Neoadjuvant PI3 Kinase Inhibitors for HR+ Breast Cancer
Panelists: Adam M. Brufsky, MD, PhD, University of Pittsburgh Cancer Institute; Michael Gnant, MD, Medical University of Vienna; Joyce A. OShaughnessy, MD, Baylor University Medical Center; Hope S. Rugo, MD, University of California San Francisco Helen Diller Family Comprehensive Cancer Center; Michael Untch, MD, Helios Klinikum Berlin-Buch
Hope S. Rugo, MD: At this meeting, we saw some very interesting neoadjuvant data with taselisib in the LORELEI trial, where there was higher clinical response. No pCR, good self-cycle suppression, but I think it was encouraging and there will be data with alpelisib at San Antonio in a neoadjuvant trial as well.
Adam M. Brufsky, MD, PhD: So, Michael, do you want to talk about LORELEI a little bit for us, as an investigator?
Michael Gnant, MD: Yes. I may be biased, but we contributed a lot of this trial and we conducted it, and I think that’s also significant information that needs to be made. This is an academic network trial together in a fair and transparent partnership industry, which I believe is a role model to go for the future. So, it’s SOLTI and ABCSG and other groups under the big umbrella who have conducted this. It’s a randomized phase II trial, 334 patients. And to me, as it was presented here, this is really very, very good news because for the first time, a trial in the neoadjuvant setting could prove, first, that it can make its primary endpoint closely and at a significant level of 0.049. So, we see improved response in basically luminal B disease when you add an alpha-specific PI3-kinase inhibitor to letrozole. And second, that effect is even better in the molecularly defined mutant subpopulation, which I believe is a type of a proof principle in a neoadjuvant setting that we have been waiting for, at least that I have been waiting for. I believe that’s the significance. There are a lot of additional results from this trial coming up to be presented in San Antonio and probably later, including translational status. But there’s also that primary endpoint result, a 10% difference in the overall response rate for the combination therapy for the whole population and 17% in the mutation subcohort.
Let’s say, guiding us toward the availability of a biomarker, of another not too easy to use, it’s probably not a very cheap track in the sequence and together with the results from the metastatic trials, that are SOLAR and SANDPIPER, as you have mentioned. I think in the next 6 to 12 months, we are going to have a much better picture on how to utilize PI3-kinase inhibition in our multi-pathway approach.
Adam M. Brufsky, MD, PhD: And LORELEI will be the final question that we’ll have for this section. I want to ask again, Michael: is there any role for these in the neoadjuvant setting? Is the Gepar group doing trials with these agents, targeted agents for luminal A/luminal B disease?
Michael Untch, MD: So, it was quite an interesting piece of proof that we go straight on, from bench to bedside, with these kind of trials—CDK4/6 inhibition. I can remember when the first data were presented; quick clinical data and then first clinical data, up until the approval now with multiple drugs—palbociclib, ribociclib. It was less than 10 years, a huge development, and good for the patients. Same story maybe with one of the alpha-specific PI3-kinase inhibitors, very nice story. And obviously, yes, we would like to see more data from the neoadjuvant setting as a proof-of-concept in the clinical setting because the usual way was always second-, third-line metastatic patients treated and then moved forward to adjuvant, and then we move to neoadjuvant. Now we see that we go the other way. And I’m very happy that this year, at the St. Gallen meeting that occurred in Vienna, it was stated that neoadjuvant treatment is the better way to go, especially in high-risk breast cancer populations like HER2-positive—we’ll come to that in a moment probably—and triple-negative.
Now, what about luminal disease? pCR as an endpoint in luminal cancer was not the best way to look. And that is the reason why all those endocrine treatments, targeted treatments, didn’t make it to the neoadjuvant clinical setting. As a proof of concept, I like very much what Michael Gnant just presented that we look for the target, look how it works, and even find a subgroup of patients where it’s working even better: PI3K mutations. And also with abemaciclib, Sarah Hurwitz’s data from the neoadjuvant setting were very reassuring that it is the way to see how to develop a drug, whether these drugs are going to make it in the clinical setting.
Adam M. Brufsky, MD, PhD: Right, that’s the point. Will you ever use it clinically?
Michael Untch, MD: Well, I would say again, and I come back to the statement from St. Gallen this year, I’m very happy to hear that we have it now in the clinical setting. For high-risk breast cancer populations, yes. For luminal B or luminal cancer at the moment, neoadjuvant treatment is not a clinical standard.
Joyce A. O’Shaughnessy, MD: I just want to say, the NeoPAL trial was interesting to hear. It was comparing preoperative chemotherapy to letrozole/palbociclib, and the objective response rates were the same. But in terms of the median of which residual disease, path CRs were very rare. But with regard to the median Ki-67 on the residual disease, it was like a median of 8% with chemotherapy and 3% with the palbociclib. So, could that be a substitute for some patients? That’s an interesting question.
Hope S. Rugo, MD: That’s what I was thinking. We are really struggling. We screen with MammaPrint for I-SPY 2, or a neoadjuvant phase II randomized trial, and the people are low-risk by MammaPrint. They often have tons of disease and low risk—they have a high risk of recurrence and death. If we could give them targeted agents with endocrine therapy, I think it would be super.
Adam M. Brufsky, MD, PhD: Right, that’s what I’m thinking of. There must be a population out there with a big tumor that’s luminal A, luminal B-ish that maybe we ought to be treating with neoadjuvant endocrine therapy with or without a targeted agent instead of chemotherapy.
Michael Gnant, MD: But the problem with that, though, is that the bar is set by neoadjuvant chemotherapy. Suppose, in terms of the surgical endpoint, breast conservation in terms of pCR, which—and I completely agree with Michael—on this, pCR is just a wrong endpoint. We keep picking it because the regulators want to see pCR, which I think, for these agents, is as rubbish as overall survival is for luminal disease in the advanced setting. And in all these trials, for example, LORELEI, you have 16 weeks of treatment. The alpelisib trial is going to have 24 weeks of treatment. When you talk to people who use neoadjuvant a lot traditionally, like Mike Dickson from Scotland, they have treatment durations in an elderly population; sometimes nonoperable historically—9 months or 12 months. Do we dare to do this in a situation where a quick anthracycline and taxane combination could get rid of half of the tumor, even in luminal disease or where breast conservatory is done? So, that’s the problem for clinical standardization of this approach, also for surgeons. Because we got used to these wonderful shrinkages, you know?
Adam M. Brufsky, MD, PhD: Well, there are surgeons, interestingly enough—at least the ones who I work with—who are now really very cautious. They really are now doing more and more neoadjuvant therapy. They’ve decided, someone comes in with a luminal A-ish like tumor, big, doesn’t want an operation…
Hope S. Rugo, MD: Ki-67 or 2%, they don’t get tumor shrinkage with chemotherapy.
Adam M. Brufsky, MD, PhD: They’ll try the interleukin therapy option.
Michael Gnant, MD: That’s good, but you have to see that in the LORELEI trial, fortunately it was shown that it was the 5 days on, 2 days off schedule. Taselisib is quite manageable. It’s less than 10% of patients stopping the treatment, which is a little bit different from continuous therapy in the metastatic setting. But still, these drugs are not easy. You have diarrhea, you have the rash, you have all of these. LORELEI’s showing us the way, I guess. If we can identify a mutant population where you can see we can have response probabilities—which are similar to chemotherapy in triple-negative, for example—then I think for large tumors, why not put the patient on 6 to 9 months of neoadjuvant endocrine therapy plus targeted? So, this is going to increase in clinical practice in the future, but it’s going to be a slow development. That’s what my prediction would be.