John Marshall, MD: Paul, our lung cancer colleagues are very used to doing liquid biopsies. It’s one of their standards, even as a primary diagnosis for molecular abnormalities. Good turnaround time—you don’t have to wait on your pathologist. Should a general oncologist be doing liquid biopsies on our patients with pancreas cancer?

Paul Oberstein, MD: I would say it’s even more challenging than doing the tumor biopsies. In lung cancer, where we have well-identified mutations, you can select for those when you look in the blood. We don’t exactly know what we’re looking for. It’s to know if we’re missing it or if we’re not, because we don’t know what it is. In experimental settings, we do it all the time. But so far, we don’t know if we’re going to capture enough information from the blood biopsy.

John Marshall, MD: Yes, a positive one is very helpful. A negative one doesn’t absolutely rule it out, though, so you’re then having to do tissue-based biopsy. Is everybody tissue-based yet? Allyson?

Allyson Ocean, MD: I do choose to do a liquid biopsy if I cannot get enough tissue.

John Marshall, MD: Had you been doing this?

Allyson Ocean, MD: I have.

John Marshall, MD: I’ve been doing this. It’s the second or third pass at a biopsy, and you’re still not getting it.

Allyson Ocean, MD: Right, yes.

John Marshall, MD: Could we just find a RAS mutation and call it a day?

Allyson Ocean, MD: Honestly, it’s sometimes the first question. The patient is like, “Can’t you just get it from my blood?”

John Marshall, MD: Right. The newspaper tells them they can get it from their blood.

Shubham Pant, MD: The thing is, it can be more challenging to find blood. It just doesn’t shed that much, so it’s hard to find that mutation. In pancreatic cancer, that’s a tougher one. The other thing about coming back to germline, and I completely agree with Allyson, is that you can really identify them for frontline platinum therapy. Even for classic chemotherapy, you can choose to treat patients with classic chemotherapy based on germline data. Now, somatic—is it actionable or not? What are the exact mutations? There can be different mutations that you pick up on somatic. I think the data are much muddier for somatic. That’s not in prime time right now.

John Marshall, MD: All right. We’re sort of ivory tower dwellers, and so we say everybody should be tested. Are you not giving treatment, Ed, until you know the results? Are you sending and treating? We’re talking about platinum based versus not.

Edward Kim, MD: The challenge is the turnaround for what we treat. We don’t have time to wait, so there is that appeal for liquid biopsy of shorter turnaround time. As far as using the information from the actual tumor tissue to make decisions, where I’m incorporating it right now is not as the very first step. It is important to capture all these patients. To get to Shubham’s point about whom we test and when to test, I think the goal is to test everyone, to get that information at some point. Whether you can use that information before you start a first treatment for a metastatic patient, it’s still challenging.

John Marshall, MD: Allyson, you know I think you’re really good in this space. How do we get the word out? How do we make sure the oncologists are ordering it and patients are aware?

Allyson Ocean, MD: Right. It’s been my goal to get the word out. I think social media is such a powerful platform. We use Twitter. Twitter is the biggest platform for the dissemination of medical data, and if anyone is on Twitter, you can share the data that have come out. All announcements about screening studies—the GENERATE study just opened. The GENERATE website is www.generatestudy.org.

It’s a large study sponsored by Stand Up to Cancer and Lustgarten Foundation where family members of someone who is known to carry a genetic mutation can get screened and then find out more if they’re at risk. I have created Let’s Win!, which is an online platform to disseminate information about pancreatic cancer. We have subjects each month, and one of our subjects is genetic testing. We try to get the word out every month with information about why it’s important to check.

John Marshall, MD: I almost think it’s easier on the patient caregiver side because they’ve got 1 disease, right? They’re going on Pan Can [Pancreatic Cancer Action Network], which is very effective.

Allyson Ocean, MD: Right.

John Marshall, MD: They’re going on your website, and they’re watching Twitter. I’ve already gotten an e-mail from 1 of my patients because of the CNN report about the new PARP [poly ADP-ribose polymerase] inhibitor data. They’re fairly aware, but our oncology partners are getting 40 new drugs a year and new genes that need to be tested. Let’s say your pancreas cancer patient is now in second-line therapy and you didn’t do it. Is that a patient I should now be sending down to genetics, Paul, and getting tested?

Paul Oberstein, MD: I would say, for the germline genetics, yes, because the best person to test is the patient. It gives information about them potentially for their disease treatment, and as we talked about, for their families. Their family members will want to know if they have a mutation or not.

John Marshall, MD: Yes. One of the things I have always been a little envious of in breast cancer—one of the many things—is that every HPI [history of present illness] of a breast cancer patient begins, “48-year-old, ER/PR [estrogen receptor/progesterone receptor], HER2,” right? In GI [gastrointestinal] cancer, we still don’t have that culture where that first line is this. I think the more we do that, the more we teach our fellows, the more we share that information as we talk to one another, I think that comes forward.

Paul Oberstein, MD: The guidelines have just changed.

John Marshall, MD: They just changed, right?

Paul Oberstein, MD: I think that’s going to have an impact in doing exactly what you said. The guidelines are a powerful group of experts saying, “Everyone should be considered for genetic testing.”

John Marshall, MD: It is a lot. Sending somebody for germline testing is not just the tick box on the LabCorp sheet, right? You need to have genetic counselors, often, to support this.

Allyson Ocean, MD: It’s another appointment for the patient when they’re sick already. I worked with my genetic counselors to have the kit in my office so that they don’t have to go. We have our genetic counselors come over and meet with the patient. I encourage practices to do that—work together, because it makes it easier on everybody.

John Marshall, MD: When we tweak the guidelines, we have to remember this does have some downstream effect and ask if it is really worth it. Is it worth it? What percentage of patients are going to come back positive with something like this?

Shubham Pant, MD: In the germline data, it’s about 5%, so more BRCA2 than BRCA1. When you look at different data—when they tested in the POLAR trial, it was about 7.5%; 5% got on trial. When you look at it across different trials, it’s in the 5%-to-10% range—maybe more toward 5% than 10%.

The somatic is about a 2% to 1.5%. Then you’re getting into the minutia of the BRAF, FGFRs, and MSI [microsatellite instability], which is about 1%. When you get it together, it’s about 20% or 25% now. The other information we should get out there is, when you look at the whole DDR subset, that’s about 24%. However, it’s not clear right now if a PARP inhibitor really helps with the ARID1A mutation or PALB2. The jury is still out on that. Our focus right now is BRCA1 and BRCA2, and we have data. For ATM, we have some data that you may use a PARP inhibitor. Again, we’ll discuss the POLAR trial later. I’ve been less enthusiastic about it than I was before, and I was really enthusiastic about it until yesterday. I was still happy, but I think what happened is that caution went down a couple of notches.

Transcript Edited for Clarity