Debu Tripathy, MD: Let’s talk about the next set of discussion questions, which is, what do we do for patients progressing on CDK4/6 [cyclin-dependent kinase 4 and 6] inhibitors? This is really a huge issue, and we have a lot of patients in this situation. There are many trials looking at this. Of course, we’ve got the recent approval of alpelisib with a PIK3CA-mutated group. Adam, what are your thoughts on the options, and are you routinely doing biopsies at this point? Are you doing biopsies up front? How are you managing both diagnostically and therapeutically?

Adam M. Brufsky, MD, PhD: Let’s start with what we think the mechanism may be, and we’ll go from there. It’s important to note that resistance to an aromatase inhibitor and CDK4 could be either resistance to the CDK4, or resistance to the aromatase inhibitor. I think what a lot of us are doing because alpelisib has been approved, is doing the PIK3CA mutation testing. I think what a lot of us are doing, at least we do in my practice, is we don’t really use the FDA cleared or FDA suggested biomarker… diagnostic, but in fact we use something like a Guardant. We do ctDNA [circulating tumor DNA] testing. Whatever we get out of that ctDNA testing will guide what we do.

For example, if there’s a PIK3CA mutation, I think that we probably would use fulvestrant and alpelisib based on the SOLAR trial. In that trial obviously you had women who got alpelisib and fulvestrant versus fulvestrant alone and had a median PFS [progression-free survival] of about 11 months versus about 5.5 months. Clearly, it’s beneficial.

In that trial about 6% to 7% of the patients I believe had prior CDK4/6 inhibitors. It wasn’t huge, but in that limited subset they also benefitted as well. Understanding it was small numbers. It is a reasonable option after. My go-to therapy before that had been probably fulvestrant and everolimus. There were very nice randomized phase II data from the PrECOG 0102 study showing, again, about an 11-month PFS with fulvestrant and everolimus. The problem being that none of those patients had prior CDK4/6 inhibitors, so it’s hard to know what to make of that.

I would be guided by that, and say that if someone has, for example, an ESR1 mutation, obviously we’ll go with fulvestrant, maybe fulvestrant and everolimus. If they have a PIK3CA mutation, maybe they’ll get fulvestrant and alpelisib when we go forward. Just to mention, my experience with alpelisib has been a little bit better that I thought it was going to be. I do fasting glucoses every week, starting metformin very quickly if they’re glucose is over 140, 150 mg/dL. I start them on Zyrtec [cetirizine] or Claritin [loratadine] to help with the rash. For me, they’re actually working well. But nonetheless, that’s how it would go.

There has been an experience, in fact, a group of performing cancer specialists at this meeting also gave their experience with just changing the CDK4/6 inhibitor and continuing the hormonal therapy. Joyce and I, and Kevin Kalinsky, MD, MS, and Aditya Bardia, MBBS, MPH, have groups that put together an experience of about 80 patients who were given abemaciclib after progression on a palbociclib-containing regimen.

In that trial, if you immediately got abemaciclib afterward, you had about an 8.3-month PFS. Clearly, you’re getting a benefit from that.

Hope S. Rugo, MD: Without changing endocrine therapy?

Adam M. Brufsky, MD, PhD: Well, just seeing abemaciclib as a single agent.

Hope S. Rugo, MD: Oh.

Joyce A. O’Shaughnessy, MD: It was a case series.

Adam M. Brufsky, MD, PhD: It was a case series. It wasn’t a randomized trial. It wasn’t controlled. If you had at least 1 intervening regimen, your PFS went down to about 4 months.

Debu Tripathy, MD: I have a question for the group as well about ESR1 mutations. Many of you listening out there are obviously ordering next-generation sequencing [NGS]. We generally think of ESR1 as certainly a mechanism for AI [aromatase inhibitor] resistance. Maybe they should be on fulvestrant. The plasmaMATCH study actually did not show much in the way of responses there. Any insights on how one should be handling that therapeutically…?

Hope S. Rugo, MD: Don’t pay any attention.

Adam M. Brufsky, MD, PhD: Really, don’t pay any attention to it?

Hope S. Rugo, MD: We’re already using the order of AI, fulvestrant to AI, right? So why does it matter if they have an ESR1 mutation or not? I do NGS, but it’s certainly not to find an ESR1 mutation.

Adam M. Brufsky, MD, PhD: But, Hope, if you have someone on AI, say letrozole and palbociclib, you do NGS looking for PIK3CA, and it shows an ESR1 mutation, would you consider putting him on abemaciclib?

Hope S. Rugo, MD: If they’re not progressing?

Adam M. Brufsky, MD, PhD: No, they’re progressing. They’ve now progressed.

Hope S. Rugo, MD: But then I’m going to put them on a fulvestrant and a CDK or I’m going to put them on something that includes fulvestrant. I don’t know that abemaciclib….

Adam M. Brufsky, MD, PhD: But would you continue the CDK4 is the question.

Hope S. Rugo, MD: No, but I don’t because there are no data to continue to progression.

Adam M. Brufsky, MD, PhD: That’s what we’re talking about.

Hope S. Rugo, MD: I don’t think that there are data to support abemaciclib as targeted against ESR1 mutations, but I think we think we should switch to fulvestrant, which I do, with progression.

Ian E. Krop, MD, PhD: Right, but these are these encouraging sets of data that are now coming out suggesting that CDK4/6 inhibitors may have benefit after prior progression. You just mentioned the multicenter study that you guys looked at that showed 36% of patients had disease control for 6 months. That’s significant. There was another poster presented from a Florida group.

Adam M. Brufsky, MD, PhD: Right, the Florida Cancer Specialists & Research Institute, right?

Ian E. Krop, MD, PhD: Yes, it showed 25% clinical benefit in that same patient population. What we need to do is establish who are the patients who benefit and who are not. I actually was wondering in your study, did length of time on the first CDK inhibitor predict benefit in the second?

Adam M. Brufsky, MD, PhD: It did in that abstract that was presented. The poster that was presented here at San Antonio, it did. What they did is they characterized the long-term responders with the therapy, and almost all of them had had a prolonged response to their initial therapy.

Hope S. Rugo, MD: But I will caution that this is retrospective.

Adam M. Brufsky, MD, PhD: Of course it is.

Hope S. Rugo, MD: You’re picking out the best people, but it is encouraging, and we have prospective trials that are going on.

Adam M. Brufsky, MD, PhD: But they’re very small trials, Hope, they’re 200 people….

Hope S. Rugo, MD: If we bring it back to ESR1 mutation, they don’t help us with this question. It’s really the question about if you are going to use fulvestrant. I think I tend to use abemaciclib alone when I don’t have an endocrine partner to give because they’ve run through them all. But again, I still don’t think, and you know, ESR1 mutations, unlike PIK3CA, and Rb and FGFR, they come and go. Right? It’s one of the only ones that goes.

Joyce A. O’Shaughnessy, MD: I don’t know if you guys will agree, I think data are emerging that if they have an ESR1 mutation, that your chance of benefitting from any endocrine therapy, including fulvestrant, is quite low. I’m hearing about median PFS of 2 months or less. But I would also say the good news is that when you partner these agents, which is what we’re increasingly doing, that they seem to be more agnostic to the ESR1s.

Adam M. Brufsky, MD, PhD: That’s the question. The whole idea behind oral SERDs [selective estrogen receptor degraders] or 1 of the many ideas around oral SERDs, is that some of them have, at least preclinically, in preclinical cell lines, responses to the oral SERDs that you don’t have with the fulvestrant. That’s part of the apparent theoretical benefit.

Debu Tripathy, MD: We really await the results of some of these trials.

Transcript Edited for Clarity