Panelists: John L. Marshall, MD, Georgetown University Hospital; Dirk Arnold, MD, PhD, Instituto CUF de Oncologia; University of Hamburg; Fortunato Ciardiello, MD, PhD, Seconda Università di Napoli; Paul R. Helft, MD, Indiana School of Medicine of the Indiana University Melvin and Bren Simon Cancer Center; Tanios Bekaii-Saab, MD, Mayo Clinic Cancer Center in Arizona
John L. Marshall, MD: Fortunato, let’s discuss HER2. We talked about it earlier, and we all agree that we want to know HER2 status. It’s not typically done. And so, we, in academic centers, are picking this up and asking for the testing, right? Why would you care?
Fortunato Ciardiello, MD, PhD: Actually, this is very interesting. This started a few years ago. This was an observation clinically done by some of my friends in Italy—Alberto Bardelli and colleagues. We have to give them a lot of credit because their submission started with some PDX (patient-derived xenograft) models of patients who were KRAS exon 2 wild-type, and they tried to see if mice were responding or not to any drugs. They found a group of them that were resistant to cetuximab or panitumumab. So, they went on to do an extensive evaluation of any potential growth factor receptor gene that would be amplified. And the bottom line is, that about 3% of metastatic colorectal patients have a truly amplified HER2 gene. Now, if you go to the all-RAS wild-type patients, this goes up to maybe 4% to 5%, let’s say 5%. And, you have another couple, 2% or 3% of patients, who have some activating mutations. For some of these mutations, we know that this is activating. For some of them, we don’t know if they have any biological role or not in the disease.
In the patients that have a real amplified HER2 gene, what they’ve done is they set new standard techniques for immunohistochemistry. In these patients, according to a proof of principle phase II trial, the combination of trastuzumab and lapatinib was highly effective in terms of response rate and long-term survival.
I can tell you, this is anecdotal. One patient who we put in the trial was a very young male who was diagnosed, around 41 or 42 years of age, with very aggressive disease. He came to our clinic for adjuvant therapy. We found lots of cervical lymph nodes. So, we did more workup and found that he had, already, metastatic disease after 3 or 4 weeks following surgery—stage III disease. Then, he was treated with first- and second-line therapy. And, then, we had this trial. We found that he was amplified, and then we did immunohistochemistry, and then we did gene amplification. And this patient was living for about 31 months by sequential treatments, without chemotherapy. So, we did everything you can imagine. We started with trastuzumab plus lapatinib, and then trastuzumab and pertuzumab, and then T-DM1. And so, he had 31 months of survival after we started.
This is just a clinical case, but it teaches you, or teaches me, a lot of things about if you find HER2-dependent tumors in colorectal cancer. We spend all these efforts for MSI-high situations, and all these efforts for BRAF, but HER2 is also important. In this way, we can change the history of the disease.
John L. Marshall, MD: In right-sided versus left-sided tumors, we know HER2 tends to be more useful on the left-side—so, higher in rectal and it goes that way, down the other way. Are there any other key molecular characterizations? We’ve got clinical data. We’ve got trials going forward. Tony, I know you’ve built a little program around HER2, and that you try to always have studies for HER2-positive patients?
Tanios Bekaii-Saab, MD: Yes. We’re building a whole platform around liquid biopsies and selecting targets. We’re looking, actually, at even the less common targets like FGFR and RET to try to integrate that in a big platform through our network. And HER2, of course, is a pivotal pathway, as well as EGFR re-challenge, and a number of METs, etcetera. So, there’s a lot.
Fortunato Ciardiello, MD, PhD: Another potential target, also in colorectal cancer, is rearrangement of NTRK genes.
John L. Marshall, MD: That’s still around 1% or so.
Fortunato Ciardiello, MD, PhD: It’s 1% or 2%, but you just found a patient…
John L. Marshall, MD: If you found it.
Paul R. Helft, MD: If you find that patient, the response is amazing.
John L. Marshall, MD: It’s like imatinib.
Tanios Bekaii-Saab, MD: The point is, now we have proof of principle with BRAF, MSI, and with HER2. We’re learning, more and more, how to refine our EGFR selection. But, now we get busy, also, looking at the less common ones—FGFR2, NTRK, MET, EGFR amplification after exposure, clonal differentiation, etcetera. So, all these things are making this disease actually much more molecularly-based than ever. We’re learning more and more about this, and we’re going to integrate those therapies into the mix in a more intelligent way.
John L. Marshall, MD: And, at least in the United States, both ASCO and NCI have tried to develop access tools—the NCI MATCH trial, for example, and the TAPUR study, which is being run by ASCO. These are ways where, instead of begging for the drug, there are clinical trials that you can enroll patients in throughout the country. Is there a similar sort of structure in your world to have these kind of basket studies available?
Dirk Arnold, MD, PhD: We do, but this depends on the heterogeneity in the countries, of course.
Fortunato Ciardiello, MD, PhD: We have a lot of academic initiatives in different countries.
Tanios Bekaii-Saab, MD: But, I do think that the key is to test patients early. One of the lessons you learn from the NCI MATCH trial, for example, is that if you wait until it is too late, you can detect an actual mutation, but you may not be able to act on it.
Fortunato Ciardiello, MD, PhD: For example, for the rearrangement of NTRK, where the patient starts in the first-line setting, we at least know that it’s there or not. And patients are very happy if you explain that.
John L. Marshall, MD: Absolutely. This has really been an extremely informative session. Before we end, I’d like to give each of you a chance to give your closing thoughts. Dirk, you’re first off. That’s the easiest position.
Dirk Arnold, MD, PhD: Really, all throughout our discussion, we have learned a lot on this amazing new information we have to integrate into our decision-making with patients with metastatic colorectal cancer. We have predictive factors which, now, really drive treatment with RAS status, which potentially drives treatment of MSI and HER2, positively. So, there is a great point in this. And, on the other hand, we have to have 2 caveats. One caveat is, clearly, that we should be cautious that we tend to over interpret positive results. We have to carefully weigh the results. We have seen many case reports, also discussed here, about the Lazarus effect of anti–PD-L1s with a limited number of patients with nonhereditary cancers.
In HER2, we have a good series, but these are localized series’. What I’m missing, and what we still have to find out—and this is, for me, the beauty of the disease—is that we are not only treating patients with drugs, but we also have local ablative treatment. We also have surgery in patients with mCRC. We have to learn, from this prognostic information, if it is worth resecting a patient with BRAF-mutant mCRC, for example? Is it worth it to send this patient to liver-met resection, or is he recurring immediately? If the chemotherapy doesn’t work, potentially, he’s an ideal candidate? Or, potentially he’s not a candidate? We also have to ask this multidisciplinary question in the light of the biological information, which we do have.
John L. Marshall, MD: He’s left us nothing else to say. Fortunato?
Fortunato Ciardiello, MD, PhD: I would say that colorectal cancer, in general, and metastatic disease, even more, is a very complex disease. That are so many diseases, altogether. It’s our responsibility, from the beginning, that when we see a patient for the first time, we need to consider the most important prognostic factors. What do we do in terms of strategy? In terms of multidisciplinary approach? In terms of we’re using? How do we consider all of this information that can be so extensive in an intelligent way? There will be confusion. So, it’s our responsibility to define which is the best for the patient. And, again, as I said before, we have to go back to, also, being doctors—taking all the information, discussing it with the patients, and using it with, in a humble way, our colleagues as surgeons and radiotherapists as local treatment, or whatever can be useful in the strategy. Because, if we don’t do it in a strategy before we do the first step of therapy, maybe 1 year later it will be less effective or too late.
John L. Marshall, MD: Paul, to win the Tour de France, you need a team, right? You can’t do it by yourself.
Paul R. Helft, MD: I guess I find that it’s amazing that everything that’s old is new again. In this era, we’re really beginning to understand the complexity of a disease that we thought was relatively straightforward. Something like tumor-sidedness has become the most important story of our disease, this year. I also appreciate that oncology perspectives are truly global, and we have a tremendous amount to learn from our colleagues around the world.
John L. Marshall, MD: We should have other members of our world take a lesson from that, from your lips to their ears. Tony, you get the last word.
Tanios Bekaii-Saab, MD: I think the most important key point, from what comes out from this discussion, is that we should always interrogate the tumor, and interrogate early, and even consider interrogating the tumor on multiple points to uncover its molecular and genetic composition, because that actually allows us to more intelligently target the tumor, whether in clinical practice or in clinical trials. For the longest time, we’ve been observing the tip of the iceberg, knowing that there’s a lot underneath. I think we’re seeing more and more of the iceberg, and it’s not because of global warming, but because we’re getting a little better vision of the iceberg. I think we’re uncovering more and more of these small baseline abnormalities that can be targeted in colorectal cancer, which will soon become a lot of diseases, not just one.
John L. Marshall, MD: Thank you very much for all that you’ve done, and I want to thank you guys for joining us. On behalf of our panel, we thank you for joining us and we hope you found this Peer Exchange® discussion to be useful and informative.