Precision Medicine in NSCLC: Ramifications of Recent Data - Episode 4

How Are We Using Liquid Biopsy in NSCLC?


Benjamin Levy, MD: Speaking of plasma, getting back to the real utility of plasma and liquid biopsies of identifying genetic aberrations that may be missed in tissue, we’ve come a long way in the past 5 years having liquid biopsy, the clinical utility, identifying T790M, but more recently looking at the role of layering in liquid at the time of the first treatment or before treatment in the treatment-naïve setting.

Josh, do you want to talk to us about some of the data we’ve seen recently about the role of liquid biopsy for treatment-naїve patients in lung cancer?

Joshua Bauml, MD: Absolutely. As you mentioned, liquid biopsy is really emerging as a good way to obtain molecular information. It is not 100% sensitive. The sensitivity is somewhere around 70% to 80% in most of the series and thought to be a little bit lower for fusions. The issue is that when you find something on a liquid biopsy, it is actionable. You can take it to the bank and do something with that. As you were mentioning before, Ross, the turnaround time is often faster, so there are a lot of benefits.

One of the issues that many people have spoken about in the community as well as in academia, is not having enough tissue with a new diagnosis of lung cancer. Our group did a series like my colleague, Dr Charu Aggarwal, MD, where we looked at tissue biopsies versus plasma biopsies, and we found that a good portion of patients were not able to have enough tissue to do molecular profiling. And that’s at a center where very good interventional pulmonologists pride ourselves on getting this information. And what they were able to show was that when you cross molecular targets, not just EGFR, when you identify a target on a liquid biopsy, you could act on it, and the treatment response is similar to what you’d expect from a tissue-based biopsy.

Benjamin Levy, MD: Based on that and then we’ve got the NILE study data, too, that have been recently published showing essentially the same thing, that if you layer in liquid at the beginning, you’re more likely to pick up on actionable mutations that can drive treatment decisions. Let’s do another vote. I like the votes. Who uses liquid biopsy on every patient when they walk through the door, every patient? Let’s do this. Who does liquid?

Zofia Piotrowska, MD: Can we say who thinks we should be using this, as a first step?

Benjamin Levy, MD: As chair, I’ll go with this question I went with at the beginning. Who uses a liquid biopsy on all stage IV lung cancer patients who have had a biopsy that has confirmed adenocarcinoma where NGS [next-generation sequencing] is pending?

D. Ross Camidge, MD: I don’t know.

Joshua Bauml, MD: Pending or...?

Benjamin Levy, MD: Don’t know if you’re going to have it or not.

Joshua Bauml, MD: Oh, I don’t know if I’m going to have it or not.

Benjamin Levy, MD: It’s been biopsied. It is CK7 [cytokeratin 7], TTF-1 [thyroid transcription factor-1] positive. It’s been sent off to the in-house assay, you’re waiting for the result, the patient is in your office, you’re telling them, as you usually do, we’ve got about 2 to 3 weeks to wait here. Does everyone at that point use a liquid biopsy?

D. Ross Camidge, MD: For me it’s, you know how you can do ALK IHC [immunohistochemistry] and it gives you an answer in relatively few days or some people have a rapid EGFR test, and yet you’re still sending off that big next-generation sequencing panel? It’s like that. You’re spreading your bet, so that if you get a result back earlier you can act on it.

Benjamin Levy, MD: Yes.

Robert Doebele, MD, PhD: I think the most common situation that I use it, and it’s very helpful, is a patient comes to me and they have insufficient or depleted tissue, and I know I’m going to have to order a new biopsy, so I’m ordering that biopsy at our institution that’s taking 7 to 10 days. I’m likely to get the ctDNA [circulating tumor DNA] back that I’m drawing in the office that day before the biopsy happens. I can cancel the biopsy and remove any risk of that.

D. Ross Camidge, MD: But do you cancel? Well....

Robert Doebele, MD, PhD: I cancel the biopsy. The NGS testing is already running. It’s harder to cancel, yes.

Joshua Bauml, MD: I do that. The other setting in which I’ve done it is if I have a second opinion, the patient had a biopsy at their local institution, I need to get the block. I have no idea what the block is going to look like or where things are, then I will do a liquid biopsy. But if something is internal and our pathologists have seen it, they say there is adequate tissue for molecular, then I don’t do a liquid biopsy in that setting.

Benjamin Levy, MD: Go ahead, Zofia.

Zofia Piotrowska, MD: My perspective is that in an ideal world, I think, absolutely, we should be sending both of these on this test. I believe the data that these can be complementary, and if you are able to send and get both, you’re going to pick up more than if you send either of the other. I think, Ross, you brought up the issue of health economics. I think that there is a concern here about insurance coverage and sending them off. And so that’s been my hesitation. I would like to send it for everyone while the tissue is pending, but I want to reserve that for the patients where I really feel like I need it just in case there are any issues with coverage.

Benjamin Levy, MD: I tend to use it on just about everyone who has a biopsy done. If I know and the pathologist can ensure that there is adequate tissue, then maybe I’ll hold off on an asymptomatic patient. But for patients in which there are second opinions where the tissue block’s in the never-sphere somewhere out there that I’m trying to track down, that’s certainly an opportunity. And then we learned from your data, from Aggarwal’s data from the University of Pennsylvania, we’ve learned from NILE, we’re just not as good as we think we are at completing tissue genotyping. In the NILE data, less than, I believe, 25% of patients had completed biomarker testing for all targets, and that was under the umbrella of a clinical trial where they’re mandating tissue and concording it with plasma. I think this is a discussion. I’m being somewhat controversial here just to kind of elicit discussions, but I don’t think we’re as good as we think we are even in the halls of academic medicine. And I think oftentimes we think we have enough tissue, and then we don’t. Health economics I think are a big deal. Can we make this cost-effective for patients? I think that will be important.

Joshua Bauml, MD: One thing that is helpful is that our pathologists in the pathology report will write, this is the diagnosis, adenocarcinoma of the lung. They’re like, there seems to be adequate tissue to do molecular testing, and we’re going to do that.

Zofia Piotrowska, MD: That’s nice.

Joshua Bauml, MD: That’s been very helpful. And they’ll also write, this is not a big biopsy. There is scant tissue. We might be able to do it, and so that helps guide my decision making.

Zofia Piotrowska, MD: Yes.

D. Ross Camidge, MD: I want to pick up on one thing that Josh said in the sense that everything you see on the blood-based thing you can believe, you can take to the bank, I would say with the exception of gene amplification. I do not believe that NGS in the blood really is a good way of calling it gene amplification.

Joshua Bauml, MD: You think if you see gene amplification in the blood, you don’t believe it?

D. Ross Camidge, MD: No.

Zofia Piotrowska, MD: It maybe over calls it?

D. Ross Camidge, MD: I think it over calls it. Because, first of all, it’s like 1+ 2+ 3+, it’s like what does that mean, and it all depends on the bioinformatics and it’s a black box. And certainly, we’ve had people have that called, and then you do FISH [fluorescence in situ hybridization] and there’s no amplification at all.

Joshua Bauml, MD: The other place where there is, it’s not a false positive, but the other factor that needs to be discussed is CHIP, this clonal hematopoiesis indeterminate potential.

Benjamin Levy, MD: Can you explain that?

Joshua Bauml, MD: Absolutely, I can explain that. We have blood, and inside the blood there’s the plasma. There’s DNA floating in the plasma, the cell-free DNA [cfDNA]. But we’re doing this analysis. You can also get DNA from the white blood cells, the red blood cells, and they can get mutations. That’s what MDS [myelodysplastic syndrome] is. Patients can have this clonal growth of mutated cells, and they can be identified with recurrent mutations. Some of those mutations are the same mutations that happen in cancer. For instance, p53, pretty common mutation in both CHIP and in cancer. Luckily, for the purposes of our analyses, there is no CHIP mutation that is associated with a targetable alteration, which was true, up until today. Because one mutation that is seen in CHIP is KRAS.

Benjamin Levy, MD: It’s rare, but you see it.

Joshua Bauml, MD: It’s rare and it can be seen. The one thing that I will say is that if I have a patient who their liquid biopsy only comes back with KRAS, I have tended to regard that as a not informative assay, and I will repeat it with the tissue.

Zofia Piotrowska, MD: Are they the same KRAS point mutations in CHIP as in lung cancer?

Benjamin Levy, MD: I don’t know if that was described, and Josh wrote a very nice editorial last year on this looking at CHIP. It is rare but KRAS is there, and I think it changes our algorithm because if we see a KRAS in the plasma, we could essentially say, you know what, it’s probably in the tissue, it’s not druggable. I don’t even need to follow up on the tissue biopsy.

Zofia Piotrowska, MD: The person probably doesn’t have EGFR.

Benjamin Levy, MD: But there is a rare chance that is a CHIP mutation and the patient harbors a totally different alteration in the tissue, and that may have been missed.

D. Ross Camidge, MD: Did we get a clue from the percentage? Is CHIP like 30% in the cfDNA?

Joshua Bauml, MD: What the percentage can do, is when I see that, I tend to think more of germline changes. It’s not as reliable because not all the cells are going to have it. So, you can still see it. I will say that the editorial that I wrote, the coauthor on it was also very good, Ben.

D. Ross Camidge, MD: Did you just promote yourselves?

Benjamin Levy, MD: I never prompted him to say that either.

Zofia Piotrowska, MD: The one other key point about plasma that I think for most people is something that we all know and accept but always worth pointing out, too, is that negative plasma results are another key limitation, right? If you send off the plasma test and you don’t find anything, that does not mean that you can stop and that the patient does not have an actionable mutation. I think that’s always an important point to remember.

Benjamin Levy, MD: Not everybody sheds.

Zofia Piotrowska, MD: Exactly. You may have someone who is an EGFR patient or KRAS or anything and is just a non-shedder, and that plasma may come back negative. I think that it’s always really important in the face of a negative plasma test to still go back and do tissue testing.

Transcript Edited for Clarity