Breast Cancer Update for Surgeons, Issue 1, 2017 (Video Program)MINDACT trial: Utility of the 70-gene signature in selecting patients with BC and 0 to 3 positive nodes for adjuvant chemotherapy
5:05 minutes.
TRANSCRIPTION:
DR GRADISHAR: So this was a large trial, node-positive, node-negative, HER2-positive, HER2-negative across the board, lots of different patients. And that point is important too, because it’s somewhat difficult to draw conclusions you’re absolutely confident about in all these subsets, because they were not huge subsets of patients. So predicting that you don’t need chemotherapy in HER2-positive disease, which might be suggested from the results here, again is based on a relatively small number of patients. So at the present time, I would not defer giving chemotherapy/HER2-directed therapy to somebody based on a low score from MammaPrint, just as an example. But this was a large undertaking — thousands upon thousands of patients. And it was basically looking at a clinical assessment of risk of recurrence using Adjuvant! Online, which is more often than not, not available and looking at a genomic evaluation of risk of recurrence using the MammaPrint and then really focusing on the population where there’s discordance between those 2 assessments: patients who had high clinical risk/low genomic risk and vice versa and then looking at the outcome of patients who got chemotherapy or not in those populations. And they basically make the argument that you can use this tool as an adjunct to clinical assessment, particularly in that discordant group, and by doing so avoid giving chemotherapy unnecessarily, where there’s little added advantage, at least based on prognosis. It’s not really predictive at this point. I would view this as a test of the test. And I think it could be argued, how much additional information do you actually get over the Recurrence Score or other similar tools? But again, it is a molecular assay. And I would point out that about, I don't know, a decade ago, there was a paper in The New England Journal of Medicine that looked at each of these different tools and, number 1, recognized they’re all looking at different genes, but they all came down to the same conclusion about prognosis. So I think that we have a number of these molecular assays that are currently available that can certainly be used for making prognosis assessments. I think at the present time, the ability to predict is still not uniform across all these tests. And the test that has the most validity for predicting the benefit of chemotherapy, in my view, still remains the Recurrence Score. DR LOVE: Are there clinical situations where you utilize the MammaPrint, or the 70-gene assay? DR GRADISHAR: Not yet. We haven’t adopted that as adding a whole lot right now. Now, that said, I think it is Level 1 evidence. This is a big trial. It’s got a lot of patients in it. I think their prognostic information is valid. I think, in order to replicate what was done in this trial, you have to use Adjuvant! Online. That’s number 1. Adjuvant! Online is not available. So making the case that you can use other things to arrive at the same information, I think maybe that’s true, maybe it’s not. But I think it is a tool that could be used to suggest the prognosis of patients. And there are some things that I still don’t understand, having read this paper 6 times, if not more. DR LOVE: I know. DR GRADISHAR: So maybe I’m a simple person, but I just struggled through this. I’m not kidding. DR LOVE: The other thing that I wasn’t really clear about, that I think Cliff Hudis and Maura Dickler got into with the New England Journal editorial, is whether or not it’s really clear that there was not chemo benefit in the patients with the low 70-gene assay, because even though it wasn’t statistically significant, the hazard rate in all these subsets is always lower. It’s not statistically significant, but it’s like 0.7. You know what I’m talking about? DR GRADISHAR: Right. And I agree, because I think, although the benefit might be small, you can’t rule out that there could be a benefit here. And there’s also the subset that didn’t make any sense to me, I think it was the clinical low and genomic high. And again, there was no benefit from — it discounted the value of the test. DR LOVE: Right. DR GRADISHAR: So I don't know what to do with that. |