Case 4: Treating EGFR+ NSCLC After Osimertinib

EXPERT PERSPECTIVE VIRTUAL TUMOR BOARDBenjamin P. Levy, MD:Before we get to IMpower 150, and before we get to what we would do post-osimertinib in a patient with or without a C797S mutation, the PD-L1 on this rebiopsy is 100%. Is there an opportunity to just give immunotherapy in this patient? Is this even a consideration?

Anne S. Tsao, MD:I wouldn’t do it.

Benjamin P. Levy, MD:OK.

Paul K. Paik, MD:The retrospective data make it relatively clear that even in the setting of high PD-L1 expression, there’s not very good response.

Benjamin P. Levy, MD:Yes.

Paul K. Paik, MD:And not to sound sort of direct about this, but to sort of waste that option of immunotherapy and potentially change the immune microenvironment—I think this is potentially a disservice to the patient.

Benjamin P. Levy, MD:Yes. I think we know that, and earlier we mentioned that treating a patient with a TKI [tyrosine kinase inhibitor] can induce PD-L1 programmed-cell death ligand 1]. In these resistant settings, they may potentially be high. At this year’s ASCO [American Society of Clinical Oncology] Annual Meeting, we’ve learned that the response rates are 0% with single-agent pembrolizumab in anEGFRtreatment-naïve patient whose PD-L1 is high. I think that the message needs to be clear—if the PD-L1 is high, a single-agent checkpoint inhibitor is probably not the best drug for this patient.

Anne S. Tsao, MD:There’s no reason not to give them chemotherapy plus immunotherapy. We know that these mutant patients respond beautifully to systemic therapy. So, in my mind, you should never use a single-agent checkpoint inhibitor.

Benjamin P. Levy, MD:Let’s talk about this patient. This patient’s tumor has progressed on a TKI—osimertinib. You have the mutational analysis that we’ve reviewed here. What are people routinely doing? Paul, I’ll start with you. What do you do, off a clinical trial, for a patient like this—who’s symptomatic and you need to get something going?

Paul K. Paik, MD:Here, it really boils down to an individual patient approach. There is a third option that’s not mentioned, which is continuation of osimertinib with chemotherapy. We don’t really have any data on this. We do have data regarding continuation of a TKI with a first-generation TKI, but we don’t for osimertinib. I think it’s fair to say that the biology is sufficiently undefined, but it is also different from that resistance after a first-generation TKI. It’s a question that we do need an answer to, and we should try to answer it.

Benjamin P. Levy, MD:What chemotherapy?

Paul K. Paik, MD:I still go with carboplatin/pemetrexed/bevacizumab as the standard in these patients. The IMpower 150 data are interesting. At this point, I still have not utilized that for the osimertinib progressors. I don’t know, at this point, what it’ll take to get me to utilize it.

Benjamin P. Levy, MD:What about the KEYNOTE-189 regimen?

Paul K. Paik, MD:Again, I think this is interesting. Part of my concern about this is conceptualizing, in the long run, the role of immunomodulation and what the best strategic approach is going to be. You have to be very careful about that. It’s there as a standard. There might be better options. I think it’s just something that we have to think about carefully. That’s the best answer I can give.

Benjamin P. Levy, MD:Anne?

Anne S. Tsao, MD:Right now, the only data that we have, of course, is IMpower 150. With the quadruplet regimen, that is the only trial in the frontline setting that allowed forEGFRmutants who were previously treated with anEGFRTKI. So, that is the only data. In theory, that would be the regimen that people would want to move toward if they fail osimertinib.

That said, in my clinical practice I give carboplatin/pemetrexed/pembrolizumab because the taxane is hard. I used to be just like Paul. I used to give carboplatin/pemetrexed/bevacizumab to myEGFR-mutant patients. For me, it’s not much of a leap to switch out the bevacizumab. With that caveat said, there is emerging data that suggests that antiangiogenics and immunotherapies interact on the tumor microenvironment. There’s a synergistic effect and there may be something additive there, that we’re not yet aware of, that specifically benefits this population of patients who have oncogenic driver mutations.

Benjamin P. Levy, MD:That was very nicely laid out. I think there’s a lot of scientific rationale for combining a checkpoint inhibitor with an antiangiogenic drug. I hope that we’ll soon have regimens for routine use, and that we can get some generation of data post-osimertinib. At times, I have a crisis. Do I use the carboplatin/pemetrexed/bevacizumab regimen, or do I go with carboplatin/pemetrexed/pembrolizumab? KEYNOTE-189 didn’t allow forEGFR-mutant lung cancer patients. We think that bevacizumab may have some activity inEGFR-mutant lung cancer. I think it’s dealer’s choice. It’s not really ironed out. I hope that we’ll see more and more data with the quadruplet regimen, based on the scientific rationale, Anne, that you just laid out. I think we still struggle with what to do.

Paul K. Paik, MD:One of the concerns I have, that plays into this, is, what happens if you have a patient for whom you take off osimertinib? You start them on a chemotherapy-based regimen and it becomes evident that they’re one of those patients who ends up having some degree of progression that you think may be because you took them off osimertinib? What is the ability to add back osimertinib?

Anne S. Tsao, MD:Oh, I’ve done that.

Paul K. Paik, MD:Right. But one of the things that’s always in the back of my mind is, if you’re giving an immunotherapy-based chemotherapy regimen, the concern about pneumonitis in that setting is much higher. How does that temper how you think about things, or play into how you think about things? There’s no straightforward answer. There are big data gaps here, in terms of how to move forward.

Benjamin P. Levy, MD:Yes. Hopefully we’ll have more data very soon. In summary, I think that we’ve witnessed a tremendous amount of progress, not only in non—small cell lung cancer but particularlyEGFR-mutant lung cancer. If we would have had this discussion a year ago, we would not have had these data points. So, who knows what we’re going to see in the next 12 to 24 months. It’s very exciting.

Thank you to Dr Riedlinger, Dr Paik, and Dr Tsao for your thoughtful case presentations and a lively informative discussion. To our viewing audience, thank you for joining us for thisTargeted Oncology Virtual Tumor Board^®presentation. We hope today’s discussion was a valuable use of your time, and that you have acquired some practical knowledge that you can take back to your clinic.

Transcript edited for clarity.