David J. Kerr, MD; Josep Tabernero, MD


October 05, 2012

This feature requires the newest version of Flash. You can download it here.


David J. Kerr, MD: Hi. I am David Kerr, Professor of Cancer Medicine from Oxford and past president of the European Society for Medical Oncology (ESMO). This is Medscape Oncology Insights in Gastrointestinal Cancer. We have an opportunity to look at some of the highlights and significant studies in gastrointestinal cancer that were presented at the ESMO 2012 Annual Congress being held here in Vienna, the city of Strauss and Mozart, a city that thrums to the beat of the waltz.

I am absolutely delighted that my great friend and colleague, Dr. Josep Tabernero, Head of Medical Oncology at Vall d'Hebron University Hospital in Barcelona, and Scientific Chair of ESMO 2012, is here to talk about the highlights, so let me quickstep to Josep.

ESMO Reports Strides in Personalized Medicine

Good to see you, as always. It is a huge job that you have undertaken as chairman of the scientific committee. Are there any broad themes that unite this important ESMO meeting?

Josep Tabernero, MD: The whole meeting is moving in the direction of personalized medicine, so we have more and more presentations that are related to the field of targeting the right populations of patients with new drugs. Obviously, all presentations are not related to treatment; ESMO is a society that considers several aspects of cancer. We also have some interesting data on screening, prevention, and diagnosis.

But for this discussion, we can focus on how personalized medicine has been evaluated in some of the clinical trials in gastrointestinal malignancies. Our colleagues from the breast and lung cancer fields have done very well with personalized medicine. They have done very well in targeting specific populations of patients and have demonstrated advantages in survival. Data are going to be presented here on crizotinib in the ALK population with adenocarcinomas, and this is the way to move.

Colorectal Cancer: Does Adjuvant Cetuximab Improve Survival?

Here at ESMO 2012, we have had a presentation of the results of the adjuvant PETACC-8 study.[1]This was a pan-European study in the adjuvant setting for stage III colorectal cancer. After surgery, patients were treated with a conventional treatment, FOLFOX-4 chemotherapy or, in the experimental arm, with the addition of cetuximab.

At the beginning of the study, when it was designed, we thought that the whole population could be included. Later on, when we had recruited more than 1000 patients, we became aware of data on how predictive the KRAS mutation status could be, so we had to change the design of the study to recalculate the sample size and focus only on the KRAS wild-type population. The objective of the study was to demonstrate an advantage in disease-free survival at the 3-year point for patients who had stage III cancer.

Dr. Kerr: So, this study was adapted in response to new knowledge suggesting that cetuximab is much more likely to be effective in patients whose tumors have wild-type KRAS? You effectively stopped, redesigned the study, and did the power calculations again. We are waiting with bated breath for the results. Does cetuximab add anything useful in the adjuvant setting in these selected patients?

Dr. Tabernero: Actually, we have concluded now, with 2 studies of cetuximab in the adjuvant setting in this population of patients with KRAS wild-type tumors, that we have not observed any advantage in disease-free survival. We did our best. As you mentioned, we recalculated the study in the face of emerging knowledge. Recalculating studies is an important thing that we will have to do more often when new data appear in the interim. It is good for science, and especially for patients. It is the way to move forward.

There are more aspects at play in treating patients with EGFR-positive colorectal cancer. It is not only KRAS that matters in predicting the response. We know more and more that there are other factors that unfortunately were not included in the design of the study. So, we have to say, at the end, the study is negative under the design that was used initially and modified later on. But I wouldn't say that this closes the possibility of moving forward with EGFR inhibitors in patients with colorectal cancer.

What Will It Take to Crack the Colorectal Cancer Code?

Dr. Kerr: It is an example of how we adapt -- good science, better medicine, best practice. It is a good example of how we would drive that forward. Do you plan to do any further translational science? You have the clinical data. You have the patients in the trial. Are you going to try and collect the tissue blocks, the tumor sections, to see whether we can ask any other questions about the biology and natural history that might allow us to define hypotheses or define new studies or new ideas?

Dr. Tabernero: Absolutely. We have samples from more than 95% of the patients -- primary tumors and some blood samples. There will be a lot of correlative studies, because the only way that we can move forward in this field of personalized medicine is to take advantage of even the negative studies. Negative studies can be very informative for moving forward.

Dr. Kerr: Knowledge is power. The more we understand, the more we invest in basic and translational science -- an area that you and I are comfortable with -- the better the chance we have of designing effective, potentially positive studies. There is a feeling that we have done our best on behalf of the science and on behalf of the drug, but above all, on behalf of our patients who were thoughtful and responsive. That is the recurrent theme of personalized medicine: uniting the science to bench-to-bedside work.

Dr. Tabernero: This is the way to proceed, and the results of this study are going to contribute much to the design of future studies. For example, we are now reaching different populations of patients with colorectal cancer, even in the adjuvant setting. At some point, not very far from now, we are going to define different populations of patients in the adjuvant setting, and we are going to offer them different treatment options.

Dr. Kerr: At one of the poster discussions, there was a very interesting piece of work from Mike O'Connell and the NSABP team[2] looking at some of these prognostic signatures. Although we are looking for predictive markers (eg, can we map the right drug to the right patients and to the right tumor phenotype?) there is also some nice prognostic work going on with gene signatures that may define the colorectal cancer patients at higher risk for recurrence who may benefit from more intensive chemotherapy. That work looked quite interesting, too.

Dr. Tabernero: Perhaps one might think [biomarker investigations] make the field more complicated, but actually they offer a real possibility of getting the best information about the science behind it -- the gene expression -- and convey a particular signature that offers first, prognostic information and, eventually, predictive information. That is what we want for our patients: to be sure that the treatment that we administer to them is the one that most likely is going to benefit them.

Esophageal: EGFR Inhibition Not Yet a Benefit

Dr. Kerr: Moving on from colorectal cancer, is anything happening in esophageal or gastric cancer? Any good or bad news there?

Dr. Tabernero: We have had the final results of some phase 3 studies. One particular study that was very important was conducted in the United Kingdom,[3]in the refractory setting of patients with esophageal cancer. According to some preliminary data coming from phase 2 studies, it was thought that EGFR-inhibitors could eventually provide a survival benefit in these patients. Unfortunately, this study failed to demonstrate an advantage in survival. We should move forward with more molecular characterization, more profiling of the patients and looking at the specific populations that are going to benefit. But for the broad population of patients with esophageal cancer, EGFR inhibition did not demonstrate an overall survival advantage.

Dr. Kerr: It is interesting, isn't it? We have made our best attempt with the molecular toolkit that we have available. We looked at EGFR-receptor expression to see if we can segregate patients that way. It's clearly not enough. We clearly need to look deeper and more narrowly at the other downstream activators, the other elements of this pathway. This is an area in which you are very strong in Vall d'Hebron. We need to understand the pathways with greater clarity, and perhaps we need not one but a number of different predictive factors that we may bring into a signature or something else.

Dr. Tabernero: Actually, 10 years ago, we thought that the expression of the target was the important thing. Now, we know that not only is the expression important, but the dependence of the tumor in this receptor or target is the most important thing. As you mentioned, we are trying to understand all the escape mechanisms that the tumors may have when a particular target is not regulated by any treatment. We are working on this approach of trying to find the different ways that malignant cells escape to the signals.

Dr. Kerr: Taking the systems biology approach, not only with biologists but with mathematicians, who would have credited 10 years ago that now we would be looking at how complex systems respond, and that you and I would be going back to school and learning mathematics to relate to some of these different formulas...

HCC: Erlotinib and Sorafenib Not Better Than Single Agent

Did any other highlights or anything else about gastrointestinal cancer catch your eye that would be worthy of mention for our readers?

Dr. Tabernero: One of the highly respected studies that was presented at the ESMO meeting evaluated the combination of erlotinib and conventional treatment with sorafenib in patients with hepatocellular carcinoma. This study had a good rationale from the beginning because there were some phase 2 studies showing that erlotinib could eventually have an effect on survival in patients with advanced hepatocellular carcinoma. Building on sorafenib, the next step was trying to combine the 2 drugs.

A phase 3 study called SEARCH[4]tried to demonstrate the advantage of the combination, but unfortunately, erlotinib added to conventional sorafenib was not better than sorafenib as a single agent. This does not mean that EGFR inhibition may not play a role in patients with hepatocellular carcinoma. This means that the combination of the 2 drugs does not work, and that not all tumors in patients with hepatocellular carcinoma are dependent on the EGFR pathway and, therefore, could benefit from EGFR inhibition.

There are a lot of correlative studies looking in detail at the molecular basis of the relationship of several signaling pathways. In the next 5 years, we will have the results of other phase 3 studies done for particular populations of patients who will be quite well characterized.

Dr. Kerr: There is a sense that we shouldn't give up, but do we feel despondent or sad or gloomy? It has not been a bad meeting for gastrointestinal cancer -- that is absurd, because clearly this is good science, and we can learn from negative studies as we always do. Are there any take-home messages? Has anything been reported at this meeting that will change what you and I should do in terms of looking after our patients with a range of gastrointestinal cancers?

Dr. Tabernero: One of the messages is that we have to collect as much as possible -- tumor samples of the patients as well as blood samples -- and try to elucidate the mechanisms of the disease, and especially the different populations of patients who have one unique disease. We now know that there are either 3 groups or 5 groups of colorectal cancers, and that's a big step. We are going to differentiate the patients in less than 2 years on these particular characteristics. This offers us the opportunity to know how the tumor in a particular patient is dependent on one pathway or another.

This is the way that we are moving forward. We can illustrate this with the BRAF-mutant population of colorectal cancer. It is a small population, but now we know more about the mechanisms of the disease and the mechanisms of resistance to BRAF inhibitors that work well in other situations, such as melanoma. On the basis of these data, new studies are currently being conducted combining different drugs that target, at the same time, different pathways.

That is the best way to go. Perhaps we are a little bit behind the other specialties, but we are more robust in the science behind it.

Closing Remarks

Dr. Kerr: This is an excellent place to end, with 2 ideas for the future. One is that the cry is "no surrender." We never give up. The other, for Latin scholars, is nam et ipsa scientia potestas est, "knowledge itself is power," and we continue in that way.

Dr. Josep Tabernero, thank you very much indeed for these insights. This is Medscape Oncology Insights, a gastrointestinal cancer portfolio, reporting from the ESMO meeting in Vienna. Thank you very much, indeed.