David J. Kerr, CBE, MD, DSc, FRCP, FMedSci; Dirk Arnold, MD


October 06, 2014

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

Fine-Tuning Colorectal Cancer Treatment

David J. Kerr, MD: I am David Kerr, Professor of Cancer Medicine from the University of Oxford. I am delighted to welcome you to this edition of Medscape Oncology Insights, with a focus on gastrointestinal (GI) cancer. We will highlight some of the key presentations from the 2014 Congress of the European Society for Medical Oncology (ESMO) in Madrid—city of Velasquez, Goya, El Greco, and Picasso's Guernica.

I can think of no better friend and colleague to guide us through the art and science of medicine than Professor Dirk Arnold. Dirk is Professor and Director of the Tumor Biology Institute in Freiburg, Germany, and he has played a very important role on behalf of ESMO in terms of putting the whole GI program together. What GI highlights can you pick out for us? Was there anything that will change the art or the science of GI cancer medicine?

Dirk Arnold, MD: The meeting just started; we only have some of the data in our hands today from the oral presentation session on colorectal cancer. We are fine-tuning the treatment strategies. There is no breakthrough—we will not have many new drugs or de novo therapeutic principles—but we are gaining insights in how to optimally apply these treatments to the patients.

Maintenance Therapy: Borrowing From Lung Cancer

Dr Kerr: Small steps forward; we are moving in the right direction. When you say "fine-tuning," is this in induction treatment or better maintenance?

Dr Arnold: The principle of maintenance treatment is relatively new in colorectal cancer. We know it mainly from lung cancer, where they use the different concepts of "de-escalation maintenance," in which we only use single drugs, and "switch maintenance," in which we integrate new drugs. We are now looking at these concepts in colorectal cancer. Three oral abstracts[1,2,3] with quite interesting findings were selected to define the role of maintenance treatment.

Dr Kerr: Is it a clear message? We are worried about the cumulative peripheral neuropathy associated with oxaliplatin. In the United Kingdom, we often give drug holidays. We have some randomized evidence from the old days saying that we don't negatively affect overall survival (with drug holidays), and we improve quality of life, so we like drug breaks, or at least going from combination to single-agent therapy. Are there some new insights with biologics?

Dr Arnold: Yes. All of these trials looked at combinations of biologics, and two trials involved an induction treatment with oxaliplatin, fluoropyrimidine, and bevacizumab. One was the trial[1] presented from our group (the AIO group in Germany) where we conducted a de-escalation trial, finding that active maintenance with fluoropyrimidine and bevacizumab provides longer progression-free survival (PFS) compared with de-escalation to bevacizumab single agent or to no treatment at all. The overall survival (OS) data are still premature, but it looks as though there is no detriment in OS when you discontinue treatment. However, continuation of treatment results in longer PFS.

An interesting role came out of this for bevacizumab as single agent, which did no worse compared with fluoropyrimidine/bevacizumab (4.8 vs 6.2 months median PFS). To understand this, it is important to understand the French trial[2] that was presented, because they had the same induction treatment (combination chemotherapy plus bevacizumab), and then did a randomized comparison of de-escalation to bevacizumab single agent (which is clearly not yet a standard treatment) and tested an experimental treatment by adding erlotinib to bevacizumab as maintenance treatment. This was a switch maintenance treatment by integrating a new drug. Surprisingly, this approach led to a slightly improved OS. There was a strong signal from this large randomized trial (in my interpretation, not yet defining a new standard) that erlotinib may play a role in maintenance treatment of patients with metastatic colorectal cancer.

Dr Kerr: Let me try to summarize. We agree that following induction chemotherapy, we can de-escalate treatment somewhat. Emerging evidence suggests that a complete drug-free holiday is probably not a good idea, but providing maintenance treatment of some sort does have an impact on PFS. It looks as though there may be a role for bevacizumab as a single agent or in combination with a fluoropyrimidine (eg, capecitabine).

PFS: Enough "Eureka" for the Euro?

Dr Kerr: How do we value PFS? Let's say I am the Health Minister of Germany, and budgets are tight, times are difficult, and this is an expensive treatment. I don't see an impact on OS. It's a strong word, but how can you sell me the idea that PFS is something that I should invest in on behalf of my citizens?

Dr Arnold: That is always the question—how this contributes to the whole strategy. We took PFS as a surrogate for OS for quite a long time, because they are slightly correlated. It's not a strong correlation, meaning that prolongation of PFS does not necessarily result in a gain in OS. However, in some of these trials, it does.

There is another maintenance trial[4] from The Netherlands (presented at the American Society of Clinical Oncology meeting) that was very close to prolonging OS. The French have shown this,[2] so there may be some evidence, although it is not super-strong. If we try to optimize treatment without harming the patient, without applying extra toxicity (and these maintenance treatments are all very well tolerated), we can exploit the maximum benefit of the treatment by continuing it. This can also allow patients who want to discontinue treatment to do so, but for most patients, in our understanding, active maintenance provides some benefit.

Dr Kerr: They feel more comfortable; there is a psychological benefit as well as a real one. This is fascinating. If we are buying into maintaining quality of life, patient choice, and so on, what sort of improvements in PFS are we talking about? You and I are part of a cohort of medical oncologists who have grown up with the concept of cancer becoming a chronic disease. In terms of chronicity, are we talking about PFS benefits of 5 years? Or is it 5 months, or is it 5 weeks?

Dr Arnold: It should be in the range of months. The new drugs, which may come also into this maintenance treatment approach, reflect this. In the United Kingdom, you have the legendary FOCUS-4 trial evaluating different targeted agents in the maintenance setting. Together with the United Kingdom, we are doing a trial with immunotherapy—a TLR-9 agonist —in the maintenance treatment setting. Those may be the concepts that will help to define colorectal cancer.

From Patient to Molecular Characteristics

Dr Kerr: I like the idea of selecting patients better. Molecular phenotypes tell us that these are the right drugs for the right patients, but to do it prognostically. We give our induction chemotherapy, and we have response rates as high as 60%. The field has moved forward, but how do we identify those patients who will have a relatively indolent path to progression in whom we could say that a drug holiday might be reasonable? How do we identify those with a more rapid progression? Apart from looking at predictive markers to choose drugs, can we identify prognostic factors for the more aggressive tumors?

Dr Arnold: That's what we have done, and the Dutch also have done. We looked at different tumor and patient characteristics to identify the patients who would benefit most. We failed. We could not see anything from the clinical characteristics that we have. So now we are looking at the molecular biology. We know that RAS or RAF status does not predict whether it's a good idea to put patients on active maintenance therapy, but this is an evolving field.

The future of colorectal cancer is to define an optimal maintenance treatment rather than to define an optimal first-line induction treatment. My interpretation is that whatever you do as induction, it doesn't matter; it doesn't differ that much.

Dr Kerr: I'm intensely interested in this. You moved from clinical characteristics to look at phenotypes, and now to the molecular biology. Is that on the basis of tumor samples that you have from the primary tumor, or are a rebiopsy of metastases?

Dr Arnold: The new generation of trials is using the rebiopsy approach, and I think they do well in doing so. Current trials mainly use the initial tumor sample, which was taken from the primary tumor, or from the metastasis before induction treatment.

Dr Kerr: We must have a discussion over coffee about the plasticity of colorectal cancer and how much changes between the primary tumor and the metastasis, because there is biology to be understood. We are going to collaborate more and work together. Next year when we come back, we will have conquered this. This is extending the concept of the right patient, right drug, right dose, and so on.

Extended RAS Testing: EU vs US

Dr Kerr: Where do you stand on KRAS testing, because there is a huge controversy just now about what combinations we should use up-front? I'm thinking about the biologics—the bevacizumab vs cetuximab argument. Where are you in terms of RAS testing?

Dr Arnold: We come from healthcare systems that have adopted RAS testing as standard, which is not a global understanding of the disease. For example, in the United States, they are not forced to have expanded RAS testing as a prerequisite for treatment.

Dr Kerr: That is astonishing. That's a good point to bring up.

Dr Arnold: In all European countries, this is a prerequisite. We now have to identify all patients who bear a RAS mutation, which predicts those who are very unlikely to benefit from anti-epidermal growth factor receptor (EGFR) treatment or may potentially be harmed by it.[5] The remaining question, is what should we do with patients who have KRAS wild-type status? Should they be treated instead with an anti-EGFR or with an anti-vascular endothelial growth factor (VEGF) treatment?

We may have an answer tomorrow in the session where we hear the expanded RAS analysis[6] from the large US trial with overall survival as the primary endpoint. We received some signals from European trials that it may be beneficial for those patients to be treated with anti-EGFR upfront. But from the design of the trial, we do not know whether this is really the activity in induction treatment at the first line, or whether it's the treatment sequence.

Treating People, not Diseases

Dr Kerr: This is a very thoughtful response. You are right—we need to understand better in those first-line trials the relatively crude markers that we use to look at response. They may not be sophisticated enough to allow us to look at PFS differences. But the sequencing question I find fascinating, because there is some biology suggesting that the EGFR first may sensitize to the effects of bevacizumab. There is much to be uncovered, and we have colleagues who spend their lives untying the Gordian knot. But in terms of practical advice for our colleagues working in the community, is your needle swinging for the KRAS wild-types more toward using first-line cetuximab/panitumumab?

Dr Arnold: The key question is whether the information that we get from the RAS status valid enough to be a strong predictive marker, which should drive our treatment decision toward A or B. The clinical practice guideline[7] in colorectal cancer from the ESMO guideline group maintains that this is an important tool for decision-making, but we should not forget patient characteristics, treatment characteristics, and anticipated toxicity. Patient preference may play a role, because whatever the results of these trials, there won't be a world in between.

Personally, I treat most, but not all, of my patients with anti-EGFR antibodies in this setting, and the decision is always made together with the patient. I tell the patient that there may be a small overall survival gain, but we have to consider acute symptomatic toxicity and other factors.

Dr Kerr: This is a perfect note on which to end—we see that fusion of art and science. We will use the science to drive, but we never lose track of the individual patient, the clinical determination that you and I use all the time to personalize treatment .

To all of you listening or reading, this has been a fantastic session. I would like to thank Professor Dirk Arnold for his extraordinary insights into GI oncology and this update from ESMO 2014.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: