Prof Kimberly Blackwell – Duke University, North Carolina, USA
Prof Hope Rugo – University of California, San Francisco, USA
Prof Peter Schmid – St Bartholomew Hospital London, UK
Dr Mark Verrill – Freeman Hospital, Newcastle, UK

KB: Hi, I’m Kim Blackwell and I’m here at this year’s 2017 ESMO meeting on behalf of ecancer. We’re going to be discussing, as a panel, the role of CDK inhibitors in the treatment of metastatic hormone sensitive breast cancer. So I’m lucky enough to be joined by my colleagues Hope Rugo, Dr Hope Rugo, Dr Peter Schmid and Dr Mark Verrill. We’re going to have a discussion about not only new data, old data, how we’re incorporating CDK inhibitors in our day to day practice and we’ll finish up with some practical concerns about how we incorporate CDK inhibitors. So with all that let’s get started. I’m going to ask each of you what you’re most excited about, maybe we’ll start with Mark. The data in regards to all three CDK inhibitors, maybe in combination with the aromatase inhibitors, just a brief perspective of how you’re using that and the studies that led to you using CDK inhibitors in combination with aromatase inhibitors.

MV: It’s very exciting that we’ve got something new in hormone therapy. The last time I was involved in a new trial of an anti-endocrine agent it was CGS20267 before it even got a name. I think it’s amazing that we’ve now got a new class of drugs, completely new, we’ve got three drugs that are all performing really well and we’re virtually doubling the time over which we can control endocrine sensitive disease. From a patient perspective that means that people are coming up, they’re picking up their tablets, they’re getting very little in the way of clinical side effects and we’re pushing back the time at which we have to abandon endocrine therapy and move on to chemotherapy and the inevitable diminishing returns of chemotherapy in that patient population.

KB: Peter, do you just want to give us a very brief overview of the data that looked at the use of CDK inhibitors with each of them? Maybe start with the PALOMA series and move on to the MONALEESA series.

PS: Yes, and I think it’s probably best to look at them by treatment indication. So we have a number of randomised trials that combine CDK4/6 inhibitors with aromatase inhibitors in what’s perceived to be hormone sensitive disease. Again, as we heard, the progression free survival time has doubled with the combination compared to single agent aromatase inhibitors. What I find remarkable there is that in the studies we consistently reach a median progression free survival of two years which is unique in metastatic breast cancer. If you look at the next series of studies based on fulvestrant, these are patients who have progressed on previous endocrine therapy so have some degree of hormone resistance. Again we see a very consistent result in terms of the relative benefit, again a more than doubling of the time to progression in those trials. What we also see across all randomised trials is that the objective response rate is higher for the combination of CDK4/6 inhibitors and hormone therapy compared to single agent hormone therapy which is really important from a patient perspective because response often leads to improvement of symptoms. Again, if you look at symptomatic control, symptomatic improvement, that’s an important result. As we heard before, the side effect profile, apart from blood-based side effects from myelosuppression, is relatively favourable and we add very little what I would call subjective toxicity from a patient point of view. So it’s a treatment that can be given for two years to patients without having the effects, the fatigue and all the other cumulative effects, we see with chemotherapy if you try to give chemotherapy for 2-3 years. So looking at all this together it’s fantastic new treatment options that’s changing the way that we are treating ER positive breast cancer.

KB: What amazes me is across the studies the hazard ratio for adding the CDK inhibitors is right on top of one another. So, both in MONALEESA-2, which is the ribociclib study, and PALOMA-2, which is the palbociclib study, the hazard ratio is 0.55 - identical. And we saw a similar thing with the phase II PALOMA-1 study. Speaking of toxicity, one of the things that’s great across the studies is we’ve not seen a patient reported outcome decrease in quality of life when we’re adding another drug. So those two consistency things, activity but also really little to no impact on quality of life, that makes the therapeutic index for these CDK inhibitors quite attractive. Hope, you’re always on the cutting edge of what’s going on, what are you looking forward to over the next couple of months besides the large phase III studies that have already been presented? Anything that you think might shape using one CDK inhibitor over the other or just how you use this group of agents.

HR: Maybe I’ll start with the second question first which is how we use the drugs in practice. It’s been actually very interesting to try and think about who we would not give a CDK4/6 inhibitor to in the first line setting and what the right partner is in terms of hormone therapy. With the FALCON data looking at fulvestrant versus an AI and showing this marked improvement in patients without visceral metastases, it’s been an interesting question about whether that’s good enough now that we have approval for fulvestrant, really a remarkable achievement given starting at a not great dose; we had dose change plus line change. Whether we should be adding a CDK 4/6 inhibitor, is 24 months enough for your median PFS? Maybe we want it to be really a lot longer. Trying to balance that out versus the economics and also just the intensity of follow-up for patients is always a question. You have to come in for an injection versus getting a pill so how does that manage in terms of the quality of life for our patients? I don’t think we pick up those differences in our quality of life tools yet. When we give CDK4/6 inhibitors what I’ve found is that patients who are taking the medication who are older tend to get more cumulative fatigue over time but we can manage that really well. Recognition of side effects is incredibly important, even when they’re not at a grade that brings up any red flags. So we find over time some of the older patients in particular need to be dose reduced. We’ve actually used a lot of the steroid dental paste for patients who get a single aphthous ulcer, it’s not terrible but it’s a little uncomfortable and you can make it go away overnight using this steroid dental paste so that’s a good thing too. The neutropenia I’ve found really not a big challenge at all. We have patients going in locally to get their blood checked, the blood count results are sent to us and the nurses give feedback to the patient so they don’t have to even come in very frequently which is very nice. So that does give a lot of weight to using an all-oral regimen. So we’re learning about that right now, what’s the optimal hormone partner for a specific patient? The FALCON patients never had any prior endocrine therapy, we tend to use fulvestrant more in the people who are exposed to a prior AI. So I think we’ll have to ferret that out.

One of the exciting things that has come out from these trials, we have the huge wealth of data from these randomised phase III trials so we can go into the databases and try and look at which are the patients who didn’t do as well. It turns out we can’t figure out who doesn’t benefit from a CDK4/6 inhibitor, that’s a whole other story, but even looking at hope on two different markers. But we can figure out who does relatively more poorly and has a shorter PFS. So maybe those are the patients we could select out to add additional therapy. So some of the things that have been looked at are adding PI3 kinase inhibitors, of course, and balancing the toxicity versus benefit. There are some really exciting early results adding the alpha-specific PI3 kinase inhibitors and also adding a PI3 kinase mTOR combination drug which is quite interesting as well. I think we’ll see a lot more information about that in the next year or so, early data that will be expanded. There are other combinations that are being studied, some data from the neoMONARCH study as well as data with the other CDK4/6 inhibitors suggests that these drugs may increase the CDK8 T-cell infiltrates in tumours over time. So maybe they could actually be used as a priming for immuno-oncology drugs such as the checkpoint inhibitors which have traditionally not been thought to be very effective in our ER positive, slower-growing, less immune active drugs. That’s another exciting area. One study being presented at ESMO is looking at whether or not you would be better using chemotherapy or a letrozole CDK4/6 inhibitor, in this case palbociclib, as neoadjuvant therapy for a luminal-like cancer. Most of their patients have luminal B disease so the actual pathologic complete response rate was a little higher with the chemotherapy than with the hormone therapy but actually it wasn’t hugely different. You probably could select a group of patients who might be better treated with the CDK4/6 hormone therapy combination in the neoadjuvant setting. What’s really striking, and we’ve seen this now with all three CDK4/6 inhibitors, is that you see this dramatic drop in Ki-67, very rapidly within two weeks after starting drug to complete cell cycle suppression.

KB: Wow, so that was a lot of information but it sounds like not only are you widely incorporating the CDK inhibitors but I, like you, are excited about even more confusing combinations, combining it with immunotherapy. I’m excited about the alpha-specific PI3 kinase inhibitor because there’s a lot of pre-clinical synergy at least that those will work very well with anti-oestrogen or endocrine therapy. So it’s going to now, in the next six months to a year, we’re going to get in the situation not only are we going to have two, perhaps three, CDK inhibitors but maybe now a whole other class of agents that are going to be emerging as far as the alpha specific PI3 kinase inhibitors. We’re seeing fairly rapid regulatory approval of these agents; it was pretty straightforward when we just had palbociclib, now ribociclib received approval in the United States earlier in 2017, just got regulatory approval in the United Kingdom. Even at this meeting we’ll see AI plus or minus abemaciclib data from the MONARCH-3 which is a registration study that’s been reported at least as being a positive study. So anything that you all are going to use, maybe I’ll start with you, Mark, that will drive you to using one drug, one CDK inhibitor, over the other? That’s probably the hardest question I’m going to ask all morning. So are there any differentiators that you see across the studies that might make it more likely, assuming that all three drugs get approved?

MV: One of the remarkable things about this drug class is that you can’t actually differentiate at all between the clinical data. When you look at the survival curves for MONALEESA-2, PALOMA-2 they’re superimposable. I don’t think there’s any reasonable grounds on which we can differentiate between the two licensed products that we have at the moment. My concern is that in the UK then our access to the drugs will be driven by price and we won’t be sitting in the clinic saying, ‘Well this patient I think would benefit from ribociclib, this patient would benefit from palbo, this would benefit from abemo.’ What will happen is that the companies will go into a closed room with NHS England and will come out with the cheapest one and that will be the one that we’ll be encouraged to use unless there’s a very good reason to do something different. But that’s a shame. What I think is amazing is that we’re in a world now where over the last couple of years our mind-set for hormone receptor positive disease has changed and actually when I see a patient now I’m saying, ‘I want to give you a CDK4/6 inhibitor,’ and the decision then is which is going to be the partner, is it going to be letrozole, is it going to be fulvestrant. Rather than say, ‘OK, we’re going to give letrozole or fulvestrant, let’s decide whether or not to add a CDK4/6.’ That’s how we should be thinking, we should be thinking the first time you can use one of these drugs you should be offering that to your patient because you know they’re going to do better. The other thing, going back to Peter’s comment about response, we’re also now going to have to rip up the rule books about who we select for a first line endocrine therapy because the CDK4/6 letrozole first line combinations coming with a response rate of 50% has a higher response rate than any monotherapy chemo. So maybe we should be extending the use, using it in the more challenging cases, using it in patients with visceral compromise because we’ve got a good response rate without the baggage and the toxicity of the chemotherapy and without the risk, then, of harm to the patient.

KB: Yes, I agree with you. Some of the pre-operative data using the CDK inhibitors shows that you can see a drop in Ki-67 very early, just within two weeks, which is really no different than everyone uses chemo when they want a quick impact. But at least in the pre-operative setting you see it drop in the Ki-67 within that first two weeks and that data really helped reassure me, at least, that it kicks in pretty quickly. Peter, I’m going to ask you the same question. You have all three CDK inhibitors at your disposal, are there things that will drive your decision to use one versus the other, other than what we’ve already heard about the preferred endocrine backbone?

PS: I personally would rank it in an order where the first driver, decision driver, would be efficacy but, as Mark highlighted, there isn’t really a difference in the relative and absolute performance of each of those CDK4/6 inhibitors. The next decision point for me is safety profile and, associated with this, quality of life. If you look at ribociclib and palbociclib there is very little in between those two drugs. If you look at abemaciclib there’s slightly more GI toxicity and we will have to see how this pans out outside trials and whether that leads to a shift to one drug or another. We then need to look at the use of application, that’s sometimes depending on where you practice, depending on the label, you have to do extra tests such as an ECG or can you give the drug without those extra tests. My personal bias would be that this choice of endocrine therapy may be less important if we combine with a CDK4/6 inhibitor but ultimately we don’t have that answer yet and we need to see what the trial shows. Again, if you give treatment for two years or longer it is equally important to have an endocrine therapy that’s convenient to administer.

KB: Yes, PFS benefits across the trials are quite meaningful in terms of efficacy. I personally would love to see a survival benefit emerge. PALOMA-2, MONALEESA-2, we just don’t have enough events to even be able to say there will or will not. We could spend another thirty minutes discussing this. When you start to see the curves separating, like we saw at ASCO for MONALEESA-2 in terms of overall survival, even if it doesn’t reach that magic p-value of 0.05, it would be great. I don’t care which drug it is, it would just be great to say that we’re improving outcome in endocrine sensitive breast cancer in terms of survival like we’ve done with the HER2 positive space. So, Hope, this is a tough question but this is really what people are wanting to know. We’re going to quickly have a lot of drugs with efficacy data, safety data, anything other than what we’ve talked about that might drive your use of one drug over the other?

HR: It hearkens a little bit back to the AIs where we had a lot of great data and no real reason to choose between one versus the other.

KB: And lots of discussion around which one we would use.

HR: And tonnes of discussion, exactly. So in that case if it was Tuesday we gave one, if it was Wednesday we gave another, really there wasn’t another way to do it. For these drugs, palbociclib and ribociclib are more similar. You don’t want to forget the fact that you have to monitor for acute TC for ribociclib, there is a little liver enzyme change but these seem to be very minor issues and not really a big deal. The neutropenias are very similar so there’s no real reason to choose one or the other except for in the US the Patient Assistance Programs make a really big difference. So the easier it is to get the drug and to have patient assistance, because the costs are very high for patients, based on what the individual insurance company will pay. So that’s been our biggest battle, actually, and the weight on the staff in terms of getting that benefit, how much the patient has to provide, makes a big difference. Abemaciclib is really the most different drug, as you pointed out, the diarrhoea which was reduced by reducing the dose in combination in the MONARCH-2 studies. So we’ll see MONARCH-3 in a couple of days. In terms of that toxicity less neutropenia. So if you had a patient who was really struggling with the neutropenia and you kept having to dose reduce or had a bone marrow full of tumour, maybe using abemaciclib might be a better choice in that group of patients. That’s really the main differential that I can see. There are also going to be studies looking at abemaciclib after the other two CDK4/6 inhibitors because of the single agent efficacy but there’s nothing that we know of that suggests that these agents will work in sequence. But that’s a very important question that we’ll hear some information about in the next year or two. I’d be willing to take a bet with you about the survival differences, whether or not we’ll see any. I don’t think the trials are powered to see survival differences and at least in the trials that enrol in the US all patients will crossover, regardless of whether or not they can be unblinded. It’s very hard to be blinded for these drugs so we’re going to be hard pressed to see survival differences, you’d probably need close to 3,000 patients and we’re just not going to see that.

KB: I’m not taking that bet, by the way, I agree with you. I think the studies probably… it would be nice even just to see a large meta-analysis which I’m sure we’ll see emerge eventually.

HR: Of course.

KB: And maybe I just like that endpoint and the studies are probably too small to show that difference although one can dream and aim high.

HR: I think the interesting thing is going to be to see if I’m still working and well in the ten years or so until we see the adjuvant trial data. I’m joking a little bit because the adjuvant trials…

KB: You will still be working, I promise you.

HR: Yes, I will still be with you. But not only that, the adjuvant trials have really worked to try and include a group of patients that have a higher risk of earlier recurrence so that we do get information earlier. Whether or not that’s the case, or not, as we continue to change our therapies and incrementally improve outcomes remains to be seen. But it’s important to keep in mind that more than 15,000 women worldwide will be randomised on adjuvant or post-neoadjuvant CDK4/6 inhibitor trials in the next couple of two, three years. The trials are all going to accrue very rapidly and they are working on trying to include higher risk patients. As you go from the first to the third CDK4/6 inhibitor the risk for the patients that can be enrolled goes up as they want to try and get an answer faster.

KB: I have two more questions for this morning’s discussion. You brought up the idea of continuing or switching to CDK inhibitor past progression on first line CDK, are you in any way incorporating these drugs in later lines of therapy? We have a little bit more experience because the drugs have been available a little bit longer in the United States but are you offering this to patients with an endocrine backbone outside the first line setting if they’ve not seen a CDK inhibitor?

HR: If they haven’t seen a CDK4/6 inhibitor yes. In fact, I have some patients who stayed well on endocrine therapy for a very long time so that they have already gone through, for example, letrozole or fulvestrant and now are coming up on their third line hormone therapy and haven’t seen a CDK4/6 inhibitor. A patient who was on an alpha-specific PI3 kinase inhibitor was CR in liver for four years. So it’s going to be really interesting to see how those drugs work. We are fairly flexible in the US, unlike in many places, where we can give it almost with any agent. So, for example, I could give it with exemestane etc. I have given it in those settings, I’d say the later in line you give a CDK4/6 inhibitor the more cautious you have to be about bone marrow suppression because you can see some more thrombocytopenia in these very later line settings where patients’ bone marrows may not be quite as hardy. In terms of continuing after progression it’s an interesting question. Some insurers cover anything and others don’t in the US and I have to say I’ve continued CDK4/6 after progression in two patients, both were on clinical trials in the first line setting so it made it a little bit easier, but they had very limited progression in bone. Where you think everything else is really controlled and maybe it’s worth it in our patient, they tolerated the drug well, they’re very motivated to continue it, they really wanted it but we don’t have the data and I certainly wouldn’t recommend it. The other thing I don’t like, just in practice that we see a lot, is people are on a hormone drug and they develop progression and the doctor adds a CDK4/6 inhibitor. That I wouldn’t do, I think it’s much better to change the endocrine agent.

KB: Yes, because they’ve, in effect, progressed on their anti-hormonal therapy.

PS: Looking at some of the pre-clinical evidence we have, for a lot of these questions Hope just addressed we haven’t got clinical data. But in our pre-clinical models we see that there may be a small group of patients who could possibly benefit from continuation of a CDK4/6 inhibitor as long as an additional drug is added, for example a PI3 kinase inhibitor, that modifies some of the mechanisms of resistance. The other thing we see pre-clinically is if you look at cross-resistance we see complete cross-resistance between ribociclib and palbociclib but incomplete cross-resistance between either of those drugs and abemaciclib. Again, we haven’t got the clinical data and one needs to be very careful extrapolating pre-clinical data into the clinic but there are two hypotheses that would be worthwhile testing in clinical trials. One is is there a small group where we can continue CDK4/6 inhibitors the same beyond progression and the second is does it make sense to switch class of CDK4/6 inhibitors at progression.

KB: Yes, so even though we have all these large datasets there are still so many other questions that remain unanswered as to how we will incorporate all of these. I’ll finish up this morning’s discussion really just getting everyone’s opinion on the scope of their use of CDK inhibitors and I’ll start with myself and then maybe that will lead to you all giving us your input. At least in my practice I’m incorporating the CDK inhibitors across all first line patients. So if they’ve had aromatase inhibitor I’ll utilise it with fulvestrant; if they’ve not I’ll use it with an aromatase inhibitor. I’ve looked at the data and my perception is that, as you mentioned earlier, there’s not really that one subgroup that sticks out that says ‘A-ha! This group will not benefit from adding the second drug, this group of CDK inhibitors.’ We’ll start with you, Mark. Now that they’re available in the UK and have been available are you using them across the populations of patients?

MV: I would certainly use them and I think that you’ve got to ask yourself the question if it was your relative would you want them to get the CDK4/6? Yes. Is there a group of patients where you don’t need them? I’ve heard the argument that you’ve got people with indolent disease, soft tissue and bone only, but the truth is for those patients while you say, yes, they’re going do very well on letrozole or fulvestrant, they’re going to do even better if you add in a CDK4/6 inhibitor. I’ve yet to meet a patient who says to me, ‘I’d like you to give me the second best treatment.’ You know that if you went out to them on day 10 and did a neutrophil count it would probably be zero and we’d say, ‘Fine, just get in touch if you’ve got a fever.’

KB: I think your point is well taken Mark, it’s that we’re not meticulously checking counts after most of the stuff we do. Who knows what goes on once the patient leaves the clinic? I just want to make certain that we get Peter and Hope’s last thoughts on this.

PS: What I would say is, very clearly from my perspective, every patient with ER positive breast cancer should have a CDK4/6 inhibitor as part of their therapy. There may be a group where it doesn’t have to be the first treatment and what we’ve seen across the trials is these drugs give the same relative benefit whether we give them as the first or as the second endocrine therapy. I do think there may still be, and this is an individual choice, it’s about talking to patients about their two options and you can say, ‘I have a treatment that is clearly more effective but needs slightly more monitoring.’ Whereas if I have a patient who says… some patients may decide, they have very low volume disease, very indolent disease, and ‘I’m happy just to go on an aromatase inhibitor alone. I live 120 miles away from the centre, I don’t want to come up on a monthly basis. I’ll just come up every three months.’ I do think we have seen in the control arm of those trials that single agent endocrine therapy is still very effective for some patients. But, of course, all these patients will unfortunately progress and will, at least later, then need a CDK4/6 inhibitor combination. It’s probably a very small group and probably the majority of patients will say, ‘I don’t want to go for the second best treatment.’ But I do think there is still a small group where we can discuss that choice.

On the other hand I also think it’s important that we are widening our criteria for endocrine therapy with more, for example, substantial visceral involvement. I do think if you’re in an early line a combination of CDK4/6 inhibitors should be considered because, as Mark indicated earlier, the objective response rates and the time to response is very similar to what we see with single agent chemotherapy if we use it in ER positive breast cancer. So there’s broad indication and the bottom line for me is every patient with ER positive breast cancer should get a CDK4/6 inhibitor. The majority will get it early on, there may be a small group who decide jointly with their treating clinician to have it as a second line treatment.

KB: Hope, final thoughts on how you’re using these in everybody, offering it to everybody.

HR: Peter’s description and paradigm is a good one. Most patients in the US will receive a CDK4/6 inhibitor as first line therapy for metastatic disease regardless of the hormone partner. There are, of course, occasional patients that we will all see who maybe isn’t well suited for a CDK4/6 inhibitor in the first line. Or actually a couple of patients, I do live in Northern California, who have refused a CDK4/6 inhibitor thinking it’s too much therapy and they only want to take the endocrine therapy. But they also refuse everything, so they’re kind of unique patients. The one patient in the last couple of years who I haven’t put on a CDK4/6 inhibitor was an older woman who had a 32 year relapse and had taken five years of tamoxifen before and actually it’s difficult that we put her on fulvestrant because I could understand compliance. She had a single site of metastatic disease which went away. So there are those unique patients but most people we’ll give a CDK4/6 inhibitor. I really want to echo what Peter said about the fact that it broadens the group of patients where hopefully internationally we can feel more comfortable by using endocrine therapy when CDK4/6 inhibitors are available. One of my patients who was one of the first patients on PALOMA-2, actually a young woman who relapsed at three years with widespread bone mets on tamoxifen, and has now been on her combination therapy for four years, so longer. We just never saw that before so we can really offer patients who relapse even early on endocrine therapy a very good option for treatment.

KB: So lots of excitement, we love getting new therapeutic agents that work and make our old friend endocrine therapy work even better than it has in the past. I want to thank everyone for joining us and on behalf of ecancer I want to thank the audience for listening in about CDK inhibitors from ESMO 2017.