• Home
• CME test
• CME information and faculty disclosures
• Select publications
• Related audio program

DR BRUFSKY: This is a lady, she’s 58 years old. She had a 2.2-cm infiltrating ductal carcinoma removed by a segmental mastectomy. She had 1 out of 4 sentinel lymph nodes positive. There was about 2.5 millimeters of tumor in that lymph node. Her H score was 280 for ER. PRH score was 230. Her Ki-67 was 8%. And her HER2 FISH ratio was 1.2 with a copy number of 2.2.

DR LOVE: And she had a completion axillary node dissection?

DR BRUFSKY: No, she did not.

DR LOVE: Hope, what would you be thinking about? And what’s the question you’d like the group to address?

DR BRUFSKY: Would you order a 21-gene recurrence assay on this lady? And if so, how would you act on it?

DR RUGO: We know from the presentation at San Antonio that Ki-67 is correct about 50% of the time. So it’s not all that helpful, although you have a low Ki-67.

DR BRUFSKY: I trust my pathologist.

DR RUGO: It’s negative — that’s good. But I think that one should be cautious in trusting him — and even if you love them. But T2 lesion, HER2-negative, and I think this is a postmenopausal woman, where we don’t know what the benefit of chemotherapy is. And this is the kind of case where I think a 21-gene Recurrence Score or, if you happen to be doing MammaPrint^® more in that setting, is quite reasonable to do. And I think discussing with the patient, I would consider doing a Recurrence Score. This is just the kind of patient that I would do one in who has node-positive disease. And if the patient had a very low score, I would feel comfortable using hormone therapy alone.

DR BRUFSKY: Define “very low.”

DR RUGO: I think fitting into what the standard definition is. What I do, actually, is I look at the risk of distant recurrence and discuss it with the patient. The benefit of chemotherapy in that low-risk group, as defined by Genomic Health, not by the Breast Intergroup, is very hard to see. And if you look at some other analyses, it’s also quite similar, even if you look at NSABP-B-28. I think that all those patients got chemotherapy, so it’s a little bit hard to say.

I think you just have to discuss it with the patient. And these are the kinds of patients that are getting TC more commonly than ACT, where I think the benefit of ACT is going to be very modest. But using a Recurrence Score to try and decide on that seems quite reasonable to me.

Genomic Health will have a new tool available shortly, which is under this “Tool” section. And it’s really interesting. I mean, we’ll have to see how it all plays out. But it comes from the RSPC analysis where you try and put in the clinical/pathologic criteria into the Recurrence Score. That has prognostic value, but not predictive value, since biology matters more than your clinical and pathologic criteria. But what that helps you to see is — actually, I’ve used it, being a beta tester when people email me with patient questions.

“I have an old lady with diabetes and gastroparesis, but she’s a doctor and she wonders if she should get chemo for this low-grade, node-negative but bigger tumor.” It’s designed for node-negative disease, but it’s really interesting because you put in a second category and it shows the number of people who are at risk in purple. You can show 100 people with the number in there, which is nice graphically for your patient. But you can see that what comes out in the Recurrence Score may be substantially lower or higher based on the clinical criteria in terms of risk. Then you have to decide on benefit. You know the benefit is really small. If it’s 20% of a small number, it’s going to be a small benefit.

DR LOVE: In what patients do you not use a Recurrence Score with obviously ER-positive, HER2-negative disease, both node-positive and node-negative, Hope?

DR RUGO: Patients for whom I know from my review of the data and discussion with the patient that we are either not going to use chemotherapy or we are definitely going to use chemotherapy. A young patient with high-grade, ER-positive disease, we tend to use chemotherapy. An older patient who I know can’t tolerate chemotherapy and has a low-grade, strongly ER/PR-positive tumor, I’m not going to get a Recurrence Score.

DR LOVE: So, maybe just teasing it out a little bit more, is there a number of nodes? Obviously, you would do it in this patient, as you said. It there a number of nodes, for example — the RxPONDER trial — up to 3 nodes? Do you follow that?

DR RUGO: In patients who have truly multiple positive nodes, I haven’t sent the Recurrence Score, maybe once in an older patient. But in general, I haven’t. I certainly haven’t sent it in anybody with 4 or more positive nodes. It may have value, but we just don’t have enough data to know how to use it. In 1 to 3 positive nodes, I tend to send it more in patients who have little bits of disease in their nodes, rather than those who have, for example, a replaced node with extranodal extension.

DR LOVE: Again, before I go to Sara, I’ll throw out this question we got from an oncologist: 59-year-old patient with a 2-cm, ER/PR-positive, HER2-negative, 14 positive nodes, with an Oncotype of 6.

DR RUGO: This is one of the challenging situations we face. And that’s why he’s emailing it because probably the benefit from chemotherapy in this patient is low. However, the patient has a very high risk. We have no data in this situation. I would tell the patient that. She’s reasonably young and healthy. And I would recommend our standard ACT regimen, for better or worse, because we don’t have any data that helps us not use it in this situation.

DR LOVE: Sara, how would you approach Adam’s patient with 1 positive node? Again, what are your general parameters for using Oncotype nowadays?

DR HURVITZ: I wasn’t a huge fan of Oncotype. And then when the IHC4 data came out, I felt sort of justified in not being a huge fan of it. That said — and the reason is, because I do trust my pathologist and the 21-gene Recurrence Score is heavily weighted. The most heavily weighted feature is Ki-67.

DR BRUFSKY: It’s proliferation.

DR HURVITZ: Proliferation, but genes related to proliferation, which can be roughly approximated by Ki-67, which we know now is difficult to reproduce. What has happened in my practice is patients have come in and it’s become a standard of care just like sentinel lymph node biopsies did once upon a time because patients demanded to have it. I don’t think that it changes my practice in the majority of cases, but I don’t feel it’s inappropriate to check it in a patient like this because it can inform treatment decisions.

DR LOVE: Is there any number of nodes where you just flat out would not do it?

DR HURVITZ: I agree with what Hope said. I probably wouldn’t do it in multiple nodes, except for an unusual situation.

DR LOVE: What about Adam’s patient, assuming the patient is not pushing you, who just says, “What should I do?” Would you generally do it?

DR HURVITZ: I think it’s not unreasonable to do it. I would discuss it with them. I would bring it up if the patient didn’t bring it up.

DR LOVE: I always love that term, “not unreasonable.” Bill?

DR GRADISHAR: At the center of your crafting here is if the patient is asking for it. It gives me the impression, like most, they’re not too eager to get chemotherapy. If you’re trying to discuss with them what the merit of chemotherapy is, I think it would be rational to do it and I would think about it.

Now, the limitations, I wouldn’t do it in multiple nodes. I’d do it where there was minimal nodal involvement. And I wouldn’t do it when there’s 16 nodes.

DR LOVE: What about if there’s a large node-negative tumor? Is there a size that gets you to not do it?

DR GRADISHAR: Again, I think most of the data, when you look at the validation studies, were roughly up to 4 cm. I think people start getting squeamish before then. But the validation was roughly up to 4 cm. I think as the patient gets younger and the tumor gets bigger, the squeamishness gradient goes up.

DR MORROW: Neil, could I ask a question? We actually order the Oncotype out of our surgical practice based on an agreement made with our medical oncologists that anybody who had a greater than 5-mm, node-negative, HER2-negative, ER-positive cancer up to basically 5 centimeters in size — so, T1 and T2  — we would get the test. And we do that. But I’m very interested listening to all the people who would do this in a patient with a single positive node because if everybody’s already accepted that, then why are we doing this clinical trial, which is undoubtedly going to be very heavily subscribed by patients with 1 positive node, if people have bought this already?

DR CAREY: Not everybody’s bought that.

DR BRUFSKY: Not everybody’s bought it.

DR MORROW: Oh, I know, because we haven’t bought it at all.

DR CAREY: I don’t do it in macro mets.

DR RUGO: 2.5 millimeters? Macro mets.

DR CAREY: In truth and as we’re getting evidence based, the number of patients contributing to this particular analysis in SWOG-8814 is on the order of 150. Right? There’s a reason we’re doing a large fancy trial in the node-positive thing. I have no doubt — absolutely no doubt — that there is a biologic contribution to chemotherapy benefit and that there will be a threshold for node-positive. I just don’t know where it is.

DR BRUFSKY: We’re getting caught up in old anatomical definitions that our surgeons have given us, which have worked very well for 50 or 60 years, but we now have these tests, which really test the biology of the tumor. Now, which is more important, the biology or the anatomy?

DR CAREY: No. We’re getting caught up in magnitude of effect.

DR BRUFSKY: I agree with you.

DR CAREY: There’s 2 ways for this to play out. One is through a differential effect. The other is through a variation in the magnitude. And there is a point at which you start to see a difference simply because of —

DR LOVE: Let me just clarify to Monica as a follow-up in terms of your institutional policy, what about node-positive? What’s your institutional policy?

DR MORROW: For patients with macro metastases, we don’t order that, period. If there is a rare individual case where the medical oncologist wants it based on conversation, they will get it.

DR LOVE: Are you putting patients on the RxPONDER trial?

DR MORROW: Yes, we are.

DR BRUFSKY: Which is the right thing to do. It is. On the other hand, I suspect that RxPONDER is going to show, in the intermediate group, no benefit.

DR LOVE: Edith, again, what would you be thinking about in Adam’s patient with 1 positive node? And generally, how do you approach both node-negative and node-positive in terms of this decision?

DR PEREZ: We typically look at the data with node-positive disease as very early, kind of almost pilot data from the SWOG database. We support the RxPONDER trial, but if a patient is not on trial, routinely we would manage this patient with chemotherapy. The patient has node-positive disease. I think the stakes are so high, if we make the wrong decision in terms of recurrence and death, that I think it sways us to be a little more aggressive in the adjuvant setting.

DR LOVE: Hope?

DR RUGO: It’s interesting listening to everybody because I think that we all do things in a fairly similar manner. And then there’s this little variations on the theme. Because I think that I’m somewhat surprised, given my institution where I’m generally aggressive, that in this group I seem to be less aggressive. I do believe that biology drives benefit. And if you look at recurrence risk for patients who have a little bit of node-positive disease that’s strongly ER-positive, it’s spread out over a long period of time. And at least 50% of the recurrence risk is late. I think that we don’t see very early recurrences in extremely low-grade, strongly ER-positive, hypoproliferative disease.

I feel comfortable in those situations where patients have minimal node involvement. And I think this idea of macro versus micro, when it’s point-something millimeters, it is just rules that we made because it’s easier for us to talk to each other. I think individualizing it for me has worked. And I do think we have to use a lot of caution. I agree with that 100%. There just isn’t enough data. On the other hand, biology drives response. I guess I fell somewhere in the middle.

DR PEREZ: I think, if I may, biology may be important, and it is important. But I tell you, amount of tumor is also very important.

DR RUGO: Yes. For prognosis.

DR PEREZ: It has to be a combination of both. And we’ve clearly seen this in the HER2-positive setting. I think that’s a great setting to look at this issue. And where it differs —

DR RUGO: But that’s not the situation where you have a strong predictive factor for benefit.

DR PEREZ: Yes.

DR RUGO: In this situation, size doesn’t predict benefit. It predicts risk. So, what you’re saying is that when people have a lot of cancer, we’ve all agreed even if somebody had a score —

DR PEREZ: It’s worse.

DR RUGO: — of 1, if they had 17 positive nodes, we’re going to give them chemotherapy.

DR PEREZ: Right. Right.

DR RUGO: Because the risk is high. Even a little benefit might be enough.

DR PEREZ: Yes.

DR RUGO: But if the risk, relatively, is lower, then — and the benefit isn’t going to really change a whole lot, right? It’s a percentage — it’s got to be that. Right?

DR LOVE: Adam, what happened with this patient?

DR BRUFSKY: This lady had an Oncotype of 11, or a 21-gene recurrence assay of 11. And we gave her anastrozole.