Debu Tripathy, MD: Let’s move on to our final segment on triple-negative breast cancer. It remains a very challenging area. Of course, the development of immunotherapy and some successes in that area are very notable. I have a couple of questions to start this discussion. One for you, Ian, is the role of next-generation sequencing and using things such as tumor mutational burden, mismatch repair, and even DNA damage signatures. Hope, you can maybe follow with some of the nuances of PD-L1 [programmed death-ligand 1] testing and what we know actually from the randomized trials with the different testing that’s available.

Ian E. Krop, MD, PhD: I think we already touched on some reasons why we should be doing next-generation sequencing even on triple-negative cancers. As Joyce pointed out, there were some data suggesting that these cancers may be sensitive to drugs like neratinib if they have a HER2 [human epidermal growth factor receptor 2] mutation and drugs like capivasertib if they have AKT alterations. There are now evolving data for some of these, albeit rare mutations, that that could give us potential target.

Joyce A. O’Shaughnessy, MD: NTRK.

Ian E. Krop, MD, PhD: Right. Then there are the issues of knowing the tumor mutational burden. Is that going to be a predictor of response to checkpoint inhibitors? Mismatch repair also seems like, in and of itself, a target that confers sensitivity to checkpoint inhibitors. I think we’re getting to a point in 2020 that really we should be doing next-generation sequencing in addition to PD-L1 testing, which is a good segue.

Adam M. Brufsky, MD, PhD: When would you do it in triple-negative breast cancer?

Ian E. Krop, MD, PhD: That’s a good question. It’s been coming up lately. In general, you don’t have time to do it in a patient who’s presenting with symptomatic, first-line triple-negative metastatic disease. So I think there the primary thing to do is figure out their PD-L1 status and then, while they’re on their first-line therapy, assess their…

Joyce A. O’Shaughnessy, MD: The first-line trials with the ipatasertib—the turnaround has to be quick.

Ian E. Krop, MD, PhD: But outside of a trial…

Joyce A. O’Shaughnessy, MD: Yeah.

Hope S. Rugo, MD: I think it’s interesting because now we have to decide what we’re going to test for, and the ipatasertib trials are treating around PD-L1. So if you were going on that trial…the new trial says you do PD-L1 testing, but everybody gets atezolizumab. But our approved indication that we use outside clinical trials is for patients who have PD-L1—positive immune cells in the first-line triple-negative setting, based on IMpassion130 study. In that trial, where there’s a 7-month improvement in survival in patients who had PD-L1 IC [immune cell]–positive disease, the determination of PD-L1 positivity has been reasonably controversial because there are a number of tests depending on the specific immune checkpoint inhibitor.

There were 41% of patients who were positive, and we looked further in patients who had enough tissue to do additional testing. In that group, 46% were PD-L1 positive. Otherwise, they were fairly closely matched with the overall IMpassion130 population. What we found was that PD-L1 still seemed to be the best determinant of response to atezolizumab, PD-L1 positivity as determined by SP142. It’s been interesting because we’re going to see a bunch of metastatic trials report next year with ATEZO [atezolizumab] and with PEMBRO [pembrolizumab], and we’ll see a lot more data on PD-L1 testing. Right now we have to use the approval, SP142—at least 1% positive; doesn’t seem to make a difference how positive it is—for use of atezolizumab combined with nab-paclitaxel. But I think we’re going to see a lot more data in the future.

Ian E. Krop, MD, PhD: I think your data that you just referred to that was presented at ESMO [European Society for Medical Oncology 2019 Congress] was very reassuring that SP142, at least for now, is a good predictor of…

Adam M. Brufsky, MD, PhD: I think it’s a better predictor than CPS [combined positive score]. That’s the issue. We’re replicating the lung problems that our lung cancer colleagues did for the last 15 odd years.

Hope S. Rugo, MD: It seems as even SP263 and the CPS 22C3 antibody—CPS is a score of the antibody 22C3—are measuring something different, because in our modeling poster we presented here, they don’t completely overlap. Even if you go up on the number as you do a CPS score of 10 or 20—and 1 being positive—the populations actually move apart as opposed to come together. You’re measuring something else, which is complicated. The 1 thing you know is no matter what you test, you can find a positive population who will benefit because the SP142 population is encompassed within the CPS of 1.

Adam M. Brufsky, MD, PhD: But the question is, is CPS giving you a false positive, and then you’re over treating somebody?

Hope S. Rugo, MD: It ends up being very different in the early stage setting, which we’ll talk about in just a moment. I think it makes sense based on the immune changes that happen from early to late-stage disease.

Debu Tripathy, MD: It’s great that we have a lot of biomarker data, multiple IHC [immunohistochemistry], in the IMpassion130 study. That’s been tremendously helpful. While we’re on that topic, there were some updated survival data presented and perhaps we can comment…

Hope S. Rugo, MD: Well, they didn’t update.

Debu Tripathy, MD: Updated survival at ASCO [2019 American Society of Clinical Oncology Annual Meeting], not at the 2019 San Antonio Breast Cancer Symposium.

Hope S. Rugo, MD: It’s the 7-month that I was mentioning earlier, and I think it went from 10 to 7 months, which is still highly statistically significant. At ESMO Asia this year, there was also a subset analysis of Asian patients that showed effectiveness. Interestingly, unlike some of our drugs, like the oral tyrosine kinase inhibitors and others, there’s not a big differential toxicity apparent between the Asian subset of patients and non-Asians. There’s a mixed bag in terms of immune-related adverse events. The other thing that was interesting to me in IMpassion130 was that it seemed we were seeing fewer immune-related adverse events overall. Although it’s hard because it’s a big trial compared with smaller neoadjuvant data so far, but less than what we saw in the neoadjuvant setting. Perhaps, your immune system is not so intact for immune-related AEs [adverse events], but rash, hypothyroidism preceded often by hyper.

Adam M. Brufsky, MD, PhD: Well, that’s the hypothyroidism. You remember from medical school, 20% of women probably have subclinical hypothyroidism that will reactivate by using I/O [immuno-oncology].

Hope S. Rugo, MD: We see some life-threatening toxicity, and I would say to people the key thing is any organ can be affected. If you see something funny, it doesn’t go away. And you haven’t seen it before. It’s immune-related, so you should treat it. People are saying, “Well, the hemoglobin is 6 g/dL. Do you think I should give steroids?” The issue is that you give steroids and then think more about it. This is because you’re not going to hurt anybody for giving it unless somebody is really sick.

Debu Tripathy, MD: It’s so important to have clinical pathways for all of us that are in practice and seeing these patients. It’s impossible to keep up with all this at an organic level, and that’s true more and more now. These pathways are available through many areas, through ASCO, through a variety of other agencies. So that’s critical if you’re treating these patients.

Transcript Edited for Clarity