Using Genomic Assays for Early Stage HR+ Breast Cancer

Andrew D. Seidman, MD: In our remaining time, I want to make sure we sufficiently cover questions related to the use of multigene assays for decision-making in early breast cancer and also spend a little time thinking about the adjuvant and neoadjuvant settings for hormone receptor-positive, HER2-negative disease. Nick, talk to us a little about how you use multigene assays in early breast cancer. Do you use 1, 2, 3 assays? How do you think about using them to make optimal decisions for management in the early stage setting?

Nicholas McAndrew, MD, MSCE: They’re very important and can be very useful clinically. Prior to the data from RxPONDER, like Dejan mentioned, it’s important to stick to the data and allow prior research to help us guide specific patient decisions. I tended to use the Oncotype DX assay for my node-negative patients, and I would consider using the MammaPrint for my node-positive patients with 3 or fewer nodes.

However, in the setting of RxPONDER, there are good data that are also predictive for both node-negative and node-positive with the use of Oncotype. I tend to think of these assays, especially the Oncotype assay, as being very helpful in figuring out who definitely needs chemotherapy and who may not need chemotherapy. When I’m stratifying by high-risk or low-risk disease and I’m thinking in terms of whether the patient needs chemotherapy, Oncotype is best used as a de-escalation device.

Some studies have shown that more minds are changed when you look at the clinical pathologic factors that a patient has and you decide what you may want to give them. Then you look at the Oncotype score, and you see what percentage of clinicians changed their mind. More often, clinicians are going from recommending chemotherapy to not recommending chemotherapy, rather than looking at a tumor that seems low risk to them and these assays changing their mind to recommending chemotherapy. While we can get lost in the weeds of patients who have intermediate benefit, at the end of the day, they’re best used as a de-escalation tool.

The most important recent data that have come out from RxPONDER help us understand that patients who are premenopausal and have node-positive disease, pending further subset analyses of the RxPONDER trial, really everyone who is premenopausal should be considered for chemotherapy at this time. Until data are released from further subset analyses of just the microinvasive node population, which I don’t believe has been fully presented yet, I would not use these updates in the premenopausal.

Andrew D. Seidman, MD: Dejan, when a 39-year-old premenopausal woman shows up for consultation, and some other physician has perhaps ill-advisedly ordered an Oncotype, and she has 2 positive nodes and a recurrence score of 14, what do you tell that patient? She shouldn’t have had the Oncotype, or she doesn’t need chemotherapy?

Dejan Juric, MD: Well, let’s start another boxing match.

Andrew D. Seidman, MD: She’s 39 and has 2 positive nodes. Somebody ordered an Oncotype. You don’t want to know it, but she brings you the report and it’s 14.

Dejan Juric, MD: Let me ask you one thing. I’m curious what other faculty think. With the RxPONDER trial, there’s one problem. Chemotherapy has 2 effects: it kills the cancer and suppresses the ovaries. Unless there’s a trial design that answers both of those questions, I’m having a hard time reconciling some of the observations we have in premenopausal vs postmenopausal patients. One could argue that the effect of chemotherapy in premenopausal patients with the benefit seen in the RxPONDER trial could be related in part to the effects on the ovary. A more precise question then would be, what does chemotherapy do in that very patient you described, where we have done ovarian suppression and 10 years of endocrine therapy? I don’t have answer for that. I don’t know about you.

Andrew D. Seidman, MD: I would say until we do have an answer, as Nick alluded to, the default position is to give chemotherapy.

Nicholas McAndrew, MD, MSCE: I completely agree. We’re all suspicious, especially in patients who have very low proliferative disease who do not seem like they would respond clinically to chemotherapy, that the majority of the benefit is going to be from the ovarian suppression aspects of chemotherapy. It was a minority of patients. It was less than 20% of patients who had received ovarian suppression plus an AI [aromatase inhibitor] in RxPONDER. It certainly doesn’t match what the post-SOFT/TEXT [trials] world looks like in terms of treatment recommendations. That’s an unfortunate aspect of the trial.

However, especially in these premenopausal patients in whom you’re trying not to leave anything on the table, and especially in patients with fewer than 3 positive nodes, the benefit of an anthracycline is not necessarily clear in that setting. Doing cycles of TC [docetaxel and cyclophosphamide], while it may be tough, likely won’t leave patients with any long-term problems. There’s potentially neuropathy, but in this younger population, we all want to be a little more aggressive. While we’re not sure if there’s benefit, I agree with Andrew. The default position should be to give chemotherapy until we have more data.

Andrew D. Seidman, MD: I want to make sure we have enough time to get to other matters, but go ahead if there’s a burning comment on this.

Hope S. Rugo, MD: It’s important to keep in mind that for RxPONDER, we’ve just seen the very first analysis where they had very little time from the initial release of the data to the presentation. We haven’t seen any updates since then. Also, a lot of the premenopausal women received tamoxifen, which is quite interesting, given that most of us would choose at least ovarian function suppression and an AI, if we could, for patients who have higher-risk disease. Giving a patient who has 1 positive node and a recurrence score of 2 chemotherapy, including an anthracycline and a taxane, doesn’t really make sense to me.

It’s really important to keep in mind the long-term toxicities of chemotherapy for these patients. Neuropathy is not a small issue. A lot of patients have long-term neuropathy. With anthracyclines, the long-term serious consequences are low risk. But even a low risk isn’t worth it if you don’t need it. We have to be cautious in making global statements.

Transcript Edited for Clarity