Biomarker research in breast cancer

Bookmark and Share
Published: 4 Jul 2017
Views: 1930
Prof Martine Piccart - Institute Jules Bordet, Brussels, Belgium

Prof Piccart speaks with ecancer at WIN 2017 about issues facing translational research, and the need to look beyond individual biomarkers to characterise a whole-cell understanding on cancer.

She outlines her priorities in determining how chemotherapy regimens could be reduced or phased out entirely with immune therapies, and considers how ongoing surveillance of tumour genomics may offer immediate insight into treatment response.

The clinical utility of serial biopsies was also discussed by Dr Daniel Hayes, here.

What I explained in my talk is that fifteen years ago when we discovered the Achilles heel of a subtype of breast cancer, namely the HER2 receptor, and when we started to develop drugs targeting this receptor we were all convinced that we would be practising precision medicine in this particular breast cancer subtype today. In looking back I have examined all the efforts of scientists and clinicians in trying to get really to personalised treatment for these women with HER2 positive breast cancer and I realised that we are not there yet. So we are not delivering precision medicine to these patients but we stratify patients today according to whether they have this HER2 receptor overexpressed, yes or no, and then we treat them with drugs targeting the receptor but we have been unable to identify any new biomarker in addition to the HER2 that is telling us who benefits from these therapies and who doesn’t. So essentially we treat all the patients and then after two months if they have advanced disease we find out whether they respond or not. In the early disease setting where the treatments are given after surgery in a blind fashion we treat everybody because we are unable to tell who is really benefitting from the treatment. So the most important message to deliver at this stage is that we have to revise the way we do translational research in breast cancer and in many other cancers. We have too simplistic views of the cancer cell, we look at one biomarker in one clinical trial and then we derive a conclusion and that cannot be reproduced in another clinical trial. So the message is we need to work more closely together, we need to become smarter, we need to look at pathway alterations and not at single biomarkers and we need to integrate the data generated by genome sequencing, we need to look at DNA but also at RNA and probably also at proteins.

How do anti-HER2 targeted drugs fit into patient care?

HER2 positive breast cancer is clearly a disease where if the diagnosis is made early enough we can cure a large proportion of women with our standard treatment – chemotherapy and trastuzumab. The results are so good today that we are at the point where we need to be thinking who are the patients who could be cured with less aggressive chemotherapy. So trying to use the elegant anti-HER2 targeted drugs as a way to give less chemotherapy to these women, I think that this would be the most significant next achievement in this disease.

Could checkpoint inhibitors play an important role here?

Yes, checkpoint inhibitors could be a way to get rid of chemotherapy in some patients with advanced disease because patients with advanced disease, especially when they went through very aggressive adjuvant treatment, didn’t want to get chemotherapy any more, especially in a palliative disease setting. So they are very interested in the development of therapies that are going to be chemotherapy free. So I’m hoping that we will find a subset of women where we can obtain very long-lasting disease control with the combination of trastuzumab and a checkpoint inhibitor.

Are there any prospects for a curable biomarker in the future?

This is my personal bias, sometimes I think that there is a dimension that we completely neglect in this active search for biomarkers and this is the way the patient handles the drug. So we look at a list of potential biomarkers at baseline before the patient gets the treatment and we believe that we are going to be able to predict with this range of biomarkers whether the patient will do well or not. I’m more and more thinking that the dynamic model of looking at biomarkers could be a lot more efficient. I mean by that that we would expose the patient to the treatment and after a very short period of exposure, two or three weeks, we would then re-biopsy and look at biomarker changes under the treatment and look if that is not predicting in a more efficient way the clinical outcome. So it’s the whole way we do translational research that needs to be revisited because, as I was saying today, this type of work leads to publications but is not really addressing the needs of our patients.