CM: Welcome to this ecancer event where we will discuss the last data on TAILORx and RxPONDER trials presented during the last San Antonio Breast Cancer Symposium. I am Christos Markopoulos, I am a breast oncoplastic surgeon, Professor of Surgery in the Athens Medical School of the National and Kapodistrian University of Athens in Greece. I have the pleasure to have with me Professor Kalinsky.
KK: Hi, it’s nice to see you. I’m Kevin Kalinsky, I’m a breast medical oncologist at Winship Cancer Institute and Director of the Glenn Family Breast Center here at Winship.
CM: Hi Kevin, I’m very happy we have the opportunity to discuss all this data. So let’s start with the TAILORx and let me summarise what we knew up to date from the TAILORx trial. It was really a practice changing trial which randomised women having a recurrence score between 11-25, that were all lymph node negative, ER positive, HER2 negative patients. The results were that for women over the age of 50 and a recurrence score of 11-25 there was clearly no benefit of the addition of chemotherapy. As for younger women, less than the age of 50, those in the group of recurrence score 11-20 had a minimal benefit, almost no benefit in those who had low clinical risk, some benefit for those who had high clinical risk. But for the group of recurrence score 21-25 there was no benefit irrespective of the clinical risk.
However, we all know that we are talking about hormone receptor positive disease and we know that these patients tend to recur lately, after the following five years, with a peak at the area of 7-8 years. So we were all waiting on this TAILORx update which would you be kind to present us the results?
KK: Yes. So that really explained well just how practice changing and informative this study was. You mentioned the significant issue that we see with this subtype of breast cancer can be late recurrence. So we saw updates a little beyond ten years for patients in TAILORx and there were three main points. The first is when you look at the overall outcomes, the overall population, that the data remained the same, meaning those patients who had recurrence score 11-25 if you look at the overall population and you look at the various outcomes, whether it’s invasive disease free survival, overall survival etc., there were really no differences whether patients received chemotherapy or not.
That was point number one. Point number two is we had the exploratory analysis looking at those patients who were age 50 or less. We had seen from the initial results from TAILORx, as you mentioned, that there could be some benefit for patients who had a recurrence score between 16-25. With additional updates, again this is an exploratory post-hoc analysis, if you look at those patients who had low clinical risk and recurrence score 16-20 there was really no benefit for chemotherapy for those patients. If you had recurrence score of 16-20 and your clinical risk was high, your absolute benefit of chemotherapy is about 3% but also when you look at those data the standard error is pretty wide, it’s about 5.4%. So that’s helpful just for explaining to patients what the potential benefit of chemo could be. As you mentioned, the patients who had a recurrence score of 21-25, regardless of their clinical risk, there was benefit even though numerically there was a slightly higher benefit if the clinical risk was high in that subgroup.
Then the third point that I was going to make was about when we looked at race and ethnicity data because originally when TAILORx was reported, these data were published by Kathy Albain on behalf of the TAILORx group, we’d seen that non-Hispanic black women seemed to do worse compared to other ethnicities like non-Hispanic white women. But with additional data and follow-up that finding was no longer the case. So when you look at those patients after five years there was no difference based upon race and ethnicity.
CM: It’s very interesting and the last point you mentioned I think we will have the opportunity to come back at the end when we discuss the RxPONDER. So, following this update, could we say that we feel now more confident dealing with ER positive disease to involve recurrence score in our treatment decisions? Also I absolutely agree with your point for the low clinical risk women with a score 11-20 and probably even with a high clinical risk between 16-20 we could discuss and find the balance between short- and long-term side effects of chemotherapy in relation to the small benefit they get.
KK: Yes, overall it’s nice to see that these findings remain consistent. Based upon the Early Breast Cancer Trialists Group we had seen that the benefit of chemotherapy oftentimes was within those first five years. So to see that the trends remained the same after five years, that is informative for our patients. Your point about those patients who have a recurrence score of 16-20, even those patients who are high clinical risk, we do have to ask ourselves whether that potential 3% benefit is worth chemotherapy.
Of course, the other question that we really have is what about ovarian function suppression and hormonal therapy – can you get the same sort of benefit? There is a study that should be opening in the Cooperative Group relatively soon through the NRG that will be looking at patients, including in the subgroups we’re talking about, including in the RxPONDER population, comparing ovarian function suppression and hormonal therapy with or without chemo. Because this remains an important question of whether the benefit is just due to the fact that the patients are stopping having their periods.
CM: Yes, and this is very, very important, especially for women at the age of 48, 49, very close to menopause, if they can skip chemotherapy. Let’s move now to the RxPONDER sub-studies. There you asked a very interesting question about race and clinical outcome and this is important because we know that although black women have a lower incidence of breast cancer, they had about 40% higher mortality. So exploring why this is happening is very, very important in clinical practice. So would you like to present us what you found?
KK: These were data that were presented by one of our junior colleagues, Yara Abdou, who is at the University of North Carolina. What we saw was, one, that about 6% of our patients were non-Hispanic black which is consistent across various studies, that this remains under-represented in our clinical trials, this population remains under-represented. We saw that despite the fact that non-Hispanic black women, compared to non-Hispanic white women, were more likely to accept their randomisation. Despite the fact that the rate of adherence at one year to endocrine therapy remained the same between the two subgroups, we saw a worse invasive disease free survival as well as distant relapse free survival if we compare non-Hispanic black to non-Hispanic white women.
What we discussed in the study is that we don’t fully understand why this is existing. The magnitude of how worse it was slightly diminished when we incorporated body mass index into those analyses. But this does ask the question of whether this is related to social determinants of health or whether there are differences in tumours between races and ethnicities. We also had mentioned in the presentation that non-Hispanic black women were more likely to have, for instance, higher grade tumours. But that really remains exploratory but this is something that was also seen, as I mentioned, early on in TAILORx and something that is critical for us to understand why we are seeing these differences in various outcomes.
CM: Yes, this is very important because they did follow even better treatment and everything compared to non-Hispanic whites. Of course, the higher percentage of grade 3 tumours could be an explanation and probably leading to a hypothesis of biology behind all this explanation. In my opinion, it also emphasises the fact that we should focus research on this population to explain because the difference is very, very important.
Let’s go now to the other sub-study when we had patients’ self-reported outcomes with regards to cognition. Because there you also studied a very interesting parameter which is closely related to quality of life of patients, especially for premenopausal patients who have more implications. Could you give us also your results on that?
KK: So this was also presented by one of our junior faculty colleagues, Irene Kang, who reported there was a sub-study within RxPONDER where we followed different patient-related outcomes longitudinally. So we had baseline and then up to three years. Again, it was a smaller population of the overall study and we looked at cognition. Cognition was based upon some PROMIS score analyses. We saw when we compared baseline to three years that premenopausal patients, when you look at their baseline compared to three years, those who received chemotherapy as opposed to just endocrine therapy, you can see the decline over six months and then 12 months. But you can also see that compared to baseline, two, three years, even though it gets a little bit better from 12 months to three years, that still patients are still having cognitive effects, worsened cognition, at three years. When you look at the postmenopausal population the same sort of trend holds true but the magnitude of the detriment is not quite as bad.
But to me, really the take home from these analyses is we know that patients feel different cognitively when they receive chemotherapy. It’s something that has been described as “chemo brain” that patients experience. What is important, the important take-away from this analysis, is that those effects can remain three years later. So that is something that as a field we need to be really aware of and then, too, come up with some mitigation strategies about how to help these patients. Then also it reinforces that we should be really thinking about who needs chemotherapy or not.
CM: I would agree with you but the only thing is allow me to say that these are based on a small number of patients, especially for premenopausal. Probably we should focus also on two important parameters related to this – adherence to therapy and the other thing that we don’t know how many patients through the three years of follow-up from premenopausal became postmenopausal and probably we have another added effect.
KK: If we look at the overall population compared to this population, the rate of ovarian function suppression was about the same. But you’re right, these were some limitations that we had highlighted in the presentation, just about the small sample size, there were some dropouts that had happened along the way. Then we don’t have long-term adherence data for these patients and that is something that we will be looking into.
CM: The last thing I would like to ask you, because I was also not surprised but I was expecting it to be different, we had parallel sub-studies like those you presented in TAILORx and RxPONDER. We have slight differences between the results – do you have an explanation for that?
KK: In terms of the race ethnicity or in terms of cognition?
CM: Actually we have in both, if I remember well, in race and ethnicity and for premenopausal.
KK: Yes, for race and ethnicity we had seen those differences between non-Hispanic black and non-Hispanic white women early on. The question is going to be within RxPONDER when we look after five years does that difference maintain itself or not. The truth is it’s not entirely clear to me in TAILORx why that difference was no longer the case. That will require some further explanation and we will see whether that holds true in RxPONDER.
In terms of the cognitive differences, our colleagues in TAILORx also have looked at differences in cognition and at three years they didn’t see the same sort of effect. We did look at different patient-reported outcomes so that may help explain some of this, that these were questionnaire based. Also, as you highlighted, these are smaller populations and sometimes it just happens due to chance but that may be another explanation.
CM: Yes, thank you very much. I don’t know if you have anything else to add? I really enjoyed our conversation. However, all these results are very encouraging, especially from the TAILORx. From sub-studies of RxPONDER I believe that they show us the way of what to study next because quality of life and outcome are equally important. So thank you very much for the discussion we had and I would like to thank again ecancer for giving us the opportunity to discuss it today.
KK: Thank you.
