Understanding Tumor Sidedness in Colorectal Cancer

Transcript:John Marshall, MD: Hello, and thank you for joining this OncLive Peer Exchange® titled “The Divide and Conquer of Advanced Colorectal Cancer.” The flood of information regarding the clinical subtypes of colorectal cancer is bringing the field closer to a precision medicine model. This OncLive Peer Exchange® panel of experts will discuss tumor biology, guideline updates, liver-directed therapy, and sequencing while providing the latest evidence within a context that allows for practical application in the clinical setting.

I’m Dr. John Marshall, and I’m chief of the Division of Hematology-Oncology at Georgetown University and the director of the Ruesch Center for the Cure of GI Cancers at Lombardi Comprehensive Cancer Center in Washington, DC. Participating today on our distinguished panel are: Dr. Johanna Bendell, director of the GI Cancer Research Program and associate director of the Drug Development Program at The Sarah Cannon Research Institute in Nashville, Tennessee. Johanna, welcome; Dr. Tony Saab, the co-leader of the GI Cancer Program and Research and associate consultant for the Division of Hematology and Oncology at The Mayo Clinic in sunny Phoenix, Arizona. Tony, welcome; and to my side, Dr. Alan Venook who’s the Shorenstein associate director for Program Development at the Helen Diller Family Comprehensive Cancer Center and professor in the Department of Hematology-Oncology at UCSF. Alan, great to see you.

Alan P. Venook, MD: Thanks John.

John Marshall, MD: Alright, guys, we’ve got some work to do. There’s a lot going on. We’re going to start with this whole thing about right versus left colon cancers. Alan, you went through every chart of all the patients on the CALGB/SWOG 80405 study and finally confirmed for us the data that we had known for a while. Take us through how you did it and what you found.

Alan P. Venook, MD: Doing it was the time-consuming part. There had been a few small publications suggesting that there’s a difference in outcome. Then, I happened upon a report from about 15 years ago in the JCO. Scott Wadler was one of the investigators, and that study, which is all 5-FU—based chemotherapy, showed a 5-month difference in survival for left versus right primaries. And it occurred to me that that’s something we should have looked at. There are a few new series suggesting that. We hadn’t captured that information prospectively though, so I went back into all the medical records of the patients. It was about 2300 patients in total in the CALGB/SWOG 80405 trial. It took me a while.

John Marshall, MD: It took you a weekend, right?

Alan P. Venook, MD: I had a couple of long trips. I basically determined sidedness and really looked for definitive evidence of sidedness. I sent off a spreadsheet to the statisticians and forgot about it, and then 6 weeks later, I got the analysis back. It just blew my mind.

John Marshall, MD: For our audience out there, how did you end up dividing what the sidedness means?

Alan P. Venook, MD: So, to me, we had right-sided colon cancer, which was the cecum to the hepatic flexure, hepatic flexure to splenic flexure was transverse colon cancer, and then splenic flexure down to the anus was, in fact, left-sided.

We found it was very difficult to determine what was rectal and what wasn’t, and so I just lumped rectal cancer with left-sided cancer. And then, about 100 patients we couldn’t categorize. No matter how I looked and searched, I couldn’t figure out where the primary was. And so, we stuck with that. As we did the analysis, the transverse colon really fell intermediate between the right and left side.

John Marshall, MD: There weren’t that many of them either, right?

Alan P. Venook, MD: Correct. The transverse colon constituted maybe 60 cases, so we just elected not to report on them and just for simplicity sake, took right versus left and did that comparison. And the comparison was eye-popping. There was literally more than 1 year’s difference in survival, left versus right, which was stunning.

John Marshall, MD: And all of these people were still getting all the drugs, right? So, they were being treated the same way?

Alan P. Venook, MD: Correct. It was a randomized study. The 2 arms did exactly the same thing. In fact, every patient was suitable for analysis for the sidedness question. And then, as we dug into it, we discovered some really stunning things that have now been confirmed in a number of subsequent analyses. We found, for example, that right-sided patients—these were all RAS wild-type—had no benefit from EGFR antibodies. A difference in survival in a right-sided patient who got an EGFR antibody was 20 months between right-sided and left-sided. So, left-sided patients who got EGFR antibodies had a 20-month better survival than right-side, which is just stunning.

John Marshall, MD: But did you know that they didn’t have benefit, they just didn’t have as much, right?

Alan P. Venook, MD: It’s hard to say. There’s only one study that informs whether there’s benefit or not, or harm even, and that’s the CRYSTAL study. The CRYSTAL study, which was subsequently analyzed, showed that it was probably not harmful and maybe a little bit of benefit on the right side, but much more benefit on the left side. In our study, I can’t say how much the benefit was. And the other issue in our study was some patients got FOLFOX and some got FOLFIRI—three-quarters had FOLFOX, one-quarter had FOLFIRI. That may be another variable that will play out.

John Marshall, MD: Tony, help me with this. Alan gets up there and says that right-sided colon cancers do not seem to really have much benefit from EGFR therapies, even in the right molecular profile. Why is this happening? Do you have any sense about the biology here or why this might be so?

Tanios Bekaii-Saab, MD: I think we still have a lot to learn about the biology, but there’s certainly a biologic factor or molecular factor. Alan will be presenting more data, which hopefully will clarify a little bit to us why that group of patients, in particular, may not benefit as much from EGFR inhibitors. It is interesting. Of course, it’s a first-line phenomenon. It changed the NCCN guidelines. There’s a statement in the NCCN guidelines that essentially refers patients with RAS wild-type right-sided disease to avoid or not use EGFR inhibitors. I think the problem with this is, how does that apply to later lines, not as much frontline? In the frontline setting, the majority of United States practices, including mine, still overwhelmingly use bevacizumab. So, that doesn’t really change what we do. For some practices, it may affect what they do. But for later lines though, the data are less clear. You have a patient who is RAS wild-type on the right side, then you use chemotherapy plus bevacizumab in the first-line.

John Marshall, MD: Let’s come back to later lines in a minute because I really want to talk about this frontline question. When I’m seeing a new patient with metastatic disease, I know there are a bunch of different ICD-10 codes for where the cancer is, and sometimes I’m trying to pour through to figure that out. So, it matters?

Johanna Bendell, MD: I think it does for now. I think that right now, sidedness is probably a surrogate for some type of biomarker profile, but we don’t completely understand yet. And as Tony alluded to, Alan is going to be presenting some more data to try to hone in a little bit more on the biomarker profiles that might be associated with these outcomes that we’re seeing. But I think we don’t know the whole story yet.

John Marshall, MD: Yes. And so, I’d need to know where it came from. What if I’m treating somebody, do I need to go back and figure out where that primary was?

Alan P. Venook, MD: Well, the answer is maybe yes, maybe no. As Tony said, our study really only looked at first-line use of therapy. Now, there are others studies that have looked at EGFR antibodies in the same population, right versus left side, and also don’t appear to show much benefit. But I think the answer is, it’s important. The main question though, is, again, are we harming patients with EGFR antibodies? Probably not, although there’s the opportunity loss and there’s a toxicity that you’d rather not face. One fascinating subset that we had, and we will hopefully explain soon in our study, is the cohort where we started out treating all-comers. In fact, a number of patients with RAS mutations were included in the study. In those patients with right-sided primaries who had RAS mutations and got cetuximab, they had an 8-month better survival than patients with right-sided primaries that were RAS wild-type who got cetuximab. Now, that’s an upside down finding, but it makes you really think there’s something else going on because clearly RAS is not the dominant driver on the right-sided cancers. And there are other factors that are involved.

John Marshall, MD: How does the data change your thinking about molecular testing in a brand new metastatic colon cancer? Do you need to know right away still or if you have a right-sided, you don’t care?

Johanna Bendell, MD: I think in the first-line setting when I see that patient for the first time with metastatic disease, the thing I want to know right away is actually more microsatellite status, and we can talk a bit more about that in terms of decision making in the first-line. Right now, I don’t know that there are enough data that I believe are solid to say that it makes a difference when my go-to regimen is a FOLFOX/bevacizumab regimen. I think it would make more difference if I’m considering an anti-EGFR therapy. So, I’ll go ahead and get people started, but I will send off molecular profiling on those patients right away.

Transcript Edited for Clarity