Role of Adjuvant Osimertinib in Early Stage NSCLC

Benjamin Levy, MD: Josh, clearly your enthusiasm is there for potential use of the drug in the adjuvant setting. We’ve all been familiar with using this drug in the advanced-stage setting and the toxicity is manageable. It’s not like a first-generation TKI [tyrosine kinase inhibitor]. Is there a patient in whom you wouldn’t consider using this at stage II or III? Then maybe we can talk about the stage IB patients and I can get Rebecca’s and Lyuda’s input on this, as well.

The first question is to you, since you so eloquently went over the data. Is there a patient in stage II or III in whom you couldn’t consider it? Is there any scenario in which you’re not going to think about using this drug?

Joshua Bauml, MD: I don’t think so for stages II and IIIA. I would say that 1 of the major limitations of this trial is the fact that, for patients with stage IIIA disease and nodal positivity in the mediastinal, or those with positive surgical margins, postoperative radiation was not allowed. To me, that is a major barrier. I don’t think that would prevent me from using the drug. I would probably give the radiation and the drug, but I don’t think I would withhold it in stage II or IIIA disease.

Benjamin Levy, MD: I’m going to play contrarian here. The drug did not show an overall survival [OS] advantage. You can’t answer affirmatively when patients ask, “Will I live longer with this drug?” There is financial toxicity associated with this. I throw that in there just for balance and discussion. Lyuda, what are your thoughts? We can get back to you, Josh, to get your thoughts on this. Lyuda, what are your thoughts on this drug in the adjuvant setting, and where we should go with this?

Lyudmila Bazhenova, MD: I’ll be honest with you. It took me 2 weeks to come up with my own answer. For the first time in 15 years, I did not know at the time of the presentation whether I’m going to do it or not. Based on exactly the facts that Josh pointed out, in the early stage disease, we are looking for a cure, and the study design was not perfect. In the end, I’m in the yes camp. It is practice changing to me because my practice will now be talking to patients about the study.

When I talk to patients, I will definitely let them know that we don’t know what’s going to cure or not. But it is hard to put a price on not giving bad news to your family members. The way I decided for myself was to think, what if I were diagnosed with early stage lung cancer with EGFR mutation? Would I take the drug? My answer today is yes, because I have young kids who still need me. My answer in 10 or 15 years might be no, because then I really wouldn’t want to. I’m would not be as upset about being faced with a diagnosis that potentially will take my life.

At this point, for stages II and III, I will be discussing that. In stage I, it is more controversial. I will bring it up, but it’s not a primary end point. It’s a secondary end point of the study. If you look at the absolute difference in DFS [disease-free survival] at 2 years, it is only 10%, compared with the difference in stage III, which is almost 50% improvement in 2-year DFS. I will be discussing it and I will be offering it to my patients, but with the underscoring that we don’t know if it’s going to cure the patient. I will guide the patient through the decision.

What swayed me, to be honest to you, is the Jill Feldman publication on Twitter. She is a patient with early stage lung cancer with EGFR mutation who was receiving erlotinib when we had no data behind it. I agree with the thought that went through her brain, putting it in my perspective and in my life right now.

Benjamin Levy, MD: Interesting. I saw that Twitter exchange between you and Jill. It was interesting to get your thoughts based on your hearing her out. Patient advocates have a voice that needs to be heard. Becca, what are your thoughts on this and where we head with this? Which camp do you belong in, or do you subscribe to, for this trial?

Rebecca Heist, MD: I agree with nearly everything that’s been said. There is a real financial toxicity issue. I do think that financial toxicity issue should have been best addressed with the design of the study, rather than how we actually implemented findings now. At the design of the study, I would have designed it with an OS end point. That’s what all of us wish had happened, because that’s what we really care about. We want to cure more patients.

That being said, it wasn’t designed in that way, and now we’re left with a finding where we definitely have a disease-free survival advantage that’s quite significant, and the OS advantage is unknown and will take years to know. What do we do with our patients in the meantime? We have to discuss this. The data are very compelling, and you would want, at least in my mind, to offer this to patients as a potential option.

Different patients will come down differently on it, to be honest with you. People will have different interpretations of what it means to have a DFS advantage with an unknown OS advantage. I have certainly been surprised when I discuss, for example, the ALCHEMIST study with patients after surgery and after chemotherapy. There are a fair number of patients who say, “I’m done. I hear what you’re saying, but I would like a break from all medical treatments for a while. Just follow me.”

Different patients will make different decisions, but there are definitely compelling data that we have to be discussing with our patients.

Benjamin Levy, MD: Rebecca you raise an interesting point about ALCHEMIST. Where do we go with ALCHEMIST, and do we enroll patients on this trial? Is it equipoise now that the trial is 2 years of erlotinib versus placebo, in a very similar setting, with a primary end point of OS? We all want OS, and that’s the primary end point of the study. Can we reasonably expect patients to go on ALCHEMIST if we have these data from ADAURA?

Rebecca Heist, MD: It’s hard to know what’s going to happen with ALCHEMIST. I would be challenged to offer it with equipoise to patients. In general, I would say we were a bit like, “I wish that the drug were OSI [osimertinib] rather than erlotinib.” If we want to use our best TKI in this setting, and then with the ADAURA data, it’s hard to know how we would enroll robustly to that study.

Benjamin Levy, MD: Yes. Are there any other thoughts?

Lyudmila Bazhenova, MD: It is challenging, Rebecca, that we can’t enroll robustly. Also, the patients who now randomize to observation, because this is not a blinded study, might seek erlotinib or osimertinib in an adjuvant setting, and we might never know the OS outcome.

Joshua Bauml, MD: One of the concerns that I have in terms of the discussion of financial toxicity is that when I have a patient in front of me, I view my job as an advocate for that person. Our health care system has major problems. There are barriers. Your point about the design and how things should have been done are all valid. But at the end of the day, I have a patient in front of me who is entrusting their care to my judgment, and I don’t think it’s my role at that time to say, “Listen, this costs a lot of money, and I don’t want Person X’s premiums to go up.” What am I going to do to help this person in front of me? To me, the answer from ADAURA is pretty clear. I’m going to give them this medicine. If we as a society want to reevaluate how we structure our system, that’s fine. But for now, I have to work in the system that exists.

Rebecca Heist, MD: I completely agree with you. If we’re actually serious about financial toxicity, we cannot be having that conversation at the doctor-to-patient individual level. We have to be having that conversation when thinking how we are going to design the study. What is it going to take for the FDA to approve this drug? Those data will then be generated. But financial toxicity at the level of the individual patient decision is not appropriate and not ethical to make those decisions.

Benjamin Levy, MD: Yes. These are all really important points for this trial that raise a lot of questions that are going to have to be addressed in the future. I agree with what everyone is saying. The magnitude of benefit in DFS is compelling. We have not talked about at the potential CNS [central nervous system] protection that this drug offers. We know, at least from data, that some of these patients who are receiving adjuvant TKIs do relapse in the brain. The drug is reasonably well tolerated. It is 1 of the cleaner targeted therapies we have. We all have this existential crisis in which we have this allegiance to OS and cost. But at the same time, as Josh mentioned, what would we want to do for our patients, and what do we think is in the best interest of our patients? That is a discussion with patients about the utility and the benefit of this drug, even in the absence of OS benefit.

Interestingly, I just had a patient right before ASCO 2020 who we screened for ALCHEMIST. She has come back to me and says she’s no longer going on ALCHEMIST, and I can’t blame her because there’s a 50% chance that she’ll get nothing. It’s not a placebo trial anymore. They just get nothing, and who’s to say that patient won’t go on, as Lyuda said, to osimertinib at some point and compound the benefit that may be seen with the experimental arm in that study?

Transcript Edited for Clarity