Suresh S. Ramalingam, MD, FASCO: Cho, as you think about patients getting frontline osimertinib, what are your thoughts on understanding resistance to osimertinib in patients who develop acquired resistance after they have been through first-line osimertinib? Can you also talk about strategies that you recommend people use in their clinic and some directions for research?

Byoung Chul Cho, MD, PhD: Acquired resistance mechanism to osimertinib in the frontline setting was already presented last year, and there are a lot of diverse mechanisms, including loss of T790M mutation, small cell lung cancer transformation, and most commonly c-Met amplification. In our routine practice, there is no approved drug that has shown a clinical benefit in any of the resistant mechanisms after progression on osimertinib. Currently, cytotoxic chemotherapy, usually pemetrexed and platinum combination chemotherapy is the most commonly used in our setting. But there are many clinical trials targeting the c-Met amplification, small cell lung cancer transformation, and other genomic alterations. We are eagerly waiting for the mature data of these clinical trials, especially the AstraZeneca-driven study that is being tested for osimertinib and c-Met—targeted agents in the setting of the c-Met amplified or c-Met–altered, EGFR-mutated non—small cell lung cancer.

Pasi A. Jänne, MD, PhD: Just to build on that a little bit, I agree with you that the resistance patterns are quite diverse. But there are ones that can be targeted. I think in clinical practice we do test individuals who progress. If they have a targetable alteration, we can enroll those individuals on clinical trials. If they don’t, then oftentimes they get chemotherapy. It is an evolving landscape, and we will continue to learn. But I do think it’s still worth testing patients who progress because there may be things there that you can do for your patients today.

Suresh S. Ramalingam, MD, FASCO: Pasi, a common clinical scenario is you have a patient newly diagnosed with lung adenocarcinoma, stage IV disease. You already have the PD-L1 [programmed death-ligand 1] results back. Let’s say the PD-L1 comes back at 80%. Molecular testing results are not back. What is your approach in that situation for that particular patient?

Pasi A. Jänne, MD, PhD: I agree with you, it’s common because PD-L1 testing comes back by immunohistochemistry, and next-generation sequencing may take longer. I think it depends a little on the patient. Certainly an individual who clinically would be likely to have a targetable alteration—someone who, for example, is a never-smoker or a limited smoker—in those individuals we tend to wait for molecular testing because the likelihood of finding a targetable-alteration EGFR, ALK, and others—is quite high. We would tend toward this as long as there’s no immediate clinical need to treat that individual. I think if there are other scenarios to present, for example, if you need to treat somebody, then that may drive a treatment decision earlier. But I think we want to be most informed about how we treat our patients, and incorporation of molecular testing in that regard is important.

Suresh S. Ramalingam, MD, FASCO: So we should wait until the molecular testing is fully done. You have all the results before you act on the situation.

Pasi A. Jänne, MD, PhD: Right, that is the ideal situation.

Suresh S. Ramalingam, MD, FASCO: Whenever possible.

Pasi A. Jänne, MD, PhD: Whenever possible. Obviously, not everybody fits into that, but I think that’s the optimal from a therapeutic standpoint.

Suresh S. Ramalingam, MD, FASCO: Does the fact that the panel here also agree that for a patient with targetable mutation, like EGFR, the frontline therapy should be a targeted drug as opposed to immunotherapy or chemotherapy?

Johan F. Vansteenkiste, MD, PhD: Just adding to what was said, indeed, you should give the best therapy first and you should make the best choice as well. It means it’s better to wait for the molecular data before you make your choice. Most of the patients can wait a bit. I agree there are exceptions, but in general it’s best to have everything at hand before you make your decision. The question that you’re raising now, I think there’s no doubt that these are 2 different worlds. If you have an oncogene addiction, your treatment is TKI [tyrosine kinase inhibitor], not immunotherapy.

Byoung Chul Cho, MD, PhD: Yeah. We also remember that predictive power of tumor PD-L1 expression is quite low, especially in EGFR-mutated lung cancer, so we should be fairly careful in treating EGFR-mutant lung cancer with immunotherapy.

Pasi A. Jänne, MD, PhD: At the moment we have sort of 5 molecular targets where we have targeted therapy, which we can begin as first-line treatment. I think that number is only going to increase as we move along in the next few years. It could be that we need to test for 7 or 8 different things, and it’s about matching the right treatment to the right patient.

Suresh S. Ramalingam, MD, FASCO: One of the things I’ve heard other colleagues say as well is that if you are compelled to start a patient on some form of therapy because they are symptomatic, their condition is declining, and you don’t have the molecular testing back, it’s better to start with chemotherapy alone—not have chemotherapy plus immunotherapy until the results are available. This is because if you have to move to targeted therapy, it’s a lot easier when immunotherapy is not involved yet. Is that something you have done in your practice?

Pasi A. Jänne, MD, PhD: Yeah, absolutely. One of the issues is there are interactions with targeted therapies and immunotherapies. So that’s an issue. I don’t think you do a disservice to the patient that if you do 1 cycle of chemotherapy alone and then add immunotherapy later if there’s no targetable alteration. But I think if somebody needs to be treated—and I agree with Johan that it’s usually the exception as opposed to the rule—chemotherapy is a fair choice.

Johan F. Vansteenkiste, MD, PhD: Moreover, if you give chemotherapy alone, you stay within the label because these studies, at least with pembrolizumab, were done in non—oncogene-addicted patients.

Pasi A. Jänne, MD, PhD: Correct.

Suresh S. Ramalingam, MD, FASCO: One last issue I want to cover on this topic before we move on is a clinical scenario where the plasma was sent for testing and plasma cell-free DNA testing reveals EGFR exon 19 or exon 21 mutation. Is that enough information for you to proceed with osimertinib as frontline therapy, or would you still wait for the tissue, Cho?

Byoung Chul Cho, MD, PhD: It’s a really hard patient. There is a lot of temptation to use osimertinib over other EGFR tyrosine kinase inhibitors just with plasma liquid biopsy testing based on fairly high sensitivity and specificity of the plasma liquid biopsy for EGFR mutation. But it’s always a regulatory issue, so without the tumor tissue examined by a pathologist, it’s really hard to give any kind of cancer therapy in our routine practice. But I do believe that kind of practice will be available and feasible in the very near future.

Suresh S. Ramalingam, MD, FASCO: In the United States, that is already available.

Pasi A. Jänne, MD, PhD: That is already available and approved, right. As long as you have a diagnosis of cancer on a pathology, I think if the molecular testing comes from blood as opposed to tumor and it comes back quicker, that’s a perfect scenario to use targeted therapies.

Suresh S. Ramalingam, MD, FASCO: Great. That’s been a great discussion on EGFR-targeted therapies. We have osimertinib as the preferred frontline therapy agent with a median survival of approximately 39 months.

Transcript Edited for Clarity