Dr. Heather Wakelee:
Welcome to The ASCO Post Roundtable Series on New Directions in Nonmetastatic Non–Small Cell Lung Cancer. I'm Dr. Heather Wakelee, Professor of Medicine and Chief of the Division of Oncology at Stanford University and Deputy Director of the Stanford Cancer Institute. Joining me today are two world experts in early-stage, non–small cell lung cancer treatment, and I will allow them to introduce themselves.
Dr. Marina Garassino:
My name is Marina Garassino. I am a Professor of Medicine at the University of Chicago, and I am the Director of the Thoracic Oncology Program.
Dr. Tina Cascone:
Hello, everyone. I'm Tina Cascone, Associate Professor in the Department of Thoracic and Neck Medical Oncology at MD Anderson Cancer Center in Houston. Thank you so much for having me.
Dr. Wakelee:
Thank you, Tina and Marina for joining me today for these discussions. Today we will be discussing the treatment of receptable, nonmetastatic, non–small cell lung cancer with three patient case studies. Our second installment will focus on the treatment of PD-L1 negative, early-stage, non–small cell lung cancer.
Mr. TK is a 75-year-old man who presented to his primary care physician for evaluation of worsening cough and some hemoptysis. His past medical history is notable for hypertension. He smoked for 45 years, worked in construction and is now retired. His ECOG performance status is a 1, and he's still relatively active. He is worked up for this hemoptysis and gets a chest X-ray, which shows that he actually has a large left lung mass, unfortunately. The biopsy shows that this is a squamous cell carcinoma. PET and brain MRI are negative, but there is this large, lower-lobe mass, which is 4.3 cm in size, and he has a 4L lymph node involved. Endobronchial ultrasound did not confirm that there was a disease in 4L or any other station. So the final staging is a T2BN0, though there was that question, and this is per the eighth edition staging.
At a multidisciplinary tumor board, the team felt this was a lesion that was potentially resectable. And so the question is, how would your approach differ based on the PD-L1? So this is a relatively large squamous cell tumor. We are not able to confirm any nodal involvement, though there was that one area that was suspicious. And so as you think about this case, Tina, how would you think about this differently if the PD-L1 was 0%, 15%, 25%, 50%, 85%, you can pick other numbers as cut points, too, but how would you go about approaching this from your perspective?
Dr. Cascone:
Yes, thank you, Heather. This is actually, indeed, a challenging case to discuss based on the availability of data at this moment in time. So for a stage II, deemed resectable, so again, medically operable as you mentioned, based on the performance status and comorbidities and resectable, I think it's important to remember that at this time, we have an option with neoadjuvant chemo immunotherapy if, in this case, there has been, of course, basal molecular profiling, not the presence of a driver. Here we are in the squamous cell, so heavy smoking history. So I think the likelihood to have a driver is very minimal, but again, as a pearl, I think it's a good practice, if we can, to always have the molecular profiling in this setting.
With a PD-L1 at 0%, we have neoadjuvant data based on the CheckMate 816 trial, where definitely the benefit is greater as the PD-L1 expression grows. At the 3-year follow-up, we had 0.29 as a ratio for the population of PD-L1 50% and greater, with the neoadjuvant chemoimmunotherapy and the CheckMate 816.
I also want to remind everybody that with the perioperative trials we have seen, and this is, of course, looking at the data and the confidence interval, we want to take this with caution because we don't have true comparisons between subgroups. So these are exploratory analysis in key clinical subgroups. But we have seen ratios in the order of 0.72, 0.73 with a perioperative approach in the PD-L1–negative population with CheckMate 77T, with Neotorch and toripalimab, but so far we have only learned about the stage III data. So I'll leave that trial on the side for this specific patient. But we have seen with a perioperative approach that perhaps the continuation of therapy might be an option with PD-L1 0%.
In the presence of a PD-L1 positive, and I'll wrap up because I really want to hear what Marina has to say, I think as I mentioned, we have the option of a neoadjuvant, of a perioperative approach, and if we decide to go with upfront surgery followed by adjuvant therapy, then we have the IMpower010 for the patient with PD-L1 positive, depends of course on the regulatory agency and where we are in the world, 1% vs 50%. And then we have of course the PEARLS trial with the pembrolizumab in the adjuvant setting, all-comers regardless of PD-L1 expression. But happy to hear Marina's thoughts here as well.
Dr. Garassino:
Yeah, yeah, you summarized already everything very well. I think that this is the case where the multidisciplinary tumor board can really make the difference because this patient is symptomatic, but he has hemoptysis that can be something that in my opinion, should be considered. He has COPD, he has hypertension, so he's not a patient that is also—if he's active, he is a fragile patient. And so, here, I think that the decision must be between surgery from the beginning or the adjuvant. As we said, I think it's important to have the PD-L1 from the beginning because maybe your decision to go with the neoadjuvant and then clearly can influence. Also the adjuvant can be influenced by the PD-L1 level. Because if you are 0%, you don't have the possibility to do the adjuvant.
So I think that knowing the PD-L1 status, I think for the audience is crucial because you can go in one direction or you can go in the other direction. And then I think that should be very carefully evaluated. Also, all the pulmonary symptoms, and in particular the hemoptysis, because we have always to remember that across all the trials, 20% of the patients who do not receive the surgery for multiple reasons, the depression, but also sometimes the side effects. So I think that the multidisciplinary tumor portal should address how can be important to resect this patient from the beginning.
Dr. Wakelee:
Yeah, no excellent points made by both of you. I mean, I really think about it as if there's PD-L1 expression, I don't think pure adjuvant works. I know that we have the one outlier trial, the KEYNOTE-091. But that just is sort of, it's here and doesn't fit with what we know about pembrolizumab and metastatic or what we know in the KEYNOTE-671 perioperative. So I don't really believe that. And so I'm very nervous about the idea that IO is helpful in a PD-L1–negative patient if it's pure adjuvant. And then when I look at the neoadjuvant and the perioperative trials, the trends are pretty strong in the PD-L1–negative in most of the trials though, not as exciting as when there is PD-L1 expression. So I'm less enthusiastic, but I'm not, not enthusiastic. So if there's a PD-L1 0%, I feel like that neoadjuvant is more likely to be helpful.
But I think, Marina, your points about if this is someone who we know needs surgery early, if we're really worried about that hemoptysis getting worse and potentially leading to a very negative outcome, if surgery's not done first, that kind of is more important than that PD-L1 level in the decision-making. But it becomes an issue, I think about PD-L1 levels from it's 0% or positive, and then I look at that 50% mark just because that high level, and it may be 48%, but we've kind of arbitrarily picked that 50% because we've seen obviously in IMpower010 and the pure adjuvant atezolizumab trial, the greater than 50% are really benefiting quite well. And those with less than that, it's questionable. And with 0%, there's really no benefit. And across the trials where they've done the PD-L1 breakdowns, the higher the level, the more the benefit is pretty consistent across the trials.
So I do get more enthusiastic with those higher levels. But as much as we're talking through this, we always have to think about that patient first. And as you brought up, sometimes you just have to do the surgery first because of the risk. So why don't we keep going with this case. Because it turns out this is a patient with PD-L1 of 0%. So he has started on the KEYNOTE-671 regimen, which is the perioperative pembrolizumab with chemo plus pembrolizumab surgery, and then additional year of pembrolizumab where we did see benefit in that trial, even in the PD-L1 0%, but less than when there is PD-L1 expression. But now we've got a twist on this one.
So, here, this patient has developed hypothyroidism. So an immune toxicity, and we think about that. We think about what if there was lung toxicity that resolved after steroids? What do you do about immune related adverse events in this perioperative setting? And I think we can think about, well, what about if it's before the surgery? And then what about if it's after the surgery in the adjuvant setting, and how do you think about those potentially differently? And how do you rank order of those immune related adverse events? So I'm going to ask you Marina first on that.
Dr. Garassino:
Yeah, so I think that the toxicities are totally different. And so it totally is different if you have an hypothyroidism, but also the rare adrenal insufficiency that you can have with the anti–PD-L1 inhibitors. And it's totally different maybe if you have the lung toxicity. So for the hypothyroidism, we know that it can happen. It's the most frequent side effect, and you replace just the thyroid and you continue your treatment. If you have a lung toxicity, again, I think that what I do is I do the PFTs, I repeat the pulmonary function. I really try to understand how quick the patients recovered from the toxicity. If it is grade 1, grade 2, and then with this information, I decide whether to restart the treatment, again. I would say, to your point about the adjuvant part, that may be. Clearly, looking at this case, it’s PD-L1 0%. So I would say absolutely no. But thinking that if this case was PD-L1 50%, I would consider also the adjuvant part, in any case.
Dr. Cascone:
I think Marina very nicely summarized from a medical management standpoint, we want to be very prompt. I think, probably, that's an important message that we want to send to the community and reach out to each other. Many times we might have seen cases of immune-related toxicity that are very rare. And I think to Marina's point is very important. In this case, we have the usual suspects, the hypothyroidism, the potential pulmonary toxicity. But I think a key message is to have a team of sub-specialists on board with us in managing these patients. The hypothyroidism, the lung toxicity, to me, are very common or very important to understand, first of all how symptomatic our patients are, what's the grade, of course, of the toxicity, repeating the test, as Marina said, have a very low threshold also for CT scan to rule out pneumonitis, discuss the case with your pulmonary colleagues.
And then as far as continuing the treatment, the grade, to me, matters a lot. The grade 2 is that area where we can start probably the steroids vs waiting. It depends really on the clinical scenario. But in a PD-L1 0%, I would be very cautious if we're in the grade 2 setting and continuing. And I think if we are in the adjuvant setting, adding to your point, also the response is something that we want to think about it to tailor. The next step is therapy. So perhaps a patient who has achieved a response to surgery, and has a developed toxicity, we probably should think thoroughly about whether are we gaining any benefit, or are we bringing more financial and clinical toxicity, in this case, for our patients.
Dr. Wakelee:
Right? Yeah. It's always weighing what's the potential benefit, what's the potential harm? And if you got the harm right in your face, because the patient's clearly having an immune-related adverse event, then it behooves us to step back. And what is the potential benefit, right? And if the patient's had a couple of cycles of perioperative, it often makes sense just to get them to surgery then, and not risk losing the kind of key part of the treatment. And in the adjuvant setting, there's a question of: "Are we really doing additional benefit?" It's always intriguing too, I think we've all had those cases in the metastatic setting. The ones who have the really bad immune related adverse events, often their cancer's gone. So they're often having the strongest responses. So we don't want to over-treat. All right, well this case, let's come back to it.
Patient had a very good radiographic response, went to surgery, ended up having a complete pathological response. And then in the perioperative surgical regimen, these trials all had neoadjuvant chemo immune therapy, surgery. And then in the perioperative trials, that adjuvant component, and it was just the CheckMate 816, that's that pure neoadjuvant. We've seen subanalyses with the path CR, those patients often do well. And I guess just a quick answer from both of you. If you've started on a perioperative regimen and the patient has a complete pathological response, do you ever not give that adjuvant treatment?
Dr. Cascone:
I probably will be very fast and say that if the patient has not suffered moderate to major immune-related or overall toxicities, and there is the opportunity to continue treatment from a logistical standpoint for the patient, I tend to continue therapy.
Dr. Garassino:
I really decide with the patients. Last week I had two cases and one patient had a complete pathological response and I explained that we didn't know what was the benefit of the adjuvant component. And the patient was very happy to say, I want to see you in 4 months. I don't want to come to the hospital anymore. There was another patient, 3 days after, and this patient said, "I want to maximize as much as possible my opportunities also, it is tough to find the parking close to the University of Chicago." So I think that depends really on a lot of factors that sometimes I'm not just medical factors also, for example, the presence of a caregiver or these kind of things. I think that can matter. And I think that as researchers, we should try to address as soon as possible this question: Who are the patients that don't really need anything? Or: Who are the patients that maybe really they will need more than one drug and they will need a combination of drugs. So work in progress, I would say.
Dr. Wakelee:
Yeah, really, really important that we keep trying to get those questions addressed because right now we just don't know. Well, thank you both very much. I think we'll just emphasize a few key clinical takeaways. Again, always complete the staging, really important to involve multidisciplinary discussion, ideally as part of a tumor board. The level of PD-L1 can really help in some of those decision-making along with the molecular drivers. So this was a squamous patient, so that was a little bit less relevant, but always can be important to look at as possible. And then we have to be mindful about those immune-related adverse events. Really important to know who you can call if something like that develops in an organ that you don't treat as often. And so that, again, the teamwork of these sorts of approaches being so, so important and still a lot of unanswered questions, but a lot of excitement with the new data that's been coming out.
This brings us to the end of this case. Please see the other segments for further discussion about the latest data in non–small cell lung cancer or visit ascopost.com.
