Practice Changing Trials in Advanced Non-Small Cell Lung Cancer - Episode 5
Benjamin P. Levy, MD: Let’s move on to the next section, which is looking at immunotherapy in advanced-stage disease. This is obviously an area where immunotherapy got its name in lung cancer. Checkpoint inhibitors with pembrolizumab, nivolumab, atezolizumab all are cemented in stage 4 paradigms in the refractory setting, initially. But we’ve had more and more data looking at these drugs in upfront settings, both as single agents in an enriched group of patients who were PD-L1—positive and in combination with chemotherapy, which has created a lot of enthusiasm coupled with a lot of confusion. So, let’s start out with the frontline data with a single-agent pembrolizumab from the KEYNOTE-024. Roy, you want to talk about that?
Roy S. Herbst, MD, PhD: Sure. It’s funny, patients who were treated in some of the early phase I trials, I still remember one coming up to me and asking, “Dr. Herbst, if you knew it was going to work this well, why did I have to have all of that chemotherapy first?” And, certainly, the earlier we can bring these drugs to patients, the better. But again, for that we want to have predictive markers to know that they work better. So, the KEYNOTE-024 trial really was historic in the fact that it took the biomarker that had been developed in the refractory setting, that being PD-L1 by immunochemistry, a score or percentage of cells positive, looking at greater than 50% versus not. In fact, this trial only included the high-risk patients. And it took that very select group that we knew from the phase I and the phase III in refractory disease, that really had the best response rate and the best outcome, and compared it to chemotherapy. Now, this was a trial that you would imagine is harder to accrue because only by definition, 20% to 25% of patients are going to be eligible and then there’s all the other screenings. Remember, it was a selected trial, but in this head-to-head comparison of chemotherapy—and it was a dealer's choice chemotherapy versus the pembrolizumab, the PD-1 agent—I believe the hazard ratio was quite good. It was like 0.5 or something like that, 0.6. And even better, the response rate was about 45%. Who would have thought we would have 45% response rate without chemotherapy or targeted therapy?
And then the PFS was 0.5. So, this was a really amazing result, and it really showed that in selected patients, one could use immunotherapy without chemotherapy. And this became immediately the standard of care for patients who were PD-L1—high. It also then brought the profile of PD-L1 to the forefront. I think probably all of us at our centers want to have a PD-L1 on our patients. And these data have held out. When first presented at ESMO a couple years ago, it was rather early, but now there have been subsequent viewings of the data and the benefits continued. And the toxicity, while different, obviously the toxicity of immunotherapy, grade 3/4, is less than what you get with chemotherapy. So, this really changed the entire landscape of how we treat lung cancer.
Benjamin P. Levy, MD: It’s interesting. They had to screen 1900 patients to get those 300, but nevertheless, the data, as you’ve mentioned, are compelling. The update from World Lung shows a median overall survival of 3 years in those patients who were getting single-agent pembrolizumab.
Hossein Borghaei, DO, MS: Which is amazing.
Benjamin P. Levy, MD: And that is unheard of in a nontargetable population without a genetic alteration that we can act on with a targeted therapy.
Naiyer A. Rizvi, MD: I think a really important part of the update, that I think Julie Brahmer of Hopkins Medicine presented last year, was that patients who received chemotherapy were randomized to chemotherapy and then went on to receive PD-1 after the chemotherapy.
Benjamin P. Levy, MD: Yes, 60% crossover.
Naiyer A. Rizvi, MD: Which is part of what was allowed. The response rate was much lower with pembrolizumab…
Benjamin P. Levy, MD: It was 20% to 25%.
Naiyer A. Rizvi, MD: Post chemotherapy, and also the overall benefit was less. So, I think for physicians who argue, “Well, I’m going to give chemotherapy first and give PD-1 second, and I’m not going to need to PD-L1 test,” I think this would really argue otherwise that these drugs are more effective positioned early. And so, you really got PD-1s caught up front.
Hossein Borghaei, DO, MS: I think that’s a really important point to emphasize about the study. It was amazing to see that. For this particular patient population, immunotherapy was definitely better up front than in the second-line setting.
Benjamin P. Levy, MD: Anyone on this panel is an enriched group of folks here, but is anyone having trouble getting PD-L1? And we need to be mindful that this is both for adenocarcinoma and squamous cell patients. Are we having challenges at all getting this done now? Are there any issues?
Roy S. Herbst, MD, PhD: Oh, no. In fact, at my hospital, Yale New Haven Hospital, we can get the PD-L1 result before we get the mutational tests. The problem we have is waiting for the mutational tests to come back so that we can move forward. But the PD-L1 is an immunochemistry test, and we do it up front now as a reflex. And many places probably send it out as well, but I think I see that almost everyone comes in with a PD-L1 now. Now, which test is used is a different story, and I think the data from the IASLC would show us that almost all the tests are the same.
Benjamin P. Levy, MD: I’ve heard from the community folks that given how many genetic alterations we now need to test for, in addition to PD-L1, there may be tissue exhaustion and problems with tissue stewardship as we begin to look at the underpinnings of the tumor. We’ve had no problem with it. I know there’s some confusion out there on what to do first. Roy, you mentioned PD-L1 comes first. You don’t act on it until you know what the genetic alteration is?
Roy S. Herbst, MD, PhD: In the frontline setting, I still believe that an EGFR mutation or an ALK mutation should trump that. That said, it’s almost 20 years now, I’ve never seen anyone cured with a targeted therapy. But still, I would usually start with that because I know the response rate is going to be upwards of 60%. So, I do want to have that, but sometimes it gets a little bit dicey. I also don’t want to give an immune therapy to someone if I don’t know that they’re PD-L1—high up front.
Benjamin P. Levy, MD: Yes, that’s a good point. I think the point that EGFR- and ALK-rearranged lung cancers may not be the best candidates for immunotherapy, even if they are high PD-L1 expressers, is an important one. I think we’re going to have some data on this that are going to look at immunotherapy for EGFR-mutated lung cancers with PD-L1—high disease coming out of UCLA in the next few years.
Roy S. Herbst, MD, PhD: But we need to do that because I’m sure we’ll talk later about what do you do when someone fails an EGFR inhibitor. And we’re going to have to figure out how to get immunotherapy to work, whether it be checkpoint inhibitors or agents that target other aspects of the immune regulatory pathways.
Benjamin P. Levy, MD: I don’t know if we have a lot of data on this, but the utility of pembrolizumab up front in a PD-L1—high patient, the utility of eliciting response in the brain, have you seen that, a patient with untreated brain metastases that you would use this on?
Roy S. Herbst, MD, PhD: Oh, absolutely. That was actually a paper that Harriet Kluger and Sara Goldberg published. They had to fight that point years ago to get the drug in an investigator-initiated setting to use it in patients with brain metastases, and actually, they’re about to present their update on that. But in Lancet Oncology a few years ago, they showed that the responses in the brain were consistent with the responses seen in the systemic disease. And that’s really important because those people are going to live for a long time—7, 8 years—dare I say even be cured. To avoid that, radiation to the brain would be a very important thing. So, I think now some of the phase III trials are beginning to incorporate patients with treated brain metastases and so forth.
Naiyer A. Rizvi, MD: I think that, in general, even for PD-L1—positive patients if they have brain lesions that are amenable to SRS (stereotactic radiosurgery) Gamma Knife, I am still radiating them up front with focal therapies because the response rate is still 45%. And I think, in general, I don’t know about you, but I’m still treating my brain metastases up front.
Benjamin P. Levy, MD: So, competing strategies here.
Roy S. Herbst, MD, PhD: It depends how big they are, right? You’ve got a couple and they’re very small. Remember, the trial that we’re talking about really was selecting people. In fact, when I look for good examples to show in a lecture, I can’t find one because they all start out with very small lesions.
Naiyer A. Rizvi, MD: Right, if they’re millimeters in size. But if they’re centimeters in size…
Hossein Borghaei, DO, MS: Asymptomatic, left or the right, it makes a difference.
Roy S. Herbst, MD, PhD: If they’re close to your fascia.
Naiyer A. Rizvi, MD: Right, exactly.
Suresh S. Ramalingam, MD: It’s fair to say that immunotherapy is still not at the point where it can replace local brain radiation therapy. We need to study it further. But in selected patients, like Roy said, if your need to give systemic therapy is compelling, you may go forward with it.
Benjamin P. Levy, MD: I’ve seen some responses in the brain for these patients, but I agree. I think treatment decisions have to be individualized in this setting.
Thomas E. Stinchcombe, MD: Another thing, sometimes when they do SRS and you follow with immunotherapy, they can get cerebral edema. We’ve had a few of our patients do that, so the timing of doing the SRS followed by the immunotherapy can sometimes be a logistical challenge.
Roy S. Herbst, MD, PhD: Absolutely. I was going to say there is one challenge here and that’s Jimmy Carter.
Hossein Borghaei, DO, MS: Different disease.
Roy S. Herbst, MD, PhD: Different disease. Melanoma. We’re not sure exactly of his medical record, but you’ve got pembrolizumab in that setting and everyone says, “I want what he had,” so we all see a little of that. I think those patients are very informed and they read it on the internet, which is a good thing.
Transcript Edited for Clarity