I-O + PARP Inhibition for Triple-Negative Breast Cancer

Transcript: Joyce O’Shaughnessy, MD: Fascinating. It is great and so gratifying that we have this now for practice. The neoadjuvant data could be absolutely stunning, hopefully. But we have a choice of partners now that we’ll have an opportunity to utilize checkpoint inhibitors. We’ll be wondering: Does it always have to be nab-paclitaxel? Could it be a platinum-based regimen? What about our patients who have a BRCA1 or BRCA2 germline mutation, where we have data with the PARP [poly (ADP-ribose) polymerase] inhibitors in the metastatic setting? In the frontline, for example, in OlympiAD versus chemotherapy, there’s a substantial survival improvement with olaparib, first line. And so it’s tempting to want to combine the PARP inhibitors with checkpoint inhibitors. But the data are early. What’s your take on the PARP inhibitors, Hope? How are you using them, and are you going to be tempted to combine them with checkpoint inhibitors?

Hope S. Rugo, MD: Yes. I think the answer is yes, but I’ve already said to you that I don’t follow labels well. I think that’s 1 of the caveats. I think that the data from the MEDIOLA and TOPACIO trials suggest that you can safely combine a PARP inhibitor with a checkpoint inhibitor. There were responders. Maybe homologous recombination defect is a predictor, but certainly there were responses and people were getting PARP inhibitors. You had the germline BRCA mutation, so it’s a little hard to separate it out. There’s more of an effort going on in trying to combine these now with the idea that you actually upregulate, via the same pathway, interferon. As you talked about, it might upregulate increased expression of PD-L1 [programmed death-ligand 1]. And so PARP inhibitors could be synergistic. I think that’s data that we won’t see for a little while.

But the PARP inhibitors—I agree with you that the subset analysis of OlympiAD suggests that we should be giving these agents much earlier. And, of course, we’ll see adjuvant data someday from OlympiAD. Let’s say a patient you see has triple-negative disease and a BRCA1 mutation. If you have almost a 9½-month survival, or any difference in survival, then you might want to give chemotherapy, nab-paclitaxel, and atezolizumab first, right? And then you’re going to want to use atezolizumab as maintenance after you stop the chemotherapy, so you don’t make them unable to walk from neuropathy, eventually. Then do you add the PARP inhibitor? I think that’s really a cool idea because the idea of maintenance PARP is really intriguing because of ovarian cancer, for which there are huge survival benefits. So that’s actually going to be studied in a number of trials, and in patients who don’t go on those trials I think that will be a question: whether or not we should be trying that.

Elizabeth A. Mittendorf, MD, PhD: Joyce, can we go back to the question you kind of posed when you led into the PARP inhibitors and about other chemotherapy agents? I think it’s important to note that, in fact, the KEYNOTE trial that I referred to used different chemotherapy partners, different backbones. So I think that will be very informative for you all, particularly in patients who may not want to get another taxane. I think it’s important to note that the FDA label is for nab-paclitaxel, but there is a study that’s ongoing looking at atezolizumab now in combination with paclitaxel. So there are going to be data in the not-too-distant future to address the question you asked about other chemotherapy partners.

Hope S. Rugo, MD: In the early relapsers, in particular, being able to give gemcitabine and platinum.

Elizabeth A. Mittendorf, MD, PhD: The other thing I thought was quite interesting about the approval, and I might need to go back and reread it, but the trial, and Hope described it well, was in the first line. The approval didn’t say that. I had actually thought that might help inform you all, for the BRCA-mutant patients you’re talking about now, on which agent you would use first, because there would be improvement only in the first-line setting. But it doesn’t appear that was in the label. I don’t know what you all think of that.

Joyce O’Shaughnessy, MD: I think the data showed that the checkpoint inhibitors tend to work the best early on. So I suspect we’re going to do like the IMpassion130 did. But if there’s a patient who has very, very locally advanced unresectable disease, that might be another issue. Right now, it’s sort of an ACT [Adriamycin, cyclophosphamide, and Taxol]—plus-or-minus-platinum approach. But interestingly, if that patient really is completely unresectable, she kind of fits within the label, which is quite interesting. So I think that’s going to be a potential option for the patient. But I suspect we’ll use it in the first line to make sure we get that survival advantage.

Ruth O’Regan, MD: One thing I would just say is I think there are some trials going on in the second- and third-line settings. Their accrual may be impacted.

Kevin Kalinsky, MD, MS: Absolutely.

Ruth O’Regan, MD: And as you pointed out, Joyce, I think the data outside the first-line setting are not as good, and we do need to address that. Most of it is with single-agent checkpoint inhibitors rather than with chemotherapy.

Joyce O’Shaughnessy, MD: Yeah.

Kevin Kalinsky, MD, MS: Just going back to the PARP inhibitor combination with a checkpoint inhibitor, we saw some impressive responses. I think 1 of my questions is, What’s the durability of those responses? And then there are also randomized studies that are accruing now for which we’re hoping to get the answer to combinations. Like the DORA study, for instance, in patients with triple-negative disease who had some sort of response with platinum. Patricia M. LoRusso, DO, has a study to the ETCTN [Experimental Therapeutics Clinical Trials Network] giving a PARP inhibitor with immunotherapy versus a PARP inhibitor alone. So I think we’re trying to answer some of those questions, and those were in patients with BRCA mutations.

Elizabeth A. Mittendorf, MD, PhD: And if you think about it mechanistically, with the PARP inhibitors, as we’ve been discussing, 1 thing that has now been demonstrated with a growing body of literature is the activation of the STING [stimulator of interferon genes] pathway. But the other thing, of course, is that it leads to accumulation of mutations which would, in theory, have neoantigen. So both of those things together, in theory, stimulate an adaptive immune response, which now you’re augmenting with a checkpoint blockade. And that adaptive immune response, one would hypothesize, of course, would make for a longer-term benefit. Again augmenting the activity of the PARP.

Hope S. Rugo, MD: And the DORA study is specifically addressing this issue of maintenance. Maintenance, first PARP inhibition and then adding the checkpoint inhibitor in 1 arm. It’s a very cool study, I think.

Transcript Edited for Clarity