Adam M. Brufsky, MD, PhD: So, that gets me to a few other questions. The first question is, is there any role for single agents now? Fulvestrant by itself? Is there any role for that or are we done with that? Are we just now going to use fulvestrant with a CDK4/6? And we have that in MONARCH2. We have that in PALOMA-3. Is there any reason for single-agent fulvestrant anymore? You saw the 22-month PFS in some of the trials as a single agent.

Michael Gnant, MD: I think we all, at least if you have some part of the responsibility, would love to have a clear-cut definition on that lowest subgroup where this is reasonable. I have to say, and joining Joyce in her initial statement, it’s very difficult to identify that. I’m struggling a little bit with the FALCON trial because it was basically on pretreated patients. And in my environment, those patients simply don’t exist. So, I’m not so sure that these PFSs can be translated to where I practice.

In our interdisciplinary tumor center, it’s really more a backup for patients. For example, you know I’m from an alpine country so some patients live in very remote areas. Whatever you do with targeted combinations, whether that’s anti-inhibition for the side effect managements—also CDK4/6 in the beginning—frequent blood count is a logistical challenge for some of the patients, and obviously a monthly injection with fatigue as the most terrible side effect can be convenient for some low-risk patients. But I would consider the backup approach at this point in time.

Hope S. Rugo, MD: It’s interesting because a lot of our patients don’t want to do injections. They don’t like the 2 injections, especially on a very thin patient; they don’t like it at all. And I think we work really hard on that. I just had big discussions about whether you ice or heat the injection area with our nurses. But I think for a patient who gets really tired of the injections and gets bruises all the time, that now with the fulvestrant being approved first-line, it does give us some flexibility. I would take the same approach. It’s really dependent on the patient.

I have some patients who refuse to take the CDK4/6 inhibitors—I do live in California. They believe that maybe that’s too much therapy or it might do something they don’t know. I don’t understand it, but those are patients who you know you would give fulvestrant to instead of an AI if they haven’t had prior endocrine therapy in the first-line setting without visceral disease. So, that’s based on the FALCON trial. It’s helpful information I think to have, even though most of us really wouldn’t find too many patients who we would treat with fulvestrant alone.

And I think there’s a trial going on in Spain with multiple countries, as part of one of the networks, that’s looking at fulvestrant or an AI with CDK4/6 inhibitors as first-line therapy, in more of a treatment population we don’t see—people who’ve relapsed and were exposed to endocrine therapy before.

Adam M. Brufsky, MD, PhD: So, what happens on progression? Where does everolimus fit in now for people’s practice? Joyce, where do you use everolimus?

Joyce A. O’Shaughnessy, MD: I still do, and when patients benefit from CDK4/6 inhibitors first-line, I tend to go on to everolimus second-line, with the exception of liver metastases. In general, I have found patients seem to do better with capecitabine afterwards. There are the occasional exceptions, such as somebody with extremely indolent disease. But in general, I have found that patients with what I would consider to be still ER-driven sites of disease—bone, lymph nodes, parenchyma lung, pleura—they seem to do well on everolimus. And generally speaking, what I’ll do is, I’ll go on with exemestane, everolimus, getting them T’d up. Because I think if they benefit from everolimus, I think they re-express that estrogen receptor really well. And then I go on to fulvestrant single-agent next for those patients.

Adam M. Brufsky, MD, PhD: So, it’s third-line therapy?

Joyce A. O’Shaughnessy, MD: Yes, because if they benefit from everolimus, I have found that they get about 6 to 9 to 10 months from either tamoxifen or to fulvestrant after everolimus.

Michael Untch, MD: A question to my American colleagues. I fully agree with this statement, but I hear from many of my colleagues in Germany that this is not the label of everolimus, meaning using everolimus after a CDK4/6 inhibitor is not on the label.

Adam M. Brufsky, MD, PhD: The problem is that there wasn’t a label. They were supposed to give CDK4/6 when the label was developed.

Hope S. Rugo, MD: There’s no cross reactivity that we have seen. I think that the crossover activity you have is resistance to the endocrine therapy.

Adam M. Brufsky, MD, PhD: Right. It’s an ESR1 mutation, if anything.

Hope S. Rugo, MD: Yes. If you’ve developed a cancer that’s very resistant to all endocrine therapies, explosive visceral disease, that’s a different situation. But I’ve had patients who responded to everolimus who didn’t respond to CDK4/6 inhibition who presented with visceral metastases on adjuvant tamoxifen. And I had a patient who also responded to everolimus and then responded to a PI3-kinase inhibitor for 4 times longer, but still had a good response. So, I think all of that clinical experience makes you realize that we don’t really understand which resistance mechanism is playing a role and that these drugs can be used in sequence in many situations with great efficacy.

Michael Gnant, MD: From a clinical point of view, I think in general it’s fair to say that; if an inhibition has been pushed back at least 1 line by the adjuvant of CDK4/6. Having said this, clinically, I think in addition to presumed resistance in an individual patient, this says really a lot about the side effect profile. Leukocytosis can be an issue even now that we know better how to deal with it. Pneumonitis can be severe and actually is a concern also for patients. And for some of our patients, we can avoid chemotherapy as compared to earlier than your treatment with CDK. I have the same or equally looking side effect, neutropenia. And then I go to the next stage and then I get leukocytosis, which is also a traditional chemotherapy side effect. So, what is better? Obviously, this is not true, but I think in the clinical sequencing of all this, we urgently are needing better tools of finding a rational way. Because it’s really trial-and-error in many of these patients.

Adam M. Brufsky, MD, PhD: Well, speaking of trial-and-error and rational tools, let’s finish this section talking about PI3-kinase inhibitors. You think that at least the initial randomized phase III trials in the relapsed setting, or after the first- or second-line of endocrine therapy, have really been modest at best. And so, where are we going with PI3-kinase inhibitors?

Hope S. Rugo, MD: It’s so funny because I think that when we saw the data from pictilisib, which is the FERGI trial, that wasn’t too exciting in light of toxicity. Then we saw these 2 BELLE trials with buparlisib, and buparlisib is a drug that has unacceptable toxicity in this patient population. But what I thought was really exciting about those trials was the whole population; not much, a little benefit. But one of the trials actually was able to compare PI3-kinase—activating mutations in blood versus in tumor and see a very high concordance, which I think took us a big leap ahead in our ability to use blood and showed that the back fit was greater in that population. So, the 2 trials that are looking at what we think are the most promising step in PI3-kinase inhibition—mostly alpha-specific alpelisib and taselisib—they actually are powered to look at the cohorts that have PI3-kinase–activating mutations, as best as we can check it from the tumor, because that’s how they’re doing it. And we’re hoping by the end of the year to see the first one of these trials, the SANDPIPER trial, with taselisib.

Transcript Edited for Clarity