If you look to the pathway that is influenced by an EGFR inhibitor such as cetuximab but also panitumumab you see that other factors might play a role in the outcome of the patient. That is the reason why we looked for other factors outside the KRAS gene influencing progression free survival and overall survival, not only in the first line setting, based on the PRIME study where patients were randomised between FOLFOX and FOLFOX plus panitumumab, but also in the second line study where we looked to a study where patients were randomised between FOLFIRI and FOLFIRI plus panitumumab. What we have seen, and this is true in the literature that is generated not only with panitumumab but also with cetuximab, that around 17-18%, of additional mutations are found outside the KRAS region exome 2. The novel thing in this study is that we looked on the impact of the outcome and you clearly see that on overall survival in the PRIME study and on progression free survival in the second line study there is an impact in favour of better selection of the subgroup that have a RAS wild-type instead of a KRAS wild-type status.

If extended RAS is also wild-type will they do even better?

You have a clear split between the curve that is treated with chemotherapy alone. If you extend your RAS analysis and you look to the wild-type the benefit is bigger in favour of the combination and, looking to the mutated population, there is clearly no benefit of treating these patients with anti-EGFR. So we moved from a situation where it was 60% wild-type KRAS, 40% mutated, no additional benefit of anti-EGFR in the mutated to a situation of 50% wild-type versus 50% mutated. So you select the better population and you protect the population that have no benefit.

What’s the take-home message for doctors?

Already in Europe we changed the label of these drugs towards testing for RAS and not for KRAS so we need that urgently also in the US. Second, we moved, again, in the direction of real, personalised medicine, we redefined the sub-groups up front which are having benefit versus no benefit and I think this is really a step going to individualised medicine in the colorectal cancer scene and we have to invest more and more in this type of research.