Expanding Therapeutic Options for Breast Cancer - Episode 7

Patient Selection for Adjuvant Therapy in EBC


Joyce A. O’Shaughnessy, MD: Thank you for that great discussion about ER-positive metastatic breast cancer. I think we hit on all of the high points. Let’s now turn to ER-positive, early-stage disease, particularly node-negative disease, based on the enormously long-awaited TAILORx data—the highlight of the ASCO Annual Meeting. Hope, would you summarize that for us? What’s your take on this? How will it impact our practice?

Hope S. Rugo, MD: Joe Sparano presented the TAILORx data at the plenary session. Of course, it’s always a big thrill for us, in breast cancer, when we get plenary time. And the data published on the same day in the New England Journal of Medicine. It was very exciting to see that data, which I think impact clinical practice right away. TAILORx was a very large phase III trial that randomized patients with intermediate-risk recurrent scores to receive endocrine therapy or chemotherapy and endocrine therapy. The unique characteristic of the trial is that the scoring system is different, although that will probably change based on TAILORx’s data. Less than 11 have already been reported and published. Those patients did great with endocrine therapy alone. Remember, these are node-negative, ER-positive patients. These are patients with node-negative, ER-positive breast cancer.

Patients who had a score from 11 up to 24 had intermediate-risk, and those patients were randomized. The 25 and higher group were high risk, and those patients were suggested to get chemotherapy. Obviously, people could do what they wanted in those groups. They weren’t randomized.

The patients in the intermediate-risk group had really low-risk tumors. The median tumor size was 1.5 cm, with a range of 1.2 to 2. So these were small tumors. All except 13% had grade 1 or 2 disease. Almost 90% of the patients had grade 1 or 2 disease. The patients fit into a category that was defined by the MINDACT trial, with a MammaPrint assay as clinical low-risk. So there’s this clinical scoring system where you say, “OK, the tumor’s clinical low-risk.” Three-quarters of the patients in the intermediate-risk group were clinical low-risk.

When we think about this trial, we have to think that the patients are patients who we would categorize largely as low-risk patients who will do very well, in general, with endocrine therapy alone. In fact, that’s what they showed. There was no benefit in the overall trial population regardless of the endpoint and base of disease-free survival since recurrence-free survival, etc. Any disease-free survival subset that you looked at was identical.

It was very encouraging to see that the freedom from distant recurrence was about 95% in both arms, which is great for our patients, and that the current treatment works really, really well.

Reverse of what we might have predicted, 18% of patients who were randomized to receive chemotherapy chose not to get it, which is interesting. They must have had small tumors and have said, “No, I’m not getting it.” Then, about 5% of patients who were randomized to endocrine therapy alone decided to get chemotherapy.

Then they looked, in a step analysis, to see if they could figure out a subset of patients who could benefit from chemotherapy. There was some suggestion that if you looked at the higher end of intermediate risk and you looked at patients who were younger, there may have been a benefit. So they looked at the combination of recurrent score and age. Women under the age of 50, who had a score of 16 to 25—that sort of higher end of intermediate—seemed to have some benefit. Again, that wasn’t a prospectively defined subgroup, so it’s just a suggestion. It’s hard for us to take that back to the clinic right now. It’s fascinating data. It’s practice changing. In general, women who have that newly defined intermediate-risk score, who have tumors that are low in intermediate grade—2 cm or lower—don’t benefit from chemotherapy.

The younger women may have had a possible benefit, in part because of the ovarian suppressive activity of chemotherapy in young women. Based on the SOFT and TEXT trials, we know that does seem to be very important for young women.

We just don’t know. I think that we’re in the same situation. For a patient who has discordant clinical features, clinical pathologic features, and a recurrent score, who seems to have what we would feel is high-risk disease, we can still give chemotherapy. But for the vast majority of patients, we shouldn’t give them chemotherapy.

Joyce A. O’Shaughnessy, MD: Thank you. What a great summary. Thank you so much, Hope. What do you think, Komal and Sara, about this?

Komal Jhaveri, MD, FACP: We were certainly awaiting this data, because they have the most practice-changing implications that we’re going to bring to clinic. But again, I think it proves that patients with a low-risk case that we thought was low—those patients having an intermediate value that was defined by TAILORx—will not really derive benefit from chemotherapy.

We worry about those patients who potentially have those characteristics, and then they have the score. In the less than age 50 group, the younger women, those 16 to 25, that’s where chemotherapy will perhaps have some benefit. So, I think it’s reassuring that way.

Sara A. Hurvitz, MD: But it leaves some questions unanswered. I hate to be devil’s advocate here. As Hope pointed out, very low-risk patients went on this study. I’m not sure that this really changed our practice, because I think we are pretty good at identifying based on grade, Ki-67, coexpression of PR (progesterone receptor), and other factors that have been shown, in a much less expensive way, to the patient or insurance, to be predictive of outcome. It doesn’t tell us what to do with a 4.5-cm tumor. It doesn’t tell us what to do with lymph node—positive disease.

Community oncologists and many of us will say, “Let’s just do the score. I really feel like this lymph node—positive disease is a well-behaving disease. If it’s a score of 18 but is lymph node–positive, I’m going to try not to give her chemotherapy.” We can’t interpret that data there. We have to be very careful how we interpret the data here, and I think it’s tricky. I don’t want to overtreat. The power of these data is that they give us more support to avoid chemotherapy in disease that’s not chemo-sensitive. But I would also caution against overinterpretation.

Debu Tripathy, MD: The big question is, do you believe that gene profiling trumps the other characteristics—size, grade, and age? When the original retrospective data was presented from Oncotype from the NSABP (National Surgical Adjuvant Breast and Bowel Project) trials, it appeared that all of the independent variables that had predicted outcome in the original analyses of these trials no longer applied when you factored in the Oncotype score. Also, when they reported the low-risk group (under 11), those patients did very well. The distant disease-free survival was 99%. Ten percent of those patients actually had grade 3 tumors, but the ones with low scores also had an equally good outcome. That would suggest that it does trump grade. But I share everyone’s concerns that oncologists and patients acted the way they normally would. They sort of selected out the lower-risk patients to go on this trial because they knew that they could get randomized. They had to feel comfortable that their patient was just going to get endocrine therapy.

Hope S. Rugo, MD: I always put a fair bit of stock into the clinical pathologic criteria, and I was interested in the RSPC (Recurrence Score-Pathology-Clinical) work that went on. It was not perfect, but adding the pathologic and clinical criteria changes your prognosis, as we know. You could have a low score. If you have 9 positive nodes, you still have a high risk of recurrence. It doesn’t mean that chemotherapy is going to change that. What we’re learning is that prognosis and benefit are obviously a little bit different, as well. It was interesting. In the recurrent score high-risk group that largely got chemotherapy, those patients had a worse outcome than the intermediate-risk patients. So, it does have a big prognostic value. We just need to learn how to best use it for determining who needs chemotherapy.

Sara A. Hurvitz, MD: One of the real powers of it is that it’s a centralized analysis of tumors. It’s leveling the playing field for pathologists. It’s putting less pressure on them to interpret in a uniform way. I think that gives us some power in helping our patients make decisions.

Joyce A. O’Shaughnessy, MD: And for those patients who are in that grade 2 area, where we really do have some uncertainty, I would historically split the difference. If they were above 20, I’d give them some chemotherapy. Now, I have more data. I have to do that. If they fit the criteria of the majority of the patients in the trial, and they come up to 25, those curves are on top of each other for all of those key endpoints. That is helpful. It’s not every answer. No trial gives us every answer, but it really does help. We do struggle. We see those larger grade 3, very high Ki-67 cases. We’re getting chemotherapy. We’re not sending a recurrent score on those. But for the ones that we’re sending, I’ll bet that we are questioning the value of the chemotherapy. It’ll be really interesting. I hope we get a deep dive into those less than 50s. Maybe they were young? How old were they? What was their endocrine therapy?

Sara A. Hurvitz, MD: Right.

Joyce A. O’Shaughnessy, MD: Yes, exactly.

Hope S. Rugo, MD: The other exciting thing is that we’re going to see data. Node-positive disease was included in MINDACT. It was designed a little bit differently, but RxPONDER is a really interesting study looking at node-positive disease. It’s a big question now. I saw a patient who had a 4—node-positive lobular cancer with a recurrent score of 10. What do you do with that patient? It’s very, very difficult. Most of us would give some chemotherapy, but we really have no idea if it benefits those patients.

Joyce A. O’Shaughnessy, MD: Yes. Thank you. Thank you for that discussion.

Transcript Edited for Clarity