 Let’s finish up here. This has been an incredible conversation. It’s been an informative, rich discussion. Let’s provide some parting shots before we conclude. Dr Leal, I’ll start with you. What are your take-home thoughts as we wrap up this discussion?

 I think it’s really encouraging to see these improved outcomes in early stage non–small cell lung cancer, and now the locally advanced non–small cell lung cancer space as highlighted by the data that we discussed. I really hope this will translate into improved cure rates for our patients. I’d like to thank everyone for being here today. It’s been great to see everyone, and I hope to see you all in person in the near future. Thank you.

 I wanted to thank you, Dr Levy, for such great leadership in moderating all of us and keeping us in line; and Dr Leal, Dr Naidoo, and Dr Zhang for brilliant discussions. I’m going to start out by saying we’ve come a lung way, pun intended. It’s really important that these therapeutic options advance for all of our patients, and it’s also our job, as we learn more and more now and become the disciples of using these therapeutics, that we teach the strategies to everyone. So these educational symposia are incredibly important. Thank you everyone, and thanks for inviting me. This is great for our patients.

 It was a great pleasure meeting all of you. This was well organized. The way I seduce our fellows to get into thoracic oncology is that I keep telling them that we have 2 waves of revolution in lung cancer therapy. One is targeted therapy and the other is immunotherapy. With all of these wonderful studies with the beautiful data, we are going to see the same revolution happen in early stage nonmetastatic lung cancer. It was a fantastic discussion. I hope to see you soon, and I hope the pandemic is over soon as well.

 Very good, Dr Zhang. Dr Naidoo, your final thoughts?

 Thank you, Ben, for an amazing discussion. I thought this was really educational for all of us. I was contemplating becoming a thoracic oncologist 10 years ago, and my colleagues said, “Why would you want to do that? There haven’t been any major advances. All we have is platinum doublet therapy. You should be a breast oncologist.” I remember saying, “No, I think lung is the next big thing.” This conference, among others, really highlights the advances that have been made in that time, and these great advances that are coming to our patients.

 I’ll layer in my brief commentary. It takes a village. This is getting complicated, but it is very exciting. All disciplines, I say it takes a village, not only from medical oncology, radiation oncology, surgical oncology, pulmonary, but also the village within medical oncology—our social workers, our navigators, our nurses. It takes everyone to take these patients through this journey that has been altered dramatically over the past 10, 15 years. What an exciting time it is to be in this field.

Thank you all, again. And to our viewing audience, we hope you found this 

 discussion to be useful and informative. Thanks again.

Videos

Takeaways on Treatment of Early-Stage NSCLC

 Let’s discuss the risk of pneumonitis for patients on durvalumab post-chemoradiation. Dr Naidoo, I’ve heard you’ve done some work on pneumonitis before, so I’ll turn to you first. What is your take on the pneumonitis risk with immunotherapy, particularly durvalumab, post-concurrent chemoradiation? Is it any more risky to do that, independent of whether radiation is given or not?

 There have been a couple of presentations on this topic, led by Johan Vansteenkiste, MD, PhD, at the World Conference on Lung Cancer a couple of years ago, updated recently, and now there is a paper in press on this particular topic. I think we all expected that the pneumonitis rate with a consolidation immunotherapy approach after chemoradiation could be horrific, and I think we were all pleasantly surprised when it was not much of a departure from the placebo arm in the PACIFIC study.

I would say that the rates of pneumonitis are pretty encouraging, or not unacceptable, and are really a surprise in that regard. In many of these subset analyses looking at the rates of pneumonitis, we did see that patients of Asian ethnicity appeared to have higher rates of pneumonitis—patients with 

 mutations—than what we might expect. There have been some subsequent real-world data that have been published showing that patients who receive higher doses of radiation, radiation that involves a greater proportion of the lung field than was included in the PACIFIC study, may also be at increased risk. I think these are all intuitive risk factors for the development of pneumonitis.

Touching on your other question regarding how we differentiate between RT [radiation therapy] and I/O [immune-oncology] pneumonitis, and what is the gray area in between, this is a hugely important area for us to understand a bit better. There are some assumptions and some barriers that are preventing advances in the area. In differentiating them just routinely at the moment, we understand that I/O pneumonitis can have a variety of radiographic appearances. It can be pretty esoteric in terms of its time to development—anywhere from 9 days to 19.5 months after the start of immunotherapy in our 

 paper. RT pneumonitis is a bit more predictable in this sense. Also, it generally follows the curvature of the radiation field, whereas I/O pneumonitis is usually a bilateral phenomenon. That is something that we use in a colloquial sense, but certainly hasn’t been formally looked at and really does require a formal look in the stage III population.

I think the other thing to consider is we don’t know that pneumonitis can so cleanly segregate into I/O-related or RT-related cases, and there probably is a middle zone where both factors are coming into play in the development of that pneumonitis. We need to study that. The only data that have been published in that area was an abstract presented at AACR [the American Association for Cancer Research] looking at a 7-cytokine signature that was associated with pneumonitis in this setting. So I think those kinds of biomarker studies are very important to do. We did a BAL [bronchoalveolar lavage] study in patients who developed I/O pneumonitis and identified some immunologic signatures in the BAL. I think we need to have a low threshold for doing bronchoscopy and participating in these kinds of studies, in helping to understand who’s at risk for this very important toxicity.

 Yes, I think it’s also important for future trial development. As you’re aware, Dr Naidoo, we’ve got some ongoing trials looking at immunotherapy given concurrently with definitive radiation. Soon, I think we’ll find out the optimal timing of immunotherapy. Is it consolidation? Can we safely give it with radiation? Can we do dual checkpoint blockade post-concurrent chemoradiation? I know you’ve been a part of some of these studies. Do you want to give some parting shots on the future trials for this?

 Sure. Again, I feel like my whole discussion today has been about believers and nonbelievers. But I think that’s where some of these concepts come from. Honestly, we don’t know about timing and optimal dose. This is something that has evaded us, in terms of a clear answer, when combining radiation with immunotherapy. It’s something that requires a dedicated study. Regarding the concurrent group, the believers obviously support the 

 approach—more treatment, more neoantigens. It’s the same sort of argument that we had on neoadjuvant immunotherapy. If you add chemotherapy, and then now if you add radiation, you increase that swell of neoantigens and potentially improve outcomes from immunotherapy. I think that’s the basis of the concurrent argument. But in the Salma Jabbour, MD, study that was highlighted recently, 69% of patients had grade 3 toxicity. So while it might be scientifically interesting, it might be very difficult to administer.

In terms of the consolidation approach, the Hoosier Oncology Group study randomized patients to nivolumab, or ipilimumab and nivolumab in the consolidation setting. Again, this showed a higher rate of toxicity for the combination arm. This is exactly what we expected. We are awaiting survival outcomes from that. And then the last study we know of looked at giving immunotherapy pre- and post-chemoradiation. This was a study of single-agent atezolizumab.

I think with all of these, we’re going to have to see the data in their entirety and balance these approaches. I do foresee that in the same way we have 9 options in first-line lung cancer now, we may have several options that we don’t have a clear comparator for. We will probably be having these similar discussions in the coming months or years.

 Yes, an incredible time. Confusing, but what a gain for our patients to have these options. Who would have thought?

Risk of Pneumonitis Post-Chemoradiation in Stage 3 NSCLC

Dr Zhang, trials are fine and dandy, but real-world data can also be important to look at. There were recent updates from ESMO [the European Society for Clinical Oncology annual meeting] looking at some real-world data with durvalumab. Do you want to go over that data and tell us your impressions? Any take-home points?

 Yes. They actually had 3 studies related to real-world use. One abstract that was presented was on a retrospective review of practice in British Columbia. About 196 patients were identified, and they found that 50% of patients did not receive durvalumab. This was partly due to physician’s choice as well as patient preference. They argued that the benefits need to be better communicated so that more patients can benefit from the treatment. Now, their median time to start was about 43 days after chemoradiation therapy, which is a little bit different from the PACIFIC trial. But their 1-year overall survival [OS] is pretty good—91% in the durvalumab arm versus 72% in those who received surveillance. This is more impactful then what was reported in the PACIFIC study.

There’s another study that presented data from a German expanded access program. This study looked at 126 patients, and 43% of them finished 1 year of treatment. The 12/18-month overall survival rate was 79%/75%. This is equivalent to what we saw in the PACIFIC study.

I think the more interesting thing is Nicolas Girard, MD, in fact, also had this study called PACIFIC-R. “R” stands for 

. They actually described patients’ clinical characteristics, but more data regarding the PFS and OS, as well as the duration of durvalumab are still pending, probably for the next meeting.

Nice summary. I’m encouraged to see that the real-world data presented in some ways mirrors, at least in outcomes, what we’ve seen from phase 3 data. Hopefully, there will be more uptake of durvalumab across the world as the drug gets approved across countries for unresectable stage III disease. This is the standard in the US and is one that should be considered, I think, for all patients.

. There are subsets of patients in the PACIFIC trial who maybe didn’t derive a benefit. There were the PD-L1 0% patients, the confidence interval for cross unity…. And then, the 

-positive patients. I get a lot of, and I’m sure we all do, questions from community oncologists. “What are you doing with the patient who is unresectable, who is 

-mutated, who has gotten concurrent chemoradiation?”

We’ve got this great ADAURA data, but the drug is not approved in the unresectable stage setting. Are you giving immunotherapy? Dr Naidoo, you are the unfortunate analyst who has to talk to us about what you do for these patients.

 Yes, I think it’s tricky. When PACIFIC was first presented and published, there was also the controversial European Medicines Agency approval of durvalumab only in patients who were PD-L1–positive. That caused a stir in the beginning, but I think highlighted some of these issues and the fact that there are subsets of patients who may or may not derive benefit from this approach. I think as Dr Leal said before, it requires a careful discussion with the patient. They biologically probably have a different cancer. Also, particularly for the 

-mutated patients, what are we going to do next? Or, what does this imply for their whole cancer journey?

From there, I think the Adam Schoenfeld, MD, data published in the 

 about sequential EGFR-TKI [tyrosine kinase inhibitor] therapy after PD-1/PD-L1 therapy is really important. As a recap, that was a retrospective analysis from Memorial Sloan Kettering Cancer Center that looked at patients with 

-mutated lung cancer who got PD-1 or PD-L1 therapy followed by osimertinib, versus the other way around. That sort of sequence was associated with severe immune-related toxicities, whereas the opposite sequence was not associated with immune-related toxicity. And similarly, other EGFR TKIs were not associated with that toxicity. So, sequencing really matters. For those patients, I often will have a frank discussion of whether we think durvalumab may make sense. In terms of the PD-L1 subsets, again, I think it’s a tricky situation. I would probably offer durvalumab regardless of the PD-L1 status, but recognizing and saying to the patient that, “These are the factors that we take into consideration.” How do they feel about it?

What are other people’s thoughts on delivering immunotherapy for the 

 subset and the PD-L1 0% groups? Are people extrapolating from ADAURA and giving osimertinib as consolidation therapy even though it’s not approved, or are people sticking by the data? Dr Zhang?

 I would probably not consider PD-L1 as the factor to make a decision as to whether to give durvalumab. I’m aware that this post-hoc analysis is basically suggesting that if the patient’s PD-L1 is negative, there is not much value or benefit from durvalumab. But we have to remember this is only a post-hoc analysis, number 1. And number 2 is, this is a pretreatment biopsy. The reason we are giving chemoradiation therapy right before durvalumab is we hope that chemoradiation therapy can change the immune profiling of the cancer. I don’t think completely relying on the pretreatment PD-L1 is the way to go. And number 3 is, in the PACIFIC study about half of the patients did not have the biopsy. So based on that, I think our decision purely based on PD-L1 expression is probably not ideal.

 mutation, I think it’s very tempting for me to consider osimertinib in this setting. However, I would not go to this after immunotherapy, as Dr Naidoo mentioned. That study reports about 15% incidence of immune-mediated toxicities, and some of them are pretty severe. I would not take that risk. But having said this, it seems like this is only relevant to osimertinib. When they look at the patients who are receiving immunotherapy followed by erlotinib, they do not see the same incidence. It could be something just relevant to osimertinib, but not necessarily the EGFR TKI, the whole category.

Now, whether there’s any value, let’s say a patient has not received immunotherapy and has finished chemoradiation therapy. Is there any value or reason to consider osimertinib? I think there is a study, LAURA, that is ongoing. I very much anticipate the results. I think the chance of success will be high.

 Look at the ADAURA study. In earlier-stage disease, we already see a good response. For this LAURA study, the patient population, regarding stage of disease, is much higher. My educational guess is there is a high chance that this study will be successful, but we’re waiting for that.

 The LAURA trial randomizes osimertinib to no osimertinib. There is no durvalumab arm in that trial.

 Right, it’s osimertinib indefinitely. It’s a global trial, and I think one important point is that if you do have a patient in that setting, they’re enrolling in the trial. This is a difficult trial to complete accrual to get meaningful end points. So I think it would be important to consider the trial if you have a patient in that setting, instead of trying it off trial because otherwise we won’t learn and we won’t know if that’s really the way to go.

 I think you can extrapolate that a little bit more. In our clinical practice, we don’t tend to give durvalumab to the 

-positive patient population. But then we have to think about the 

 translocated population as well, and the 

 patients. We’re seeing a large number of those patients. Since they don’t respond as well to the immune checkpoint inhibitors, it requires a discussion with the patient. It is important to think about clinical trials, if you have them. If not, you have to individualize the therapy.

Yes, I think it requires shared decision-making. I’m not convinced that consolidation durvalumab is best for 

-mutated patients, but I think the jury is still out and we’ll learn from LAURA.

Considering PD-L1 and EGFR Status With I/O in Stage 3 NSCLC

 Let’s move on to the final section, where we will discuss the role of immunotherapy in unresectable stage III disease. We’ve talked about the role of targeted therapy in the adjuvant setting and the role of neoadjuvant immunotherapy in the resected setting. But in the unresected setting, even for cure, we’ve obviously got new treatment paradigms that have evolved over the past 2 years. We have an FDA approval for durvalumab based on the PACIFIC data.

The PACIFIC trial has been presented multiple times. It has been published twice in the 

, once for the PFS [progression-free survival] end point and then for an OS [overall survival] end point. We’ve all attended these historical meetings in which we showed up in person. It was presented at ASCO [American Society of Clinical Oncology Annual Meeting] and at the World Lung World Conference on Lung Cancer in Toronto in 2018, and we have some data from ESMO [European Society for Medical Oncology Congress]. Dr Leal, do you want to talk about the PACIFIC data at high level, and then share some updates from ESMO too?

Yeah, definitely. It is great to have this update regarding the 4-year survival for the PACIFIC data. It’s very encouraging. It brings with it a lot of optimism for patients with resectable stage III non–small cell lung cancer, especially if you compare it with historical controls. We had 5-year overall survival rates, despite multiple randomized phase 3 trials, of about 10% to 15% at best.

The PACIFIC trial was a phase 3 trial that randomized patients after completion of concurrent chemoradiation with at least 2 cycles of platinum-based chemotherapy to durvalumab every 2 weeks or placebo. The initial report was an improvement in progression-free survival. The hazard ratio was 0.52. There was an improvement in overall survival, with a hazard ratio of 0.68. As you said, it’s been presented multiple times, and the good thing about having the updates is that it’s been reassuring to see that the data are holding up, and we’re seeing consistent improvements in PFS over time, as well as OS. 

This leads us to think that at the 5-year overall survival update, perhaps we’re actually going to be talking about cure rates. Are we curing patients with stage III disease with consolidation durvalumab?

The update now is a 4-year OS update. We have a median follow-up of 34 months with durvalumab. This is the first time there is a median overall survival in the durvalumab arm. We’re seeing a PFS of 17.2 months vs 5.6 months, with a hazard ratio 0.55. We’re seeing an overall survival benefit as well—a median of 47.5 months vs 29.1 months and a hazard ratio 0.71. There are really impressive results.

The tolerability of durvalumab has been well known to all of us, and it is well demonstrated in the trial. It really is a game changer, and it is reassuring to see these updated 4-year overall survival updates.

 Great summary. How do you put this data into clinical practice? I’ll put you on the spot. The PACIFIC trial did not allow consolidation chemotherapy, which we were generally doing; at least I was for some patients. A third of patients got induction chemotherapy, which no one really ever does, at least in our institution, or at least until recently. Which chemosensitization regimen are you generally using for these patients? Do you always start durvalumab right after the radiation is done, or do you give consolidation treatment?

One of the things I think about, even seeing this update, is that half the patients were alive at 4 years. When I start a patient on concurrent chemoradiation, I think about how I get them to the finish line. How do I get them to durvalumab? I want to make sure we don’t rock the boat during chemoradiation with extra toxicity. In general, I’ve preferred to use the carboplatin-paclitaxel regimen on a weekly basis during radiation. There are patients for whom I may alter the regimen for 1 reason or another. Say somebody has really bad neuropathy. If I feel they can’t handle Taxol, I may change the regimen. This is all based on comorbidities and certain clinical characteristics of that patient. But in general, I favor carboplatin-Taxol with concurrent radiation.

I have gotten away from using consolidation chemotherapy. I don’t think we have strong data for that. We did that in select patients because we had intuitively thought that more is better. But really, we have not seen that pan out in phase 3 clinical trials. So I don’t use consolidation chemotherapy. I am happy to let that go, and so are patients. That added toxicity and might take patients away from being able to get to the consolidation with durvalumab, which is really the main goal. That seems to be the main treatment that has really improved survival and outcomes for patients.

 Yeah. Are there any differences in how the rest of you manage the chemosensitization regimen and the timing of durvalumab for patients, or is that generally what everybody is doing—weekly carboplatin-paclitaxel with radiation, followed by durvalumab within 2 to 4 to 6 weeks after the completion of radiation?

 On the PACIFIC trial, a specific chemotherapy regimen wasn’t necessarily mandated. I do use carboplatin-paclitaxel, as Dr Leal explained. I also use platinum-pemetrexed for patients with adenocarcinoma because it is very well tolerated. I am less of a fan of the classic SWOG regimen. I’ve been burned a number of times with that regimen. However, there are purists who use it and use it well, and there would be no reason not to use that regimen. But generally, patients do find it to be a tougher regimen.

In terms of the timing of beginning durvalumab, there was a lot of discussion regarding optimizing the timing—a change in the protocol to allow for a wider window in commencing treatment. It changed our paradigm slightly, in requiring a scan to rule out progression. Beforehand, this was something we weren’t necessarily doing until later. The timing has become more of an issue and requires us to have seamless communication with our radiation oncology colleagues. That might change the workflow for many oncologists in settings where they don’t have real-time access to their radiation oncologists.

 Are there any other thoughts on the management of this, in how we apply these data into clinical practice?

 The every-2-week dosing is a bit tougher for patients, in regard to coming in each time. Of course, we look forward to the q4 week dosing. The other question always becomes, why only 1 year? Should it be 2 years? Is it going to be 3 years? Or is it going to be 6 months vs a year? All those are important questions. We want to make it as palatable for patients as having a great overall survival but make it as less chemotherapy, less immunotherapy, less surgery, less whatever we can get away with by not having this toxicity. Those are all important questions as we go forward. But this is 1 of the first landmark studies that has revolutionized our practice.

 Great point, Dr Salgia. Receiving 1 year of durvalumab treatment, every-2-week dosing vs every-4-week dosing, and in this era of the coronavirus pandemic, what treatment considerations are there in terms of how you’re timing this and how long you’re going to give it?

 Yeah, that’s very nerve-racking, right? Our colleagues at Memorial Sloan Kettering Cancer Center published a beautiful paper in terms of cancer discovery, and there are a lot of other papers out of China and other institutions as well. We’re very worried about the coronavirus pandemic. You have this acute lung injury that can happen with COVID-19 [coronavirus disease 2019]. You can experience pneumonitis with immune checkpoint inhibitors. At what point do you worry about concomitant infections? It’s very clear that you have to be able to be cognizant of the pandemic, try to test your patients as best as you can, or at least have symptom checks. Then stick to the regimen, because that’s what FDA has approved. At the same time, for those who get COVID-19, I’m not necessarily sure we should give them this immune checkpoint inhibition. That data is not really out there to help us, and we need to be able to gather more information.

 It’s a lot of medicine as an art and not a science with COVID-19, in trying to understand timing, frequency, and monitoring of patients. That’s a whole other discussion for a whole other time.

PACIFIC Trial in Stage 3 NSCLC

 Where do we go from here? What are the trials that we need, or what trials are currently going on in this space that hopefully will help answer these questions? Do we have phase 3 trials coming out in this space? If so, when will we know answers? Does anybody on the panel want to take the lead on general approaches for the neoadjuvant setting? We’ve got a lot of phase 1/2 trials here. Where do we go from here?

 There are certainly large, international phase 3 studies that are looking at this. The differences in the approaches between countries will matter. Obviously, patients are generally sold on immunotherapy and the activity of immunotherapy in lung cancer. These studies, where available, are going to be taken up with a great deal of enthusiasm.

It’s not the same cancer, but there’s a cooperative group study of limited-stage small-cell disease using immunotherapy. You know, a rare beast. But that study is over accruing. It is accruing faster than they thought was possible. So I think there will be a lot of excitement for these studies.

Obviously, only 20% to 30% of our patients, on a good day, have early stage or resectable disease. The other factor is that we need to do better in identifying earlier-stage lung cancer patients if this is really going to be an effective treatment. That goes hand in hand with this. The majority of our patients have stage IV disease. The advances here are going to need to be coupled with advances in lung cancer screening and those kinds of studies.

 Yeah. I want to see—and there are trials ongoing—neoadjuvant chemotherapy vs neoadjuvant chemotherapy with I/O [immuno-oncology]. We can consider neoadjuvant chemotherapy the standard of care if you’re going to consider neoadjuvant therapy. Having randomized trials such as ones that explore chemotherapy vs chemotherapy with I/O would be the most relevant to me to help answer questions about toxicity, major pathologic response, complete pathologic response, and outcomes. We’ve run into this in our own centers, and we don’t like to talk about it, but sometimes it takes the surgeon’s buy-in. Is this something that you guys experience at your institutions? Dr Leal, you’re smiling a little. Do you have to have the surgeon’s buy-in here?

 Yeah, for sure. It requires multi-D [multidisciplinary care] to the max. The reason why I’m smiling is we have a recent neoadjuvant trial in which the surgeons are coinvestigators. We actually persuaded 1 of the surgeons to sit through an SIV [site initiation visit] with us and be a part of every step. Obviously, they buy into it. We have a great multi-D team. We have a great group of thoracic surgeons. It’s not something that has been historically on their radar. It’s really a shift in how you think about clinical trials and really work together to advance the field in a meaningful way. We need buy-in from the thoracic surgeons, and we need buy-in from the pathology group. This has really been an incredible effort to get everybody on board to get these trials up and going.

 It takes a village on so many levels. Dr Salgia, your thoughts on this and getting the surgeons to buy in to this strategy?

 I like what you said, Dr Levy. It takes a village. And I like the new term, Dr Leal. I’m going to say multi-D rather than multidisciplinary. It absolutely takes that communication. What I love is that we’ve come a long way. In the old days, it used to be 1 specialty vs another specialty. It really is a combined effort now. Our surgical colleagues see that early stage patient, and then we all do clinics together. It makes it easier. Our surgical colleague might say, “Hey, Ravi, are you able to see this patient today? You’re in clinics and so forth.” It makes it so much easier.

Those days where it took weeks and weeks to see a patient are gone. With this cross-communication, having the cross–multi-D clinics altogether, it makes things so much easier. That’s really what we envisioned as the future in the past, and it’s here. I’m very happy with that.

 That’s a great summary. Just going back to Dr Leal’s comment, I would propose that the surgeon who attended the SIV should get an award. I’ve never heard of such a thing. That’s just emblematic of the need to communicate and all be on the same page with not only the clinical care but the science. Everyone is right on target with what we must do moving forward. Dr Zhang, what is your experience with this? Discuss your interface with the surgeons and how to pull this off.

 Our surgeons have had concerns about whether immunotherapy induces some toxicity, so that a patient would no longer be a perfect candidate for surgery or that there’s some delay. But based on those clinical trials, they are now more convinced. We need to do something in early stage lung cancer. But for surgery, there’s a big change, and they’re really into it now. In fact, we also have clinical trials using neoadjuvant immune checkpoint inhibitors.

 For those patients who are going on neoadjuvant trials, is there patient reluctance? Or are patients saying, “I just want this out. I don’t want to wait to get the treatment.”? We’ve run into this issue before. We thankfully got the surgeon and the medical oncologist on board. But the patient said, “I don’t want to wait. I’ve been taught that this needs to come out. I’ve always been told cure is achievable through surgery.” Is that happening at other institutions? Dr Zhang, go ahead.

 Yeah. In fact, I have patients like this. I have a guy who just wants to get this done as soon as possible. Of course, when you discuss neoadjuvant clinical trials, mention potential risks, etc, they don’t want to lose any chance that their cancer can’t be resected. Certainly there is, but I would say the majority of patients are interested in participating in a clinical trial that hopefully offers longer survival or a chance for cure.

 I agree completely, Dr Zhang. In the old days, it used to be much more of an issue. If it were some cytotoxic therapy, or given in combination with bevacizumab, patients were reluctant. But this doesn’t happen as much now because patients are seeing this information flow through on social media. You even commercials on TV. Patients want to be part of an interesting and important clinical trial, and I certainly think it will become standard of care for us in the next few years.

 One last question. For patients who are 

 mutant, is this the right modality? Let’s say we know that they’re 

 mutant from the biopsy, not necessarily from the surgical resection. Do we pump the brakes on neoadjuvant immunotherapy given that we’ve got a great adjuvant drug? Do we not have enough data? This is a tough one. Dr Naidoo, I see you’re thinking hard about this one.

 Yeah, no data. We encountered this a couple of times in various different patients who were being considered. In many of the neoadjuvant studies, these patients are not necessarily excluded. We know this was the case for PACIFIC as well, which I know we’ll get to. It opens up a bit of a quagmire for what to do next in the patients who have received immunotherapy. It’s an evolving experience. Obviously, we’ll discuss that in the context of PACIFIC. But right now, these patients are definitely being enrolled, and we will have to see whether they derive equivalent benefit, don’t derive the same benefit, or have different toxicities. Until we have the data, it wouldn’t behoove us to exclude these patients from this opportunity in case they may benefit.

 Any parting words on the role of neoadjuvant immunotherapy for 

 Yeah, I’m not a fan, to be honest. I think these tumors are cold tumors. Are we doing a service for our patients, knowing the data for stage IV disease and then applying it to early stage disease? I’m not necessarily sure. It’s clinical-based evidence rather than evidence-based medicine. Without haing that trial, I just don’t know if it’s appropriate to put your 

 or ALK patients on these trials. We have to be very concerned about that. That’s why we do next-generation sequencing for everyone. Then we have that discussion with our patients.

Future of Neoadjuvant I/O in ES NSCLC

 Let’s talk about some of the data, Dr Salgia, that came out of ESMO [European Society for Medical Oncology Congress]. Although maybe not as promising of activity from the trials that Dr. Leal went over, but let’s get your take on the ESMO presentations.

 Thanks, Dr Levy. Beautifully summarized, Dr Leal and Dr. Naidoo. I enjoyed that discussion. There were 2 abstracts that were presented. One was by Dr. Marie Wislez, and this was the IFCT-1601 IONESCO study looking at durvalumab. This was given on days 1, 15, and 29, and it was relatively very well tolerated. The partial response rate was 8.7%, with a major pathologic response [MPR] rate of 18.6%. There were about 46 eligible patients, but the issue was it was stopped early. The trial was stopped early because 9% had observed deaths. This was most likely related to not the immune checkpoint inhibition but comorbid disease, and that’s what the author stated.

The second study, PRINCEPS, was presented by Dr Benjamin Besse and colleagues. This study looked at using atezolizumab. Only 1 dose was given, and then surgical resection was 4 weeks later. That’s an important end point. They looked at 30 patients. Four of 30 patients had MPR. That was 14%.

This summarizes that for these 2 abstracts that were presented, I don’t negate what’s been done in the past, but we always tend to come back, with the pendulum coming back to normalcy. The initial trials may show a great response rate, but this one is showing major pathologic response. But these ones, not necessarily. I agree with all of you that it may be about not just using an immune checkpoint inhibition strategy by itself, but is it a combination? Let’s say nivolumab with ipilimumab gives you better outcomes, or is it chemotherapy plus immunotherapy?

My only editorial comment here is that we have to be cautious. We have seen patients who have gotten immune checkpoint inhibition or chemotherapy combinations and have had hyperprogression. We have to really consider whether we should do next-generation sequencing on those patients up front to see if there’s a negative predictor, like STK11 or MDM2 or something else, that will help us prevent that. It would be a terrible shame, even if 1 patient had hyperprogression in our clinical trials or in our clinical experience. That would really negate a lot of things. I do think we need some predictive values as well.

 Yeah, I think you’re right. Understanding the underpinnings of the tumor as well as trying to understand end points for these trials and harmonization of end points is going to be important. I’ve kept a tally in my head as everyone has gone over all this data. My head is spinning. Dr Patrick Forde’s data are around a 40% to 45% major pathologic response. The other single-agent immunotherapy neoadjuvant trials elicit a major pathologic response rate somewhere between 15% and 20%. From the NEOSTAR data, it looks like dual-checkpoint blockade elicits a major pathologic response, around 18% to 20%; maybe a little bit higher—30%. And with chemotherapy and I/O [immuno-oncology] as neoadjuvant therapy, we have a pathologic complete response of close to 60%.

Dr Zhang, you’ve been quiet, but I know you’re taking notes. I see you’re ready to distill all the confusion and simplify this. What is your take on all these trials? What do you think about them? How do we put this all together?

 These are really good summaries. I have learned a lot from everyone’s comments. My take-home message is that we’re seeing a similar trend like what we’ve seen in the metastatic setting. Single-agent immunotherapy has some response but is pretty limited. When we combine that with chemotherapy, the response is higher, especially when we use the major pathologic response. When we combine with another I/O therapy, I/O plus I/O, it seems like the response is higher as well. I would say that if we’re really serious about considering this in the neoadjuvant setting, if we are going to use MPR as 1 of the end points, then the combination of I/O plus chemotherapy might be something that I would favor.

Having said this, what we also know from the NADIM trial is that when we combine chemotherapy and immunotherapy, we have a higher incidence of grade 3 and 4 toxicities. Whether those things would really delay surgery in the real world is another question.

Another point Dr Salgia mentioned is that we actually need to know the molecular profile of those cancer cells as well. For example, if a patient has an MDM2/MDM4 application, 

 application or mutations, there’s a chance of hyperprogression. In clinical trials this would be rare, but we cannot really afford for any patient in our lives to get to hyperprogression. Certainly, these are things that need to be investigated as well.

 Yeah, it’s getting exciting and complex at the same time.

Neoadjuvant I/O for ES NSCLC: Key Trials from ESMO 2020

 Dr Leal, we’ve had several neoadjuvant studies that have been published or presented that are starting to look promising. Maybe you could go over some of the ones that we’ve had up to ESMO [European Society for Medical Oncology Congress], and Dr Salgia will talk about some of the neoadjuvant immunotherapy trials presented at ESMO. Can you discuss the current state of affairs for neoadjuvant immunotherapy? Just provide a high-level overview of some of these trials.

 We’ve had really promising trials that have been completed on the use of preoperative nivolumab in patients with early stage non–small cell lung cancer who underwent resection following 2 doses of nivolumab. The primary end point of this study was really safety and feasibility. This was a small study. There were about 21 patients who had tumors removed. These was very interesting data because they demonstrated a very high response rate in these patients. Nine of 20 patients were resected, and they called out a major pathologic response [MPR] of 45%. That really caught everybody’s eye and interest, in regard to this strategy. Interestingly, in this study, this was independent of PD-L1 expression. In the pilot phase of the study, there was no delay in surgery. This was concluded to be something that was feasible and safe, and it really opened up the path for the up-and-coming neoadjuvant strategies we’re seeing.

Another promising strategy was the LCMC3 study, which looked at preoperative atezolizumab in resectable patients. In this study, adjuvant chemotherapy was also allowed. Again, we saw similar safety and tolerability with atezolizumab in the neoadjuvant setting. We also saw that it did not delay patients going to surgery. The major pathologic response was lower, at 19%. This was encouraging but lower than what we saw in the nivolumab study. The pathologic complete response rate was 5%. Again, the major pathologic response was independent of PD-L1 expression.

We also saw presented data on the NEOSTAR study. This randomized patients in the neoadjuvant setting to nivolumab vs nivolumab plus ipilimumab. Again, this study showed the feasibility of this approach.

As one can expect, there were higher MPR rates in the nivolumab-plus-ipilimumab arm—33% vs 17% in the nivolumab arm. I will say this study reported out that there were no unacceptable toxicities. However, there was a death on the study because of complications in the postoperative setting. This leads us to understand that, so far, there haven’t been any unacceptable toxicities. But as you roll out surgery in the post immunotherapy and immunotherapy-chemotherapy setting to general practice, the safety needs to be demonstrated better in this patient population.

Most recently, the most provocative results that have been presented are from the NADIM study. The NADIM study actually looked at a chemotherapy with I/O combination for patients in the resectable setting and showed a very high overall response rate of 78%, with a complete pathologic response in about 60% of patients. Again, it’s going back to looking at safety and tolerability and feasibility of rolling this out to practices as the data come out and really validate this end point. Certainly, the MPR end point is a very good 1 to use if it really is validated because you can get results quicker. Overall survival would be looked at in a longer study and would take more patients. So I can understand why we’re using this end point, but I would like to see that it’s a validated end point that we can trust to roll out to our patients.

Moving forward, just to summarize, neoadjuvant strategies are here to stay. I’m excited to see the results from ESMO. Chemotherapy and I/O is probably the way to go, as long as we can demonstrate safety, feasibility, no delays getting to surgery, and no increase in postoperative complications.

 That was a very nice overview of very complicated studies. These trials are really landing front and center. I did a double or triple take with a pathologic complete response of 60% in the NADIM trial. That is quite high with just 3 cycles of chemotherapy and I/O followed by surgery. I’ll be very interested to get people’s take on that, because that’s 1 that really stood out to me, especially coming from our institution, which has done a lot of neoadjuvant studies.

Key Trials in Neoadjuvant I/O for ES NSCLC

Let’s move on to our next section, which is the role of immunotherapy in early stage disease. All of us are aware of the immuno tsunami that has occurred in non–small cell lung cancer, specifically in the advanced-stage setting.

We now have 9 different approvals and different indications for advanced-stage disease without an actionable mutation. It’s incredible to think about how many options are currently available as single-agent I/O [immune-oncology], dual-checkpoint blockade, chemotherapy with I/O, dual checkpoint blockade with 2 cycles of chemotherapy. Of course, all these are approved for advanced-stage disease and lead to improved outcomes.

Similar to the targeted therapy paradigm, we’re trying to look at the role of these strategies in early stage disease. There is a role that we’re starting to learn about for perioperative immunotherapy and immunotherapy combinations. Clearly, 1 of the ways it has been explored first has been in the neoadjuvant setting, and it makes a lot of sense to look at this in the neoadjuvant setting.

Jarushka, you’ve worked a lot on this topic. Talk a little about the biological rationale for immunotherapy in early stage disease, and maybe even the clinical rationale for neoadjuvant immunotherapy.

 I’d be pleased to talk about that. I obviously worked with Dr Patrick Forde on his neoadjuvant study of nivolumab, and I have my own study that is looking at combining immunotherapy with radiation in the neoadjuvant setting for stage III disease. So I have some understanding as to why these approaches are interesting both clinically and biologically.

Biologically, when we think about how PD-1 or PD-L1 inhibitors work, the idea is that releasing this inhibition allows our immune cells to kill the tumor cells. But this requires antigen presentation by the tumor cells to generate a group of what we call 

Therefore, it makes sense to use immunotherapy in the neoadjuvant setting when the tumor mass is already in place. Imagining a large tumor mass with lots of different tumor clones, this presents us with an opportunity for a large volume of potential antigens to be presented and to generate antigen-specific T cells.

From there, the idea is that you generate this compendium of antigen-specific T cells, almost like a cancer encyclopedia. After you’ve had your treatment, these antigen-specific T cells can then act as your means of immunologic memory, so if a tumor clone does rear its ugly head, you have a specific T cell that can come and take that out and act as a silent and constant killer without necessarily needing repeated dosing.

That is something that has been looked at in tumor models comparing neoadjuvant vs adjuvant immunotherapy approaches. Investigators have looked to see if the generation of T cells is different when given pre- or postoperatively. This has certainly been corroborated in multiple different tumor models and different tumor types, not just long cancer. This has also been studied in breast cancer. So from a biological perspective, it makes a lot of sense.

From a clinical perspective, we talked a little about the downsides of giving a neoadjuvant chemotherapy approach. One was the toxicity of chemotherapy that may limit the ability to get to the definitive stage of surgery or incur certain surgical complications.

That’s not the case with immunotherapy. We have been really positive about the improved tolerability of anti–PD-1 and PD-L1 agents. Thus far in the neoadjuvant studies, which I know the rest of the panel will go through in due course, we haven’t seen any unacceptable safety signals in some of these early studies. When Dr Forde and my study were being developed, there was a concern for pneumonitis. That was the highest on our list. Thankfully, we have not seen that as an unacceptable rate of toxicity that has limited the ability to get to surgery.

None of those things appears to be panning out with immunotherapy—the tolerability question, the ability to get to surgery, the complications of surgery. If anything, it seems to be tentatively easier than giving neoadjuvant chemotherapy. So there are many clinical pluses associated with this approach.

A beautiful explanation. I’m sold. I’m ready to do it on all my patients. Let’s talk about some definitions that keep coming up on these trials: major pathologic response and complete pathologic response. Dr Naidoo, maybe you can walk us through what these are and how they have correlated with outcome, or if they have at all.

 With regard to pathologic complete response, way back when all of us treated breast cancer, a long time ago I’m sure, that was the Holy Grail that many patients were going for. Patients were treated with neoadjuvant chemotherapy. For breast cancer, we know that pathologic complete response correlates with survival outcomes. Was this the same or true for neoadjuvant chemotherapy in lung cancer? Unfortunately, the answer is no.

One of the reasons is that the incidence of a pathologic complete response—basically achieving complete tumor eradiation for patients treated with neoadjuvant chemotherapy—was very low. In the major published studies, this ranged for 0% to 16%, with a median of around 4%. Statistically, the incidence of this is too low to really be used as a real-time end point for neoadjuvant chemotherapy and, therefore, presumably other therapies as well. Therefore, it was not selected as a pathologic end point of interest in some of the early studies. Major pathologic response was based on a seminal study published in 2012. I think this was based on 192 surgically resected lung cancer samples from patients who had undergone neoadjuvant chemotherapy. A pathologist looked at all these samples and identified that this end point, major pathologic response—defined as less than 10% viable tumor cells—was significantly associated with overall survival in this compendium of studies.

That was the data for which, for example, Dr Forde’s study in the 

 used major pathologic response as the pathologic end point of interest. But this was with the understanding that this was based on a clearance of chemotherapy in the setting of, or lack of experience in, neoadjuvant immunotherapy.

Since then we have learned a lot, and the pathologists who were involved in this study and others have really come to the table and try to understand how we could potentially do better from a pathologic end point standpoint.

I would refer you to a great paper in the 

 that was published in 2018. In that study, the 20 patients who were treated on the 

paper with neoadjuvant nivolumab were assessed very carefully for their pathologic features.

What they identified was that certain pathologic features of the regression bed of the surgically resected specimens harbored specific features that could be combined in a score called the immune-related pathologic response criteria score. This included a number of factors, including immune activation, cell death, tissue repair, and a regression-based score. They were assisted by multiple pathologists and found that it was reproducible.

This has been looked at in a couple of different neoadjuvant settings. It has been looked at not just in lung cancer but also in melanoma and other cancers. It looks to be something that is going to be studied into the future. So there is a lot to be done here. I’m pretty sure it’s not going to be where we end.

Yeah. As with everything else, we’re just beginning to scratch the surface of all this. Identifying the right end points takes a lot of scientific, deep work, which you’ve gone over. I’d be curious to see what that end point is at 5, 10 years down the road vs what it is now, and whether these end points will lead to drug approvals.

Role of Neoadjuvant I/O in ES NSCLC

 Who gets this, and for how long? Dr Leal, you’re not considering doing it yet, but let’s say you’ve decided that you’re going to give it routinely. Whom would you give it to, and for how long would you give the drug to them?

 One of the things I want to add, in terms of adjuvant vs early metastatic, is that there are challenges with NGS [next-generation sequencing] testing. We’ll discuss this later on, but this is also something to look at. We know that the rates of NGS testing, even nowadays, across practices vary quite a bit. So yes, we could potentially use this at progression with metastatic disease. Are we treating patients who would never recur? Those are important considerations, and that’s why I’m cautious about giving it a bit more time. I want to see if we are going to see a bigger signal for which we’ll feel more comfortable exposing patients early to prolong therapy with potential adverse effects. Ultimately, they’re very well tolerated. But we do see a lot of adverse effects. I was actually quite impressed to see the data from a trial that was done with 

 patients from a patient perspective. We can look at toxicity grading on clinical trials and look at higher-grade toxicity as how we report toxicities. But with patients who are young and are otherwise going back to work and trying to live their lives as normal as possible, you have to think about long-term toxicity, chronic low-grade toxicity, and how that affects their quality of life. So I do want to see some of that as well.

Whom would I give it to? I definitely think it is appropriate for the stage II/IIIA population. This is the population of higher risk of recurrence, in which we saw the most meaningful benefit. Regarding the duration of therapy, I would try to mirror what the study is doing.

 Dr Zhang, what are your thoughts on this? Would you give it to a similar patient population? Are there people you wouldn’t give it to?

 Based on the study, stage II and III patients benefit the most. But the number also suggests, at least from the hazard ratio, that even stage IB patients get benefit as well. So these are patients for whom I also might consider this treatment if there’s another contraindication or if the patient is able to tolerate therapy.

Back to your question regarding whether we should start now or wait for the time when the patient develops metastatic disease, I would like to add that although there are probably not much clinical data, from a cancer biology perspective, if you let cancer continue to grow there will be clonal evolution. And yes, those patients develop metastatic cancer. They are more likely to respond to osimertinib. But at the same time, they may have some other concurrent mutations that might potentially affect the response from osimertinib. That is the reason why I would like to start to mirror the study.

 How do we put this all together in terms of testing? We’re taught that we need to find these driver mutations in advanced-stage disease. If they have a particular genetic alteration, we want them to go on genotype-directed therapy. Now we have these data in the early stage setting that would suggest that mutations matter in terms of treatment decisions, adjuvantly. How do we put this together, in terms of testing? Should everyone be tested? All early stage patients? If so, how? Is it just 

, or is it all mutations? Dr Salgia, what are you guys thinking about doing, or what are you doing at City of Hope Comprehensive Cancer Center? What’s your take on this?

 That’s an absolutely great question, and I have enjoyed the discussion. I strongly believe that we should test everyone. This is really important if it can help us make a clinical decision, if we can make a potential predictive value. As Dr Zhang said, are there any concomitant mutations? If you have 

 and RB, is that a negative predictor for osimertinib, as compared with 

 mutations without any concomitant mutations? The second thing is, are there nongenetic mechanisms involved as well, such as EMT [epithelial-mesenchymal transitions]?

All those are important, from the perspective of having this predictive value. I’ve been in practice for over 30 years. I’m so excited that we’re seeing long-term survivors. We even have contemplated and are doing survivorship clinics for our patients. How great is that, right? So why not take advantage of all the information we can get and make the best clinical decision. Also, we have research behind the scenes and in front of the scenes too. Learn more so that we can add to the equation. I’m actually very excited about the DFS [disease-free survival] data, no question about it. But I do think we should do NGS testing on everyone.

 Dr Leal, what are your thoughts on this?

 I agree. I would echo those statements, and I really don’t have much to add. You said it very well, Dr Salgia. I definitely agree with testing everyone with NGS testing.

 Dr Naidoo, give us your early experience in Ireland, in Europe. What’s being done there?

 Great question. I think doing NGS testing is the ideal approach. It provides information that you can save in your back pocket and can hopefully act on if needed in the future. In countries where we have a national health service and all this is not necessarily reimbursable by an individual payer, the discussion is a little different.

Certainly here, and in different places in Europe, there are standard panels that are done. The panel is usually completed, but the results are reported only for actionable alterations that are reimbursable therapeutic targets. Usually the data are there, but it’s going to be mined only if there’s something we can do about it.

I might disagree in some way, but I understand the thinking behind that. It doesn’t allow for the bit of the oncologic cookbook that can occur occasionally, where we see an alteration and might have a hunch or phase 1 or phase 2 data. It limits a bit of that and approaches things in a more standardized fashion.

Having said that, if I had my choice, I would get NGS testing done on everybody. I would like to know that information so I can inform my patient and myself.

 Dr Zhang, what is your experience, and what you guys are doing?

 At this moment, we are not routinely doing an NGS panel for early stage lung cancer. However, I am personally in favor of it. There are a couple of reasons why. No. 1 is because of neoadjuvant immunotherapy. We know that for patients with 

 mutations, for example, if they receive immunotherapy followed by osimertinib, they’ll be concerned with immune-mediated toxicity. It’s something that I would like to understand and try to avoid as well.

The second thing is, we don’t know whether other new immunotherapies—for example, other immune checkpoints—will have some value for this 

-mutant non–small cell lung cancer population, especially in the neoadjuvant setting. Having that piece of information will really answer this question. So yes, from my personal perspective, I would strongly recommend this approach.

 This has been a great discussion. Before we move on, I’ve gotten some second-opinion consults recently. I have a patient of resected 

 lung cancer. Should I just be using alectinib or osimertinib for 

? What are people’s thoughts on that? Again, we have a high rate of relapse. We have a known drug with great activity. Is this where we stop and pause, or is there a role for adjuvant targeted therapies for other genotypes in the absence of data?

 I’ll take that. My answer is, show me the data. We can extrapolate in certain circumstances. But that is a very slippery slope, and I certainly wouldn’t advocate for that approach until we have data in the 

 What about ALCHEMIST? Enroll them in the ALCHEMIST trial then. Would that be ideal?

 Yeah. This is an opportunity where we’re testing patients more commonly in the resected phase. We can boost accrual to trials like ALCHEMIST.

 Yeah. And just to call for trials in this space, ctDNA [circulating tumor DNA] at some point may also be leveraged in the adjuvant space to understand monitoring for minimal residual disease. I look forward to seeing some of the data coming out, hopefully in the next year or 2, looking at ctDNA. Scans are scans. We think there are things going on underneath those scans that we could detect in plasma. I don’t know if you guys have experience with this? Dr Salgia?

 Yeah. I love, Dr Levy, what your institution has done. There was a beautiful paper from Johns Hopkins and collaborative studies in terms of the science paper for screening for cancer, looking at it not only from blood biomarkers. I was very happy to see things such as CEA, CA 19-9, and CA 125 come back to fruition, in some sense. But then cell-free DNA as well, and PET scanning at the same time.

I’m always the biggest believer that you should use all the science and tools you must be able to arrive at an answer. We’re actually looking at ctDNA like there’s no tomorrow. No question about that. I hope we can get it approved through the payers—repeated ctDNA analysis. If somebody has a 

 alteration and you give them a specific therapy, and then the 

 goes away, that’s actually a very positive sign. But you have to do it in terms of clinical trials too, of course.

Patient Selection: Adjuvant Therapy and Testing for ES NSCLC

 Let’s move on to a strategy that has worked recently, at least from a disease-free survival [DFS] end point, and that is adjuvant osimertinib. There clearly were very thought-provoking and potentially practice-changing data presented at ASCO [American Society of Clinical Oncology Annual Meeting], and we’ve had updates at ESMO [European Society for Medical Oncology Congress] that are also clinically important.

Dr Naidoo, can you talk to us a little about ADAURA’s study design and outcomes and then highlight some of the updates we found out about at ESMO this year?

 Sure. So ADAURA is causing a stir on every possible airwave. This was a phase 3, double-blind, randomized study. Theoretically, this was a simple study design of adjuvant osimertinib vs placebo in patients with stage IB, II/IIIA 

-mutant non–small cell lung cancer. This was a simple study where just over 600 patients were randomized in equal parts to placebo vs osimertinib. 

The study was stopped early because of efficacy seen on the investigational arm by an independent data-monitoring committee. This was based on the outcome for the disease-free survival end point for patients with stage II to IIIA disease who had been enrolled. It showed a hazard ratio of 0.17, which was highly statistically significant. And then in the overall population, there was a DFS hazard ratio of nearly 0.21. Again, highly significant. 

This equated to an 89% 2-year disease-free survival rate, compared with 53% on placebo. We saw the full publication about this in the 

Controversially, the overall survival outcome was immature—only 4% maturity. Of course, this is because the study was stopped early, which obviously complicates our ability to interpret some of these things and has created quite a war between the believers and nonbelievers regarding the DFS vs OS [overall survival] end points.

Moving on to the second part of your question about ESMO, what is really interesting is Dr [Masahiro] Tsuboi and colleagues presented an abstract at ESMO looking at the sites of recurrence in patients treated as part of ADAURA. This was an exploratory end point within the study.

All patients enrolled in the study were mandated to have some form of brain imaging, but only in the absence of symptoms, and this could have included MRI or CT scans of the brain—whichever was available at that particular institution. Very interestingly, 45 patients had CNS [central nervous system] events. Only 6 CNS events were noted in the osimertinib arm compared with 39 CNS events in the placebo arm. 

The median CNS DFS was not reached. It was 48.2 months for the placebo arm, and a very impressive Kaplan-Meier curve of the CNS DFS presented showed that the osimertinib arm nearly looked parallel, with a hazard ratio of 0.18.

Obviously, as we know, brain metastases is a very common outcome in patients with oncogene-addicted lung cancers—

-mutant lung cancers, and others. Patients can be very symptomatic. This can be a very difficult complication of this disease. It is difficult to manage. The idea that 1 could actually prevent CNS disease from occurring with adjuvant osimertinib is clinically important.

Obviously, there are some issues here in terms of the interpretation of these data. Not every patient was necessarily required to have baseline imaging, and that baseline imaging may have been different. As we know, CTs of the brain may miss about 10% of brain metastases compared with MRIs of the brain.

This may complicate our ability to interpret these results and may sort of say that these results may actually be telling us that osimertinib was treating subclinical disease in some cases. But regardless, a very impressive CNS DFS, and different and interesting data.

 That was a very nice summary, highlighting the pros and cons of both the original presentation and now the ESMO update with this potential theoretical protection of the CNS that we’re seeing but without mandated MRIs.

Dr Leal, what is your take on this? Is DFS enough? Do we need OS? We’ll probably never have it. The study was stopped prematurely. We’ve talked about how challenging and unforgiving lung cancer can be, even with resected lung cancer, and high rates of relapse. How do you put this all together, and what are you doing in your clinic?

 The results presented are very impressive, and there’s room for a lot of optimism. Obviously, we’d love to have OS results. That’s the gold standard. If we’re not going to get that data, I’ll take DFS. For patients who have this disease and for whom we worry about their outcomes, these results are meaningful. They are meaningful enough that if we don’t get OS data I would definitely still take action on it, with approval on DFS end point.

That being said, I have not yet changed my practice. I’ve been waiting to see the data evolve. I’m waiting to see an FDA approval. We have the ALCHEMIST trial at our institution. We are still enrolling patients to ALCHEMIST. There is an EGFR targeted therapy arm. But I think it will come soon, and I am having the discussion with patients. I’m explaining to patients, “These are the results. This is what’s coming.” It’s just a matter of time before I really incorporate it into my clinical practice.

 Dr Zhang, what are your thoughts on this? Does this change things for you?

 Yes. When you look at the initial design of study and their defined hazard ratio, they still ended up with 0.17. This is of dramatic significance. And I agree with Dr Leal. Even if there’s no OS data, I definitely will still prescribe osimertinib once it’s FDA approved for those patients based on these significant, unprecedented results of DFS.

Jun Zhang, MD, PhD, University of Kansas Medical Center

Articles

Takeaways on Treatment of Early-Stage NSCLC

Risk of Pneumonitis Post-Chemoradiation in Stage 3 NSCLC

Considering PD-L1 and EGFR Status With I/O in Stage 3 NSCLC

PACIFIC Trial in Stage 3 NSCLC

Future of Neoadjuvant I/O in ES NSCLC

Neoadjuvant I/O for ES NSCLC: Key Trials from ESMO 2020

Key Trials in Neoadjuvant I/O for ES NSCLC

Role of Neoadjuvant I/O in ES NSCLC

Patient Selection: Adjuvant Therapy and Testing for ES NSCLC

Phase 3 ADAURA Study: EGFR-Mutated NSCLC

Role of Postoperative Radiotherapy in Early Stage NSCLC

Neoadjuvant Chemotherapy in Early-Stage NSCLC

Treatment of Early-Stage NSCLC: Current Standard of Care