John L. Marshall, MD; David J. Kerr, MD


June 18, 2012

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


David J. Kerr, MD: Hello. I am David Kerr, Professor of Cancer Medicine at Oxford and past president of the European Society of Medical Oncology (ESMO). This is Medscape Oncology, and we are giving insights in gastrointestinal (GI) cancer. Today we have an opportunity to look at some of the significant GI data emerging from the 2012 annual meeting of the American Society of Clinical Oncology (ASCO®).

I am delighted to say this morning that joining me is my friend and colleague, John Marshall, who is Professor of Medicine at Georgetown and Director of Clinical Research for the Lombardi Comprehensive Cancer Center in Washington, DC. John, good to catch up, as always; absolutely fantastic, as always.

John L. Marshall, MD: Dave, good to see you.

CRC: Managing the Disease

Dr. Kerr: Let's start with colorectal cancer. Has this been a good meeting for you? Are we seeing anything of real interest, something that set the fire to light and warmed the cockles of your heart?

Dr. Marshall: I wouldn't go quite as far as warming the cockles of my heart. Some important, possibly practice-changing papers are being presented here at the meeting. They come in 2 categories: managing the disease and the concept of maintenance therapy and biologics beyond progression. There are some interesting data in those areas. Some new medicines are emerging, finally. It has been since 2005 that we have had a new medicine in colorectal cancer, so finally a couple of new medicines are emerging, mostly in the second- and third-line settings. Things have been presented that all oncologists need to hear about.

Dr. Kerr: We are understanding more about the natural history of the disease and perhaps taking a step forward in terms of management. Any specific examples? What really caught your eye, John?

Antiangiogenics Move Into Maintenance

Dr. Marshall: If you thematically put this together, data now support antiangiogenic therapy essentially at all lines of the treatment process. We all know about frontline use of antiangiogenic therapy. Separately, we have second-line studies showing antiangiogenic therapy worked, but not in the continuum. This year, an important study[1] looked at what we would call bevacizumab beyond progression (first championed by Axel Grothey[2] and some retrospective analyses). We now have a prospective study where patients were given frontline chemotherapy of choice with bevacizumab, and once they progressed, they were randomly assigned between continuing bevacizumab with their second-line therapy or not.

That study was actually positive. It suggested that continuing -- or "herceptinizing" -- bevacizumab into the second line was positive. That goes very nicely with some new studies of a drug called aflibercept -- a cousin, if you will, of bevacizumab -- a different molecule that has a broader mechanism of action. It is not only vascular endothelial growth factor (VEGF)1, but VEGF2 as well. It is, in theory, a different antiangiogenic therapy. Aflibercept was tested in an important study (VELOUR)[3] in second line with FOLFIRI [leucovorin, fluorouracil, irinotecan] plus or minus aflibercept. This study, too, was positive.

In a sense, we have 2 second-line antiangiogenic therapies: either continuing bevacizumab through progression or making a decision to switch over to aflibercept, this new agent, mostly in the KRAS-mutated patient. In the KRAS wild type, we have a third choice of course in the setting of bringing an epidermal growth factor receptor (EGFR) to the table: either cetuximab or panitumumab. This second-line biologic question is in play now, and we have many more choices for those choosing to incorporate a biologic in second line.

Dr. Kerr: We have some clear detail at last. In a way, it was counterintuitive for you and me, being old lags and being used to chemotherapy. We are used to the genetic instability of cancer. When you get tumor progression, you stop. Whatever the treatment is, you switch to second line because of the emergence of resistant mutations. However, with drugs like bevacizumab and aflibercept, their targets being genetically more stable endothelial cells, it looks as if there is continuation of hard data -- good randomized evidenced for continuation in that way.

Dr. Marshall: We can't forget that we also have 5-FU beyond progression. If you are going from FOLFOX [leucovorin, fluorouracil, oxaliplatin] to FOLFIRI, you are continuing that drug too. The only variable here is going from oxaliplatin to irinotecan or vice versa.

Dr. Kerr: You and I were talking a while ago about the VELOUR study and aflibercept, which is an interesting drug, a genetically engineered VEGF trap. It sucks a whole variety of these factors out. With these new data coming out on bevacizumab continuation, how do we see aflibercept playing a role? Are we going to have to do some head-to-head studies, or is there a niche for aflibercept?

Dr. Marshall: It will find its niche with its reputation of having a slightly higher side-effect profile. If I have a very stable frontline metastatic colon patient, who has been 12 months on frontline therapy with subtle progression, I am probably going to keep that patient on bevacizumab.

On the other hand, in a patient who is not having quite a good response with more rapidly progressing cancer (particularly in the KRAS-mutated patient in whom I don't have an EGFR drug to use), I might consider trying aflibercept as a different approach. Maybe I will get a little bit more mileage from that approach.

Regorafenib, Yes; Perifosine, Not So Much

Dr. Kerr: You mentioned that there are some new drugs coming along, some new kids on the block. Is anything of interest there?

Dr. Marshall: One that is on everyone's tongue is a drug called regorafenib. For us, it is an important new kind of therapy because we have tested many tyrosine kinase inhibitors (TKIs) in colon cancer over the years and we haven't found one that clicked. Van Cutsem and colleagues[4] randomized regorafenib against placebo in refractory disease. It is a multitargeted TKI, a bit of a dirty bomb. It showed some improvement in overall survival. Honestly, there has been a lot of stagnation around drug development in colorectal cancer. Where will I put this new drug? Where will I test it?

In a way, this positive refractory study has reopened the gates to a legitimate place to test a new drug. That drug likely will gain approval in the United States. We are already doing extended-use types of programs. I am looking forward to getting that drug in my hands to sort of see if it's a real drug. As you know, it can look good on paper, but in the clinic...

Dr. Kerr: Perifosine (the AKT inhibitor) was also out there. Any good or bad news there?

Dr. Marshall: Unfortunately, it is bad news. This is an interesting process because we had a very strong randomized phase 2 that strongly supported testing this agent. They did the same study, but bigger, and all of a sudden the study is negative.[5]

This has been true in GI cancers in general. Data that emerged from randomized phase 2 trials hasn't translated into the larger studies. We need to understand better how to enrich and how to sort for those patients.

Dr. Kerr: It's a recurrent theme, isn't it, that we hit upon something in phase 2, but it doesn't seem to translate. It's the old notion that biology is king. There must be some element of patient selection going on in phase 2 that is ironed out in the larger phase 3 study.

5 Subtypes of CRC Emerge

Dr. Marshall: This is another element that has emerged from this meeting. We have been talking about KRAS wild types in mutated tumors. We have been talking about microsatellite instability. What seems to be emerging from these large retrospective tumor bank analyses is as many as 5 different subtypes crossing between molecular profiling (which includes KRAS and MSI and so forth) and histopathologic characteristics.[6]

It was the first time I had seen these data presented, and they were impressive -- and hopefully true. If we can, in fact, take your "vanilla" colon cancer and say that "you belong in one of these groups," it will help us tremendously in therapeutic decisions and in prognostic guidance and counseling. Stay tuned. We will hear more about these subtypes in colon cancer.

Dr. Kerr: One of the areas in which there is fantastic potential for international collaboration is in being able to show colorectal cancer flavor A, type B, color C. It does require some very heavy lifting in terms of international cooperation. This is a sort of transatlantic endeavor that we should be making. Watch this space. Biology is king, so let's see what happens with that.

Erlotinib Heading to Maintenance

Dr. Kerr: Did anything else in colorectal cancer catch your attention?

Dr. Marshall: An exciting study that caught us a bit by surprise is the DREAM study.[7]We often joke that it was a dream for Roche/Genentech because it included 2 of their medicines.

It goes back to the question of maintenance therapy and how best to do it. Starting with an oxaliplatin-containing base regimen, patients were randomly assigned between continuation vs "optimoxing," if you will. In a subsequent randomization, the groups were bevacizumab by itself vs bevacizumab plus erlotinib. We were all relatively surprised to see that erlotinib actually performed a little bit better.

We have done many studies with oral EGFR inhibitors in the past and they have been negative. In this setting, this drug might have some benefit. We have also been nervous that VEGF and EGFR together might be negative. This was reassuring. It is a different window. A patient whose tumor is stabilized and maintained and a patient who is progressing probably have different biology. This interesting, positive study might be practice-changing by bringing erlotinib to the table in that maintenance window for metastatic colon.

Dr. Kerr: It was a fantastically well-designed trial. Some positive results are coming out, but I wonder where the value might come in. We are looking at stabilization of disease, which we think will bring improved quality of life for our patients, but it comes at a cost. The health economic analyses haven't been done yet, but do you think cost might influence whether this is practice-changing? We have the results of the trial. It is clear, clean, and well designed. Will you and I as practicing clinicians have to consider cost in deciding whether this drug will come into our daily practice?

Dr. Marshall: Absolutely -- cost in terms of both money and toxicity. Particularly in the United States with oral agents, these are things that our patients consider more. They are paying for all these drugs. Will this be something that they readily go off to the pharmacy and buy for what is a relatively small absolute benefit? That remains to be seen.

Upper GI Cancer: Slow Progress

Dr. Kerr: We have had a really good opportunity to catch up in colorectal cancer, but there were other things happening in esophagogastric and other GI cancers. What were the highlights?

Chemo Bests Panitumumab -- But Why?

Dr. Marshall: A lot is happening, but no highlights. That is the problem. Yesterday, SOG[8] presented a very large study based in Great Britain but conducted in a few other countries where patients with esophagogastric tumors were randomly assigned between the epirubicin, oxaliplatin, capecitabine (EOC) regimen that came out of the REAL study, plus or minus panitumumab. We know that upper-GI cancers don't have high KRAS mutation rates. There was no known biological enrichment at the time and no obvious one to include. It was an unenriched, large, randomized study.

The discouraging point was that the panitumumab was inferior to chemotherapy alone. Why would that be? Yes, there is toxicity. Yes, the panitumumab group had to have lower doses of oxaliplatin and had to have modified capecitabine doses to counter diarrhea. There was a lot of talk that the dose changes were responsible for the inferior outcome, but I am not convinced by the whole dose-intensity argument. If you look at some of the colorectal cancer data with EGFR drugs in various cocktails, I don't know of a capecitabine plus EGFR-positive trial yet. The studies with infusional 5-FU have been positive, such as the COIN study.[9] The capecitabine subtype was negative. The infusional 5-FU subgroup was positive. That may be an oxaliplatin interaction. Before we decide that this is a dose-intensity argument, we need to also think about our combination.

Dr. Kerr: I agree, John. With post-hoc reasoning, it is always easy to go back and put the finger on the wrong suspect -- an "It was the butler in the library with the candlestick"-type thing. We may need a deeper look at the biology, rather than immediately going to some very superficial quasi-clinical pharmacology.

Dr. Kerr: Time will tell. Should we be giving more dose-intense chemotherapy? We have been arguing back and forth about adjuvant gastric cancer for a long time. Is anything of interest popping up?

Use More Drugs Perioperatively? Maybe Not

Dr. Marshall: Should we add more drugs perioperatively in these very nasty [gastric] tumors? A big study was 5-FU leucovorin vs a complicated cocktail of FOLFIRI followed by a taxane and a platinum.[10] In the good ol' days, it was ProMACE-CytaBOM [acepromazine, prednisone, cyclophosphamide, doxorubicin, etoposide, cytarabine, bleomycin, vincristine, methotrexate and leucovorin] vs CHOP [cyclophosphamide, doxorubicin, vincristine, prednisone] for everyone. These are all active drugs in the metastatic setting, so the logic is that they would be also good in the adjuvant setting. Yet again, there is more toxicity and no added benefit.

We have seen the same thing in colorectal cancer. We have drugs that work in the metastatic setting that have not brought any benefit to the adjuvant setting. Is the adjuvant disease that we are treating fundamentally different from the metastatic disease, making metastatic ground not good proving ground for adjuvant drugs?

We need to understand this better. It's not fair. In other diseases, like breast cancer, whatever works in metastatic works even better in the adjuvant setting. In our world, we sure haven't seen that.

An Option in Anal Cancer

Dr. Kerr: A nice, pretty study in anal cancer[11] had some important messages about the timing of chemoradiation. Do you have any comments about that study?

Dr. Marshall: In a mitomycin-short world, this is an important study. It has been presented before. It is essentially a question of mitomycin vs cisplatin, along with 5-FU and radiation. The initial report of this trial suggested that the mitomycin response rate was higher -- not much, but enough to say that we should stick with mitomycin. Of interest, they followed the patients to longer time points, and over time, the complete response rate comes together.

Dr. Kerr: They converge.

Dr. Marshall: It basically supports the use of a cisplatin-based regimen or a mitomycin-based regimen. They were the same. It showed us that complete response clearly correlates with outcome. We felt that was true, but it is nice to have it confirmed. Be patient. Let patients have their complete response many weeks after completion.

Dr. Kerr: That seemed a key thing, waiting for the 26 weeks. We are not always patient by nature. Sit, watch, wait, allow it to shrink, and then step in.

Dr. Marshall: An important point is that it was a single dose of mitomycin, not a repeat dose of mitomycin. It is helping us refine our treatment options.

Tivantinib Gets Thumbs up in HCC

Dr. Kerr: Something that you and I have been fascinated by is the whole c-MET story: Can we select patients? How do we use these fantastically potent inhibitors coming out? And what about some new data in hepatocellular cancer (HCC)?

Dr. Marshall: It was a positive trial[12] of second-line HCC treatment in patients who were randomly assigned to tivantinib, which is a c-MET inhibitor. In this c-MET subset (they did biological enrichment retrospectively) they actually showed positive data. We thought this was going to be the case in HCC. We hoped it would. These are nice, early randomized data supporting tivantinib in c-MET-positive HCC patients.

Dr. Kerr: Many thanks. It has been a great discussion, as always. It is a mixture of news, the good and the bad. (There is no ugly this time.) Thanks for that very clear description of what is going on in GI cancer.

From John Marshall and me, thanks for listening to a really important update in GI cancer coming from the 2012 ASCO® meeting.


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.