Upfront Immunotherapy for PD-L1+ NSCLC

Transcript:Mark A. Socinski, MD: Hello, and thank you for joining us in this OncLive Peer Exchange® titled “Nuances in Treating Advanced Lung Adenocarcinoma.” As we continue to see progress in development of new therapies for advanced lung cancer, we also endeavor to refine the use of systemic therapy. In this OncLive Peer Exchange® panel discussion, I am joined by a group of experts in the field. Together we will help you to navigate the ever-evolving landscape of treatment options and the nuances of integrating newly approved agents and new indications into current practice patterns.

I am Dr. Mark Socinski, and I am the executive medical director at the Florida Hospital Cancer Institute in Orlando, Florida. Participating today on our distinguished panel are Dr. Ross Camidge, professor of medicine and oncology and the Joyce Zeff chair in Lung Cancer Research, and director of thoracic oncology at the University of Colorado Cancer Center in Denver; Dr. Mark Kris, the William and Joy Ruane chair of Thoracic Oncology at Memorial Sloan-Kettering in New York, New York; Dr. David Spigel, the chief scientific officer and lung cancer research program director at the Sarah Cannon Research Institute in Nashville, Tennessee; and Dr. Jared Weiss, associate professor of medicine and section chief of thoracic and head and neck oncology at the UNC Lineberger Comprehensive Cancer Center in Chapel Hill, North Carolina. Thank you for joining us. Now, let’s begin.

I want to start with an evolving therapeutic paradigm. About 2 years or so ago, we had the initial approval of nivolumab as the first immunotherapy agent, and then approximately 8 months ago, we saw some very exciting data selecting patients by PD-L1 expression who were very high expressers, greater than 50%, in a very well-conducted phase III trial comparing pembrolizumab to standard platinum-based chemotherapy. So, Dr. Kris, give us your perspective on that.

Mark Kris, MD: This trial was really a game changer. I think the first thing is that it led to the need to test every single patient with squamous cell lung cancer and adenocarcinoma to see what their PD-L1 expression is. But the next thing it did, it showed that people who had a high degree of PD-L1 expression—about one-third of the patients—had the trifecta plus. They had better rates of response, greater progression-free survival and overall survival, and a substantial decrease in the degree of serious side effects. Never before have we seen that. The other critical piece of new information is that these patients not only had good responses, they were very durable responses. And some people did not seem to recur ever, which is really amazing. Now, that’s the good news—that’s roughly one-third of the patients. Then there are the two-thirds, and we’ll be talking about them, as well. But this trial was really a critical one. It’s a fundamental change in care, and I think every oncologist now has to do PD-L1 testing for every patient with squamous adenocarcinoma.

Mark A. Socinski, MD: I completely agree with you. We do it reflexively at our place. I think that’s the way it should be done. It’s a simple immunohistochemistry test, once you set up the platform. Many oncologists are sending it out. We do it internally, so I think you have to work that out. Let me ask you this: The KEYNOTE-024 trial enrolled, or screened, 1925 patients. Only 305 were randomized. If you do the math, that’s 15.7%. So, even though about one-third of those, or 30%, were greater than 50% positive, there are lots of contraindications to using that. Is this really applicable to 10% to 20% of the average oncologist population in lung cancer?

Mark G. Kris, MD: It’s hard to say. I think the magnitude of the benefit is so compelling that it’s vital to offer the option to people who have it. As for how it relates to the whole population, I think that we’re going to see many, many more trials that try to tell you what the real-world situation is going to be. But I think testing it is important, and, just as you say, doing it reflexively. Just a shout-out to the oncologists: You have to meet with your pathologist and make sure that it’s clearly explained why you need this test. This isn’t some kind of crazy prognostic marker that may mean something. This is a very important test.

Mark A. Socinski, MD: This is actually one immunohistochemistry test we want them to do.

Mark G. Kris, MD: They’re unbelievably skilled people, and they’re really skilled in doing those tests. I think it’s important, though, that they want to be part of important decisions in care. This is a great opportunity for them to do that. It adds to their armamentarium of things that really make a difference for patients. But you’ve got to talk to your pathologist.

Jared Weiss, MD: We now do this as a reflex test at my place. Any non—small cell lung cancer, squamous or nonsquamous, at the time of diagnosis gets a PD-L1 test, and the cell to achieve that was that this is like the ER testing of lung cancer—you need this for treatment decisions.

Mark A. Socinski, MD: Yes. One of the issues that we struggle with is that many of our patients are diagnosed initially with an EBUS (endobronchial ultrasound bronchoscopy) or something like that, and they have FNAs. We know that PD-L is really only validated on core biopsy material. I actually don’t know what the status is of doing PD-L1 testing on something other than a core biopsy. Any experience or thoughts?

Mark G. Kris, MD: It can be done. Just as you say, it’s an evolving field. There are much less data out there, but you can do it. It is part of our standard armamentarium now, and every patient gets it. By the way, every patient that we have also gets ALK IHC and EGFR L858R IHC testing.

Ross Camidge, MD, PhD: To be clear, whilst you can do it on all sorts of specimens—because it’s a proportion score—the more of the tissue you’re looking at, the more accurate it is. But for ALK IHC, 1 cell being positive could conceivably be enough. So, I do think in cytology—where it’s not just positive or negative, it’s positive above a certain level—that that is still a work in progress.

Jared Weiss, MD: Are any of you pushing for core biopsies over FNAs in patients for this reason? Has it changed your biopsy practice at all?

Mark G. Kris, MD: I’ll confess it hasn’t changed mine at all, despite recognizing the lack of real validation in FNA.

Mark A. Socinski, MD: Well, I’ve had a couple of patients over the past couple weeks in which the initial material attained by FNA was less than optimal—let’s leave it at that. So, when you go back and have to do a second biopsy, we’ve been a proponent of trying to say, “Can you do both? Can you a get core biopsy safely, and can you get some FNA so we have plenty of material to do this?”

Mark G. Kris, MD: But the EBUS, too, it’s very often a collection of cells. So, it’s not just a few single cells. They collect it in a medium; they spin it down. It’s not physiologic, as it were, but they’re able to do what you said there, Ross. They’re able to give a proportion score.

Mark A. Socinski, MD: I want to explore one issue. I believe at the time that they reported the KEYNOTE-024 trial data—and I do agree that the survival was very impressive in favor of single-agent pembrolizumab—only about 45% of patients on the chemotherapy arm had crossed over at that point. Obviously, this will evolve over time. We may be seeing some updated data soon. Any thoughts about that in terms of sequencing? Is there something special about giving immunotherapy in the strong expressers first? Because the data in KEYNOTE-010, in the high expressers, was actually also pretty good in the second-line setting after chemotherapy. Any thoughts?

Mark G. Kris, MD: The tenet of oncology, when I went to oncology school, is that you give your best drug first. And even better, you give your most efficacious drug and your least toxic drug first. That’s the reason to do it. Does it mean that the people couldn’t have benefit later on? Of course not. But you want to give the best drug first.

Jared Weiss, MD: And crossover in the real world is not perfect—certainly not.

Transcript Edited for Clarity