Breast Cancer Update for Surgeons, Issue 1, 2017 (Video Program)Results of a prospective registry of patients with early BC for whom treatment decisions in clinical practice were made incorporating the 21-gene RS
5:45 minutes.
TRANSCRIPTION:
DR SPARANO: This was a prospective registry conducted in Israel of a healthcare system that included 930 patients who had ER-positive, node-negative disease and who were prospectively followed. You see that the results of the Oncotype did drive decisions about the use of chemotherapy in that, if the Recurrence Score was less than 18, only 1% of patients received chemo. If it was 18 to 30, 28% received chemo. And if it was at least 31, 85% of patients received chemo. The figure shows the distant recurrence rate by Recurrence Score group at about 5 years. And what it shows is that, if the Recurrence Score was low — and again, all these patients, the vast majority of these patients, just got endocrine therapy alone. And this is a higher Recurrence Score cutoff than what was in TAILORx. There was only a 0.5% risk of distant recurrence at 5 years, very low, not likely a recurrence rate that’s high enough that one would benefit from chemotherapy. If the Recurrence Score was in the intermediate range — and again, this is a higher range than TAILORx — of 18 to 30, the distant recurrence rate was about 2%. And that did reflect a mixed population of patients, most of whom actually — about 70% of whom got endocrine therapy alone, about 30% of whom got chemo/endocrine therapy. And there was about a 4% risk of distant recurrence in the high Recurrence Score group despite the use of chemo and endocrine therapy in the vast majority of those patients. DR LOVE: Now, these patients were node-negative? DR SPARANO: These patients were node-negative. So this is another prospective study confirming the prognostic utility of the Recurrence Score, particularly low Recurrence Score. DR LOVE: But it is interesting, because people have commented on the New England Journal paper of the TAILORx that you had with the low Recurrence Score, but there, it was what, 10? The cutoff was 10. And here you’re seeing the same thing with the more typical Recurrence Score of less than 18. DR SPARANO: Yes. There’s been considerable discussion and questions about why was the Recurrence Score cutoff set at less than 11 for the low-risk registry? And we chose it for a couple of reasons. Number one, we wanted to design the trial in a way that minimized the potential that we were going to be undertreating patients. And I think there was a real question for patients who had a midrange Recurrence Score about the benefit of chemo. Remember, this is back in 2004 or ’05, where chemotherapy was the standard of care. So how did we pick 11? A Recurrence Score of 11 in the B-14 data set was associated with about a 7% risk of distant recurrence. The upper bounds of 95% confidence level is about 10%. So that’s a threshold that most clinicians would recommend adjuvant chemotherapy. So that’s why we set less than 11 as the threshold and didn’t pick a higher threshold. So I think that’s a key point. We’ve described the rationale for that in a publication, a review article by myself and Soon Paik, in JCO in 2008. And I think it’s important to understand what the rationale for that is. The other point that’s also raised and discussed in that JCO paper, review article, is that when the NSABP reanalyzed their data using these cutoffs that we used in TAILORx, less than 11, 11 to 25 and greater than 25, that the 10-year disease-free survival was about 95% in that midrange Recurrence Score group using that cut point, irrespective of whether they got tamoxifen or tamoxifen plus CMF chemotherapy. This is the B-20 study. So that provided some additional evidence and reassurance that this range that we chose for the randomization was appropriate. DR LOVE: From your own personal perspective in your practice outside a trial setting right now, what do you consider a low Recurrence Score? DR SPARANO: Definitely less than 11. But I would say that I don’t use and I don’t think clinicians should use a Recurrence Score alone in making clinical decisions, that one factors in other decisions such as the age, the tumor size, the grade, even the patient’s concern about recurrence. There is a model that integrates this information, the RSPC score, which is a tool that’s available, where one can integrate the other clinical features in a model that has been validated — it’s been developed by NSABP and validated — that can provide more accurate estimates about what the true risk of recurrence may be, because it takes into account not just the Recurrence Score but the clinical features as well. |