AACR 2016

MINDACT breast cancer clinical trial: First results and impact on future trial design

Dr Martine Piccart - Jules Bordet Institute, Brussels, Belgium

This year I have presented this very important European trial, MINDACT, the results of which are quite exciting.

Perhaps it’s good to start by saying that breast cancer oncologists know that they are over-treating women with early breast cancer and probably the reason for that is the fact that advanced breast cancer is an incurable disease by and large. So we feel that we have a huge responsibility at the time we see women with early disease because we know that this is when we have a chance to cure them. Maybe it’s because we are so anxious that we over-treat women and over-treatment with chemotherapy is particularly worrisome. It happens when the benefit that you can expect from prescribing chemotherapy is a 2-3% improvement in survival, a small benefit which happens when the woman has a relatively good prognosis and that is going, of course, to be counterbalanced by the risks of some of the severe toxicities of chemotherapy like secondary cancers and congestive heart failure.

For some women, of course, the benefit of chemotherapy will be much bigger, if it’s up to 10% then you are certainly willing to take some risks but not if the benefit is small. That is a problem that we wanted to address back in 2004 and we became very excited when a new tool emerged to help us with treatment decision making and this tool was the 70-gene signature discovered by a group of outstanding laboratory scientists at NKI in Amsterdam led by Laura van 't Veer. So this signature, which was published in The New England Journal of Medicine was really impressive to us because it appeared to really predict extremely well the women who would not relapse, who would do very well without chemotherapy, from the women who, on the other side, would have a much worse outcome. We decided that we would work on this biomarker and try to validate it and that’s also another important concept. There are a lot of biomarkers published in the medical literature and very few are validated. What that means is that the clinicians really have very few biomarkers at their disposal to help them with treatment decisions. So what we really did with MammaPrint was all the work that you need to do to validate a biomarker, starting with an independent validation of the work of the Amsterdam group. We did that, that was the very first concrete step forward. Then we also demonstrated inter-laboratory reproducibility of the signature and then we really got excited and confident and we were convinced that the signature was going to outperform clinical criteria that we use routinely to decide whether we should give chemotherapy or not. The hypothesis was that the signature would reduce the rate of chemotherapy prescription.

So the trial is really unique, there is not another trial of this kind because it really confronts tumour biology and tumour anatomy. So we embarked then on this large trial which recruited 6,693 women across nine European countries, that was between 2007 and 2011. In the middle of the study we amended the protocol to also allow women with one to three positive nodes to enter the trial and that’s because we were able, again in an independent study, to validate the signature in these women. Of course after that we had to wait for mature results and now that we have a median follow-up of about five years on these women we are totally confident that our initial hypothesis has been verified. What is important to know is that the primary statistical test in MINDACT focusses on one particular group of women who are considered high risk from a clinical point of view and low risk by the signature. So this group included about 50% of women with node positive disease, many of the women, about 29%, had grade 3 tumours, and in these women we could demonstrate that if we trust MammaPrint, we don’t give chemotherapy, the outcome at five years is really very good. So 95% of these women have not developed distant metastasis and that is independently from whether they got chemotherapy or no chemotherapy. That is the really important finding of the trial because it means that if we are now going to use MammaPrint, and I hope we will, in the many countries where it’s not yet approved that for the women who are clinically high risk if it happens, and that’s going to be the case in half of them, that the MammaPrint test is good risk we will no longer give chemo. So that’s a 46% reduction in chemotherapy prescription. Very important progress.

What are the challenges you’ve faced?

I can tell you it is the story of treatment de-escalation and it’s so difficult to demonstrate that de-escalation of treatment is completely safe in oncology. It’s much easier to escalate, to add on the drugs, but of course then we over-treat a lot of patients and the cost to society of these expensive complicated treatments is pretty significant and problematic. So these are very good news but again I want to stress the fact that this required a huge collaborative effort between clinicians, patient advocates, scientists and it took us twelve years in total. But what is very exciting is that we have also a fanstastic biobank and we are also going to be able to look at the full gene expression profile of these tumours, not only the 70 genes, in fact we assayed 30,000 genes per tumour. That is going to be a goldmine for future research in breast cancer.

What does the future hold?

News are really very good for women in Europe because there are very few European countries in which these prognostic gene signatures, whether MammaPrint or other signatures, are reimbursed. In fact, MammaPrint is currently paid for in the Netherlands, Germany and part of Spain and that’s it. In my country, Belgium, it is not.

Oncotype-DX is currently reimbursed under certain conditions in the UK but you can see that in many European countries none of these tests can be prescribed free of charge. That’s going now to change, hopefully, with the results of MINDACT.