This transcript has been edited for clarity.
Jacob Sands, MD: Hello. I'm Dr Jacob Sands, a thoracic medical oncologist at the Dana-Farber Cancer Institute. Welcome to Medscape's InDiscussion series on lung cancer. Today, we'll be discussing neoadjuvant and adjuvant therapy for non–small cell lung cancer (NSCLC). First, let me introduce my guest, Dr Benjamin Levy, a thoracic medical oncologist and associate professor at Johns Hopkins School of Medicine. Ben, welcome to InDiscussion.
Benjamin Levy, MD: Thanks for having me, Jacob. A pleasure to be here.
Sands: Let’s talk a little bit about your personal journey. How did you end up in the position you're in? What led you into being a specialist in lung cancer?
Levy: I think it's quite surprising to some people that I didn't choose lung cancer. Lung cancer chose me. When I finished my fellowship, I had been in the lab doing prostate cancer research, but at the end of the day, I got interested at the very last minute because of all the new targeted therapies that were coming out in lung cancer and ended up taking a job as a thoracic medical oncologist right out of fellowship. So I've been in the field for the last 12 to 15 years and it's been very exciting for me.
Sands: Now we're going to discuss perioperative management of NSCLC. And there has been a ton in the last 10 years in the NSCLC space with targeted therapies and immunotherapies related to metastatic disease. And we're finding that moving into the perioperative space, too. But maybe let's just start at the base. Adjuvant chemotherapy really was the standard for a long time. Maybe you can just discuss that starting point.
Levy: As you mentioned, I think we've learned over the past 10 years that adjuvant platinum-based chemotherapy for resected stage IB to IIIA patients does provide a survival advantage. We've got multiple phase 3 studies that were done in the 90s that show that there is a survival advantage when patients get adjuvant chemotherapy, specifically cisplatin-based chemotherapy. Now we have the Lung Adjuvant Cisplatin Evaluation (LACE) meta analysis of all the studies that essentially showed that, adjuvant, platinum-based chemotherapy does provide a survival advantage, but that survival advantage is modest at best, that essentially it’s a 5% improvement at 5 years for those patients who got adjuvant chemotherapy versus those that did not get adjuvant chemotherapy. It’s a modest benefit but it created the foundation for some of the newer therapies that we’ll talk about in the perioperative space.
Sands: The further discussion on that in the perioperative space is essentially targeted therapies and immunotherapies. So let’s start with targeted therapies. So we have the ADAURA data, which was specifically related to EGFR-specific treatment. Can you walk us through that?
Levy: The ADAURA study was a practice changing, phase 3 double-blind study that looked at patients with resected stage IB to IIIA, all harboring an EGFR-sensitizing mutation, so exon 19 deletions or L858R point mutations. Patients were given chemotherapy at the discretion of the treating physician and then randomized to osimertinib 80 mg/day for up to a plan of 3 years versus placebo. Asking the question does adjuvant osimertinib, a next generation TKI, provide a benefit for those patients with resected EGFR-mutant lung cancer? And I think all of us were surprised at the DFS (disease-free survival) that we saw. DFS is not overall survival (OS), it is a composite endpoint of either recurrence and or death. And the hazard ratio was 0.21, a hazard ratio that we haven't really seen in a long time or ever. These curves diverged very early on. The hazard ratio continues to go down the higher the stage, not surprisingly. And then we've seen early cuts on other endpoints like CNS recurrence. Not surprisingly, there are less CNS recurrences in patients who received osimertinib versus placebo. We haven't seen OS yet. I think that just based on the early cuts of CNS, we will see an OS advantage. It's also hard for me to believe that a hazard ratio of 0.21 in DFS won't translate into an OS. Nevertheless, without seeing an OS advantage, I think all of us have been adopters of this. We’ve been begging for better therapies in the adjuvant space based on the LACE meta-analysis just showing a 5% improvement in 5 years for adjuvant chemotherapy. So this was a game changer for me and it's exciting to be able to deliver this therapy for patients with resected lung cancer harboring an EGFR-sensitizing mutation.
Sands: The separation of the curves on this was really impressive. And you mentioned the intracranial CNS DFS. That's really a subset kind of a chart there. But that graph is what we want cancer treatment graphs to look like. It is impressive how much across the top essentially that osimertinib arm looks like on that graph.
Levy: Yes. it's a welcome change and one that we now leverage in our clinic every day for our patients.
Sands: Similar to you, I would expect osimertinib to look good in a DFS analysis. I am curious as to what will happen with the OS analysis as the data matures. But the degree of separation I did find quite shocking, and I think this is actually that perfect kind of a scenario where the classic question is, "if it was you, what would you want?" And academically we say, "well, we want to wait and see the data." But if it was me in this moment, I think in talking about 7th edition staging, stage II, stage III, the curves really separated in a way that was compelling enough that I think this is something to use currently. In the stage IB setting, it was a little less clear to me. What are your thoughts in parsing through the scenarios in which you would use osimertinib, where you may not, and is there any distinction between the stages? How do you have that discussion with your patients?
Levy: I tend to differ a little bit. The hazard ratio for stage IB was 0.50. And so, there's a 50% reduction in the risk of recurrence and or death. If this were breast cancer, everyone would be all over it. So I tend to consider my stage IB patients eligible for osimertinib. There's a question about whether patients should get both chemotherapy and osimertinib. Remember, chemotherapy is the only strategy thus far that has improved OS in the adjuvant space. So I tend to offer that to patients. So, if they’re stage IB through IIIA, I tend to offer this. And there are some pros and cons to this trial, as you mentioned. We haven't had OS yet. There are a high percentage of patients who didn’t get adjuvant therapy. I don't think we have time to talk about cost and the financial toxicity here. But there are considerations that that don't make this an absolute for all patients, but certainly one that I consider for my stage IB's through IIIA's.
Sands: That is an important statement as to the adjuvant chemotherapy portion. And if I remember correctly, there are more patients, or at least numerically within the stage IB group that did not get adjuvant chemotherapy. I still see adjuvant chemotherapy as being extremely important to this group. It does tend to be chemotherapy sensitive. Can you just speak a little bit about that? Do you still see chemotherapy as being important?
Levy: I do. I think that chemotherapy is still the backbone. And again, it's been the only strategy that improves survival in our early-stage lung cancer. So I tend to push for adjuvant chemotherapy and then tag on the osimertinib for 3 years planned. There are a few patients that are stage IB that maybe the benefit of adjuvant chemotherapy is a little bit questionable where I'll just offer osimertinib, but I think that's the exception, not the rule in my patient population.
Sands: So now transitioning to immunotherapy, we have in the adjuvant setting IMpower010. Do you want to take us through that?
Levy: This was a very large study once again trying to translate the meaningful benefit we've seen with immunotherapy in the advanced stage and leverage it in the early stage, similar to what happened with ADAURA. So this was a phase 3 study, an international study looking at completely resected stage IB to IIIA NSCLC and all patients, different to ADAURA, received adjuvant chemotherapy and then were randomized to 1 year of adjuvant atezolizumab every 3 weeks versus placebo. The primary endpoint here was progression free survival and the bottom line is there was a benefit for adjuvant atezolizumab in this study, but that benefit was more modest than we've seen versus the adjuvant osimertinib for the EGFR-resected patients. We saw in the patients that were stage II/III with a PD-L1 greater than 1%, there was a hazard ratio of 0.66. And again, we are desperate to find therapies that work in the adjuvant setting. We are desperate to move the needle in the adjuvant setting. Similar to ADAURA, we don't have OS yet. And the question is, is DFS enough in this case? Again, this hazard ratio wasn't 0.21 or whatever the exact hazard ratio was in ADAURA. Nevertheless, the FDA looked at this data and specifically looked at those patients that were stage II greater than 1%, and that was the most meaningful benefit and approved this. So I would say again that if patients have a PD-L1 greater than 1%, and they're stage II/III resected, I would consider offering this even though we don't have OS yet.
Sands: As you point out, it was approved for greater than 1%. And that was really a preplanned analysis. That being said, the greater than 50% really showed a pretty clear separation. And there's some question as to whether that greater than or equal to 1% is really being pulled entirely by that greater than 50% group and how meaningful the 1% to 49% group is. What are your thoughts on that and does that shape your discussion in any way with patients?
Levy: If you look at the forest plot univariate analysis here, in the PD-L1 greater than 50%, you saw the most robust hazard ratio roughly 0.4. If you look at the 1% to 49% group, it was 0.87. So the question really is — is it similar to KEYNOTE-042? In the advanced stage, did the greater than 50% truly drive the benefit here? I think the 1% to 49% is a discussion. It's not a slam dunk for me. I think the greater than 50% is certainly one that I would pull the lever almost every time. But for the 1% to 49% group it's more of a discussion. One that we do for a lot of scenarios, which is this shared decision making with the patient and explaining to them in a way that they can understand the benefit versus the risks of adjuvant atezolizumab.
Sands: And just some background for our listeners. In KEYNOTE-042 what you're pointing out is that trial really was driven. This is metastatic NSCLC. The greater than 50% group really showed a clear benefit to single agent pembrolizumab, the 1% to 49% group was a little more questionable. But the endpoint of greater than 1% with single agent pembrolizumab was positive and led to FDA approval as a single agent. But that being said, to your point, in the less than 50% PD-L1 expression, I think a lot of people are using chemotherapy plus pembrolizumab in the metastatic setting and not really swayed by the endpoint showing essentially greater than or equal to 1% as being a single agent pembrolizumab group. And so it sounds like you're highlighting some complexity to this and in the adjuvant setting now with IMpower010 after finishing chemotherapy in that 1% to 49% PD-L1 expression, you're saying, okay, it was positive but not as clearly positive. And therefore, this is a discussion point with patients in whether or not to include this, but not a clear cut, definitely should do it. Am I getting that right?
Levy: Yes. Another discussion where we don't have a lot of clear-cut answers, where patient decisions should be individualized or nuanced and it's a shared decision making with the patient. They have to participate in this and understand the risks and benefits.
Sands: Now, in the KEYNOTE-091, we don't have as much details on this yet, but this is pembrolizumab after having completed adjuvant chemotherapy. This is a positive trial. But any thoughts on KEYNOTE-091?
Levy: Yeah, it was a little confusing. We saw a benefit, I believe in DFS, but some of the subset analysis were a little bit wacky. The benefit didn't look that great in the PD-L1 group that was greater than 50%. Patients with EGFR mutations had a benefit which we don't normally see. I think this still solidifies the use of adjuvant immunotherapy. But I got to say that atezolizumab is still to me, the standard, given the benefit we saw, and benefit in subgroups that have scientific rationale to have benefit. So I would say that adjuvant atezolizumab remains the standard for me.
Sands: It does make me wonder a bit about PD-L1 expression in itself. I mean, this has really been somewhat of a messy and not entirely consistent biomarker across different immunotherapy studies. I think a lot of people were a bit surprised that the greater than 50% group in the KEYNOTE trial did not show a far more impressive benefit than the overall PD-L1 expression group. What are your thoughts on the PD-L1 expression as a biomarker for all of this? We've discussed how it is impacting the discussion within the IMpower010 data within the KEYNOTE data. Does this alter anything about your thoughts on PD-L1 or is it more that the KEYNOTE seems like an outlier?
Levy: I think the KEYNOTE seems like an outlier. You bring up a good point that PD-L1 is not a perfect biomarker but I think it's the only one we have right now. You have to remember that PD-L1 has heterogeneous expression among metastatic sites. So some sites may be more likely to have a PD-L1 expression, whereas others may not. Within the same patient, it can be heterogeneous, depending on where the needle is. I don't think it's perfect, but it's what we have right now and what we use. So I still think it should be integrated into treatment and specifically in the stage IV setting and at least based on IMpower010 it should.
Sands: Now let's transition to the neoadjuvant space. So we had the CheckMate 816 data. What were your thoughts on that?
Levy: I think we know a couple of things I want to say. Number 1, before CheckMate 816, we knew that neoadjuvant chemotherapy pretty much provides the same benefit that adjuvant chemotherapy does. And so we also know that you're more likely to receive all planned cycles of neoadjuvant chemotherapy versus adjuvant chemotherapy, at least based on some studies. So CheckMate 816 is building on this neoadjuvant strategy and looking at once again the addition of immunotherapy. So this was again a practice-changing study looking at resected stage IB to IIIA lung cancer with no EGFR or ALK alterations, and they were randomized 1:1 to either neoadjuvant chemotherapy versus neoadjuvant chemotherapy plus immunotherapy. They were asking the question, does the addition of immunotherapy with nivolumab provide any benefit with chemotherapy versus chemotherapy alone for resected lung cancers? The endpoint here was complete pathological response, which is defined as no residual viable tumor cells in both the primary tumor and the sample lymph nodes. And interestingly, and quite surprisingly, those patients who got neoadjuvant chemotherapy-IO had a complete pathological response of 24%. And in the neoadjuvant chemotherapy arm, it was only 2.2%. There are some other nuances here that I think are important, that patients that were receiving neoadjuvant chemotherapy-IO were more likely to have a lobectomy, less likely to have a pneumonectomy, and there didn't seem to be increased toxicities, even surgical morbidities in the patients who got neoadjuvant chemotherapy-IO. And finally, I would say we're starting to see early cuts of event-free survival. Once again, a composite endpoint of recurrence and or death. And we're seeing that that complete pathological response is starting to translate into outcomes. And so I think complete pathological response is an accurate surrogate here. It's probably some underrepresented data. In the past, we've seen, at least with neoadjuvant chemotherapy, that complete pathological response and even major pathological response does translate into better outcomes. So once again, we don't have OS here. Quite exciting based on this data. And we were fortunate to lead this study at Johns Hopkins. The FDA approved it. And I think we're in good shape here. We're going to move the needle once again. And this will take some altered conversations with the surgeons once they receive these patients to send them to the medical oncologist first before they do surgical resection.
Sands: I was impressed by the data, and I think that the pathological response in particular was very impressive. You've highlighted a lot of the strengths of this and the benefits that have been demonstrated. I think one question within that data was the number of patients that had surgery delayed or didn't get to surgery. Now, there may have been a bit of a difference between countries. That wasn't a statistical analysis, but just numerically. But can you take us through that? Is that a concern at all to you? Is there any nuance to that?
Levy: Jacob, correct me if I'm wrong, I don't think there were more delays in the chemotherapy-IO arm versus the chemotherapy arm. I'm willing to be corrected here. Even if there was based on the complete pathological response and based on what we're seeing here in terms of surgical morbidity, no difference. A better operation, more likely in the chemotherapy-IO arm versus the chemotherapy arm. I'm all in on this. And we were one of the early adopters of this, and we've already seen complete pathologic responses now that this has been done. We've done it several times. And I would say that the real-world data is starting to look consistent with what we saw in the study.
Sands: I completely agree with you. In someone who's getting neoadjuvant chemotherapy, including nivolumab, to me is absolutely recommended, barring the rare scenarios of not being able to get a checkpoint inhibitor for whatever reason. But in patients who are surgically resectable, where a lobectomy seems absolutely feasible, early-stage II type of a cancer, are you using neoadjuvant chemotherapy/nivolumab in that setting or are there patients where you're having the surgeon operate and then you're talking about adjuvant chemotherapy? And that's really the basis for my question about delays in the surgery. I think someone with stage IIIA, those people should absolutely be getting neoadjuvant treatment and therefore nivolumab should be a part of that. I wonder a little bit more, particularly in the earlier stages.
Levy: I think the challenge we have is with those never-smokers where you think they are likely to harbor an EGFR and ALK, those patients weren't included in the study and, you know, biologically may not have the best benefit with immunotherapy, even with chemotherapy. And you've got to wait for the molecular results to come back. So those patients, I will tell you, we recommend going to surgery and then hopefully we unearth an EGFR mutation and can deliver adjuvant osimertinib. Because we led the study, because we had a lot of experience with this and both our medical oncologist, Patrick Forde and our surgeon Stephen Broderick are on the paper. I mean, we've got a lot of experience in this, and we are trying even for our stage IIs and certainly our IIIAs we are recommending neoadjuvant chemotherapy-IO.
Sands: Before we get into the workflow of it, the Neoadjuvant Chemotherapy and Nivolumab in Resectable Non-Small-Cell Lung Cancer (NADIM) studies, how do those add to the picture?
Levy: There was a NADIM phase 2 study and then a NADIM-2 phase 3 study. Just briefly the NADIM study that looked just at resectable IIIA and got 3 cycles of carboplatin/paclitaxel plus nivolumab and then went on to surgery, were allowed to get adjuvant treatment. They received nivolumab for 4 months every 2 weeks and then for 8 months they received nivolumab every 4 weeks. The complete pathological response for IIIA was north of 50%. I think this was probably a little bit of an outlier here because we have the NADIM-2 study that was presented at ASCO 2022, a randomized study looking once again at potentially resectable stage IIIA and IIIB by the 8th edition staging. They excluded EGFR and ALK. Randomization 2:1 to the experimental arm of carboplatin/paclitaxel plus nivolumab for 3 cycles versus carboplatin/paclitaxel alone. Patients in the experimental arm also got adjuvant nivolumab every 4 weeks for 6 months. And interestingly enough, the complete pathological response which was the primary endpoint was 36%. And again, we're seeing that from the CheckMate 816 study that this may translate into event-free survival. The pathological complete response rate for just chemotherapy was 6.9%. So consistent data and again, this just solidifies the neoadjuvant chemotherapy-IO story. It does recapitulate some of the data we saw with CheckMate 816. So I'm glad we're seeing consistent signals across the board here for these studies.
Sands: So I'm hearing you say that essentially patients with a new diagnosis, you're doing a quick analysis and trying to get rapid genomic testing. Or if you suspect something likely actionable, you're then having those patients get surgery first and sorting that out potential actionable based on the tissue, is that right?
Levy: So I would say if it's an actionable mutation, specifically EGFR, ALK or if it's a never-smoker where we suspected we would probably operate first and then offer chemotherapy plus or minus targeted therapy or specifically osimertinib depending on whether we identify an EGFR mutation.
Sands: Now in the final little bit, I just want to look ahead. There's a lot going on now. We essentially have validated both targeted therapies based on the ADAURA data. We've validated immunotherapy based on the trials we've discussed. What are some trials happening now that you're particularly interested in?
Levy: I think there are a lot of trials. One looking at adjuvant ALK directed TKIs for ALK-resected lung cancer, RET fusion directed TKIs for adjuvant RET fusions. And then, perhaps more interesting is the neoadjuvant genotype directed trials that these basket studies that compile some of the Lung Cancer Mutation Consortium efforts looking at neoadjuvant targeted therapies before surgery, EGFR, ALK, RET, the list goes on. It'll be interesting to see where these targeted therapies land in the neoadjuvant space or the adjuvant space and what's the best way to use them.
Sands: And I'll just take a moment to highlight the ALCHEMIST trials. This is something I'll say I'm involved in. There is an ALK-directed adjuvant study related to that and then I am the principal investigator of ACCIO or A081801, which is a randomized adjuvant chemotherapy followed by pembrolizumab versus chemotherapy plus pembrolizumab followed by pembrolizumab. In both cases, pembrolizumab is for 1 year. And one thing about this trial is it is the only adjuvant study that is enrolling patients prior to any adjuvant chemotherapy. In a lot of these other trials we've discussed, there is essentially a population of patients who still qualify for a trial after finishing adjuvant chemotherapy, which is a sub selected group and probably a better than what we'll call real world outcomes in the trials we've discussed. That does wrap up quite a bit that we've covered. I want to thank all the listeners for joining us. Ben, thanks so much for joining the discussion.
Levy: Thanks for having me.
Sands: This is Dr Jacob Sands for InDiscussion.
Pembrolizumab Versus Chemotherapy for Previously Untreated, PD-L1-Expressing, Locally Advanced or Metastatic Non-Small-Cell Lung Cancer (KEYNOTE-042): A Randomised, Open-Label, Controlled, Phase 3 Trial
Pembrolizumab (Pembro) Versus Placebo For Early-Stage Non-Small Cell Lung Cancer (NSCLC) Following Complete Resection and Adjuvant Chemotherapy (Chemo) When Indicated: Randomized, Triple-Blind, Phase III EORTC-1416-LCG/ETOP 8-15 – PEARLS/KEYNOTE-091 Study
Medscape © 2022 WebMD, LLC
Any views expressed above are the author's own and do not necessarily reflect the views of WebMD or Medscape.
Cite this: Managing Lung Cancer With Neoadjuvant and Adjuvant Therapy - Medscape - Oct 13, 2022.