H. Jack West, MD: One of the hesitations about people using first-line osimertinib or maybe favoring an approach that doesn’t is the issue of, “What do I do next if I’ve used osimertinib?” So if you have a T790M-negative patient or you have a patient who progresses on osimertinib, where do you go from there? And let’s presume that you can give the IMpower150 regimen. I mean, we can just categorically say, “Well, it’s not approved,” and so dismiss it. But if you have that available, how compelling is that as an option for you right now? Or would you favor chemotherapy alone or something else?

Hossein Borghaei, DO: If there’s no T790M, or MET amplification, or other stuff that we can target, I’m actually very inclined to use the quadruplet combination that you just mentioned with bevacizumab, and atezo [atezolizumab], and carbo [carboplatin], and paclitaxel. I think in the absence of a good viable option, looking at the subgroup analysis with a hazard ratio for survival of 0.52, granted there were 100 patients in that study, I’m very much in favor of trying that kind of a regimen in somebody who’s progressing on osimertinib if I don’t really have another option. Because what is my option beyond that? It will be carbo and pemetrexed. And yes, this patient population has a wonderful response to pemetrexed, and it’s a very viable treatment option. Again, in the absence of a head-to-head comparison, using just a simple carbo-pem or the quadruplet therapy, I’m still favoring the use of the quadruplet therapy if I have access to it.

H. Jack West, MD: Charu?

Charu Aggarwal, MD, MPH: I agree. I don’t think we have randomized data to distinguish which doublet I would use. I would use a doublet. I would prefer to use a pemetrexed-based regimen, but I think IMpower150 is the only data that incorporate bevacizumab, which I think is relevant in the EGFR space, and immunotherapy, and it included patients who had previously been treated with a TKI [tyrosine kinase inhibitor] that had an EGFR mutation. So I would very strongly consider the quadruplet in that setting.

Hossein Borghaei, DO: I want to put a plug in here. We have a study coming out. It’s going to be a multi-institution study looking at carboplatin, pemetrexed, and bevacizumab, and atezo versus just the triplet without the atezo. So this will be a 2-arm phase II randomized study. We’ll get some more information, but we’ll see.

H. Jack West, MD: So this brings up a question. We have what looks quite encouraging. In fact, Dr Tony Mok presented at ESMO [European Society for Medical Oncology] Asia very recently some data on the EGFR-specific population that looks especially good but only in that arm B with carbo, paclitaxel, bev, and atezo. We’ve gotten a little data on carbo and paclitaxel, and atezo. That doesn’t seem to have that benefit. IMpower130 with carbo and nab-pac, and atezo also doesn’t show a clear signal of benefit in the small population with EGFR-ALK. So the data are certainly limited, but we’ll get more information. Merck [& Co], I believe, is going to be running its KEYNOTE-789 trial, which is basically going to be like KEYNOTE-189 but in the EGFR mutation-positive post-TKI patients. Is the bevacizumab the secret ingredient?

Charu Aggarwal, MD, MPH: I think so. I feel that’s one of the escape mechanisms for resistance, and I feel like bevacizumab should be included when you’re talking about the second-line setting.

H. Jack West, MD: So outside of a trial, you would not feel as comfortable using KEYNOTE-189 if you could do IMpower150?

Charu Aggarwal, MD, MPH: Outside of a clinical trial, let’s say I could not use the quadruplet, I would probably gravitate toward carbo, pem, and bevacizumab.

H. Jack West, MD: But if you had IMpower150 available, you would favor that over KEYNOTE-189?

Charu Aggarwal, MD, MPH: Correct.

H. Jack West, MD: OK. Hoss?

Hossein Borghaei, DO: I think the question for VEGF is interesting, and I think it is relevant, particularly in combination with I-O [immune-oncology]. My only concern is, what if pemetrexed is such a better drug in this group that you don’t need to add the I-O, necessarily? Again, when you play these kinds of games, you have to consider all the issues, right? So maybe pemetrexed plus bevacizumab would be sufficient to overcome the resistant mechanism and give you just as good of a result. At this point I agree with Charu. If I have the option, I would probably gravitate toward IMpower150 as opposed to KEYNOTE-189 in this patient population, because I believe the VEGF interaction with I-O is real, but that’s just the way I’m interpreting the data.

H. Jack West, MD: I think that’s quite fair. Obviously, we’re going to learn more. We’re going to get more data, but these are important studies. Are there any data to speak to there being different patterns of resistance that develop among the EGFR TKIs, other than osimertinib versus the first- or second-generation agents? Or as far as we know, there’s no real difference between patterns, between afatinib, dacomitinib, and the first-gens?

Hossein Borghaei, DO: Not that I’m aware of.

H. Jack West, MD: I don’t think it’s well studied, but to my knowledge there are no data to speak to that.

Hossein Borghaei, DO: Right.

Transcript Edited for Clarity