Selecting Agents in CDK4/6 Inhibition for HR+ mBC

Transcript:

Joyce A. O’Shaughnessy, MD: With regard to the endocrine agent that we use to combine with CDK4/6 inhibitors, there are some new data from the MONALEESA-3 looking at first-line fulvestrant with ribociclib. And, of course, it was superior to fulvestrant alone—the same excellent hazard ratios that we’ve seen basically in all the first-line trials. The fulvestrant was also really an excellent partner for the first-line setting. Of course, in the second-line setting, it’s outstanding. What do you think?

Andrew D. Seidman, MD: Yes. So, for the endocrine therapy—naïve patient, we’ve learned lessons from trials such as FALCON that fulvestrant deserves to be considered right up front. For the patient who has had a prior aromatase inhibitor who has struggled with arthralgia, with any side effects in the adjuvant setting, it’s a no-brainer for me. I would lead with fulvestrant as the partner. But for the woman who has had an aromatase inhibitor for their early breast cancer for whom years have passed from their prior AI exposure, I’d probably still lead with an AI, more because I think patients tend to want an oral agent rather than to have intramuscular injections. It’s likely that, hopefully, with current development, we’ll see oral-selective estrogen receptor downregulators in the future, but it’s a discussion. But for someone who struggled with AI toxicities, I generally lead with fulvestrant.

Joyce A. O’Shaughnessy, MD: Yes, I agree with you. And we don’t have the head-to-head data with regard to an AI versus fulvestrant, both with a CDK4/6. That could, of course, inform our practice. But I agree, I think the preponderance of the data are with the AIs in the first-line setting, then we could always go on to fulvestrant in the second-line setting with pairings of fulvestrant with various mTOR inhibitors—hopefully, a CDK4/6 inhibitor. Hopefully, alpelisib in SOLAR-1, which we’ll talk about a little bit later, will be positive. So, we’ll have that pairing in the second-line setting. I agree, I pretty much go with the AIs for first-line therapy. Well, how about choice of CDK4/6 inhibitors?

Andrew D. Seidman, MD: The $64,000 question. It’s interesting. We have 3 aromatase inhibitors. We have anastrozole, letrozole, and exemestane, and we have 3 CDK4/6 inhibitors. So, 3 seems to be the magic number. And when I talk to physicians—if I ever ask the question “Have you used all the AIs?”—well, yes, anyone who treats a lot of breast cancer has used all the AIs. They’ve looked at the data. The data are more similar than different, and they tend to vote or use them based on their experience. When I ask doctors now—and it’s admittedly earlier in the evolution of CDK4/6 inhibitors—how many doctors have used all 3, it’s surprisingly a small proportion. And I’m sure that will change. But I encourage my colleagues to get hands-on experience and vote with their feet.

Having said that, in terms of efficacy, these hazard ratios are just so consistently on the mark—0.55 across the board with all of these agents. So, in terms of efficacy, it’s hard to argue that we’re seeing any big differences. And I don’t think we are interested in the Coke versus Pepsi trials anymore. I don’t think we’re going to see head-to-head comparisons. So, then it kind of comes down to practicality, ease of use, and toxicity. I think there are some within the limitations that one can have without direct comparisons, state that there are some differences that do appear. Abemaciclib does appear to cause more gastrointestinal toxicity and diarrhea than palbociclib and ribociclib. There is, at least for now, the need to do EKG monitoring with ribociclib regarding QT prolongation. I do wonder whether that’s going to disappear someday. In terms of myelosuppression, I would say they seem to be more similar than different in terms of the risk of febrile neutropenia. So, my long-winded answer to a very direct question is: Gain experience with all these drugs, and let your experience guide you, because the data look quite similar.

Joyce A. O’Shaughnessy, MD: Yes, I totally agree, they really are. In the premenopausal population, I’m tending to follow the bulk of the data there with the ribociclib from the MONALEESA-7 trial that now Debu Tripathy has published with the use of the LHRH agonist, of course.

Andrew D. Seidman, MD: Yes. It’s interesting that that trial actually ever happened, because we have subsets of patients in other trials, like PALOMA-3, where premenopausal women were given ovarian suppression. Do we really need to do large trials in hundreds of women where we just add on ovarian suppression when we’re asking a targeted question? I’m not so sure we do.

Joyce A. O’Shaughnessy, MD: Yes. I think we knew the answer going in and probably for access reasons, globally, I suspect there was the need for the trial. But you’re absolutely right. In fact, we just saw an update on the MONARCH 2 trial, also looking at the premenopausal subset, and the PALOMA-3, as well. These premenopausal subsets with the LHRH, they did just as well as the postmenopausals. So, in fact, with all 3 agents, we do have premenopausal data, although the big phase III data were with the ribociclib.

Transcript Edited for Clarity