Overview of CDK Inhibition for Metastatic Breast Cancer

DR LOVE: Welcome to “Individualizing the Selection of First-Line Therapy for Patients with Hormone Receptor-Positive Metastatic Breast Cancer.” This is medical oncologist Dr Neil Love. I met with Dr Joyce O’Shaughnessy from the Sarah Cannon Research Institute in Dallas, Texas, and to begin, I asked her for her global perspective on the clinical and real-life impact of perhaps the most important advance in the field: the use of CDK4/6 inhibitors plus endocrine treatment in metastatic disease.

Now that we’ve had CDK integrated into the first-line metastatic setting for a while, you’re a very astute observer of your patients and what happens to them, I’m kind of curious whether you can actually think you’re seeing the benefit in your patients. Do you think that a typical first-line case that you see nowadays is living longer, living better without disease? Do you think you see the difference compared to let’s say 10 years ago? I guess that’s pre-CDK?

DR O'SHAUGHNESSY: Yes, dramatically, Neil. In terms of the introduction of the CDK4/6 inhibitors in terms of first-line metastatic setting there are 2 things I would say. First, the patients who have an extremely long duration of response, who stay progression free for more than 5 years, that percentage of patients has increased — it used to be the very rare small-volume bone-only or small-volume lung-only patient who had that prolonged progression-free survival, and particularly, Neil, the de novo metastatics.

I would say that it still is the de novo metastatics that are much more likely to go beyond 5 years of progression-free survival, but we just see so many more of them, and they just are getting out to be 7, 8, 9, 10 years. Because we had the CDK4/6 inhibitors introduced in 2015 we’re almost at the 10-year mark, and so we’re having patients go out. I have a patient who brings me roses each year adding 1 for every year she’s NED from her de novo metastatic lung metastasis, and she’s on palbociclib, and she brought me 10 roses most recently.

DR LOVE: Wow.

DR O'SHAUGHNESSY: So we’re seeing that more and more. Number 2 is that patients who are not de novo metastatic, those patients are having progression-free survival much, much longer than they would with single-agent or even the old doublet endocrine therapy of an AI plus fulvestrant. That used to be kind of a go-to regimen in the first line metastatic before we had the CDK4/6 inhibitors. So it does really seem to translate that 50% of patients have progression-free survival longer than 2 years, as is seen in all the first-line trials. So we’re seeing both of it, Neil. It’s a fabulous, fabulous time for patients.

Not all patients respond to all CDK4/6 inhibitor. We still have to suss out who’s going to respond to what, you know what I mean, who’s going to tolerate what, et cetera. So there still is little bit of the art of medicine in there, but we see in practice what we see in the clinical trials.

DR LOVE: Well, that’s great to hear. And a related question is we’ve had in the last years a lot of discussion about the survival benefit, particularly that was seen with ribo, maybe not quite as clearly with abema. Again, do you think you can see the difference? Do you think you’re seeing people stay disease free and live longer with ribo compared to other CDK, or is that something really more buried in the trials?

DR O'SHAUGHNESSY: With regard to whether we can see in the clinic the impact on survival and specifically the differential impact on survival with the various CDK4/6 inhibitors, I don’t think I can, Neil. I can’t tell. I think they’re all effective for different subsets of patients. So of course the data are strongest on survival now for ribociclib, but both abemaciclib and palbociclib also have very important impacts on survival in my practice. Because if patients get into these categories where they’re progression free for 5 years or longer that clearly impacts their survival. So I see survival impact, but I see it across the board, Neil. I can’t really differentiate one from the other in practice.

DR LOVE: That’s what I would have thought.

The other thing I’m kind of curious about, before you kind of get into the slides, because you’re such a great teacher in so many different ways, is whether you could just take a shot without any slides and kind of talk about how you maybe even explain to a first-year fellow how CDK inhibitors actually work biologically, the simplest explanation.

DR O'SHAUGHNESSY: Sure. Well, CDK4 and 6 are proteins in the nucleus that bind together and bind to cyclin D, which is that complex of CDK (cyclin-dependent kinase) 4 and 6 and cyclin D. That’s the complex that bind together and drive the cell cycle. Specifically, they phosphorylate RB, the retinoblastoma protein, and when it’s phosphorylated it falls off the DNA and allows the cell to proliferate.

What’s really important about CDK4 and 6 is that these proteins are in the nucleus, and they’re downstream from all of the proliferative stimuli for breast cancer. There’s many, many ways breast cancers can have proliferative stimulus. The luminal As will get it through the PI3 kinase pathway. The luminal Bs can get it through HER2, FGFR1, c-MYC, a lot of other more aggressive ways, faster ways to drive the cell cycle, but they all impinge on and phosphorylate CDK 4 and 6, which then bind to cyclin D and drive that cell cycle.

So the important thing about them is they are really nodal proteins. They integrate the signal of so many other drivers of the cell cycle, and if you just get those 2 proteins you shut down cell division no matter what it is that’s driving that cell. So they’re super, super important targets.

DR LOVE: So that’s really helpful, and I think that the CDK is just beginning to crash in prostate. I know there was a negative trial at ASCO. I don’t know if they’ve given up on it or not, but it’s just always amazing to me. I always think about prostate. I know it’s totally different than breast, but it is kind of interesting.

DR O'SHAUGHNESSY: Kind of similarities though.

DR LOVE: Yeah. And there I think they’re into more — it looks like the PROTACs are going to be more helpful, where I guess that’s also something that looks good in breast too.

DR O'SHAUGHNESSY: Yes. It sure does. It looks very good in breast, yes. Interesting. That’s very interesting that they wouldn’t work there. I’d like to see the data on that and see which patient populations, et cetera, in prostate they looked at.

DR LOVE: Yeah. It was an abema study. I think there’s another one that’s cooking as well.

Anyhow, let’s take a look at your slides and go through some of the data. Whenever you’re ready you can go ahead and get started.

Features of High-Risk, HR-Positive, HER2-Negative Localized Breast Cancer

DR O'SHAUGHNESSY: Okay. Well, what I’m going to focus on here today is how I select first-line therapy for hormone receptor-positive, HER2-negative metastatic breast cancer patients. I’m going to talk a bit about risk, as well, that is, who develops first-line metastatic breast cancer who needs therapy, who are the patients that really are at risk for developing metastatic disease.

So I’ve listed here some of the characteristics of high-risk early breast cancer that’s HR-positive, HER2-negative, and one of the key features of this high-risk disease is that they tend not to be endocrine therapy-sensitive in the adjuvant setting. They can either be completely resistant to endocrine therapy, or they can basically be initially sensitive and then remodel, relatively rapidly remodel so that they can drive cell division without dependence on hormonal therapy.

So some of the characteristics of the cancers that lead to more likelihood of developing metastatic disease are large tumor size, nodal status, and high grade, and these all impact outcome for sure. Now, being a little bit more specific, it turns out that higher proliferation basically predicts for less sensitivity to endocrine therapy and very, very importantly, perhaps most importantly of all, is lower estrogen receptor levels, which I’ll show you here in a moment in more detail.

Though we don’t get the PAM50 analyses done routinely on our early-stage patients, if we order MammaPrint^® we will get the BluePrint^® as well. And we know that luminal B breast cancers, HER2 enriched and basal-like ER-positive breast cancers — now HER2-negative, not amplified, HER-positive, if they’re luminal B or HER2 enriched or basal like, they are more likely to be endocrine therapy resistant.

The other thing we can tell, if we have a poor prognosis in early breast cancer, is before that patient goes to her definitive surgery it we give her a window treatment prior to therapy of 2, 3 weeks of endocrine therapy, and if that ki-67 does not suppress to less than 10% within 2, 3 weeks of beginning endocrine therapy, that is a poor prognostic sign, and it means that the cancer is less sensitive to endocrine therapy. And basically what defines our most challenging and still today lethal HR-positive, HER2-negative breast cancers is genomic instability that leads to multiple subclones, clonal heterogeneity. It’s much easier for the cancer to develop endocrine therapy resistance because there’s so many different subclones because this leads to the therapeutic resistance virulence, which means high proliferation and metastagenicity. They all go together. So it’s this genomic instability that drives all of this endocrine therapy resistant biology, and these are the patients that develop that metastatic breast cancer.

We still have a long way to go. These are data that go back to 2011, and you can see on the left the risk of recurrence over 15 years. And this is just with 5 years of tamoxifen. Half of these patients had had chemotherapy, 44% were node positive. And so we know that the use of an aromatase inhibitor, for example, or 10 years of endocrine therapy rather than 5 years, will improve this recurrence-free survival somewhat; let’s say another 5%. But we still are looking at about a 30% risk of recurrence with optimal chemotherapy and optimal endocrine therapy prior to the days of the introduction of CDK4/6 inhibitors into the adjuvant setting, so that’s still about close to one third of our patients developing metastatic disease who are HR-positive, HER2-negative.

This is just another paper that looked at some of these risk factors for developing metastatic disease. And you can see in the bottom right, for example, grade is really, really an important driver. And we can see on the bottom left, again, that size really matters. If patients are getting into a T3 or a T4 HR-positive, HER2-negative breast cancer they have a much, much higher risk.

And then the top left, it’s interesting. The HER2-positive and the triple-negative patients have a much worse prognosis very, very quickly, within the first couple of years. And you can see the darker line, which is the HR-positive, and it’s just kind of slowly trending up there, continuing to go up at 10 years. And if we just keep going and follow these curves out 20 years eventually those 3 curves will meet.

The HR-positive, HER2-negative breast cancers, they just continue to recur year after year. They’re just slower. It just takes them longer to get to the same 30% risk of recurrence that we see with other breast cancers as well. So clearly we have more work to do. We still have way too many of our patients who are destined to recur with metastatic disease.

Another way we can identify patients at very, very high risk of recurrence are those who don’t have a pathologic complete response with preoperative therapy. Heretofore it’s been chemotherapy, but increasing it can be endocrine therapy with a CDK4/6 inhibitor in practice. And the more residual disease a patient has the higher her risk of recurrence. We can see here in the light gray, patients who don’t have a path CR, about 25% of those have had a recurrence event at 5 years, and it continues. Their prognosis continues to worsen over time. So we know that not being responsive to initial therapy also predicts for a poor prognosis.

And these data here from the 21-gene recurrence score, the Oncotype, is very, very insightful in terms of the biology of breast cancer. This is node-negative, HR-positive breast cancer, the NSABP B-14. And these patients shown here have all had tamoxifen for 5 years, no chemotherapy. And we can see over on the right the high-risk recurrence score patients. They’re not sensitive at all to the tamoxifen. Their risk of recurrence is about 30% in the first 5 years or so. And then we can see intermediate risk, and then the low-risk patients do very well. Their breast cancers are very sensitive to tamoxifen.

So the question is what’s really driving this recurrence score, this high risk versus low risk. And you can see in the bottom left the — in the red circle, the strength of ER signaling in these breast cancers. This is the estrogen module within the recurrence score, it’s the most important by far, and you can see that the lower recurrence scores have the highest ER signaling and the high recurrence scores have the lowest ER signaling. So on the top left that the proliferation module is not linear. You see how linear it is with the estrogen module. It’s not linear. It is not proliferation that’s primarily driving the recurrence score, it’s strength of ER signaling.

So in short, the recurrence score tells us the degree to which that breast cancer is going to be endocrine therapy sensitive or not. Very, very practical. Very, very useful insight, but it’s all about how much the estrogen receptor is in charge of that breast cancer.

And then I wanted to lastly mention another way we can tell who’s really at risk for recurrence and whose breast cancer is very unlikely to be endocrine therapy sensitive, and that is the newly defined MammaPrint ultrahigh risk, the MammaPrint high 2, which is this upper 50 percentile of the high-risk MammaPrint. We’ve got the upper 50 percentile that’s high 2, the lower 50 percentile of the high risk, that’s high 1.

And it turns out they’re a very, very different biology, which is just absolutely fascinating. The MammaPrint high 2’s are very similar to the basal-like ER-positive breast cancers. And in this I-SPY trial that we’re looking at here by Lajos Pusztai, which has been published, we’re looking at the control arm was preoperative weekly paclitaxel followed by AC. And over on the left we can see in the MammaPrint high 1 the path CR rate with paclitaxel followed by AC is 10%.

And when you add the checkpoint inhibitor durvalumab plus olaparib, and these patients were not germline BRCA patients, this was testing the hypothesis of the DNA damaging effects of olaparib, we can see that adding olaparib and durvalumab didn’t make any difference at all in the MammaPrint high 1. So these are high-risk patients, but it’s the lower 50 percentile, conversely on the right with the high 2 patients, we see that the control group has a higher path CR rate. Weekly paclitaxel followed by AC we have a 22% path CR rate in the high 2s, but now adding olaparib and durvalumab it looks like triple-negative breast cancer. We’ve got a 64% path CR rate, very similar to the KEYNOTE-522 regimen.

So this is all HR-positive, HER2-negative breast cancer, but we have such a spectrum of biology, to those that really are very, very indolent luminal A, all about the estrogen receptor, all the way up to MammaPrint high 2 basal-like that are not dependent at all on the estrogen receptor. So it’s quite the large spectrum of biology. Nonetheless, these patients all develop metastatic disease, the high risk 1s, and we have to figure out how to treat them in the first-line setting.

DR LOVE: Actually one of the things I’m starting to see papers about, kind of surprisingly, that I’ve been starting to think about, is so-called ER-low.

DR O'SHAUGHNESSY: Yes. Yes.

DR LOVE: It’s just kind of like what you’re starting to get at. And the thing that I’ve been thinking about is these ER-low patients. If in fact you view them as triple-negative the treatment nowadays is very different, right?

DR O'SHAUGHNESSY: Yes. Yes. That’s right, Neil.

DR LOVE: So are there patients in here who we’re calling ER-positive but really I think that’s kind of the — what you’re saying, but I’m putting it another way, is ER-low.

DR O'SHAUGHNESSY: Right. Right. Yes. Historically, Neil, the ER-low patients were excluded from clinical trials focusing on triple-negative breast cancer patients, such as the preoperative KEYNOTE-522 with pembrolizumab, which of course is the standard of care, and the ER-low patients were included in clinical trials of estrogen receptor-positive patients. Now we know it’s a very, very different biology. We now have 3 data sets that tell us that ER-low patients have the same path CR rates with preoperative chemotherapy and checkpoint inhibitor as do triple-negative patients.

DR LOVE: Wow.

DR O'SHAUGHNESSY: So we now have much — yeah, 3 data sets that show us that ER-low looks like triple-negative.

DR LOVE: So it’s one thing to talk about MammaPrint, et cetera, et cetera, but for a doc in practice can you just look at the ER level and see if it’s less than 10% to raise a flag?

DR O'SHAUGHNESSY: Yes, for sure. Yeah. If it’s less than 10% then that patient really should be treated as a triple-negative patient. I think I can very confidently say that now, Neil, that if it’s 1% to 9% ER positivity that patient, if it’s a Stage II or III, should get the KEYNOTE-522 regimen or the neoPACT regimen if they’re not an anthracycline candidate. But we have the KEYNOTE-756 with preoperative pembrolizumab plus paclitaxel followed by AC.

We have the CheckMate 7FL, which was nivolumab added to preoperative chemotherapy, and we have the neoPACT trial of docetaxel, carboplatin and pembrolizumab. All 3 of those trials included ER-low patients, and they had the same path CR rate as triple-negative patients, whereas those that were more strongly ER-positive had a much less path CR rate than the ER-low, which had the path CR rate that you’d expect from a triple-negative population. So those are the 3 data sets, Neil, that really tell us if ER is less than 10% we should treat that patient as a triple-negative patient.

DR LOVE: So I’m trying to decide if I’m the last person to hear this. Is this like in the guidelines at all?

DR O'SHAUGHNESSY: I don’t know if the guidelines have changed yet. I know for sure the FDA label has not changed yet for preoperative pembrolizumab, for example. There may be a footnote in NCCN to this effect, Neil, probably so. I don’t know whether it’s a firm recommendation, but I think that’s really well worth knowing because the issue that you’re raising is what about insurance coverage for this. Sometimes we do get pushback, Neil, but we have these 3 data sets now that we can say, “Hey, look here. Look at these path CR rates. This is like triple-negative breast cancer.” But there may be a footnote to this effect in the guidelines.

DR LOVE: Because I mean you don’t see this kind of path CR rates with ER-positive disease.

DR O'SHAUGHNESSY: No, not even close. In the KEYNOTE-756, for example, the path CR rate with weekly paclitaxel followed by AC with the pembrolizumab added was in the mid 20% range. So that’s about what you get with our best chemotherapy, as well as checkpoint inhibitor. That was Grade 3 disease as a surrogate for chemotherapy and IO therapy sensitivity. But we don’t. This is just remarkable, 64%. Spot on exactly what we saw in KEYNOTE-522.

DR LOVE: Well, I know what one of our pre/post test questions is going to be for this program. But anyhow, let’s keep going.

DR O'SHAUGHNESSY: Good. Good. No. That’s a very, very good point.

DR LOVE: Woah. Yeah, scary. That’s scary.

DR O'SHAUGHNESSY: Yes.

DR LOVE: That’s a big difference. Big difference.

DR O'SHAUGHNESSY: It is. It’s huge. It’s huge. Absolutely huge.

DR LOVE: I mean chemo/IO. That’s a big difference.

DR O'SHAUGHNESSY: Right. It’s a totally different therapeutic pathway, absolutely.

DR LOVE: Alright. Let’s keep going.

Sequencing and Selection of CDK4/6 Inhibitors

DR O'SHAUGHNESSY: So with that, we’ve talked a bit about risk and which of our early-stage patients are at risk for developing metastatic disease, and of course we always want to biopsy those patients and document their breast cancer remains ER-positive, HER2-negative. Or we still have about 20, maybe 30% of our patients in the first-line setting who will be de novo metastatic patients, and of course we biopsy them as well and ascertain their ER and HER2 status.

So now we have to make a recommendation to these patients about their first-line therapy, and about 90-plus percent of patients are really excellent candidates for a CDK4/6 inhibitor in combination with endocrine therapy, almost always an aromatase inhibitor unless patients have recurred on adjuvant endocrine therapy or within a year of finishing their adjuvant AI therapy.

And as we know, there are 3 CDK4/6 inhibitors, palbociclib, ribociclib and abemaciclib. And in the red box on the top we can see a summary of these data, shown here, which is that all of these inhibit the CDK4/6 and cyclin D1 complex. Ribociclib and abemaciclib are 4 and 5 times more selective towards CDK4 over CDK6, and CDK4 is the main proliferative driver of ER-positive breast cancer.

Abemaciclib, in addition to inhibiting CDK4 and 6, which palbociclib and ribociclib inhibit in a very focused targeted way, abemaciclib also inhibits CDK1 to a small extent, CDK2 to some extent and CDK9 to a large extent. So abemaciclib has a broader mechanism of action, whereas palbociclib and ribociclib are very, very targeted towards CDK4 and 6. So they are somewhat different entities for sure.

However, they’ve all been tested now in multiple first-line metastatic trials, and they’re summarized here with the PALOMA trials being palbociclib, the MONALEESA trials being ribociclib, and the first-line monarch 3 being abemaciclib, and all in combination with an aromatase inhibitor except for the MONALEESA-3, which was first line with fulvestrant. And in the red box we can see the hazard ratios for progression-free survival, and it’s about a 50% reduction in the risk of progression with these agents. All the same across the board.

And down in the very bottom you can actually see progression-free survival results, and it’s about 10 to 12 months of progression-free survival that the CDK4/6 inhibitors are adding on average to the addition of endocrine therapy. But there’s a big, long tail on the curve with some patients going out many, many years, as we’ll see in a second.

So with regard to progression-free survival these agents are all very, very similar. The actual numbers of PFS just really reflect some variation in the patient characteristics in these first-line trials, the patient population. There’s some variability among the trials, and that’s why the numbers differ a little bit between the agents.

Now, with regard to overall survival this slide looks at all of the Phase II and III, large Phase IIs, the PALOMA.

These are all the Phase III trials with the CDK4/6 inhibitor, and we see with regard to hazard ratios that the only one that’s really kind of right on 1, if you will, so really no benefit, is the top one, the PALOMA 2, which is the first-line palbociclib trial. The rest of the point estimates are all kind of convincingly to the left of 1, so benefit. Some of them just sneak over the line, the dotted 1 line, with regard to statistical significance. But nonetheless, the point estimates are all in the right direction.

I do want to point out down at the bottom the 3 MONALEESA trials. All of those trials in the first-line setting, in MONALEESA-2 the postmenopausal, MONALEESA-7 the premenopausal, and MONALEESA-3 with fulvestrant, all of those point estimates of about 0.7 are highly statistically significant. And the 2 palbociclib trials are not quite statistically significant.

The MONARCH 3 trial is not quite statistically significant, but we’re going to walk through some of the most important trials here. Let’s look first at the MONALEESA-2 in postmenopausal patients, an aromatase inhibitor plus/minus ribociclib, and we see about a 12-month improvement in overall survival in favor of ribociclib, and that is statistically significant. We can notice going out here about 7-plus years we still have 35% of patients progression free with the ribociclib.

Now this is the PALOMA-2, the first-line letrozole plus/minus palbociclib, and in the intent to treat population over on the left there’s no benefit, but this was a study that had a very long follow up, 7 and a half years median. Survival data were missing in 21% of the control patients and 13% of the palbociclib patients. 27% of patients on placebo later crossed over to get a CDK4/6 inhibitor. But we’re kind of concerned that the missing data, as follow up went on and on and on with this trial, may have impacted this trial’s results. But nonetheless it is a negative trial for us in practice today.

Now that’s in contrast to the real-world evidence with palbociclib/AI first line versus AI first line. Now of course real-world evidence goes to EMRs, it’s not randomized, of course, but you can still adjust for the key prognostic variables in the first-line setting. And you can see over on the far right that’s propensity score matching. This is a way to take the raw data with regard to survival with first-line AI versus first-line AI plus palbociclib and statistically adjust for differences in prognostic variables within those 2 groups of patients.

And you can see when you do that you actually do have an improvement in survival. This is a 13-month improvement in survival and is statistically significant. So the real-world data with regard to palbociclib suggests a survival improvement. So this is important because I still — I do use palbociclib if I have a patient who has a history of heart issues, for example, and she’s got comorbidities, I really can’t give her any medication like abemaciclib that would cause diarrhea, she’s asymptomatic, she’s got small-volume metastatic disease, lots of comorbidities, that’s the patient I recommend palbociclib for. I usually use mostly ribociclib based on the 3 MONALEESA studies in the first-line setting, but I do use palbociclib for the patient that I just described.

And then we have abemaciclib in the first line. This is the MONARCH 3 trial, AI plus/minus the abemaciclib, and the final overall survival data are shown here. There is a 13-month improvement in median overall survival. The hazard ratio was 0.8 and the p-value 0.06. So this trial just misses statistical significance, but I am personally not concerned at all about this. I think this is a numbers issues if you will. There were 493 patients in the MONARCH 3 trial, and there were 666 patients in the MONALEESA-2 trial and the PALOMA-2 trial.

So this trial was just simply underpowered for overall survival, and I think that had it had more patients and more events we would have had a statistically significant result. So to me this 13-month improvement in overall survival is really what drives my practice. And we can also see here going out almost 10 years that we have 30% of patients still alive with the abemaciclib.

In patients with visceral disease, and I tend to use the abemaciclib in those with visceral disease or very, very virulent and destructive bone disease, we have a 14.9-month improvement in median survival in patients with visceral metastasis. We can see the p-value again there borderline at 0.07.

The final progression-free survival data in MONARCH 3 are shown here, we have 29 months median progression-free survival versus 14.8 months, and that is a 14-month difference in progression-free survival. And going out here about 8 years we have about 20% of patients still progression free, so about 1 in 5 patients going out progression free at about 8 years. So these are very good data. We can see what a nice advance we’ve made over an aromatase inhibitor alone. And so this is what we see makes a big, big difference. We can see this in our practice, patients doing much, much better for a much longer period of time, and then we have a subset that go out many, many, many years progression free on the CDK4/6 inhibitors.

Now both the MONARCH 2 study, which was the second line with fulvestrant, and the MONARCH 3, the first line with an AI, had a prespecified evaluation of subsets. And we’re looking here in the dotted red box of patients with more poor prognostic characteristics, such as liver metastasis, progesterone receptor negative, high tumor grade and a short treatment-free interval from their adjuvant therapy. And we can see that with regard to abemaciclib there’s really convincingly no difference in progression-free survival in those patients with regard to their benefit from abemaciclib if they had these poor prognostic characteristics compared to the more favorable prognostic characteristics.

And so these data here inform my personal choice of abemaciclib in the metastatic setting; patients with concerning, rapidly progressive liver metastasis, PR negativity, high tumor grade, highly proliferative or those patients that did not really benefit from their adjuvant endocrine therapy, they’ve recurred on adjuvant endocrine therapy or within a year of stopping their adjuvant endocrine therapy. This is the clinical manifestation of endocrine therapy resistance. That’s where I choose the abemaciclib. I believe the data show that the more unlikely it is for patients to benefit in the metastatic setting from endocrine therapy that’s where they benefit from the abemaciclib, compared to the palbociclib and the ribociclib, where I believe those patients do best with those 2 agents if their breast cancer is more endocrine therapy sensitive.

DR LOVE: Could I just get you to elaborate on that a little bit more because one of the things that I’m starting to realize, the issue in the second-line setting, because we’re doing another abstract for San Antonio on that, and there we see such a difference in how people manage patients based on how long they’re on CDKi before going to second-line therapy? And it looks like you’re kind of talking about a similar thing here. Now this is only people who’ve had adjuvant therapy, correct?

DR O'SHAUGHNESSY: Yes. The treatment-free interval part, yes, but not those with liver mets or PR and tumor grade. That’s all —

DR LOVE: Yeah. I know. Right. But what I’m talking about is the treatment-free interval, and what they look at is 36 months.

DR O'SHAUGHNESSY: Yes.

DR LOVE: So in other words, for example, someone who’s been on an AI adjuvantly 36 months or more versus they relapse within 36 months, that’s the way they looked at it here, and they didn’t see as much benefit. I’m just kind of curious how you look at time on adjuvant therapy as a predictor. Are you focused more on people who relapse within 3 years, within 1 year? What’s the breakpoint clinically for you?

DR O'SHAUGHNESSY: And it’s a very important question, Neil, because we’ve known for decades that the number 1 most important prognostic factor for patients who develop metastatic disease is what their disease-free interval or treatment-free interval is. That dictates more than anything what their prognosis is.

And in MONARCH 3 here treatment-free interval means how long was it that they were off of treatment. And so these are patients who could have recurred on their adjuvant endocrine therapy, so they’d be less than 36 months, they could have recurred 1 year after finishing their adjuvant endocrine therapy, but treatment-free interval less than 3 years, meaning they recurred on it or within 3 years off of it, and then more than 36 months means they were off of their adjuvant endocrine therapy for at least 3 years. So this is actually treatment-free interval not disease-free interval.

But I think in terms of the standard in practice, I think it’s what — we utilize what has been used in the first versus second-line CDK4/6 inhibitor trials. If they had recurred on adjuvant endocrine therapy or within a year of finishing their adjuvant endocrine therapy, we consider those patients endocrine therapy resistant, and we usually treat them with fulvestrant/CDK4/6 inhibitor. If they were off their adjuvant endocrine therapy more than a year then we consider them somewhat sensitive to endocrine therapy, and we usually will go back to an AI first line. So it’s basically 5 years plus a year, Neil, is kind of what we mainly look at.

If someone’s recurring on their adjuvant endocrine therapy, kind of like by definition they have developed endocrine therapy-resistant disease. ESMO differentiates primary endocrine therapy resistant, which means the patient got zero benefit from endocrine therapy, that’s recurrence within 2 years of starting your adjuvant endocrine therapy, and then they call secondary endocrine therapy resistance is if they recur more than 2 years on their adjuvant endocrine therapy but within 5 years. And then they actually call it endocrine therapy sensitivity if patients recur after they stop their adjuvant endocrine therapy. So that’s how ESMO breaks it down. So again, resistance for them is recurrence on adjuvant endocrine therapy. That’s kind of what we do in practice.

DR LOVE: So I guess the reason I’m bringing this up again is I notice in the second line in some situations people will skip second-line endocrine therapy if the patient has been on CDK a very short period of time. My question to you is if you see a patient who’s been on let’s say an adjuvant AI less than a year will you consider not using first-line endocrine therapy and going to chemo, particularly if the patient’s having symptoms?

DR O'SHAUGHNESSY: So for those patients who recur within a year or 2 of starting their adjuvant endocrine therapy, obviously they have endocrine therapy-resistant disease, the strongest variety, I will almost always recommend a CDK4/6 inhibitor for those patients. That’s clearly where I would recommend the abemaciclib for those patients because of its broad mechanism of action and the fact that you don’t have to take a week off. That’s very virulent, aggressive disease, and so I want to just not take a week off. I want to treat them continuously. But I rarely, rarely would choose chemotherapy for such a patient.

The only time I would recommend chemotherapy, Neil, and this is going down and down as years go by, if I have a patient who has — really she’s right on the borderline of complete respiratory failure, lymphangitic spread, highly oxygen dependent, in the hospital, or someone who comes in with small bowel obstruction because she’s got big time peritoneal disease, or someone who has complete diffuse liver metastases, and the bilirubin is rising, those are the only times that I actually recommend chemotherapy. I know how to dose the chemotherapy. I tend to utilize careful doses of paclitaxel/gemcitabine for those patients because I’ve learned how to do it over the years, day 1, day 8 on a 21-day cycle. So I’ve learned how to titrate that, et cetera. So that’s the only places I’m still using chemotherapy.

Neil, the only reason is because I don’t have enough experience using the CDK4/6 inhibitors in those particular situations. But they probably would work, particularly abemaciclib. But sometimes with bowel obstruction, for example, patients can’t reliably keep down oral agents. But I tend to use chemotherapy just because of my experience with that.

DR LOVE: It’s really interesting what you’re saying because I’m not sure we looked at that carefully enough in the survey because I wasn’t really thinking about it. But if I hear what you’re saying correctly, I mean I know that there are some docs who are going to see this in a survey, I think you’re one of them, who go to abema for liver mets and visceral disease, but the idea of thinking more about abema in somebody who has a short disease-free survival on endocrine therapy, I’m not sure that we’ve teased that out, and I’m not sure that message is — not that everybody necessarily would agree, but even that thought I’m not sure oncologists have in their head right now. So I’m glad that we can bring it to them.

DR O'SHAUGHNESSY: It’s funny. My colleagues in Texas Oncology, and I would even say our US Oncology Breast Committee, which is now the Sarah Cannon Research Institute Breast Committee, we’ve done a lot of work with the clinical trials in abemaciclib over the years, and I would say particularly my Texas Oncology partners, who I’m in relatively close contact with many of them, abemaciclib has become a go-to agent, Neil, kind of in general. I think that docs feel like they’re not confident picking out which patients are going to really benefit from ribo or palbo. They know that they can use abemaciclib for the more endocrine therapy-resistant cancers. I don’t think they necessarily feel that confident that they can tell one from another. And so I find that more of my Texas Oncology partners are just utilizing abemaciclib.

I do think the ribociclib survival data has made a difference in that, though. I think many docs are looking at that ribociclib survival data and are switching some patients over to ribociclib, but I find that docs do tend to utilize abemaciclib for the more endocrine therapy-resistant patients. But I do think there’s variability. I do think there is.

But it’s interesting, of course, with the postMONARCH data now. I think we’ve all had experiences where we start patients on palbociclib or ribociclib and they really don’t benefit convincingly, and then I’ve switched them to abemaciclib, and they’ve had very durable responses. So I think that docs will do that, as well, if they start with ribociclib, and patients don’t have a great response, either they progress or they don’t have a good response, just kind of stable disease, a lot of times docs will switch them over to abemaciclib. And this has been going on even before we got the post-MONARCH data.

DR LOVE: Yeah. It’s really interesting how the debate goes back and forth when you get into all these details. And then we had a video of one of the general medical oncologists at ASCO who was saying most of my patients are in their 80s, he’s in South Florida, he says I use ribo in younger people, but these older people, they live for years with palbo, I just give them palbo, and they’re happy. But I guess for patients, particularly younger patients who want to squeeze every possible benefit out, we get into some of these details. But there’s the other side, which is — I mean, and I’m curious, too, because he was like it’s just so much easier to use palbo. Do you agree with that, or do you think that if you know what you’re doing ribo and abema are equally easy to use?

Management of Toxicities Associated with CDK Inhibitors and Promoting Patient Adherence

DR O'SHAUGHNESSY: Yeah, no. I think that increasingly these agents we’ve learned how to use all of them better. Palbociclib is very well tolerated in terms of how patients feel, no question. We do have more myelosuppression, neutropenia, with the palbociclib. And it depends on how comfortable for an individual patient physicians are in terms of tolerating chronic Grade 3 neutropenia.

For my younger patients who have started on palbo many years ago I usually tolerate the Grade 3 neutropenia, and these patients have not gotten into problems with infection. But many docs are not comfortable with the chronic Grade 3 neutropenia, ANCs of 6, 700, and so they dose-reduce. Sometimes dose reduction does not make a difference for the neutropenia, and so then docs get down to 75. So it can be really quite challenging to manage the neutropenia.

For the most part patients do well on 100 or 125 of the palbociclib. With regard to ribociclib, again we can run into neutropenia, or we can run into some liver function abnormalities, but dose reduction from 600 to 400 does decrease the neutropenia, not necessarily the liver function abnormalities. And many of us, Neil, if we run into Grade 2 liver function abnormalities, certainly Grade 3 LFT abnormalities, we will not rechallenge with ribociclib. We will switch over to a different CDK4/6 inhibitor.

Now, historically abemaciclib has been the most difficult of CDK4/6 inhibitors with regard to toxicity because of the diarrhea. But what’s really interesting, as I’ll show here in a moment, we know we can dose reduce in the metastatic setting, in the adjuvant setting, with abemaciclib, and patients do very, very well. So thankfully we have great data that we can have a very low threshold to decrease from the 150 mg b.i.d. to the 100 mg BID and not lose efficacy, and thankfully it’s a night and day difference in terms of tolerability.

And many physicians now are starting, in patients they are at all concerned about their ability to tolerate the 150, they’re starting them at 100 very, very commonly now, for any kind of questions because they can always go up. So actually, Neil, like everything else in our world, we’ve figured out how to manage the toxicities. Oncologists are good at that.

DR LOVE: I agree. Please continue.

DR O'SHAUGHNESSY: So getting to your question here, Neil, of first-line chemotherapy, does it have any advantage over first-line CDK4/6 inhibitor therapy, I think this trial, the RIGHT Choice trial, really helps us understand the answer to that question. This is a randomized Phase II trial in premenopausal patients, a global trial where 62% of these young women were de novo metastatic and 66% had symptomatic visceral disease, so a very, very aggressive, virulent group of cancers here, first-line metastatic. And they were randomized to full-dose ribociclib plus an aromatase inhibitor plus goserelin versus investigator’s choice of combination chemotherapy, docetaxel or paclitaxel with capecitabine or gemcitabine or vinorelbine and capecitabine, and progression-free survival was the primary endpoint.

And you can see the curves there. There was a doubling of median progression-free survival, it was 21.8 months versus 12.8 months with chemotherapy, so hazard ratio of 0.6. So very substantial superiority of PFS with the CDK4/6 inhibitor, even in this group of most virulent cancers. And we can see that the curves are on top of each other for the first couple of months, so it’s not like even early on when you’re really striving to get control of that disease in the first 2, 3 months, there’s no difference. And then the curves split, and it’s in favor of the CDK4/6 inhibitor.

So I’m hard pressed here with regard to the patients eligible for this trial. Then their bilirubin did have to be less than or equal to 1.5 times the upper limits of normal. So these weren’t patients with bilirubins of 5, but this still was a very sick group of patients, and the CDK4/6 inhibitor was superior to chemotherapy. So I think a real message here for our practice.

Now what about toxicity? We’ve touched on this a bit. We have Grade 3 and sometimes rarely Grade 4 neutropenia with ribociclib and palbociclib that leads to dose reduction. Abemaciclib is much more rarely decreased because of neutropenia. We have the uncommon QTc prolongation with ribociclib, but dose reduction from 600 to 400 takes care of that. It completely irradicates it.

We do have LFT abnormalities with both ribociclib and abemaciclib, a little bit higher with the ribociclib, but it is something, as I mentioned, that if someone has Grade 2 or 3 LFT abnormalities with ribociclib generally we would not rechallenge because it’s not really dose dependent. With abemaciclib we’re more likely to see Grade 2 liver function abnormalities, and if we wait until resolution and then dose reduce usually it does not recur, at least in my experience.

Alopecia is all Grade 1. This is just a low-grade alopecia. And then with regard to VTEs and ILD, these are class effects. All of these agents have a low risk of a VTE or ILD. We simply have to be aware that these can occur. So if patients have unexplained cough, fever, shortness of breath, we have to be aware that ILD can very rarely occur with the CDK4/6 inhibitors. Fortunately, very, very unlikely to be high grade or fatal.

So these are the toxicities we’re all well aware of, except I didn’t mention the diarrhea. Diarrhea is the most common toxicity of abemaciclib, as we’re all aware.

So the key to managing these toxicities is patient education and seeing the patient every 2 weeks for the first 2 months. That’s where we manage all the toxicities. That’s where we get the dose correct for whichever CDK4/6 inhibitor the patient is on. We make our modifications within the first 2 months, depending on toxicity. So diarrhea in particular with abemaciclib, about 40% of patients will ultimately need a dose reduction, mostly for diarrhea. We watch the LFTs. If they’re not abnormal within the first 2 months they will not become abnormal. We watch for neutropenia. And then VTE and ILD are very rare. That we watch over the entire course that the patient is on the therapy. But the key here is this careful monitoring for the first 2 months. If we do that our patients do very, very well.

Now, this slide here shows for both MONARCH 2 and MONARCH 3, the first and second-line Phase III trials with abemaciclib with endocrine therapy, that for patients in MONARCH 2 who had 150 versus 100 or 150 versus 50 mg of abemaciclib, there is no difference here in progression-free survival. And also for MONARCH 3 first line, 150, 100 and 50 all about the same with regard to progression-free survival. So we have very good data in the metastatic setting that if we need to go down on the dose for abemaciclib patients are not going to lose that progression-free survival advantage.

We have the same data shown here for the MONALEESA trials, all 3 of the first-line trials. The blue PFS curve on the top is actually patients who had at least a 30% reduction in their dose. So these are patients who had 1 or 2 reductions in their ribociclib and then the other 2 curves are those who did not have a dose reduction or who had 1 dose reduction. And we can see that if anything patients with 1 or more dose reductions did the best because the key here, of course, is to keep the patients on the therapy. If patients can stay on the therapy, then they’re not going to lose their benefit, even if they need a dose reduction, and we can continue with their therapy, and then they can fully benefit.

And lastly, Neil, becomes the issue of adherence. This is a big, big issue. I think it’s less so of an issue in the metastatic setting. And the bottom line is if we meet with our patients, we simplify, we write everything down, if we make any modifications to their regimen we’re writing it down for them, we’re really educating patients about when to stop the therapy and when to call us. I find this is very, very important with abemaciclib because a very small percentage of patients will get Grade 3 diarrhea. I really have to educate patients about not taking their next pill if they’re getting into significant diarrhea.

It's very important to listen carefully to patients’ beliefs about their tolerance to therapy. If patients are very, very concerned that they tend to have toxicities, then I think it’s important to potentially start at a lower dose with those patients. Very important to have the family involved, as well as the patient. And I actually think reaching out to patients, if they come in for a visit, they’ve been having a hard time, we make some modifications, I think it’s important for them to get a phone call within a week, before they come back 2 weeks later for their follow-up.

And I think it’s really just then evaluating it, really asking the patients carefully. And I ask patients to keep track of side effects, write it down, if they hold a dose or miss a dose to write it down so we can really look at what they’ve been able to tolerate since they were here last.

But I really, really think that seeing patients every 2 weeks for the first 2 months has made a huge difference in finding the right dose for a patient and ameliorating the toxicities so they’re able to basically adhere and benefit from the CDK4/6 inhibitors.

DR LOVE: So just a couple of follow up questions. I was just curious. You were talking about adherence, which is something we’ve been very interested in. Any tricks or approaches to situations where people don’t take drugs continuously? So obviously, you have palbo and ribo, but also you’ve got capi now, this 4 day on, 3 day off thing. When people have to stop and start does that cause more adherence problems?

DR O'SHAUGHNESSY: Neil, there is a subset of our patients, thankfully a small subset, where complex regimens are very, very difficult for them.

I’m thinking of a patient I had. I had her on abemaciclib. It’s BID, and it’s continuous. It was very difficult for her to consistently remember to take 2 pills a day. I think there was just a lot going on in her life. She had multiple other comorbidities. She was taking multiple other medications. She was an older woman. There was a lot on her plate otherwise, outside of her metastatic breast cancer. And we talked and we talked, and we had — her family came in, et cetera, but there — it just was very — it was just very challenging, Neil.

And I think she passed away before capivasertib became available. She would not have been a good candidate for capivasertib. I just don’t think there’s any way she could have done the 4 days on, 3 days off.

Now that is unusual, although I’m thinking of another of my patients who’s also an older woman. She and her husband come in very religiously. They’re very faithful at showing up, but they never bring her list of medications with her. I always ask, and we do our best to go down the list, and it’s just so funny. It’s just the same thing all the time. It’s just very confusing to really find out what she’s on consistently, et cetera, and they’re doing the best they can. I think they both have just very slight early dementia, and it’s just difficult, and so we have to keep things as simple as possible.

And then my nurse will call and say, “Go get your medications, and let’s go over them, et cetera.” But it really is very complex for particularly older people, Neil, with multiple medications, polypharmacy that they’re on. And then what happens, we’re always changing things up. We’re changing things. It gets too confusing for patients, but we try to keep it as simple as possible for those patients.

DR LOVE: So I can say I completely relate personally to that challenge, but also, I think it’s one thing if you have somebody like you’re describing with a real adherence problem, but I guess what I’m really thinking about is for a much more common situation, where, I don’t know, maybe people are missing 10% or 20%, 10% of their medications, for example. If it’s in the adjuvant versus metastatic setting does that bother you one way or the other.

DR O'SHAUGHNESSY: Yes. It all bothers me because I think that patients should take what was proven to work for them. I’m not worried about dose reduction if patients are really having toxicities. I think we have really good data for CDK4/6 inhibitors, both in the adjuvant and the metastatic setting. If somebody’s having toxicity, we’re okay. We need to go down. We need to keep them on the medication.

But if people are missing 10% or 20% just because they haven’t really established a consistent system to enable them to take all the medications, I do worry about that because I’m worried it may be a little more than the 10% to 20% that they’re even aware of. Maybe they’re missing that second pill of abemaciclib, for example, more often than not, and they’re not even aware of it. So I really think helping patients come up with a system to be able to remember to take their medications, whether it’s once a day or 3 weeks on, 1 week off, or whether it’s twice a day with regard to the CDK4/6 inhibitors, we talk about what strategies might really work.

I always tell patients, for example, to take their morning pill when they brush their teeth in the morning. Almost everybody brushes their teeth in the morning, and to have the pill bottle right there by the toothbrush so you have to knock over your bottle to get to your toothbrush. And then the afternoon I really — or early evening, I really think it comes down to setting an alarm. I really think it’s setting an alarm or having a family member or friend call you at that time to remind you or text you, and they’re going to bug you until you return that text. So things like that are basically what we try to help people think through what might work for them.

DR LOVE: I just ordered one of those exercise kind of watches, where it does your steps and all that stuff. I guess it could like ping you twice a day to take your medicine.

DR O'SHAUGHNESSY: Yes.

DR LOVE: I don’t know.

DR O'SHAUGHNESSY: Yeah.

Review of Clinical Investigator Survey Results

DR LOVE: Alright. Let’s take a look at this survey. You kind of described your own approaches to these situations. As you know, we love to ask people what they usually do when a trial’s not available.

So here are the people that we got. As you can see, quite a prominent group of clinical investigators to tell us how they’re practicing in terms of first-line therapy. And we gave them a bunch of different clinical scenarios to test out, de novo versus presentation while on adjuvant endocrine treatment, development 2 years after completing, as you were talking about. We varied the age, particularly looking at premenopausal/postmenopausal and then very elderly. We looked at tumor grade and PR status just to see. We didn’t see too much of an effect.

Here’s the way we usually plot these things, a number of scenarios that we asked both premenopausal/postmenopausal. We picked, I think, 40 years old and 65. And for the younger patient without visceral disease or in this case asymptomatic bone mets, clearly the preference for most people is ovarian suppression plus ribo and an AI. Similarly, usually the premenopausal and postmenopausal choices are the same other than the presence of ovarian suppression, and you see the same thing, preference for ribo.

When we get into visceral, including liver mets, we started to see some differences. You expressed your preference for abema in this situation, but the majority of people are still sticking to what they usually use, which is ribo. Any thoughts about your colleagues’ approach versus yours?

DR O'SHAUGHNESSY: Well, I think the data with regard to the premenopausal patients are strongest with ribociclib with the MONALEESA-7 trial being entirely dedicated to first-line premenopausal patients. And so I think that’s why probably most oncologists shown here do favor the ribociclib. The reason I chose abemaciclib here is because of symptomatic liver metastasis. I mean that’s serious disease.

I believe the data are strongest with regard to liver metastasis specifically with abemaciclib. because we know, for example, even in postMONARCH, now we have Phase III data showing us that yes, abemaciclib is noncross resistant with palbociclib and ribociclib, at least to some extent. We know that abemaciclib has a broader mechanism of action, and so to me we want to pick the agent most likely to benefit patients. And to me the data are strongest with liver metastasis, more virulent, this patient’s very unlikely to benefit from endocrine therapy alone, and so she is somebody who’s more endocrine therapy resistant, likely, so that’s why I would use abemaciclib for her. I believe the data are strongest in that setting. So that’s how I interpret the data.

This is a patient who if we don’t get control of this disease quickly we may not be able to really treat her effectively, so that’s another thing in my mind. I don’t want to start with ribo and then have to switch her over because that may take too long. The other thing is with symptomatic liver mets she probably has elevated liver function tests, and ribociclib already in the context of elevated liver function tests, can be difficult to interpret what’s disease progression, what could be toxicity from ribociclib. So it makes me a little bit uncomfortable, and I don’t really want to reduce the dose of it to start off with, et cetera. So I just feel more comfortable with abemaciclib, but mainly it’s because I think the data in liver metastasis is strongest with abemaciclib.

DR LOVE: So here’s another scenario that we asked about, which is patients with brain mets, and in this case we said requiring whole-brain radiation but were asymptomatic. I always get a lot of grief from Harold Burstein that I create scenarios that you don’t see very often in practice, so I don’t know how often you actually see brain mets in this situation, but it looks like here you really see the faculty moving towards abema. This is premenopausal, same thing in postmenopausal. What’s the data?

DR O'SHAUGHNESSY: The data are not terribly strong. There just haven’t been a lot of data. We do have data in pretreated, somewhat heavily pretreated metastatic breast cancer patients. It was kind of a Phase I/II trial around the US. I had a patient on that trial, where patients could have had radiation therapy to the brain, they were usually heavily pretreated, and they received a combination of endocrine therapy plus abemaciclib. And there was some cytoreduction in the brain mets. Where I’ve seen most of the benefit with abemaciclib and AI therapy in patients, and we do see this, Neil, it’s not uncommon, but it is this very aggressive luminal B biology. We do see it. A lot of those patients need whole brain if they have many, many metastases, or they have SRS. Once they’ve had the radiation therapy, I have found that abemaciclib can greatly prolong the time to new brain metastasis.

I’m thinking of a woman many years ago, Neil, when abema was just being developed, she had a very aggressive pleomorphic lobular breast cancer, completely resistant to every single thing I gave it preoperatively. At mastectomy her lymphatics were completely plugged with cancer. Every bit of the cancer was in the lymphatics, 6 positive nodes. And about 18 months into her letrozole she developed too numerous to count brain mets, had to have 1 resected because it was in the cerebellum, she was very dizzy, resected, SRS, then she got whole brain radiation.

She traveled down to San Antonio to get on abemaciclib. It was 200 mg that was being tested. She stayed on her AI, and she went 3 years, Neil, before she developed her next brain metastasis. So that’s an example of where I really see the benefit. Not so much in the cytoreduction of existing brain mets the way we may see it with T-DXd or with tucatinib, but I find it more just preventing progression of brain mets is where I’ve seen the benefit from abemaciclib.

DR LOVE: So a couple other scenarios. First was whether or not PR affected treatment choice. Basically we found that — because there were a couple of things we were curious about, did PR status affect treatment choice? And did grade of tumor affect treatment choice? We were not able to discern any difference in how people answered. Does that affect your choice of treatment?

DR O'SHAUGHNESSY: Progesterone receptor in and of itself probably not, depending on the other variables, the strength of the estrogen receptor. Really, for me, the big picture is, is this cancer likely to respond to endocrine therapy or not. If it is, I’m comfortable with ribociclib or palbociclib for patients who cannot tolerate the ribociclib or who can’t have it because of cardiac comorbidities and medications.

But for patients where their cancer has manifest that it is endocrine therapy resistant, more likely to be PR negative for sure, but really that’s more about the estrogen receptor than the progesterone receptor. That’s where I tend to choose abemaciclib. It’s very simple for me. If they are not going to be endocrine therapy sensitive, if you were to consider endocrine therapy alone, that’s where I use the abemaciclib. If they may benefit then I use the ribociclib or the palbociclib.

Now Grade 3 does matter to me because grade tends to really be associated with endocrine therapy sensitivity or resistance. Grade 1/2 much more likely to be endocrine therapy sensitive, Grade 3 less likely to be endocrine therapy sensitive, so for me grade does matter. And again, with ribociclib they’ve looked at — and I was involved in that. Matt Goetz, myself and others looked at the MONARCH 2 and 3 trials with regard to grade, and that was one of the differentiators of substantial benefit from the abemaciclib was grade. So grade matters to me, but not so much the progesterone receptor.

DR LOVE: So one final issue I wanted to ask you about is dealing with people who develop problems on CDKi and particularly the issue of trying to do dose reduction as opposed to switching to another CDK. We had a case at ASCO of a patient who was started on ribo, had LFT abnormalities, they dose reduced, more LFT abnormalities, they dose — they kept sticking with the ribo, and people were like well maybe you might have considered switching rather than continuing to dose reduce in that problematic situation.

So I’m just kind of curious. Here’s a scenario we presented in this survey of a patient who’s on ribo, develops LFT abnormalities. Most people would keep the ribo going but lower the dose but others would switch to abema, also others would hold ribo until resolve and then restart again at a lower dose. Others would keep it going. How do you approach this situation with abnormal LFTs on ribo?

DR O'SHAUGHNESSY: It depends on the grade of the LFT abnormalities. If patients have Grade 1 LFT abnormalities, I will generally continue the same treatment. I won’t make any modifications. If I have to interrupt their therapy, so that’s Grade 2 or Grade 3, I have to interrupt until resolution to Grade 1 or less, I’m not likely to restart the ribociclib, because in the first-line AMALEE trial comparing 600 to 400 mg of the ribociclib there was no difference in the rate of the incidence of liver function abnormalities, both all grade, as well as Grade 3 and higher.

We also saw that looking at the NATALEE trial with 400 mg in the adjuvant setting compared to 600 mg in the metastatic setting. There really is no difference in LFT abnormalities with the 400. So it does not appear to be dose related. And we just don’t have good data that when patients have Grade 2 or 3 liver function abnormalities, and you stop until resolution, and then continue the ribociclib at a reduced dose. We just don’t have data that the LFT abnormalities basically do well, and you’re able to continue.

Most of the data really suggests there’s no difference between 400 and 600 with regard to the incidence of LFT abnormalities. And so the last thing you want to do is have to take patients off of these agents for long periods of time then get them back on it again and have to restart it, so I tend to switch. And I would switch, generally speaking, to abemaciclib because with the patients already having time off of therapy I don’t want to take a risk that they’re not going to get real benefit from the CDK4/6 inhibitor. And so I tend to feel like patients who have had kind of any kind complications I want them on the therapy that I feel they’re most likely to benefit from, and so I would go with abemaciclib versus palbociclib. But I don’t restart ribociclib once I hit Grade 2 or 3 LFT abnormalities.