Transcript:Benjamin P. Levy, MD: Let’s shift gears here. And to quote a former panelist, “Physicians may have a difficult time finding a good interventional radiologist, but probably not that hard of a time finding a good phlebotomist.” And the role of plasma and genetic interrogation for lung cancer is here. It’s not science fiction, this is something we’re doing. We have multiple platforms that are being used; of course, the minimally invasive approach. Chandra, can you give us a sense of where we are with plasma cell-free DNA and how we can use this routinely under the auspices of a clinical trial, either/or?

Chandra P. Belani, MD: Yes, Ben. I think that the liquid biopsies are right, and they will probably change the whole therapeutic landscape, and the prognostic and predictive landscape for genotyping. It’s probably of benefit in lung cancer patients when we need an invasive biopsy sometimes, and sometimes the patients actually are not willing to undergo a biopsy. So, these liquid biopsies will probably be much easier. And what is now known is that liquid biopsies have been validated in certain instances, especially for EGFR mutations and KRAS mutations. And they have a positive predictive value of close to 100% in those cases.

When you are looking at resistance mutations like the T790M mutations, the positive value of sensitivity may be close to 79% or 80%. Even if it is negative, there is a 20% chance that the T790M could be positive in the tissue. I think, especially if you are going to give these targeted therapies, and we are going to change therapies with the availability of agents which targets those resistance mutations, we probably will be better served for our patients to do liquid biopsies and send it for circulating or cell-free DNA.

Benjamin P. Levy, MD: I’ve been very encouraged by the data. There have been some nice data, as you alluded to, showing a very high positive predictive value of 100%. If it’s positive, it essentially rules it in. If it’s negative, however, you need to do the biopsy. I would say we’re 96 hours in to an FDA-approved blood-based cobas EGFR test that just was announced, so I think this is something that is coming into routine care. So, yes, you are able to identify these patients up front. Do you think it will ever circumvent the need to do tissue biopsy, or are we really wedded to doing this?

Chandra P. Belani, MD: I think the technology is advancing very rapidly and that we probably will be able to circumvent that, provided there is not a lot of heterogeneity within the tumor tissue and you’re not able to detect that in the DNA—which has shed from the tumor into the blood. So, I think we are there, but we are not 100% there yet. The sensitivity for detecting specific mutations may still be lower, but I think there are capabilities that are getting better as time advances.

Benjamin P. Levy, MD: Yes, I would agree with you. I think there are many platforms out there that are coming to a physician near you that are saying that they can do genetic interrogation off the blood. There’s certainly a need there and certainly some of these platforms are sensitive and accurate. But none of these platforms, to my knowledge, have had concordance with tissue yet. The first perspective study to do this concordance with paired tissue were the data in Gem Pharmaceuticals Oncology recently. Do you ever foresee a time where there’ll be the ability to do longitudinal samples of EGFR, and see if there’s a drop in copy number that would be like the story with BCR-ABL with CML?

Chandra P. Belani, MD: I think that we are already seeing the data. The data are being presented at this meeting where you see a 50% reduction in a copy number of the EGFR, especially in patients when we are using the third-generation inhibitors for brain metastases and leptomeningeal disease. You have started to see that in the periphery in the liquid biopsies, the reduction in the EGFR copy numbers. So, I do think that we are there, we’re not there 100% yet, but I guess it’s a good and a rapid advance.

Benjamin P. Levy, MD: Yes. Any other thoughts about cell-free DNA and the routine use of it in clinic?

Ramaswamy Govindan, MD: I think Chandra is right. We should just remember that if it is negative, it doesn’t really tell us that it’s not there. And the other advantage is that potentially you may not find it on a particular tumor biopsy, given alteration. But when we present it in the blood, it’s highly specific. This may come from a source that you’re not biopsying from. So, that’s a big advantage. I think it’s in the next few years you’re going to see all those things.

Chandra P. Belani, MD: And if you get a positive result, at least you have it. For the negative result, we may have to go and pursue it further by getting tissue. But at least if you have a positive result, we can start the treatment in that case.

Sarah B. Goldberg, MD, MPH: And it’s often really helpful, because you can’t always get a biopsy in patients. Sometimes there’s not enough tissue and patients don’t want to undergo a second biopsy. Or again, like I was saying before, time is important and you don’t have the time to do another biopsy. So, I think both at pretreatment when someone is first diagnosed and then at resistance, it can be really helpful and quick. You can get results quickly instead of having to do a biopsy. You get the tissue, send that for testing, so it can really be done quick.

Mark G. Kris, MD: And results are in days rather than weeks.

Chandra P. Belani, MD: It’s only 48 to 72 hours you’re able to get it, as you just said.

Ramaswamy Govindan, MD: There’s one problem though, reimbursement. Because you’ll be doing this again and again. If you’re doing a large panel, then they show us, but those things are being reimbursed, and that can be the bottleneck for a while.

Mark G. Kris, MD: Well, we seem to have gotten over that with PET scans. PET scans to check response are not approved by any guideline. People do them all the time. And what does a PET scan cost at your institution, $7000, $8000?

Ramaswamy Govindan, MD: It’s come down quite a bit. They have to come down.

Mark G. Kris, MD: Yes. So, these things will get worked out. And I think the important thing is if they really impact somebody’s health, accelerate care, and help you get the right drug, it’s going to eventually get figured out.

Benjamin P. Levy, MD: Yes, I agree. I think a rapid test like this that’s minimally invasive, that’s a fairly accurate molecular proxy of disease where we can make treatment decisions based on these tests. I can tell you at our institution we’ve had a tough time with tissue stewardship, of the biopsy getting to the pathologist, getting results, and then getting it back to the treating oncologist. We’ve had turnaround time days that exceed 21 to 28 days. So, for us, this has been a welcome change, and it’s a win for the patient in a lot of ways.

Transcript Edited for Clarity