We’re going to have a discussion with Edith Perez on new generation sequencing and she has been put in the position of saying that this is something that has to be done and I will defend the fact that we have no evidence to say that it has to be done. Certainly today we have many new pathology techniques and I would call NGS one pathology technique that allows us to be more precise in the way we evaluate tumours. In lung cancer, for example, we have an open discussion about the fact that as the cost of NGS has decreased a lot actually several of the different markers that we have to look for are going to be done by an NGS in a less expensive way than with standard immunohistochemistry. So then it closes the debate.

However, in breast cancer for the time being we have little in standard practice to learn from an NGS compared to standard pathology. We have, of course, the whole debate about the genomic tests that are predictive for less chemotherapy; we have also those tests that tell us about the prognosis of the patient but this is not NGS. NGS the consideration is in a patient that has no further standard treatment available are we going to find a treatment that might be useful? In the setting of breast cancer today this wide panel of molecular diagnoses has not shown to be really useful. Yes, we do find sometimes a signal that tells us that a drug can be used; the response rates to these drugs is often dismal compared to what we would expect so we need to have further data in order to convince that we have to ply outside of a trial an NGS. Trials with NGS are not complex, NGS is available in many pathology labs throughout the world, we don’t need to go to the commercial sites to do that and there are many efforts going on in different, for example, European countries. I will, of course, allude to the fact that we have in Switzerland a common platform among all the different pathology labs from the universities and the private labs in order to further understand how to best use NGS rather than doing it and spending money, for the time being, almost for nothing.

Could it not be argued that NGS provides scientists with more open source data to analyse in the future?

NGS run by single centres in most European countries will mean a limited amount of data. In many countries, and not only within the country but also supranationally, there is now an effort to share the results of NGS, of the data on the application actually of the result in order to decide for the treatment with drugs that do not have an approval for the particular indication found by the NGS in order to see are we going to find, not from a single NGS analysis but we have to understand that the tumours most often than not are not driven by one change but by several ones and we have to find a way of doing this analysis and this is feasible with so-called artificial intelligence in many settings. So this is a co-operative effort that should be sustained but, as I will defend, this should be done in studies and not by individuals doing this with commercial companies which, of course, have done a good job but are not going to be, for the time being at least, sharing the data with the initiatives of the universities and private clinics.