Benjamin Levy, MD: I don’t know if there are any meaningful differences between the checkpoint inhibitors as PD-1 [programmed cell death protein 1] drug versus PD-L1 [programmed death-ligand 1] drug. I think we’re learning that perhaps even PD-1 drugs may be different from other PD-1 drugs, and this is based on some discrepancies in the frontline trials between nivolumab and pembrolizumab. But I’m not sure, at least in the second line, that we’ve seen many differences. This is still an unanswered question. I think as single agents, I’m not sure how different they are. But certainly in combination with chemotherapy there may be some differences between these agents, whether they are PD-1 drugs or PD-L1 drugs, that may explain the differences in outcomes we’re seeing in some of these trials. But as a rule, if a patient is on a checkpoint inhibitor, and they’re no longer responding or they’re progressing, I won’t switch to another checkpoint inhibitor, with the thought being that that 1 is different or acts differently or we’d be able to elicit any sort of different response. But this is a question I think we still need better answers to as to what the real differences are.
Patients with advanced-stage lung cancer who come in without a molecular aberration and who have a PD-L1 between 1% and 49%, we’ve got a new treatment paradigm for these patients. Before KEYNOTE-189, patients were offered just systemic chemotherapy with carboplatin and pemetrexed. But we’ve learned that the addition of pembrolizumab to this platinum backbone has translated into meaningful benefits in overall survival. Because of that, I think patients…the patients who had a PD-L1 less than 50% still had a meaningful outcome with this triplet regimen compared with just platinum doublet alone.
So for these patients who have a preserved performance status, who are wanting to get the most updated regimen and want to be on therapy, I will offer these patients carboplatin, pemetrexed, and pembrolizumab, keeping in mind that not every patient is a candidate for pembrolizumab. So we have to be mindful of that.
I’ll just make 1 small nuance here, which is that we have another regimen that’s now approved in this setting, and it’s the quadruplet regimen of carboplatin, paclitaxel, bevacizumab, and atezolizumab. This is now FDA approved, so this is yet another regimen that we can consider offering to our patients, a 4-drug regimen rather than a 3-drug regimen, and I think we’ll have to see over time what the uptake of this is in clinic compared with the triplet regimen.
Roy Herbst, MD, PhD: Well, if patients have lower than 50%, those patients would not get pembrolizumab alone. They shouldn’t. Some will ask for it. They’ve seen it on TV, no? Those patients have a lower than 20%, 25% chance of responding to immunotherapy alone. They need to get chemotherapy plus immunotherapy if they get again immunotherapy or some sort of protocol. And there are protocols now looking at CTLA-4 inhibitors, ipilimumab, tremelimumab in that setting. There are protocols with other agents that activate the immune system. But those patients, based on KEYNOTE-407 and KEYNOTE-189, would get chemotherapy plus immunotherapy.
Anne Tsao, MD: In our patients who are not candidates for immunotherapy we still must rely on the mainstay of chemotherapy with or without bevacizumab. And so certainly the patients who qualify for bevacizumab are pretty standard: no prior bleeding history, hemoptysis, recent MI [myocardial infarction], recent thrombolytic events. These are the patients we would need to consider.
Now there’s a very broad spectrum of patients who may not be candidates for immunotherapy, but largely they would be patients who have significant organ transplantation or autoimmune disease. Or if we know, for instance, that they’ve received immunotherapy, and they developed a toxicity that is autoimmune-related that potentially caused organ damage. These are patients who unfortunately cannot receive immunotherapies again.