European Multidisciplinary Cancer Congress (EMCC) 2011, 23-27 September, Stockholm

Key gastrointestinal cancer research from EMCC 2011

Professor Eric van Cutsem – University Hospital Gasthuisberg, Leuven, Belgium



Eric van Cutsem, thank you very much for just dropping out of ECCO for two minutes. Lots of hot topics and you’re leading most of them in GI cancer, what’s the headline, as you see it, from your ivory tower?

Well this ECCO/ESMO meeting is very interesting, we see a lot of good things, we see some new aspects. If you look at the major aspects, they are that we learn more about best strategy in colon cancer and other GI cancers but there are more data on colorectal cancer than on other GI cancers at this meeting; on the selection of patients; on the KRAS story but especially ones beyond the KRAS story. We’ve seen this morning, at the colorectal cancer session, interesting data from different groups, for instance the UK group who presented the PICCOLO study showing that if you look for after KRAS in the KRAS wild-type patients, so for BRAF, PI3 kinase, NRAS, that this may further increase the benefit of patients after treatment with panitumumab. So just not yet a final proof that we have to do this different mutation analysis but the data, at least in chemo-refractory patients, are pointing us in that direction. And that’s important because if you administer an anti-Gr antibody we know that also in the KRAS wild-type patients that not all patients will benefit from the treatment. So we have to learn more, which patients will not benefit and which patients will benefit.

We saw also some interesting data on circulating tumour cells looking at bevacizumab from the Spanish group. It also is that KRAS mutant patients have a worse prognosis than KRAS wild-type patients, so some of these aspects are very important.

Has anybody picked up or been able to do KRAS mutations in circulating tumour cells?

Not in a consistent way.

Because that’s going to be a clever way in future of doing the mutation analysis which you’ve just recommended.

Correct and this could be much easier. But I don’t think that’s a major issue because if this method would not be 100% trustful and not 100% comparable with doing a mutation analysis in a tumour block, we should not do it because tumour blocks are available and we don’t need them only for KRAS but in the future also for other analysis. That’s not the issue as long as we can do it on paraffin embedded tissue, as long as we don’t need fresh.

But you were pointing out that things might change under the influence of a treatment.

Things may change under the influence of a treatment, and that’s where we still have a big gap and most of the analyses looking at markers are either in older tumour blocks, either a tumour block just before the treatment, we don’t have consistent data in colorectal cancer nor in other GI cancers of doing a new analysis after treatment. What is clear is that some patients, for instance with cetuximab, have a good response and then at a certain moment they become resistant. It’s going to be much more complex and it’s for sure not a change mutation that occurs, and then the mutation in KRAS, there are other factors and we don’t have an idea what are these other factors. So that’s an important aspect that we saw at this meeting and, on top of that, we see some other data, for instance it seems that we may have a new drug in colorectal cancer – aflibercept. The first data were presented at the ESMO World Congress in Barcelona and now we see an updated data analysis and further also subgroups, forest plots and so on looking at the activity and the benefit of aflibercept.  Aflibercept, which is a fusion protein that interferes with VEGF but also broader than bevacizumab with PGF, placental growth factor, and it has been shown that when we add this drug in second line to FOLFIRI, to the standard chemotherapy regimen, that we improve the outcome of patients, the survival was longer, the response rate was higher and progression free survival was longer. So that’s important in second line and what is important also, and that’s presented also at this meeting, is that this benefit was true as well in patients pre-treated with bevacizumab as patients not pre-treated with bevacizumab. 30% of the patients in the trial were pre-treated with bevacizumab, 70% not with bevacizumab. If you look at the detailed analysis and the forest plots, we see that the benefit points are in the same direction in both groups. And that’s going to be important because one of the important questions is if a patient is, for instance, treated in the first line with a chemotherapy regimen, FOLFOX, EOX, whatever, plus bevacizumab, what do we do afterwards? Do we continue bevacizumab, do we change to another new agent from the same family with broader activity? Aflibercept is active, it has also some toxicity and that’s also a little bit surprising. Although it’s from the same class as bevacizumab, in contrast to bevacizumab aflibercept increases the chemotherapy related toxicities such as diarrhoea, fatigue, stomatitis, so we will have to look into that.

And learn how to use it better.

Learn how to use it better and also look in the near future at some other data of post-prgression, continuation of bevacizumab. So the field is moving; looking from far away, of course, this year we didn’t see a new drug, we didn’t see a BRAF inhibitor like in melanoma, data were presented earlier this year, but we’re learning much more and the data and the thinking is so rich it’s going to be complex. And that was highlighted in the session this morning in colorectal cancer and that’s also highlighted through the other settings and the other sessions. Also, another aspect that becomes clear is that we see so many relatively small studies looking at markers, also when looking at the abstracts, and they have a value in itself but at the end we should try to merge different data in European consortia that we can pool and group larger sets, group databases, tumour banks, because that’s the only way we would be able to go forward. If we see something in one study it’s also seen in other studies, the sample size is not big enough, we should make consortium and group the data.

We’re very good at heterogeneity in Europe, sometimes we need a bit of homogeneity and a bit of networking and a bit of getting together. Eric, thank you very much indeed for giving me your overview of GI cancer at ECCO, first class, thank you.

Thank you very much.