Sidedness as Drivers of Treatment Decisions in CRC
Panelists: Johanna C. Bendell,MD, Sarah Cannon Research Institute; Dirk Arnold, MD, PhD, KTB Klinik fur Tumorbiologie GmbH & Co. KG ; Heinz-Josef Lenz, MD, FACP, USC Center for Molecular Pathway and Drug Discovery; Zev A. Wainberg, MD, UCLA Medical Center
Johanna C. Bendell, MD: So, we’re talking about the sprinkles on the sundae; being whole-genome sequencing or whole-exome sequencing for patients with colon cancer, and yet bringing us back to planet earth, we talk about right-sided versus left-sided colon cancer. We can manipulate and look at individual genes, and yet one of our biggest drivers of treatment decisions is whether the colon cancer is on the left side or the right side. Zev, why is this?
Zev A. Wainberg, MD: It’s just surrogate. So right and left side is just a surrogate for the biology of the disease. I mean, it’s a simple way of classifying a patient based on the phenotypic description of what’s really a genotypic disease. When you go back now, and you look at all that old data from left versus right sided, the prognostic data have been repeated over and over. Not all left-sided and right-sided diseases fall within that paradigm, of course. But really what we’re talking about is surrogates of biology. It’s a good surrogate; it’s useful. It’s something that I think we should mark in our notes when we see a patient now—if they have a left-sided or right-sided tumor—and use it.
Johanna C. Bendell, MD: And so, our general direction is right-sided probably not as much EGFR therapy, and maybe we think about it in later lines, when we’re potentially desperate.
Heinz-Josef Lenz, MD, FACP: Yeah. So I think it may be a surrogate, maybe a little bit more because the stem cell biology of the right versus left is different. What comes on top of it is the microbiota. So the environment is different, and the embryological etiology is different. We don’t know for sure what carries it. In [CALGB/SWOG] 80405 we tried to do CMS [consensus molecular subtyping], and we tried NGS [next generation sequencing] to adjust for it, and it remained independent. Now we had a poster yesterday on system biology, throwing everything we have into it, and what came out is significant.
Johanna C. Bendell, MD: Sidedness?
Heinz-Josef Lenz, MD, FACP: No.
Johanna C. Bendell, MD: No? Then what?
Heinz-Josef Lenz, MD, FACP: It was the gene expression signature.
Johanna C. Bendell, MD: Your sprinkles.
Heinz-Josef Lenz, MD, FACP: Yes. I think we’re going into that direction. The sidedness went away with the gene expression signature, which makes sense because it’s a different biology. So it’s not a contradiction. Yes, it’s a surrogate, it’s just different biology, which for me was not shocking but surprising that all our DNA mutation analysis went…
Johanna C. Bendell, MD: Out the window.
Heinz-Josef Lenz, MD, FACP: And out of the window because, at the end of the day, DNA mutation makes sense only if they’re functional and if they signal. And if they don’t signal, it doesn’t matter. So I think we will learn more and more about this. But I have to tell you, for 80405, we put a Moonshot grant in for whole-genome sequencing and methylation. Hopefully NCI [National Cancer Institute] will give us the money.
Johanna C. Bendell, MD: It’s a plea to the television.
Heinz-Josef Lenz, MD, FACP: The sprinkles come maybe into reality.
Johanna C. Bendell, MD: Now, we’re not only using the sprinkles in gene signatures to look at the tumor itself, but we’re looking at it to help predict toxicities from different therapies we may use. Dirk, we’re back to DPD testing. What is some of the latest? Dirk Arnold, MD, PhD: DPD [dihydropyrimidine dehydrogenase] testing is being presented, and we’ll see an abstract at the ESMO meeting…coming from the Dutch group, and it is one more contribution to what has been shown before. They are always following this and pushing this and say DPD testing may predict toxicity coming from 5FU [fluorouracil]. And we know that the rate of mutants is between 3% and on the upper level, 6 to 7% of patients. These patients may be exposed to more potentially lethal toxicity, so therefore they advocate strongly to do this testing as a routine test. I’m sure this will stimulate the discussion again because this is a large series which they present. I would not say that it is the definite answer, but I would like to see how the discussion will be running from this. The topic is back on the table.
Johanna C. Bendell, MD: Yes. And then UGT1A1 [gene]; that one sort of came and it sort of went.
Heinz-Josef Lenz, MD, FACP: Yes. I don’t think anyone uses it.
Johanna C. Bendell, MD: We use it in clinical trials.
Heinz-Josef Lenz, MD, FACP: Yeah. But the wording is off, and what you don’t know won’t hurt you.
Johanna C. Bendell, MD: Yes, what you don’t know won’t hurt you. These might pick out the patients who would have an increased risk of toxicity with irinotecan but doesn’t seem widely tested.
Heinz-Josef Lenz, MD, FACP: But I want to go back to the DPD, because it was developed and came up with in the US. Even a breathing machine was developed, remember, to test the functionality of the DPD. And because it’s so infrequent, it was basically forgotten or not taken up. Only when patients go after one dose in CNS [central nervous system] symptoms do we know it’s a DPD deficiency. I think with all of our understanding, these subgroup of patients more and more, 1, or 2, or 3% doesn’t sound so bad anymore, than 10 years ago and not anymore clinically not feasible. This is feasible. It’s shown to be cost effective. I think on this data, there must be a revisiting of this testing in our panels to avoid these life-threatening toxicities.
Johanna C. Bendell, MD: OK. I’m going to go back really quickly to first-line therapy. So we have a whole bunch of EGFR-inhibitor users who are sitting around me here. So in the setting of maintenance therapy, how do you approach maintenance therapy with an EGFR inhibitor in first line, and do we have data to suggest what we do there, Dirk?
Dirk Arnold, MD, PhD: We have some data from randomized trials. I think the most recent that may be methodologically best for this question was reported at this year’s ASCO [American Society of Clinical Oncology] meeting from this Italian VALENTINO trial. And they randomized patients who underwent inductive treatment with a combination chemotherapy and panitumumab [Vectibix], then into maintenance strategies being panitumumab alone—so the EGFR inhibitor alone or panitumumab plus 5FU. And the 5FU, folinic acid arm did by far better than the EGFR alone. Therefore, we can clearly say it’s unlikely that there is a role for anti-EGFR as the maintenance treatment. The combination may be seen as a standard. However, there are more trials ongoing, also one from my group, which are randomizing the combination as maintenance versus simply 5FU as maintenance. It could even be the case that this is the better maintenance, well knowing that we may use the EGF receptor antibody later to reuse this in a kind of rechallenge or reinduction approach. Also, really amazing data we have been seeing this year will help us to guide in which patients this reuse of anti-EGFR may be beneficial.
Zev A. Wainberg, MD: That’s the FIRE-4 trial, right?
Dirk Arnold, MD, PhD: That’s the FIRE-4 trial, which is ongoing, but we have seen here in the trial and others during the last years, which help us to identify which patients are still sensitive to anti-EGFRs.
Zev A. Wainberg, MD: Yes, but it’s sort of reminiscent to what happened with bevacizumab, where there was maintenance bevacizumab alone and a push to do that. That was the MACRO trial, and that did not end up producing necessarily great results. So the concept of avoiding the cytotoxins so far has not proven to be successful during the maintenance therapy, but hopefully we’ll see in the next trial.
Johanna C. Bendell, MD: Ironically, 5FU remains probably the most effective drug that we still have for colon cancer.