Immunotherapy in Breast Cancer

Transcript:Adam M. Brufsky, MD: So that gets us to the last big topic to talk about, which is really the role of immunotherapy. We’ve seen it now in melanoma, in non-small cell lung cancer, renal cell cancer, or bladder cancer, or other cancers. And now we’re beginning to apply it to triple-negative breast cancer. Any comments, Hope? You presented data.

Hope S. Rugo, MD: We know the triple-negative data. It’s so interesting because I was just talking to a patient yesterday about the KEYNOTE phase II trial that’s ongoing in unselected triple-negative breast cancer after first-line. And I told her the response rate in the phase I trial, and she was like, ‘Really?’ This is a woman who’s progressing on adjuvant therapy, so it’s not as if we have something that’s better. Both trials—the pembrolizumab, which is a PD-1 inhibitor and the atezolizumab trial which is a PD-L1 inhibitor—they looked at about 25 patients in a phase I dose expansion trial and showed almost identical rates of response in their overall response rate, which was their primary endpoint of about 18.5%, and all of those patients had to express PD-L1. The atezolizumab trial had a more strict definition, so you had to be more positive by IHC. But the pembrolizumab trial was just greater than 1%. You just had to have 1% or more, actually. I think even 1% qualified. And the rates were surprisingly identical.

We saw at San Antonio another trial—called the JAVELIN trial—with avelumab, another PD-L1 inhibitor, where they treated all patients, and they saw actually a much lower response rate. And so we don’t really know. They had treated about 150 people, and the response rate, for example, in ER-positive disease was 3% and it was low in triple-negative disease. So it may be that all these agents are not the same, and I think we all think that’s quite likely. They found tons of people with PD-L1 expression, but they used a different antibody, so we have no standardization. So now there’s an international committee working on standardization of antibodies used to test PD-L1, like HER2. So then in ER-positive disease, the idea was, okay, now we have all these trials in triple-negative breast cancer. Everybody who makes one of these agents, the immune checkpoint inhibitors, is testing them in triple-negative breast cancer. What about ER-positive disease? There’s preclinical data that the more luminal B—like, more proliferative ER-positive breast cancer does have a greater infiltration of lymphocytes and has increased expression of PD-L1.

There was another KEYNOTE trial (they call them all KEYNOTE trials), where they had like 24 different cancers that we treated in a phase Ib trial. And all of the cancers have shown some response rate variably that have been reported so far. So there have been six different diseases reported, and we just reported at San Antonio, the ER-positive group. And it’s interesting. There’s a lower rate of positivity, so a lower rate of expression of PD-L1. But there was a real response rate of about 13%, 13.5%, and some of those responses were quite durable. In the triple-negative group, median time to response is 18 weeks; in the ER-positive group it was sooner, but people went off very quickly. I think people didn’t really buy into that pseudoprogression. Some people went off faster. So now there’s an explosion of trials testing these agents and combining them with all sorts of different drugs that sort of upregulate the immune response, so that if you could take away the blockade, then you could upregulate the immune response and maybe it would work even better.

Adam M. Brufsky, MD: Right. It’s basically doing 30 years of trials, again, with checkpoint inhibitors.

Hope S. Rugo, MD: Right.

Adam M. Brufsky, MD: The issue I have with breast cancer. Though, just in general when you think about it, I mean one of the theories right now is that the reason melanoma and renal cell cancer are so responsive is that they have a lot of neoantigen diversity and breast cancer and prostate cancer don’t. And so, in fact, people have shown the more neoadjuvant diversity you have in the tumor, the more response you’ll have. Again, with breast, that kind of gets me a little skeptical.

Joyce A. O’Shaughnessy, MD: Right. But there’s the other strategy, at least in the triple-negative patients in whom you don’t know if it will work, is to try to expose the BRCA-ness of the triple-negative disease that’s been either epigenetically modified or modified by some other mechanism.

Hope S. Rugo, MD: Using a histone deacetylase.

Adam M. Brufsky, MD: I think that’s a great idea. These are all great ideas.

Joyce A. O’Shaughnessy, MD: Yeah. And there’s a whole bunch of agents that are being looked at that could do that, actually. So that could help.

Adam M. Brufsky, MD: Great. That could work.

Hope S. Rugo, MD: I think we’re in an exploratory phase.

Adam M. Brufsky, MD: We are, but it’s really interesting. Hopefully, we’ll have that sort of thing happen.

Christy A. Russell, MD: But the immune picture in the other direction is the issue of tumor infiltrating lymphocytes (TILs) and their huge impact on prognosis. Hope, I’d love to hear what your thoughts are about it.

Hope S. Rugo, MD: It is fascinating. I think the whole idea that if you don’t block the immune response—and maybe that’s expressed by TILs, these tumor-infiltrating lymphocytes—that maybe that makes a big difference, particularly in the more aggressive phenotypes of breast cancer where TILs have correlated with better outcome in patients, as you pointed out. And Sherene Loi, who’s now back in Melbourne in Australia, has done a lot of work on this in a number of different trials, very elegant work. I think the complexities of this is whether the TILs are in the stroma or in the tumor, or where they are. But it does seem like no matter where the TILs are, it does make a difference in prognosis.

Adam M. Brufsky, MD: If you have them, maybe they’re indicative of an active immune response.

Hope S. Rugo, MD: That’s one of the things. Like macrophage inhibitors, maybe you could enhance the TIL population. But you have to not cause too much toxicity with on-target toxicity. So this is a fascinating area of exploration. There’s a whole TIL group, along with the PD-L1 group, who are interested in trying to standardize the way we measure TILs and report them, so that maybe you could use it to determine prognosis. Maybe we’ll be using RCB plus TILs in order to decide who should get post-neoadjuvant therapy.

Adam M. Brufsky, MD: And that’s for the future.

Transcript Edited for Clarity