Therapeutic Strategies in Metastatic Colorectal Cancer - Episode 4
Transcript:John L. Marshall, MD: I think this has been a great discussion around that, because I do think the bar is moving. But let’s shift gears just a little bit and talk about molecular testing for colorectal cancer. There isn’t a single disease out there anymore that a microscope is good enough for; you need to do some genetic testing. I know that we keep moving the bar on this. We keep saying this is what you should do, and a year later we say, ‘Oh, we didn’t mean that, this is what you should do.’ So, Charlie, you get to start us off. What is the, this is what we should do for today for a patient with metastatic colon cancer?
Charles S. Fuchs, MD, MPH: John, you’re so right. This field continues to move. We all remember when cetuximab and panitumumab were first developed. The idea was, oh, if we test for EGFR expression on the tumor, that will tell us who to give it to. In fact, that doesn’t predict anything, and what we’ve ultimately learned is that RAS mutations predict resistance, and that has continued to evolve—because we first thought it was just a portion of the KRAS gene.
Now we know it’s all-RAS genes. What’s really helpful is the technology is keeping up; in fact, surpassing what we’re learning clinically. Whereas it cost billions of dollars to sequence one person’s genome 20 years ago, we can now do it very cheaply on clinical samples. At our center, we’re using, essentially, a next-generation sequence platform for about 300 genes, which obviously include RAS and others. And that’s been our clinical standard. It’s helpful, obviously, and practical, because you know the RAS status. But it’s also helpful because you learn about other genes that you can use to direct them to other studies.
John L. Marshall, MD: What would you say to an oncologist who does not have access to that? What tests should they talk with their pathology group about? What should they order today to manage a patient with metastatic colon cancer?
Charles S. Fuchs, MD, MPH: They should insist on all-RAS testing, and clearly, they’ve got to get it in real-time.
John L. Marshall, MD: And if you’ve done a KRAS and it’s wild-type, send it again?
Charles S. Fuchs, MD, MPH: Yes, because we need to know. Firstly, KRAS may only be codons 12 and 13, and not 61 and 146. It obviously may not include NRAS. You need to specifically ask your pathologist, are you getting all the KRAS and all the NRAS components of the genes to know whether they’re mutated? Because we know they all predict resistance. The other thing I would say is that they should have an algorithm where if they’re RAS wild-type, they should be sequencing at least for the V600 of BRAF. Because we know that’s an important predictor of outcome, and it also can direct you to important studies.
John L. Marshall, MD: So RAS, BRAF. What else?
Charles S. Fuchs, MD, MPH: Certainly MSI.
John L. Marshall, MD: So that’s now in; it used to be stage. Last year’s message was only in stage II. This year’s message is everybody?
Charles S. Fuchs, MD, MPH: Absolutely. And, as you know, it certainly helps us direct adjuvant therapy, and now we’ve learned from our studies in immuno-oncology that MSI-high tumors are likely to benefit from PD-1 and PD-L1 directed therapies.
John L. Marshall, MD: And we’ll come back to that. So now RAS, MSI. Anything else? BRAF?
Charles S. Fuchs, MD, MPH: From a practical standpoint, I’d say those three things are critical. The others are interesting, but I’m not sure.
John L. Marshall, MD: HER2?
Richard Kim, MD: I think there’s some data that one of the resistance mechanisms, the primary, secondary resistance of EGFR, is HER2 amplification. There are some data from last year that show that a patient with KRAS wild-type with HER2-positive on FISH or IHC did not respond to EGFR drugs. I’m not sure if it’s ready for prime-time. We don’t do it at our institution. I would like to see more data, definitely a preclinical support study for sure. And one of the mechanisms of secondary resistance is also that HER2 amplification may be secondary resistant, as well. Personally, I would like to see more data. I don’t test for HER2 on my patients at this time.
John L. Marshall, MD: You don’t get it in your next generation. Are you ordering it on top?
Charles S. Fuchs, MD, MPH: Well, you’re obviously referring to the HERACLES data, which is a small study looking at trastuzumab and lapatinib in HER2 overexpressed in colon cancer, which really looks interesting in terms of response. The problem is, it is not part of our routine. And what happens is you call up the pathologist and ask, can you do me a favor? Can you do HER2 because of this promising and provocative data? But it isn’t a routine at our center. I don’t know if the others are doing it.
John L. Marshall, MD: Yeah, I was going to ask.
Tanios Bekaii-Saab, MD: It’s not routine at our center. But, again, I think its placing is probably in later line. It’s a dynamic process. It’s not a process where you take the tumor at diagnosis—unlike KRAS—and you can rely on it. You’re probably better off, if you have access to HER2-directed therapies, to do that later, either through a biopsy or even a liquid biopsy to give you these answers.
John L. Marshall, MD: HER2?
Johanna Bendell, MD: Not yet, but I think it’s coming. But I think it points to something very important, which is that tissue is precious, and that’s what we’re learning more and more about. Although, in colon cancer specimens, if we have a resection specimen, we have a lot more tissue. But there are some people who are diagnosed in the metastatic setting where we have a liver biopsy. And so what we’ve been working on is education around getting not fine needle aspirates, but core biopsies. And if you’re going to send off the tissue, try to get as much as you can in one shot so you have that information for later. And then also the importance of when you send off next-generation sequencing: just an education around it doesn’t give you MSI. It’s a separate tissue.
John L. Marshall, MD: I’ve got a patient right now who’s been treated at your cross town rival and moved to the DC area, and the only biopsy they had, which is reasonable, was the colonoscopic biopsy. And so here’s a patient with metastatic colon cancer that we had no RAS status on. And so we ended up having to go back and get more tissue in order to do that. Let me just ask. I know how you guys do it. What we do at our shop is we are doing more and more partnering with a profiling company where we’re trying to figure out if broad profiling is best. I like to say, instead of ordering a chem-7, order the chem-20 and seeing what happens. Order everything and get that back to see if that makes sense. You’re doing it internally. Are you guys doing it internally or sending out?
Tanios Bekaii-Saab, MD: No, we’re doing it internally just on a limited basis; meaning just the target genes.
John L. Marshall, MD: The ones that are approved and proven?
Tanios Bekaii-Saab, MD: KRAS, NRAS, BRAF; we’re doing MSI on everyone. So we’re just doing it targeted, and if we need an expanded panel, we’re actually sending out.
John L. Marshall, MD: So that’s fine. What are you guys doing?
Johanna Bendell, MD: Commercial next-generation sequencing.
Richard Kim, MD: We have an internal kit that we use that has 20 to 30 actionable markers. It’s not the full panel. So we test those first and if you need to go beyond that, we will order and use commercially available kits.
John L. Marshall, MD: And I’ve been advising those folks in the smaller community groups to consider these commercial groups because it’s changing so much, as you say, and tissue is precious. So, get it all when you can. And then your pathology department usually was the pushback before. I don’t think they care as much anymore. It seems that they’re happy to have somebody, although some pathology departments feel stronger about that. Is there any other emerging target? Should we know other stuff on patents, Charlie?
Charles S. Fuchs, MD, MPH: One thing I would add is, as we all start to include these multi-gene panels—and clearly you want to do MSI—but one thing that we’ve periodically discovered is, before you even get the MSI or independent of the MSI, you find these patients where the pathologist report says there are a litany of mutations, and they don’t even know what the mutations mean. But there are so many they can’t count. And when that happens, and this has been my experience, those are people who may respond to immuno-oncology approaches because they’re hypermutated. So, make sure that if you use those platforms, make sure you get everything out of them.
Transcript Edited for Clarity