The Use of Chemotherapy and Checkpoint Inhibitors for TNBC

Transcript:

Joyce A. O’Shaughnessy, MD: But, Erika, why are we putting chemotherapy with checkpoint inhibitors? Why not monotherapy, for example?

Erika P. Hamilton, MD: Yeah. One, we probably don’t know, right? But there are a variety of reasons to think that that’s reasonable. One is chemotherapy and is thought to break the tumor up a little bit and help with antigens presentations, right?

Another is that triple-negative disease tends to be pretty aggressive. By having that chemotherapy backbone, patients can actually stay on therapy long enough that maybe you’re actually going to see your immunotherapy benefit, right? Third is that chemotherapy theoretically might decrease Tregss [regulatory T cells] a little bit. This makes the immune system a little bit more sensitive.

Joyce A. O’Shaughnessy, MD: Certainly when we see a survival advantage, that’s what we’re going to go with first.

Erika P. Hamilton, MD: Right.

Joyce A. O’Shaughnessy, MD: It’s an exciting new standard and there was a poster from David B. Page, MD, at the ASCO [American Society of Clinical Oncology] Annual Meeting 2019 combining pembrolizumab with capecitabine and pembrolizumab with paclitaxel. It was a small trial but the capecitabine-pembrolizumab combination looked quite good with a 45% response rate and a PFS [progression-free survival] and then that small study that was even a little bit better than paclitaxel. So, I think we’ll get additional data about additional partners. So Debu, I know you guys are conducting the ARTEMIS trial and are looking for a lot of biomarkers for prediction of response to immunotherapy. Can you tell us some of the things you folks are very interested in?

Debu Tripathy, MD: Well the ARTEMIS trial is designed to take patients who have triple-negative breast cancer, and they get 4 cycles of neoadjuvant doxorubicin, cyclophosphamide. If they’re not responding based on volumetric response, then they’re eligible for 1 of several phase II trials in which they receive paclitaxel with a targeted agent. We try to funnel the right patients into the right different arms we have open according to the phenotype.

For example, if it’s androgen receptive positive, they would get enzalutamide and Taxol [paclitaxel]. So 1 of our arms is for immunotherapy. It’s nab-paclitaxel with ATEZO [atezolizumab], and it’s for patients that have high tumor infiltrating lymphocytes. We decided to use that metric well before data from any of the current trials were available, before IMpassion was available. We have shown and are presenting this at ASCO that the TIL [tumor-infiltrating lymphocyte] count actually is predictive of PCR [polymerase chain reaction], and others have shown that as well. In the BrighTNess study that was shown. And so that brings up the issue of, what is the best predictor of immune response? Should it be PD-L1? Should it be CD8-positive cells? Should it be something related to DNA damage responsive signature which correlate with immunogenicity as well?

We have found that we’re using 2 more infiltrating lymphocytes, and it’s a little too early to know if we’re able to turn things around. We’re basically going to look at the PCR rate knowing that in nonresponsive patients it’s only 5% to 10%, and we want to bump it up to well over 20%, 25%, or maybe even higher.

So, we have much to learn about how to properly classify patients as being immune sensitive. As you know, the big picture here is that breast cancer, on its own, is generally not very immunogenic. We need to identify the ones that are to treat them and to identify better strategies to so-called make a cold tumor hot.

Joyce A. O’Shaughnessy, MD: Yeah, big time. You know I was just thinking, I believe that atezolizumab can be utilized if you would hit the 1% or more immune infiltrating cells positive with PD-L1 based on a CLIA [Clinical Laboratory Improvement Amendments]-certified antibody.

So SP142 was used in the IMpassion130, but many of our laboratories have other antibodies. They’re very well validated. The FDA gave some nice flexibility there, from a practical standpoint. Because all over the country people need to find these patients basically, but it’s not the tumor cells—it’s the immune cells. But many of the 22C3 use, quite commonly, and others. They’re very well validated, and comparative studies show that these are very specific antibodies. I think that’s personally fine as long as you’re looking at the immune infiltrating cells.

Debu Tripathy, MD: Also, the KEYNOTE-355 is going to report out and that one is using a different antibody. It’s using the Dako antibody. There may be expansions of the label depending on if that trial is positive. Of course, we don’t know if it is positive it’s going to be in the PD-L1 positive or negative, which is designed very similarly. So, I think it’s going to be an evolving story.

Transcript Edited for Clarity