Using Molecular Testing in GI Cancers

John L. Marshall, MD: The lung cancer specialists have been consistent, right from the beginning they want to know everything. Shub, from a GI [gastrointestinal] cancer world, we’ve gotten a little slower with this. What’s your current practice on when you’re doing testing, such as next-generation sequencing, plasma samples, and so forth, for NTRK fusions?

Shubham Pant, MD: I think since I was in fellowship. Of course not as long as Mark.

John L. Marshall, MD: 30 years!

Mark A. Socinski, MD: Thanks, Shub. Thanks so much.

Shubham Pant, MD: Sorry, Mark. At that time it was still the carbo-taxol [carboplatin-paclitaxel] days, but this pie chart was coming out. In all the GI meetings we go to, we always look at the lung cancer pie chart. We used to look at breast cancer when I was in fellowship. That changed to the lung pie chart, and we look at it longingly to see if we can bring the same things to patients with colorectal cancer and with pancreatic cancer. But it’s growing. Especially in the colorectal space where we have now with the BRAF mutations, we have drugs that have been approved earlier. With pancreatic cancer, where in the metastatic setting we can give olaparib for BRCA1 and BRCA2, though it’s for germline.

Still, I think the testing has been developing with the recent article in The Lancet Oncology, that if you treat patients with metastatic pancreatic cancer, also on the target agent, if they have an oncogenic driver, there you have a driver mutation—that’s the main thing. They really can do well. It’s changing the GI world more for colorectal than for pancreatic, but definitely for colorectal.

John L. Marshall, MD: Yeah, I feel like we’re about 10 years behind on that. Tony, what’s your take on plasma samples in GI cancers? What’s their role like for lung cancer folks? What’s your thoughts on that?

Tanios S. Bekaii-Saab, MD, FACP: They have a role depending on what you’re looking for. The fusions are certainly challenging with plasma, although as was said, this is getting a little better. The biggest element with plasma is you get the results quickly. If it’s there, it’s likely positive. If it’s not there, you may have missed it. So you still want to rely on tissue.

With plasma, its major utility comes in actually being able to use those sequentially during therapy at specific time points. But you do need to get your liquid biopsy before the infusion of treatment so that you know what the landscape of the tumor looks like.

Some of the interesting findings, actually, as we learn more and more in our different cancers to work with targeted therapy, which is something we haven’t done until more recently, is when you see these mutations arising, with the FGFR inhibitors, for example, that target the infusion. In biliary cancer, you see mutation rising that predict for recurrence, and actually there are agents that target most infusions and the mutations that have been developed in resistant patients.

We see with HER2-amplified tumors, where we see the high concordance rate between the liquid and the tissue, that MET amplifications are driving a lot of the resistance, and we have MET agents that are using in clinical research. I can go down the list with NTRK fusions, taking NTRK mutations. The question, of course, is how much do you use that in actual clinical practice? You don’t have an agent that targets that mutation that appears or driving pathway, then it’s mostly an experimental drug.

But prior to treatment, I can tell you multiple times where I haven’t had enough tissue for a colon cancer patient, a BRAF mutation pops up and completely changes my treatment paradigm from a doublet to a triplet. It also pushes the patients to clinical trials. We’re moving these targeted BRAF inhibitors into the first-line setting, and actually, the study allows for liquid biopsies because we understand that those patients cannot wait. It could sometimes take us 4 to 6 weeks to find the tissue and test the tissue. So it facilitates progress in that sense.

Transcript edited for clarity.