Transcript:John Marshall, MD: Tony, a patient with right-sided colon cancer with metastases—let’s say, significant tumor burden, symptomatic—had a stroke about 3 months ago. Biologic or no biologic?

Tanios Bekaii-Saab, MD: RAS wild-type?

John Marshall, MD: RAS wild-type, right-sided, metastatic, symptomatic. Are you adding an EGFR therapy in that patient or not?

Tanios Bekaii-Saab, MD: I think this is a great example of where the NCCN guidelines actually would make sense, and I would actually not add a biologic for this patient. FOLFIRINOX may be reasonable for that patient.

John Marshall, MD: But you would withhold? I think it’s very important for our audience to hear this.

Tanios Bekaii-Saab, MD: I think I would withhold.

John Marshall, MD: You would not give an EGFR therapy to this patient if their tumor was from cecum to hepatic flexure based on the data that we have right now, agreed?

Johanna Bendell, MD: Agreed.

Tanios Bekaii-Saab, MD: And the data are pretty consistent.

John Marshall, MD: Alan, this study showed us this whole right versus left thing. And then with that, lots of other studies have gone to mine this data. Tell us a little bit about what other studies showed this, and we’ll talk a little bit about how that influences practice.

Alan P. Venook, MD: So, the prospective study that showed it was FIRE-3, and that’s a contemporary study. There have also been other studies that have been looked at retrospectively. The CRYSTAL study has been looked at. There’s a large series in the JNCI (Journal of the National Cancer Institute) that came from Genentech, a variety of studies looking at bevacizumab combinations. There were 2 studies—one with panitumumab, one with cetuximab in salvage therapy—that showed no benefit to patients with right-sided disease to the EGFR antibodies. And then, other analyses that have also come through are, recently, the PEAK, PRIME, and the classic PACCE study. None of them showed anything different than what we saw. The numerical difference is greater in some rather than others, but clearly the message is in right-sided patients, if they get any benefit; it’s minimal to right-sided disease.

John Marshall, MD: And as Tony alluded to earlier, we talked about this results in an NCCN guideline change. Where it bugs me is that we are all saying that this is some sort of surrogate for some molecular thing, and yet we’ve put a dogma on our guidelines that if you have a right-sided colon cancer, you shouldn’t get EGFR therapy in first-line. Now it didn’t go to subsequent lines. And I just worry that we could end up in certain situations where physicians might reasonably want to. How are we going to replace this, if you will, with really the more fundamental understanding? Are these strong enough data for us all to say, “No, I’m not going to give EGFR therapy, and we should tell everyone not to?” What do you think?

Johanna Bendell, MD: Part of me argues that until we know better who those patients are, with the suggestion that we have, I don’t think we do it. The con question is, who would you do that for? Would you argue, like in Tony’s patient who’s not a bevacizumab patient, if right-sided colon cancer is symptomatic, why can’t we rely on good old chemotherapy? Why can’t we rely on FOLFIRI or FOLFOX or FOLFOXIRI to be able to get that benefit? We still have incredible response rates with chemotherapy alone. So, not to say that the right-sided tumor should never get an EGFR inhibitor, but, certainly in the first-line setting, I think it’s reasonable until we know that we’re doing no harm. And even with side effects, we should just hold off until we know better.

John Marshall, MD: So, it’s the best surrogate we’ve got for the time until more molecular testing comes through?

Tanios Bekaii-Saab, MD: Well, molecular testing comes through, and we may actually find that a small subgroup of patients will benefit. And at this point of time, it’s just taking the whole patient population without further granularity. I think it’s safe to say, let’s just try to avoid them in the first-line setting. John Marshall, MD: Proximal transverse colon. Is that a right-sided or left-sided?

Alan P. Venook, MD: So, remember, this is not a fluke. If you remember the embryology of the colon, the right colon comes from the midgut, the left colon comes from the hindgut, and where they join is somewhere in the transverse colon. It’s generally two-thirds of the transverse colon is right and one-third is left. I don’t know exactly. I’m sure in everybody, it’s different. In a 7’2” individual, it may be different than it is in me. For transverse colon, I don’t know the answer. I would treat that like a generic patient. I would make the decision without bias based on sidedness.

John Marshall, MD: Let’s talk about later lines of therapy. Tony brought this up earlier. What if I’ve got a wild-type patient who’s now in third-line therapy and maybe you’d even re-biopsy to confirm the wild-type. I’m going to give it all the way down as best we know. Is this a patient where you try an EGFR therapy? Would that patient be eligible for a clinical trial with an EGFR therapy, for example? Tony, are you going to give it to that patient or are you going to say, “No, I’m never giving it to you”?

Tanios Bekaii-Saab, MD: No. I say the more solid data are, right now, in the first-line. In later lines of therapy, I’d be less reluctant to give EGFR inhibitors. There are multiple reasons for that: we don’t have the data to say no and, at this stage, there are less choices. Therefore, unless you think you’re going to harm the patient, I think you would err on trying to maximize the patient’s treatment. So, until there are further data, I would still provide these patients with the opportunity to receive an EGFR inhibitor.

John Marshall, MD: You might be a little more cautious. You might scan them earlier. Johanna?

Johanna Bendell, MD: Yes. In particular, even if it was a clinical trial, that would give me even more motivation to say, “We’re doing something a little different. Let’s use the EGFR inhibitor.” But I certainly haven’t changed my practice patterns based on sidedness after first line of therapy.

John Marshall, MD: Should those patients no longer be eligible for clinical trials with an EGFR therapy? Do you see this coming, where people are writing those patients out of eligibility?

Johanna Bendell, MD: No, I don’t think it’s that close yet. I think we still have a lot to learn.

John Marshall, MD: And isn’t this good news? With EGFR therapies, we knew we weren’t perfect. We knew there were many patients who weren’t benefitting from the treatments. Is this good news from a molecular, precision medicine perspective that we’re further refining the population?

Alan P. Venook, MD: I think from many angles, it is. First of all, these are not nontoxic. It’s an opportunity loss if you use a drug that may not be helpful. There are metabolic disorders that can happen, obviously the toxicities. I would say, though, that even as we think about the history of RAS and deciding that all RAS-mutation patients should not be treated with EGFR antibodies, we don’t know that that’s a truism either. The less common NRAS mutations, for example, do we absolutely know that none of those patients benefit? The answer is no. And so, I think we haven’t studied every one of these individual mutations to know the answer. I think, on broad strokes, we need to make those calls. Again, as Johanna alluded to, sidedness is a surrogate. Soon, I can promise you, we will have an understanding of the biology, which is to say that the distribution of the types of colon cancer, the subtypes of colon cancer, are not nonrandom. That will explain much of this, and we’ll have a better handle on who to treat and how to treat.

John Marshall, MD: And I think in all the data, some of which are being presented here at ASCO GI, that this is a continuum; it’s not some hard cutoff. We’re looking at differences in genetic expression across the board. We’re looking at rectal and others, showing that those are part of the continuum. So, there is more there. You had mentioned the anatomy and embryology of the colon. My shtick on this is I think this might be a microbiome thing—that the bacteria are different, right versus left—and may, in fact, help us learn. So this lead, if we follow that line of research, might help us to explore some of the reasons why young people are getting colon cancer.

Alan P. Venook, MD: It’s a fascinating thing, and we’re studying that. We’re looking hard at that. The problem with understanding the biome in patients with colon cancers is, by the time you know they have colon cancer, they’ve had a bowel prep or you’ve taken out the right colon, and have you changed the biome? So, it’s an exceedingly difficult question, but I totally agree. I think this is one of the mysteries. And if we unlock the key to the biome, I think we’ll solve a lot of these questions.

Transcript Edited for Clarity