Benjamin Levy, MD: The treatment paradigm for patients with advanced-stage non—small cell lung cancer without a driver has rapidly evolved, and there are multiple studies now that have come out in the last 12 to 24 months that have been exciting yet confusing. For me, for patients without a driver in which we’re talking about the role of immunotherapy, I still think adding pembrolizumab to carboplatin, pemetrexed is the standard for patients with a PD-L1 [programmed death-ligand 1] less than 50%. For patients greater than 50%, I think we have single-agent pembrolizumab for our asymptomatic, or less symptomatic patients, and the triplet therapy for our more symptomatic patients.
There are other options, of course. IMpower150 is another option. Regimen is another option with a quadruplet regimen. So I still use PD-L1 to help try treatment decisions. I don’t use TMB, and TMB [tumor mutational burden] is a marker that was shown to potentially be predictive of benefit to build checkpoint blockade. This is not something that I feel we currently should be using routinely in our clinical practice. So it all comes down to PD-L1, patient performance status, and then based on those things I will make a treatment decision usually with carboplatin, pemetrexed, pembrolizumab, or pembrolizumab alone for my advanced nonsquamous patients. For my squamous patients, we certainly have new data from KEYNOTE-407 looking at carboplatin; a taxane with pembrolizumab has also shown a survival advantage.
Anne Tsao, MD: The quadruplet regimen of the carboplatin, paclitaxel, bevacizumab, and atezolizumab is now FDA approved in the front-line setting. In the subgroup analysis from IMpower150, we saw the patients who had mutations—EGFR mutation and ALK—who had previously been treated with the respective TKI [tyrosine kinase inhibitor] did extremely well with this regimen. And there’s some speculation that it could be because of a synergistic effect between the bevacizumab and atezolizumab for this population of patients who have oncogenic driver mutations. And it also appeared that patients who have liver metastases did better as well.
And so when I’m making my choices in the frontline setting, now that this is FDA approved, I have treated some patients who had prior oncogenic driver mutations with this regimen, and I’ve also treated a patient who came in de novo with this quadruplet regimen because she had extensive disease but still very good performance status.
So it is tolerable. You do have to educate the patients about the different toxicities associated with bevacizumab and also autoimmune potential effects from the atezolizumab, but it is certainly another regimen that we can use in our patients with front-line metastatic non—small cell that are nonsquamous.
Roy Herbst, MD, PhD: I think in the metastatic setting, at least the way things are right now, frontline, most people are using pembrolizumab and most people would give treatment frontline. Nivolumab, of course, was around first and approved first, and it still gets a good deal of use in the second-line setting. Nivolumab you can use in treatment without having any PD-L1 positivity. Pembrolizumab’s label, in large part based on KEYNOTE-010, which I was involved with, requires that it have some PD-L1 staining. Meaning you can’t be 0. It’s got to be greater than 1%. I think right now in the second-line setting, if someone hasn’t had immunotherapy, you give them immunotherapy. If they have had immunotherapy, then you have to think about things like docetaxel, and I would probably use docetaxel with ramucirumab, the VEGFR2 inhibitor based on the results of the REVEL trial. And I think that’s probably the best nonimmunotherapy refractory standard-of-care regimen. We’re going to have to look for more I-O [immuno-oncology and] I-O combos and other ways of doing this, and that’s all ongoing right now in the labs at Yale Cancer Center and other places.