Case 4

Van Morris:

We’ll move to the fourth and final case for today, talking about how we order and interpret BRAF mutations in treating patients with colorectal cancer. This is a 62-year-old man. He’s recently diagnosed with stage IV right-sided colon cancer with imaging studies revealing metastases to the liver, lungs, and bone. He started treatment with FOLFIRI-bevacizumab, still waiting for his molecular profiling results to return. He comes back for his second dose of treatment and is overall doing well with chemotherapy. His profiling reveals a microsatellite-stable tumor with a TMB of 3, a KRAS/NRAS wild-type status and a BRAF mutation. But it’s not a BRAF V600E mutation; it’s a BRAF D594G mutation.

I guess one of the questions I would start by asking, how do you perform mutation profiling for patients with colorectal cancer? What tests are you ordering?

Aparna Parikh:

We do just comprehensive with any de novo metastatic patient or any metastatic patient, period. We do comprehensive genomic profiling and that includes extended RAS/RAF status. We would get this information upfront for all those patients.

Van Morris:

Dr. Pedersen?

Katrina Pedersen:

We do the same because certainly if you’re focusing only on codon 600 in BRAF, you could be missing some other very important information. But additionally, take it a step forward in my personal practice, where I do liquid profiling with circulating tumor DNA. Because that’ll give me my rapid response about what the BRAF status might be because that’s the one I need most; I don’t usually have it at that first consult visit. But will send us down one treatment pathway versus another potentially. But also with NGS (deep sequencing), you’re also getting transcriptomic data from that, some immune microenvironment data from that, but it’s also only looking at one spot and one tumor at one point in time. Whereas with the liquid, you can get tumor fragments of DNA shedding from any tumor, of any clone, at any site of disease. I also view it as a more global dynamic and real-time assessment of the overall status. Usually, they’re pretty concordant.

Van Morris:

Thanks, Dr. Pedersen. And yeah, I also order rapid ctDNA profiling for exactly the reason you just said. Which is because we have trials like BREAKWATER open at MD Anderson that require that patients cannot have had even one dose of chemotherapy, which was the case in this particular patient here. We can get that back within a few enough days that justify waiting for that result while we’re getting the insurance approval to get that patient otherwise going with standard chemotherapy. I will say one question, in leading the SWOG S2107 study that we referenced earlier.

When we wrote that trial, we said, of course the patients had to have a BRAF V600E mutation. And we said by standard sequencing profiling, and I actually got a question from a site recently that completely caught me off guard. Which is, “We have a patient who has a BRAF V600E colorectal cancer, but it was confirmed on an immunohistochemistry [IHC] test and is this okay to use?” And while we didn’t write that specifically in our protocol in the inclusion criteria, is this something you guys have seen, and what are your thoughts? Can we use a BRAF V600E by immunohistochemistry to make a decision for clinical care?

Aparna Parikh:

Yeah, I mean I think going back to this case, so V600E, sure. I think the problem is that if you rely only on immunohistochemistry for V600E, you may be missing out other important alterations. I think for colorectal cancer at least we have started to classify some of these other non-V600E alterations into different categories. Class II, class III V600 alterations including E, but other V600 sort of class I. And each of these classes have different kinase activity and RAS dependence and based on this, have responses or lack thereof to actually just anti-EGFR therapy alone. There's data, for example, in the class III alterations that you could actually get responses to single agent anti-EGFR therapy. Again, these are much less common. I think for that question, for that investigator, I think they’re reasonable, but it behooves us to think about more comprehensive testing for other non-V600 alterations too, which may be missed.

Katrina Pedersen:

There is also a level of subjectivity anytime you’re doing IHC for testing. But one other practical consideration that if you have any access to these BRAF inhibitor trials, most of them specifically are saying right now that it has to be a DNA-based sequencing-based assessment rather than IHC. There’s that practical consideration. In addition to all the other practical considerations you gave.

Van Morris:

Yeah, and you’re right, I mean BRAF V600E mutations, I think account for around 95% of all BRAF mutations in colorectal cancer. But as Drs. Pedersen and Parikh are referencing, there are still the 5% of the BRAF mutations, which can be activating are aren’t the more prognostically unfavorable V600E mutation, which is the class I.

I think this is an important opportunity when we’re talking about atypical BRAF mutations to remember that there are class II and class III subcategories. Class II, the mutation, it’s a dimer. Two of the mutated proteins interact to trigger downstream signaling. And they’re working upon dimerization to their independence of upstream RAS/EGFR activation.

On the other hand, the class III mutations, these are monomer mutations and are dependent on upstream RAS activation and often have to heterodimerize between the mutated BRAF and a CRAF or something like that to subsequently trigger downstream MAP kinase signaling. Because the class III mutations are dependent on upstream MAP kinase activation, in these retrospective series have been associated with some benefit with anti-EGFR therapies. Whereas the class II, again, because they’re dimerizing together and are not dependent on upstream RAS activation, in retrospective, small, single-institution series have not seen as strong of benefit with anti-EGFR antibodies. And this is where maybe doubling down on ERK inhibitors or pan-RAF inhibitors in clinical trials are looking at this.

One question I would open to the group as well, is a patient like this who comes with an atypical BRAF mutation. If an oncologist looks at this result and says, “Oh, the patient has a BRAF-mutated colorectal cancer, Imm going to give this patient the BEACON regimen of encorafenib and cetuximab.” Is this something that’s likely to benefit them?

Katrina Pedersen:

Honestly, probably not much, just because as you’re describing, the biology is different. You’re not going to get that inhibition that you’re going to see in a truly V600E patient. You'’e also going to run into some insurance problems getting that approved because they’re going to be looking specifically for the molecular data that support that there is a V600E. I don’t know if you’ve had a different experience clinically, Dr. Parikh.

Aparna Parikh:

I think that’s exactly right. I think these are great patients, again, to think about clinical trials for. But as Dr. Morris alluded to, especially for the class III independent of a trial, have tried to just get even monotherapy and have had some patients benefit. But these are important patients to put on these trials with ALT inhibitors, ERK inhibitors and try to harness the collective data of these rare alterations to get more understanding of how these patients will or will not benefit. But yeah, my practice has been either trial or for class III, especially monotherapy with anti-EGFR.

Van Morris:

Good. This brings us to the end of our discussion on tailoring targeted therapies in the treatment of BRAF-mutated colorectal cancer. I hope the viewers have found this as informative as I personally have today, and I really just want to thank so much my colleagues, Dr. Pedersen and Dr. Parikh for participating in this educational roundtable for JNCCN. Thank you.