Genomic Testing & Adjuvant Therapy in HR+ Breast Cancer

Transcript: Joyce O’Shaughnessy, MD: Well, thank you guys for the nice discussion about the metastatic setting for the hormone receptor-positive patients. Let’s talk a little bit about the adjuvant setting now, so the ER [estrogen receptor]-positive/HER2 [human epidermal growth factor receptor 2]-negative patient coming in with breast cancer where you really are uncertain whether they need chemotherapy. Are you utilizing gene signatures such as the 21-gene signature, 70-gene signature across the board these days? Or for specific patients? How do you approach this big issue, Ruth?

Ruth O'Regan, MD: I think we predominantly use the 21-gene recurrence score assay because we have the most experience with that. I certainly use it on most patients with node-negative breast cancer, unless they really are saying, “I absolutely don’t want chemotherapy under any circumstances.” And even then, sometimes I’ll try and send it because it gives them nice prognostic information.

In the node-positive setting, I certainly would probably send it on 1 positive node, or maybe 2 positive lymph nodes if they’re older patients. Obviously, TAILORx has shown us that the majority of patients in the overall study don’t get a benefit up to a recurrence score of 25. When you do the math, it’s by far the majority of patients with ER-positive breast cancer, which is great news.

Of course, we have the subtlety of an exploratory analysis showing that you do start getting some benefit in women under age 50 as you start getting the higher recurrence scores up towards 25. But what we don’t know is whether ovarian suppression with endocrine therapy would be sufficient in those patients. So I think the test gives you a huge amount of information. I have to say, and I’d like to hear from other people, but I don’t think the most recent printout that we get is as useful as the one we used to have. I’m always pulling out the papers to show them. But I think it’s a huge advancement avoiding chemotherapy. I haven’t used the 70-gene signature as much, but when I have, I’ve found they usually are pretty in line in terms of chemotherapy recommendations.

Joyce O’Shaughnessy, MD: How about you, Kevin?

Kevin Kalinsky, MD, MS: I would say in the adjuvant scenario, we tend to utilize the recurrence score as well. I find the calculation tool to be quite helpful, and I think patients like it as well. It just seems a lot more specific when you’re able to give numbers incorporating some of those clinical features.

We participate in I-SPY sites, so we do MammaPrint for selection in terms of whether they would be eligible. If they’re high risk, then those are criteria for going on to I-SPY. So we do use the 70-gene assay. In MINDACT, they do have data in terms of this node-positive population. Also, as Ruth mentioned, we’re awaiting the results of RxPONDER.

We do have some data from the West German group of a population that had a recurrence score that was 11 or less that included node-negatives, but a third of them were node-positive and about 5 years out patients still had excellent disease-free survival of about 94%. So that was promising, but we’re waiting for additional randomized studies to help support that.

Joyce O’Shaughnessy, MD: Yes, cool. How about you, Hope?

Hope S. Rugo, MD: I pretty much take the same approach. I use the recurrence score a lot in the adjuvant setting because we started using it first, and we’re used to looking at it. I agree with Ruth that I don’t like the new form. I don’t know if it’s just being a creature of habit or if it’s actually less useful, but I think they’ve gotten a lot of feedback about that. We use the 70-gene score for selection of neoadjuvant therapy or not. I’ve been kind of interested in that. It’s not a continuous variable. It’s very hard to know exactly what the best way to use it is. What we found in I-SPY is the people who were on the low end of high risk have a lower pCR [pathologic complete response] rate and benefitted less from the addition of immunotherapy than the patients who were on the high end of high risk. So there’s clearly some differential benefit there, along that spectrum.

And being low risk, even on recurrence score, but also the 70-gene score, if you have a big disease burden, that doesn’t actually mean that you’re at low risk, particularly in young women. We see a number of recurrences in those patients who had a big disease burden. So I think we’re still trying to figure that out and determine what to do with dormancy and all of these different issues, but we do use the test quite frequently. I think it’s extremely helpful. Somebody was talking at a meeting recently about the fact that they don’t report Ki-67. Grade and Ki-67 are really variable.

Joyce O’Shaughnessy, MD: Yes. How about you, Beth? What’s the surgeon’s perspective here? What do you think about the recurrence score and the 70-gene?

Elizabeth A. Mittendorf, MD, PhD: At our institution we primarily use the recurrence score. Because of the data that you all have generated through your trials that suggest just how impactful it is in your decision-making, the thing that we’ve done is instituted a reflex testing program, which I think has been really good as far as identifying.

When the patient comes back and sees me post-operation, they’re seeing you at the same time. They can actually make some decisions related to their therapy. As a surgeon, where I’m becoming very interested in the potential utility of this, and I acknowledge this is probably best done in the setting of trials, is to actually look at that recurrence score when we’re trying to make a decision about upfront surgery versus potential preoperative systemic therapy. And at our institution, as I think Kevin and Ruth are both alluding to, we actually do have some confidence in genomic assays in node-positive patients.

So it’s not uncommon for me, when I have a patient in clinic that appears to have, as an example, a clinical T1CN1 stage ER-positive breast cancer, I’ll consider, in collaboration with my medical oncology colleague, identifying whether this is somebody who we want to put on a preoperative systemic therapy trial. We have trials that are looking at endocrine therapy, and we have trials that are looking at chemotherapy plus immunotherapy. And so, you can imagine how the score obviously would weigh into the decision about those trials. And then, occasionally the score will help determine if they should go for upfront surgery. So I think there’s a lot of potential for us, as surgeons, to work in close collaboration with our medical oncology colleagues to use these genomic assays in different ways to help guide the sequencing of treatment.

Joyce O’Shaughnessy, MD: Yes. And I think the collaboration between the surgeons and the medical oncologists is really critical to know exactly what the algorithms are for each institution. Were you going to say something, Ruth?

Ruth O'Regan, MD: We did a trial…actually using the recurrence score to determine how you would treat patients preoperatively. I think your idea is great. In the patients with higher recurrence scores, in our city the pathologic CR rate was only 20% with chemotherapy. So you know those patients need something else for sure. And then, there was no benefit in the patients with scores under 25.

Joyce O’Shaughnessy, MD: I use the recurrence score quite a bit. I am interested in the MammaPrint, as you have said. I’m really fascinated by the data coming out of I-SPY. It’s really interesting; And also, what Laura Esserman, MD, MBA, had shown with the ultra-low. That was interesting. It’s not bad to know the ultra-low risk. You know whether they need any endocrine therapy at all. Can they get away with 2 years if they’re having adverse events, etcetera? So I think that’s interesting. And then I’m really interested in what you said, Hope, about the highest end of the spectrum for the high-risk, the so-called high-2, but I guess that’s really just the highest end of the spectrum where some of the DNA damaging agents…. I guess that’s where veliparib added the most, in the higher end. I think it’s still early, but I think that’s fascinating.

Hope S. Rugo, MD: Yes, and it’s clearly a different group of tumors. I think those are the ones that act more endocrine-resistant early on, and it’s fascinating to me about the immunotherapy data. It’s obviously still a small data set, but it’s quite intriguing because it’s the only subset in I-SPY. So far, it’s the only arm that’s graduated in hormone receptor-positive disease. Christina Yau, PhD, has looked at the subsets and has found that the group who really changed the pCR, estimated pCR, were these high-2s.

Joyce O’Shaughnessy, MD: In the ER-positive setting.

Hope S. Rugo, MD: In the ER-positive setting. Yes, triple-negative was across the board because they’re mostly basal. But it’s interesting. You look at the high-2s. There’s more basals in those high-2s as well. So, and of course we see this huge heterogeneity, so it’s quite fascinating. I’ve actually recently been plagued with this group of young women who have bulky disease, lots of nodes, grade 1, ER-positive breast cancer. And in those patients, I think we really need a new approach. And to be giving neoadjuvant endocrine therapy plus something…. Because how can we decide the benefit of chemotherapy in those patients unless we see what hormone therapy does.

Transcript Edited for Clarity