PD-L1 Testing for NSCLC

Panelists: Mark A. Socinski, MD, Florida Hospital Cancer Institute; Tracey L. Evans, MD, University of Pennsylvania; Alexander Drilon, MD, Memorial Sloan Kettering Cancer Center; Vali A. Papadimitrakopoulou, MD, The University of Texas MD Anderson Cancer Center; Jared Weiss, MD, UNC Lineberger Comprehensive Cancer Center


Mark Socinski, MD:
I want to go back to a point that I had raised early on. One of the reasons those 1600 patients got excluded from a previous trial was they didn’t have an adequate biopsy sample for testing, and the eligibility criteria for these immunotherapy trials has been PD-L1 positivity on a core biopsy. Thoughts about that? We’ve talked about it in our weekly tumor boards. Is EBUS (endobronchial ultrasound) okay? Can you do PD-L1 staining on a robust FNA (fine needle aspiration)? I’m not aware of any presented or published data yet, but thoughts about that?

Alexander Drilon, MD: I think we need to consider what else you want to test, including PD-L1, apart from molecular testing. I try to recommend, if people ask in the community, not just to get a FNA, but to try to at least get something, like 2 cores if it’s safe and feasible. That way you might have enough material to do everything that you want to do.

Mark Socinski, MD: Yes, I completely agree with you. I think the paradigm and our interventional radiologists are very good at this. They’ll get as many cores as they can and also get FNA—as long as the safety is there—but that provides a lot of material for testing. It allows us to do the PD-L1 testing. However, in Main Street USA, I don’t know if they’re tuned into this.

Jared Weiss, MD: But even in academia, some lung cancer presents more centrally. We have a real pneumothorax rate when we biopsy the lung by a CT-guided route. And even in academia, we tend to have more widespread availability of interventional pulmonology. Who’s going to be comfortable getting more robust samples through a bronchoscope? And I think there’s, sometimes, a pressure in academic practice to minimize that pneumothorax risk and get a FNA. There are data showing that you can get full molecular characterization off an ample FNA.

Mark Socinski, MD: I don’t disagree with that. The issue I’m talking about is PD-L1 testing.

Jared Weiss, MD: Right.

Vali Papadimitrakopoulou, MD: PD-L1 testing can be done, as well, from FNA, as long as they’re rich in cellularity.

Jared Weiss, MD: Absolutely. You have to keep in mind how this assay is done. There are more than 4 different assays by 4 different companies scored in different ways. The one that counts to clinically consider, if you have 50%, is 22C3, and that is scored only on tumor cells and without reference to staining intensity. So, while there are no actual data to back up what I’m saying, from a common sense standpoint, you can count cells on a FNA.

Mark Socinski, MD: It never stopped you before.

Jared Weiss, MD: Never.

Mark Socinski, MD: Yes, but I think for this whole PD-L1 testing issue, we do need standardization. I think we’re moving there. It’s not acceptable, in my opinion, to have multiple tests that are out there. We need to arrive at 1 single platform. Obviously, efforts are underway to do that.

Tracey Evans, MD: Is it worth rebiopsying just to get PD-L1? You have the plasma option for molecular, and we even do this fancy thing where we can do molecular off of the FNA rinse, but we won’t do PD-L1 staining. We’ll do that off of good FNA, but not the rinse. But we can do our next-generation sequencing off that. Is it worth re-biopsying for PD-L1? What disturbs me about this impressive data of first-line pembrolizumab is, why were these patients not salvageable in the second-line setting? Why was there such a dramatic survival advantage? It clearly seems to matter that these patients are treated with immunotherapy first.

Mark Socinski, MD: Yes. That was the surprising and yet concerning thing to me, because going back to our beginning, EGFR mutations, I’m not convinced that whether a patient gets a TKI in first- or second-line settings makes a difference. The issue is that you can’t risk giving it second-line because they might not get it. That’s the issue.

Jared Weiss, MD: Well, if you go back to the 10- to 15-year-old data on immediate docetaxel versus at progression, 40% of the patients in the control arm never got it. Either their progression-free survival event was death or they were too sick for second-line.

Tracey Evans, MD: But that doesn’t happen with TKIs. You can always throw that in there; it doesn’t happen as often.

Jared Weiss, MD: That’s a fair argument with TKIs, and perhaps as well with immunotherapy, that they’re gentler therapies.

Tracey Evans, MD: But they weren’t salvageable. Why weren’t the immunotherapy patients salvageable?

Mark Socinski, MD: Yes, but don’t assume the TKI. Remember that in the EURTAC trial, on the chemotherapy arm, only 81% of people got the TKI. Stuff happens.

Tracey Evans, MD: Right.

Transcript Edited for Clarity
Printer Printing...