have occult mets.

DR KAKLAMANI: Again, this is a premenopausal patient, and what I wanted to highlight was again another ASCO presentation when we look at giving a CDK4/6 inhibitor, in this case ribociclib, in combination with endocrine therapy in premenopausal women and improving survival. This was pretty interesting data. This was the first data highlighting an improvement in survival with any CDK4/6 inhibitor. There’s going to be other trials that are going to be presented soon, hopefully showing the same improvement in overall survival.

But to me, there’s a couple points to make. One, when we look at patients with liver metastases, oncologists are very fast to think about chemotherapy. And I think the benefit of having these CDK4/6 inhibitors is that we don’t have to think that way any more. These drugs work very fast, so within 2 months you see already a separation in the curves between the endocrine-only arm and the endocrine plus CDK4/6 inhibitor arm. And so if they work fast enough — and they seem to work as fast as chemotherapy, if not faster — there’s no reason why we should be subjecting these women to chemotherapy. And then the second, obviously, talking point is the fact that we can give a CDK4/6 inhibitor and get a benefit instead of just endocrine therapy, and the benefit is pretty substantial.

Now, there was another presentation at ASCO looking at CDK4/6 inhibitors or capecitabine, and this was a Korean trial, again in premenopausal women, and it showed, maybe surprisingly to some, not to me, that the CDK4/6 inhibitor outperformed chemotherapy. And these were patients that could have received chemotherapy previously. These were not just your first-line metastatic breast cancer patients. But there was an improvement in progression-free survival. The response rate was relatively similar, numerically a little bit higher, if I recall correctly, with the CDK4/6 inhibitor. But again, you're comparing it to chemotherapy, and you’re getting better outcomes with the CDK4/6 inhibitor. To us in breast cancer, that’s something we haven’t seen before.

Benefit with CDK4/6 inhibitors versus chemotherapy for ER-positive breast cancer with visceral metastases

DR LOVE: Yes, it’s great that you brought up that point. I don’t know if you saw a program we put out with Sara Tolaney we worked on, and we did a survey of oncologists in the community, and then we had a think tank with 6 breast cancer investigators. And one of the most striking things was exactly what you’re talking about, the patient with liver mets. And we saw a fair amount — actually, the most common strategy in the community was chemo first and then hormone maintenance, and all 6 of the investigators said, “No.” Endocrine plus CDK. It was right there in the data. Still, today. I mean, that’s something we were talking about 25 years ago, I think.

DR KAKLAMANI: Mm-hmm. I agree. I think whenever we see visceral metastases, we get scared. And I understand that with just endocrine therapy, because even though in the long run it will work better; in the short run it won’t work as fast. But these drugs work very fast. Every single trial with CDK4/6 inhibitors has shown separation of the curves within 2 months, within 8 weeks of doing scans.

To me, there’s no reason why we should be afraid, and we should be giving chemotherapy to these patients. I’ve switched my practice completely to giving a CDK4/6 inhibitor.

DR LOVE: And actually, this case brings up another thing we saw in this survey, which was, I figured everybody was going to use CDK in every metastatic situation in the community, and it’s not quite 100% by any means. And, actually, the situation where we saw the least, but, I mean, it was still three quarters of people were using it but a quarter weren’t, were exactly the case of this woman — people who present with metastatic disease.

I don’t know, for some reason there’s a segment of oncologists who won’t use a CDK. I mean, again, the minority, but in that situation, any rationale for saving it? I don’t know what the strategy might be.

DR KAKLAMANI: I don’t think there’s a reason to save the CDK4/6 inhibitor for later. Now, having said that, when you look at the trial presented at ASCO, the number of patients that were not on the CDK4/6 arm that received CDK4/6 inhibitor after progression was very small. In some way there’s still a question as to whether if we sequence these drugs we’re getting the same benefit.

So, for example, giving an aromatase inhibitor first line and then giving a CDK4/6 inhibitor second line, are we really compromising our survival or not? And that question is still unanswered. But to me, these drugs are not really that toxic. Besides some neutropenia or maybe some diarrhea, depending on which one we use, the toxicity profile is pretty straightforward. I think giving our best drug as first line makes perfect sense. The progression-free survival in all these trials is upwards of 2 years. That’s something that, again, we haven’t seen in breast cancer before. To me, I try to give it early on instead of saving it for later lines of therapy.

Efficacy of ribociclib in premenopausal women; QT prolongation associated with ribociclib/tamoxifen

DR LOVE: You mentioned the ribo data was presented. What do we know right now about all 3 of the CDK inhibitors in premenopausal women?

DR KAKLAMANI: The largest trial was the ribo trial. And there’s some subsets in the palbo and I think the abemaciclib as well that were premenopausal, but it was not a large subset in any of these other trials.

I think for us, it’s a lesson that even though we think we know, unless we do a trial, we don’t know. One of the signals that came out of that trial was the fact that we should not be giving tamoxifen with ribociclib. And we had never used tamoxifen with CDK4/6 inhibitors before because of all the patients being postmenopausal. And if you used tamoxifen with ribo you’re increasing the QT prolongation, which, obviously, may be harmful. If we hadn’t done that trial, we wouldn’t have known.

Not only do we now have a survival benefit, which we didn’t have before, but now we also know we should not be using tamoxifen with ribociclib. And I personally would probably avoid it with some of the other CDK4/6 inhibitors as well, because they do seem to prolong the QT as well.

DR LOVE: Yes. I think a lot of people have had the concept of premenopausal women, you make them postmenopausal and then you treat them as a postmenopausal patient. But it is true that the data, at least in terms of CDK4/6 inhibitors, seems to be more in terms of ribo.

Comparison of the efficacy and toxicity profiles of abemaciclib, palbociclib and ribociclib for women with ER-positive mBC

DR LOVE: What about in the postmenopausal woman? Can you talk about how you go about selecting amongst the three? And first, from your point of view, even in the premenopausal patient, at this point, is there any way — this kind of reminds me a little bit of the checkpoint inhibitor — is there any way to differentiate efficacy at this point amongst the three, other than abema, maybe in terms of the monotherapy angle, which maybe is not exclusive, or CNS? Any way to differentiate efficacy amongst the three?

DR KAKLAMANI: We never like doing cross-trial comparisons, so we always do them. That’s my first, kind of, statement in the beginning before I start looking at these different trials. And when you look at all of the CDK4/6 inhibitor trials, the hazard ratio is very similar. It’s close to 0.5. Can be 0.48, 0.52, 0.53 or something, but it’s always close to this 0.5. And so it looks like all three, efficacy wise, are very similar.

Now, the premenopausal data, we have more robust data with MONALEESA-7 and ribo. That there you can make a case and say, “I really should be using it in that patient population.” And other people will say, “The other drugs have some data as well,” and that is a very valid point, too.

The toxicity profile is something that sometimes helps me decide. I had a patient who had just had acute cholecystitis. And they wanted to take her to surgery, but they didn’t. And so this patient came to see me. She needed to be on a CDK4/6 inhibitor. And in that case, what I was thinking was, if something happens and she gets some other attack and they need to take her to surgery, I’d really not be crazy about her going to surgery with an ANC of 0.8.

In that case, using abemaciclib might be a better option because of the less toxicity in the white count. But again, you’re compromising with the diarrhea, because it’ll have more incidence or higher incidence of diarrhea.

Now the other two, the ribociclib and palbociclib, are very similar as far as toxicity profile. With ribo you need to do an EKG, and you need to look at the LFTs. But to me it’s really more a precautionary measure more than anything else, because there is a little bit of QT prolongation with the other drugs, and we also want to not expose patients to other drugs that increase their QT as well.

Those are small differences. And it reminds me of the early 2000s when we had 3 aromatase inhibitors and all 3 had very similar toxicity and very similar efficacy profiles. And we were arguing as to which one to use.

DR LOVE: I know. I was flashing on the ATAC trial. I think it was, like, 2001 or something, anastrozole versus tamoxifen, and the biggest trial in the history of oncology, 7,000 people. Everybody was all wild and excited because there was a 20% relative improvement. You look back now, it kind of doesn’t seem as exciting as the stuff that’s going on right now.

DR KAKLAMANI: Exactly. And none of these AI trials are showing much survival benefit either. In the metastatic setting, they never did.

DR LOVE: But it was just huge, huge news.

Activity of CDK4/6 inhibitors as monotherapy; CNS penetration of CDK4/6 inhibitors

DR LOVE: What about monotherapy with a CDK4/6 inhibitor? What data do we have on all three of them, and do you ever use it? In what situations?

DR KAKLAMANI: The data that we have is with abemaciclib. And it’s in later line. That kind of affects our patients, not as much anymore because now we start with a CDK4/6 inhibitor, but if we have somebody who’s been on endocrine therapy, maybe has been on some chemotherapy and has gone down all the endocrine therapy lines, then giving abemaciclib first line seems to be a viable option. It is approved by the FDA.

I honestly have not used it. And mostly because most of the patients that I had had not been on all the endocrine therapy drugs. And so I was able to put them on fulvestrant and a CDK4/6 inhibitor or an AI and a CDK4/6 inhibitor. But it’s definitely a valid option for some patients.

DR LOVE: Any reason to think about using it after a CDK inhibitor? After everything else?

DR KAKLAMANI: No data. There are trials ongoing, and so that’s the other unanswered question in metastatic disease: Can we sequence CDK4/6 inhibitors, or can we give the same one postprogression like we do with trastuzumab, for example? I’ve never done this, because we don’t have any data. To me, it’s an unanswered question.

DR LOVE: The other thing I was mentioning that you kind of hear about, I’m not sure exactly how relevant it is, is the issue of CNS penetration. Is there a difference in the three? And I don’t know how often you run into a situation of a patient who has CNS disease where this might be a factor.

DR KAKLAMANI: There’s to me not a lot of data, mostly because in ER-positive disease we don’t think of CNS disease as a primary issue. The more these patients live, the more we are going to see it. And I’ve seen it in my patients. But usually I’ve seen it in patients that have been on several lines of therapy, and some of them may actually develop leptomeningeal disease. My thought is that it may be because of the bone disease that is there that eventually will invade the meninges. But who knows.

I mean, there’s some data with abemaciclib, and you think of using that agent preferentially there. But I don’t know that I’m convinced that it’s only abemaciclib doing that. I guess we’ll get some more data in the future and we’ll be able to see better.

Monitoring and management of side effects associated with CDK4/6 inhibitors

DR LOVE: Any comments in terms of your clinical experience dealing with some of the toxicities that you’ve talked about? You mentioned GI problems with abemaciclib. When do they start? What do you see? Do you do anything preemptively?

DR KAKLAMANI: I don’t do anything prophylactically. I talk to the patients, and I ask them. They usually start soon after we start the CDK4/6 inhibitor, and I usually look at loperamide as my go-to medication. And in most cases it’s relatively well controlled with that.

Now, when you look at the other two, you’re dealing with ANC decreases, and the question is, how stringent do you want to be? When we first were exposed to these drugs, we were following the guidelines, and if any patient had an ANC less than 1,000, we were stopping and rechecking. And now I think most of us have moved away from that approach, because what I got to was starting and stopping and starting and stopping and starting and stopping, because the guidelines didn’t really say you should dose reduce — they said you can dose reduce.

Patients seem to be doing fine. The risk of neutropenic fever is pretty low with these drugs. The risk of infection is a little bit higher, so maybe in the more vulnerable patient population, a little older patient, something to keep in mind. But in the majority of patients, I haven’t seen much as far as infections. And if I do it’s usually Grade 1 or 2, so nothing that will require hospitalization. I tend to be way more lenient with my ANC checking than I used to be.

DR LOVE: What do you do in terms of dose reduction, though?

DR KAKLAMANI: I will dose reduce if a patient has fever, if I see them having an infection. And the good news is that the data combined from all three of them is showing that dose reduction does not compromise efficacy. I’m very comfortable dose reducing these patients without thinking that I’m compromising how the drug is going to work.

DR LOVE: Hmm, that’s interesting.

You mentioned a patient going to surgery. If you know in advance that a patient needs minor surgery, major surgery or a tooth extraction, for example, do you take that into account in terms of the scheduling related to the CDK, particularly, obviously, with the palbo and the ribo?

DR KAKLAMANI: I do. I think that’s where you might run into more issues. If you think of a tooth extraction in somebody who has a neutrophil count of 0.7 or 0.8, that probably might get you in trouble. I usually will think of using abemaciclib or delay the CDK4/6 inhibitor if I have the luxury to do that.

Case: A woman in her mid-50s initially diagnosed with Stage II ER-positive breast cancer develops metastatic disease to the bone 8 years later and receives palbociclib and anastrozole

DR LOVE: I want to ask you about this 54-year-old lady, just specifically about 1 issue, because we’ve been thinking a lot about CDK inhibitors in terms of where the patient is in the natural history. Your first patient was at first diagnosis. A lot of patients get diagnosed with mets during adjuvant endocrine therapy, and then the other space is people who’ve had adjuvant endocrine therapy stopped and then get recurrence later, which is what happened here with this 54-year-old lady. And it’s interesting that she actually got adjuvant tamoxifen for 5 years since she was premenopausal, in her midforties. And then she was off therapy at 3 years and came in with metastatic disease.

What was the situation when she presented there with the bone mets? She had symptoms from the bone mets?

DR KAKLAMANI: She did. She came in with back pain, which is how we diagnosed her with the metastatic disease.

DR LOVE: And she had bone-only mets?

DR KAKLAMANI: She had bone-only disease.

DR LOVE: And did anything need to be radiated, or you thought you could use systemic therapy?

DR KAKLAMANI: That’s the other tricky point, whether you can radiate someone and keep them on a CDK4/6 inhibitor at the same time. And there’s some case reports that suggest that you might increase toxicity in some ongoing trials. Personally, I think most people we don’t do both together. If you want to radiate someone, you kind of wait on the CDK4/6 inhibitor. You radiate them, and then you start the CDK4/6 inhibitor.

In this case, her pain was not severe enough and her MRI was not bad enough that we needed to do anything at that point. We ended up putting her on endocrine therapy plus the CDK4/6 inhibitor.

DR LOVE: Yes. And, actually, the thing that I was most interested in is, which endocrine therapy? You gave her anastrozole, and of course she’d only had tamoxifen. But another very common situation is a patient who’s had 5 years of an AI. What do you do when they relapse after they’ve been off 3 years?

DR KAKLAMANI: There I think it’s justified to give them fulvestrant. If it’s been 3 years, I think re-exposing them to anastrozole makes sense. The one thing that people may want to be considering is, should I do a genomic profile and see if the patient has an ESR1 mutation? And if she does, should I then give her fulvestrant?

The way I look at it is, when you look at the CDK4/6 trials and look at ESR1 mutations, whether a patient had it or not, they still receive the same benefit with the AI and the CDK4/6 inhibitor. I don’t usually do the genomic assays for that reason. I think if it’s been 3 years, re-exposing her to the AI makes sense and then obviously adding the CDK4/6 inhibitor. And then second line, you’d want to think of fulvestrant plus another targeted agent.

Perspective on the use of multiplex genomic assays to guide treatment decision-making for patients with ER-positive mBC

DR LOVE: Actually, you tied into another issue that we’ve been giving some thought to, which is the issue of multiplex genomic testing in women with metastatic breast cancer. And, of course, you could view breast cancer easily is at least 3 different subsets. But in terms of that question, so for a patient with ER-positive/HER2-negative metastatic disease who’s going to go on first-line therapy, endocrine therapy and CDK, do you do some type of genomic? Do you do NGS? Do you do liquid assays?

DR KAKLAMANI: I do sometimes. It almost never helps me. There was a nice debate — 2 years ago we had it at SABCS — about whether these genomic assays help or not. And in breast cancer they rarely help us at the moment, especially in this ER-positive patient population.

Sometimes you kind of want to do it because you’re hoping it will help you. But if you think about it, how is it going to help? If a patient has an ESR1 mutation, you probably want to avoid single-agent AI. But you almost never give single-agent AI anyway. And if you combine the AI with the CDK4/6 inhibitor, you’re fine. If you combine it with an mTOR inhibitor, you’re probably okay.

Then you can find a PIK3CA mutation, which obviously is huge. Because now we have alpelisib that was approved and should be considered standard of care for this patient population, but we have a specific companion diagnostic for that. And it’s still unclear to me whether the large multiplex panels capture all the different exons that needs to be captured for the PIK3CA status.

Then you have p53. If that’s positive, then you know you’re in trouble, because these tumors are just going to be resistant to everything. But again, you don’t have any treatment options.

If the HER2 comes back positive, then you have to kind of wonder, what HER2 are you looking for? Is it just the immunohistochemistry that some of these assays are running? Is it mutations? And then you may want to be exposing them to a tyrosine kinase inhibitor. But again, that’s not standard of care. A lot of these assays will not give us great results or actionable results.

Results of the Phase III SOLAR-1 trial evaluating alpelisib for patients with ER-positive advanced breast cancer

DR LOVE: Anyhow, in this lady, she got anastrozole and palbo, and it’s now 3 years later. How’s she doing?

DR KAKLAMANI: She’s been doing okay. She has not progressed on the CDK4/6 inhibitor yet. Now, when she progresses, I think the option, if she is PIK3CA mutated, is going to be to put her on fulvestrant plus alpelisib. If she’s not, then an mTOR inhibitor is probably the next option for her, in combination with endocrine therapy. I think fulvestrant would probably be my endocrine therapy.

DR LOVE: I am curious what your thoughts are about alpelisib. I’m curious about your statement there, that if she were PIK3-positive you think that would be the treatment of choice. Maybe you can kind of go through the data and explain why you believe that. Because, actually, I think everolimus is a pretty good drug too.

DR KAKLAMANI: I think so as well. And you can definitely sequence — I mean, you will be sequencing these drugs. Now, when you look at the alpelisib data that was published in the SOLO-1 trial, which was presented at ESMO, re-presented at San Antonio with some extra information and then recently published, the drug approved in May, showing that in patients that are PIK3CA mutated only, the combination of fulvestrant plus alpelisib was superior to just fulvestrant.

Now, very few patients, less than 20%, were given a CDK4/6 inhibitor in that trial. And I can’t remember, I think mTOR inhibition again was a very small subset of patients.

That’s where we don’t really now is alpelisib going to work post-CDK4/6. Is CDK4/6 going to work postalpelisib? What do we do with mTOR? Those are again the gaps that we have to deal with. But I think, based on the data, that’s a nice targeted approach, in that 40% of the patient population in the metastatic ER-positive setting that are going to have mutations in PIK3CA, that I would definitely consider it probably as my second-line therapy.

Tolerability of alpelisib for patients with ER-positive breast cancer

DR KAKLAMANI: The limitation is going to be toxicity, and the major toxicity with alpelisib is hyperglycemia. And initially when you think about it, you say, “Oh my god, it’s hyperglycemia, I can deal with it,” we, actually, initially when we participated in the trial and there weren’t very strict criteria, we had a patient hospitalized with hyperglycemia. Hyperglycemia is not something that we want to take lightly. It happens very quickly. You want to prophylax your patients that have high blood sugar. And if they don’t, you want to watch them very closely, because they can develop it.

DR LOVE: Hmm. What about other tolerability issues? Any side effects?

DR KAKLAMANI: The rest of the side effects are pretty manageable. You may have a little bit of diarrhea. You may a have a little bit of rash. The rash is dose dependent — the diarrhea as well. Again, it’s going to be a learning curve or how to use the medication and how to deal with the toxicities, like it is with every medication.

If you see a rash, you can dose reduce and the patients should not experience it again — same thing with the diarrhea. The hyperglycemia, definitely something to keep an eye out for.

DR LOVE: It’s still kind of hard — initially, we were hearing kind of negative things about tolerability with this class of agent. Is this one better tolerated than some of the original ones?

DR KAKLAMANI: It is. The original studies were done with the pan-PI3 kinase inhibitors, and those tended to have more toxicities, including some psychological issues, with major depression. But the targeted one — so this is a PI3 kinase alpha inhibitor. Those are much more easier to tolerate. Definitely easier.

I go back to when we initially started using everolimus. We all thought it was almost as bad as chemotherapy. It caused mouth sores. It caused a rash, diarrhea, pneumonitis. And now we use it like it’s nothing. Why? Because we got to use it enough times that we’ve figured out how to deal with these toxicities. We figured that we want to give a mouthwash, and that pretty much gets rid of the mouth sores and so forth. I think there’s going to be a learning curve with that, with alpelisib as well.

Sequencing everolimus and alpelisib after disease progression on a CDK4/6 inhibitor

DR LOVE: If you have a patient who, say, who has metastatic disease, so she’s on an adjuvant AI. They get fulvestrant/CDK first line for mets. And then they have disease progression and they’re PIK3CA-positive. It sounds like you're going to go with alpelisib?

DR KAKLAMANI: I will. The question is going to be the endocrine backbone, since alpelisib was approved with fulvestrant. But I think with all of these agents there’s good data, and there’s now ongoing trials with alpelisib looking at alpelisib plus AIs, that you can partner them with pretty much whatever endocrine therapy you want to efficacy wise and you're going to be okay. I would give her second-line alpelisib plus endocrine therapy, third line everolimus plus endocrine therapy.

DR LOVE: And so I’m assuming or guessing that if the patient were PIK3-negative, you would use everolimus with endocrine therapy?

DR KAKLAMANI: Exactly. I would, yes. I mean, those are active drugs. And I think when you sequence these drugs in breast cancer, you’ve yet to prove any survival benefit. You’re definitely justified, in my mind, to give everolimus second line and to give alpelisib third line or the other way around.

I guess the reason I’m thinking of alpelisib is, I like the fact that there’s a mutation and I am attacking that mutation and that’s why I’m getting the benefit, versus everolimus, where we haven’t really shown a patient subset that everolimus works better. And that’s probably what’s driving my decision. But you’re absolutely correct that you probably can sequence these drugs perfectly fine.

Case: A postmenopausal woman in her mid-30s with ER-positive mBC and BRCA2 and PIK3CA mutations receives talazoparib in the third-line setting

DR LOVE: Let’s finish out with your last patient. This sounds really fascinating: a 36-year-old postmenopausal woman, because she had a BSO, who’s BRCA2 germline-positive. And did she know prior to being diagnosed with breast cancer about the BRCA?

DR KAKLAMANI: She found out at the time of diagnosis.

DR LOVE: Wow. She had a family history?

DR KAKLAMANI: She had a limited family history. It’s interesting how 50% of the patients that are diagnosed with BRCA mutations do not meet criteria for testing.

DR LOVE: Right. Although sometimes I wonder what the criteria for testing is going to evolve into with the use of PARP inhibitors, but I guess that’s more metastatic disease.

Anyhow, so she has ER-positive/HER2-negative metastatic disease to the liver and spine. Now, this was at presentation, or she had a recurrence?

DR KAKLAMANI: That was at presentation.

DR LOVE: Again. Okay. And so she gets started out on an AI and palbo but then has progression in 6 months. That’s pretty unusual.

DR KAKLAMANI: It looked like she had endocrine-resistant disease, and that was my concern as well. You’re expecting these drugs to have a progression-free survival, PFS, of around 2 years, if not more, and 6-month PFS is very, very low. Very, very short.

DR LOVE: But — now, she got the BSO after she found out she had breast cancer?

DR KAKLAMANI: Exactly.

DR LOVE: Hmm. That’s interesting. She’s getting an AI and palbo.

She was PIK3-positive. Is it just positive or negative? Or is there, like, a number?

DR KAKLAMANI: It’s positive whether they’re mutated or not. Again, the trick is to make sure that you’re checking the hotspot exons on the PIK3CA, but she was positive. She had a PIK3CA mutation.

DR LOVE: Hmm. That’s interesting. It’s kind of like maybe the EGFR situation in non-small cell lung cancer, different types of mutations, a PIK3?

DR KAKLAMANI: There are different types of mutations. I don’t know that we have enough data to show that one is better than the other to have. We’ve kind of looped them all together. And so if anybody has one of those mutations that compromise the function of PI3 kinase, then they’re called mutation-positive and they’re eligible for alpelisib.

DR LOVE: She actually got started on it 9 months ago, so she’s one of the early people. I guess it wasn’t even approved. And this was on a trial I guess, huh?

DR KAKLAMANI: This was on a clinical trial. Exactly.

DR LOVE: And what endocrine therapy was it combined with?

DR KAKLAMANI: It was combined with fulvestrant.

DR LOVE: Fulvestrant. Right, that makes sense. And how did she do?

DR KAKLAMANI: She actually did not do very well. She progressed relatively quickly on alpelisib, and that’s when I switched her to a PARP inhibitor, and she’s actually doing really well with the PARP inhibitor. And this was the first medication that her tumor responded to. She had some stable disease with palbo, but she never had tumor regression. And this was the first scan that she received that had a decrease in her liver lesions.

Choice of PARP inhibitor for patients with advanced breast cancer and BRCA mutations

DR LOVE: I was going to ask you about PARP. She got olaparib?

DR KAKLAMANI: She ended up getting talazoparib.

DR LOVE: Talazoparib! Interesting. You started her on it?

DR KAKLAMANI: I did.

DR LOVE: Oh, that’s interesting. And just out of curiosity, why did you choose talazoparib?

DR KAKLAMANI: The data looks very, very similar. The reason I did was it’s once a day. It’s a little bit easier to give. I think that’s why I picked it. But, I mean, those are equivalent medications.

DR LOVE: Do you think the tolerability profile is also equivalent, incidentally?

DR KAKLAMANI: A little different. A little different. A little more hair loss I think with the talazoparib. And I want to say the red cell count’s a little bit worse. But very, very similar in many other respects. I think you’re trading one thing with the other. I mean, she’s tolerated the medication wonderfully up to now. And, like I said, she actually had a response. But, I mean, I think, based on the data from both trials with PARP inhibitors in breast cancer, they’re equivalent drugs.

DR LOVE: I have heard about the hair loss. Did she have any hair loss?

DR KAKLAMANI: She did not.

DR LOVE: Hmm. Interesting.

Response and side-effect profiles of alpelisib

DR LOVE: It looks like she was on alpelisib though for 9 months. I mean, that’s not too bad, is it?

DR KAKLAMANI: It’s not too bad. She had stable disease.

DR LOVE: But she didn’t respond.

DR KAKLAMANI: She did not have a response. She had stable disease. And when you look at the data from SOLO-1, I think the median PFS was a little bit longer than that. But she responded better to the PI3 kinase inhibitor than CDK4/6 inhibitor.

DR LOVE: Did she have hyperglycemia?

DR KAKLAMANI: She had a little bit of hyperglycemia. She was lucky enough to be a very fit 36-year-old, and so we did not run into any major issues with her blood sugar.

DR LOVE: You didn’t have to treat her?

DR KAKLAMANI: Did not have to treat her.

DR LOVE: Yes. The general oncologists know about copanlisib, which is used in lymphoma, but that’s IV. And I think those patients get hyperglycemia, like, during the infusion, and then it goes right away. But here you’re taking a pill. I mean, I know they’re not the same drug, but I guess it’s sort of the same class.

Selection and sequencing of therapy for patients with mBC and germline BRCA mutations

DR LOVE: When I saw this case, and I saw the first 3 words there, BRCA2-positive, I immediately was thinking, “Hmm, where’s PARP going to fit into this story?” If you had to do it again, or if you found a situation like this again or even just this basic situation, what about hormone therapy plus PARP?

DR KAKLAMANI: I don’t know. It’s hard to know whether you’re going to get a benefit. There’s really no data that I’m aware of combining the two. You could think of starting with a PARP inhibitor and then given endocrine therapy later on. And again, there’s no data on the sequencing. Patients on both PARP trials, if they were ER-positive, they could have received endocrine therapy previously.

I think I’d probably do it all over again, because you have a patient who’s ER-positive, and so you're thinking of toxicity. She’s 36. She has 2 young kids that she’s taking to school everyday, and so making her sick is not really what I wanted to do, especially since she was relatively asymptomatic from her breast cancer. I was perfectly happy trying to maximize her endocrine therapy as much as possible before I subjected her to chemotherapy. And, thankfully, now we have PARP inhibitors as well.

DR LOVE: I wasn’t suggesting using that strategy, incidentally, because I think it might be also tricky to figure out toxicity, and then you have stop both of them. There’s no data — of course that’s another issue. But it is interesting, though, when we present to breast cancer researchers where they put PARP into BRCA-germline patients for ER-positive/HER2-negative, they almost all give hormone/CDK first.

But they do give PARP before chemo. But just out of curiosity, even though we’re focused on ER-positive/HER2-negative, what about triple-negative BRCA germline? And, of course, there you have the issue also of checkpoint inhibitor/chemo.

DR KAKLAMANI: I think there I would definitely look at PD-L1. I think that now we have a valid option. I probably would still give a PARP inhibitor first, because it’s less toxic than chemotherapy, and then upon progression I would consider giving the atezolizumab plus nab paclitaxel/chemotherapy if PD-L1-positive, and if PD-L1-negative, chemotherapy, which probably second line would end up being a platinum in a BRCA-positive patient.

DR LOVE: I think you’re reflecting what most investigators are doing, but I was not quite sure because of the atezolizumab making it a little bit more complicated. PD-L1, definitely a factor.

Just to round it out, again, out of curiosity, HER2-positive/BRCA germline? Where, if at all, do PARP inhibitors fit in?

DR KAKLAMANI: Rare, but definitely happening. The trials did not include HER2-positive patients. That’s going to be yet another gap that we have. I personally would not treat them any differently as far as whether I would give them a PARP inhibitor or not. But now we have other treatments, such as anti-HER2 therapy, which is pretty effective in this patient population.

How to sequence these, I don’t know. That’s a good question. I haven’t had a metastatic patient that I’ve had that dilemma on. It makes it even worse if that patient is also ER-positive and now you have all 3 options to consider.

I’d probably consider anti-HER2 therapy up there, especially since in many of these patients you can keep them on dual anti-HER2 therapy without chemotherapy for a long time. And then again, toxicity is pretty good. And do you really want to stop the anti-HER2 therapy to give the PARP inhibitor? Who knows. Again, no data, because the patients on both trials had to be HER2-negative.

DR LOVE: Yes. It’s interesting. It’s not that much different than what we were talking about, hormone therapy and PARP. But there you have CDK, and that’s making it even more complicated. That’s life in oncology, isn’t it?

DR KAKLAMANI: Gaps in knowledge. That’s what we deal with. That’s why I guess we’re not replaced by machines yet.

DR LOVE: Subsequent to this interview, on September 13, 2019, the FDA issued a warning that palbociclib, ribociclib and abemaciclib may cause rare but severe inflammation of the lungs and concluded that the overall benefit of CDK4/6 inhibitors is still greater than the risks when used as prescribed.