This transcript has been edited for clarity. For more episodes, download the Medscape app or subscribe to the podcast on Apple Podcasts, Spotify, or your preferred podcast provider.
Jacob Sands, MD: Hello. I'm Dr Jacob Sands, thoracic medical oncologist at the Dana-Farber Cancer Institute. Welcome to season 2 of Medscape's InDiscussion series on lung cancer. Today, we'll be discussing perioperative management of lung cancer. And I'm excited to introduce my guest, Dr Jamie Chaft, associate attending and director of early-stage lung cancer research at Memorial Sloan-Kettering Cancer Center. Dr Chaft, welcome to InDiscussion.
Jamie E. Chaft, MD: Thank you for having me.
Sands: So first, I'd like to start out with a bit about your journey to this point in your career. Can you tell us a little bit about your path getting into medicine and then oncology and lung cancer, specifically? What drove you to this?
Chaft: Perhaps I'm one of the fairytale doctors that said I knew I was going to be a doctor from the time I was three, but that's really where it came from. With no doctors in the family, it was always a dream. And then I think in medical school, I found cancer in general so interesting because it seemed like we had such a far way to go to be able to get better at treating patients with cancer. I erroneously thought I'd be a gynecologic oncologist because doing surgery and chemo and everything, you could take care of the whole patient. But I've since learned that a steak knife is about as sharp as I can go. So, I wound up in medical oncology and in lung cancer through identification of a tremendous mentor in Dr Mark Kris. And here we are.
Sands: Well, that's a great start. This has been an incredible decade for lung cancer testing and treatments. We're going to focus particularly on perioperative management, and there's a lot to discuss. Let's start with the old data. Adjuvant chemotherapy really became established as the standard of care. There are a handful of trials, some positive, some negative, although the latest meta-analysis really ended up being positive. Can you walk us through that before we get into some of the newer standard-of-care improvements?
Chaft: Sure. I think the historic data has to be taken into a big picture context. All of the foundational data for adjuvant chemotherapy in lung cancer was developed using cisplatin. And cisplatin, as we know — I describe it as a bit of a dinosaur — is a tough drug, and it's a drug that many of our patients today are not fit for. However, all of our foundational data was developed using cisplatin in patients who had completely resected lung cancer. Now, keep in mind that these patients did not universally have PET staging or brain scans before their surgery; some of them had chest x-rays only. However, the data, when taken in sum, really shows that when we add adjuvant cisplatin-based chemotherapy to a complete surgical resection, we improve survival. And that improvement in survival by scope is proportional to stage. So the higher the stage of the resected tumor, the more advantage we will see from adjuvant chemotherapy, anywhere from probably 2%-3% in a resected what is now stage II — historically stage IB — up to probably 10%-15% improvement in overall survival for a resected stage III tumor. So that's the data we come from. Now, those cisplatin-based doublets were historically paired with drugs we don't use anymore. We've extrapolated a bit based on some small subset analysis to use more modern chemotherapies. But this is what we had until just about a year ago.
Sands: That's such a great outline for how far we've come, which is what we're getting to. But as you point out, those prior studies were really with more toxic regimens. The regimens we have today are a bit easier to tolerate. And I'd even highlight the fact that some of the supportive medications we have make cisplatin quite a bit better tolerated than it was back at the time of those studies as well, which in the last decade was a really big advance at that point, a decade ago, in making these more tolerable and in the better tolerated chemotherapy. So, starting with that as a foundation now, let's start with adjuvant (chemotherapy). Now we have IMpower010. If we're talking about non-targeted therapy, we now have the inclusion of atezolizumab; of course, we all saw the KEYNOTE trial that was reported out for adjuvant pembrolizumab, although that's not yet an approved regimen. Can you take us through those in the adjuvant space now, related to immunotherapy in particular?
Chaft: I think we all knew that we could do better than chemotherapy alone in the perioperative setting, looking at how far we've come in the palliative treatment of stage IV disease. And one could argue some of that treatment is appearing more curative in a very small subset of patients. But in the early stage, we really knew that moving these drugs up would make a big difference, and we just had to prove it. And as you know, as part of the ALCHEMIST umbrella, which I'm sure we'll get to, these efforts have been going on for a really long time. It was just that motivation, that boulder was rolling down the hill with immunotherapy that led to these drugs being whisked into the adjuvant setting and one of which, likely too soon, will be approved. So the data is all quite comparable. Patients who have had a complete resection are either offered or mandated to get chemotherapy, depending on the study. Now, the drug we have available in the United States is atezolizumab and, in the atezolizumab IMpower study, every patient was mandated to get cisplatin-based adjuvant therapy, so it was a very fit patient population. However, they didn't have to finish it. Patients without progression were randomized to atezolizumab vs best supportive care, which is the standard of care in that setting. And what we saw was a really impressive improvement in disease-free survival in patients whose tumors had high PD-L1 expression. Now, the pre-specified analysis in this study was actually all PD-L1–positive and then all-comers. And then there are some various analyses based on stage. And while the drug is FDA-approved in all patients with PD-L1–positive tumors, we really thought in this study that the benefit was being driven by those with PD-L1 high expression. The hazard ratios are impressive. The toxicities were fairly minimal. So absolutely, atezolizumab in the US is standard of care for resected PD-L1–positive disease, definitely PD-L1 high. In that study it was 50%. We have a little bit of a shakeup. The PEARLS/KEYNOTE-091 study performed internationally is a study of adjuvant pembrolizumab, where, oddly, patients with the PD-L1 high tumors did not have such a remarkable advantage, whereas all-comers, irrespective of PD-L1 expression, did. So (there was) a hazard ratio of about 0.76 in that study showing an improvement in disease-free survival. Now, the PEARLS study is a little different in that patients were not forced to get cisplatin-based chemotherapy. So it's more of a real-world study, but the data doesn't make a whole lot of sense to me. I don't know … what do you think about it?
Sands: Yes and that's part of why I pair them together, even though pembrolizumab is not currently approved in that setting, although based upon this data, I wouldn't be surprised to see that. In prior pembrolizumab studies, we've certainly seen better outcomes with a higher PD-L1 expression. So, we've seen that paradigm related to pembrolizumab as a drug. But for me, this highlights the uncertainty from PD-L1 expression in some ways. We know that biopsies from multiple sites can have different PD-L1 levels. Is this a matter of heterogeneity and just something that is oddly occurring but describable, or is there something truly different about pembrolizumab in early-stage disease? It's hard to draw any real conclusions from that, but I think then that leads to the question of how do you manage your patients? So, they're not enrolling on a trial. You have someone who's undergone resection already and is now coming into your clinic for evaluation. What do you do as far as biomarker testing and how does that guide your management as to what you would consider the standard of care?
Chaft: I'm a really big believer in comprehensive biomarker testing, irrespective of stage. And I think in the neoadjuvant setting, when we have small biopsies, we're even able to get the data there and we should be able to get it quickly and it really should guide our therapies. We know that patients who have tumors that are driven by the canonical oncogene driver mutations, particularly those associated with never smoking such as ALK, EGFR, ROS — they don't benefit from immunotherapy. We have never seen a robust study demonstrating benefit. Now you'll see some subsets in the adjuvant setting of the PEARLS study, which again are a bit confusing. However, these patients don't benefit and they may have better options in the adjuvant setting for EGFR, absolutely, and in the neoadjuvant setting, perhaps in studies. So I test everyone and we actually do it reflexively, irrespective of stage, with next-generation sequencing, including RNA sequencing if the DNA testing is negative; every patient's tumor is tested for PD-L1 expression. Now, how do we manage these patients? If they have an EGFR mutation, they will absolutely be offered chemotherapy and then osimertinib. And the discussion around atezolizumab is a bit more complicated. As I alluded to earlier, I will certainly recommend it for everyone with resected tumor and high PD-L1 expression and stage II or III and discuss it with PD-L1–positive tumors. Presently, based on the available data, I'm not offering immunotherapy to those with PD-L1–negative tumors. However, I do anticipate this discussion will have to change with the PEARLS data and the anticipated approval of pembrolizumab in the setting.
Sands: So you've highlighted some of the complexity of this space, which is wonderful, because in the real world this is what we're seeing. From what I'm hearing, it sounds like someone with a RET fusion, for example, or some of these nonsmoking-related driver alterations would not be patients that you would be particularly enthusiastically treating with immunotherapy. But if there is high PD-L1 expression, then it is something that you would consider in those patients. Is that correct?
Chaft: It is. I think it's more of a discussion. If someone's a never smoker and has a known RET fusion or ROS1 rearrangement, we really know that these patients don't benefit from immunotherapy, at least in the advanced disease setting. So it's a bit more of a discussion. A lot of these oncogenes can upregulate PD-L1 expression and probably not in the classic sense that sensitizes to checkpoint inhibitors; the biomarkers are dynamic and it's imperfect.
Sands: It's a challenging scenario; it's hard to know exactly what to do. It's certainly a discussion to have, as you point out. So in those with an identified EGFR-sensitizing mutation, we do have the ADAURA data. Can you dive into that a little bit further and how that's guiding your management?
Chaft: Yes. I'll highlight the big differences in the adjuvant osimertinib population vs adjuvant immunotherapy. To date with adjuvant immunotherapy, we've really only seen benefit in the higher stages of disease. So in the 8th edition, that is stages II and III, so tumors that are at least 5 cm or lymph node–positive. ADAURA actually enrolled patients with classic 7th edition stage IB, whereas most of the NCI studies have historically limited it to larger tumors. So that means a benefit was seen in tumors as small as 3 cm. So a population we are not even referred as medical oncologists, most of the time: the 3- to 4-cm or 3- to 5-cm tumors without lymph node involvement. These patients, I think, are a clinically meaningful improvement in disease-free survival. Now, the study randomized patients after chemotherapy and they did not require chemotherapy to osimertinib or placebo for 3 years. This study was only the classic oncogene driver mutations. That's EGFR exon 19 and EGFR L858R point mutations. However, in clinical practice, we're certainly seeing patients with atypical EGFR mutations and considering them for this therapy. And the improvement in disease-free survival was truly phenomenal. We do anticipate to see these curves coming closer together as patients discontinue drug at 3 years, and we are starting to see that in the early follow-up data from the ADAURA study. However, I think it's clinically meaningful to be years without disease recurrence with a drug that is quite well tolerated.
Sands: This was an expected but still surprising improvement. The hazard ratio was really impressive.
Chaft: 0.2 — it's unbelievable.
Sands: Particularly in stage II, stage III. This is the separation of the curves at a level that we don't really tend to see in oncology — very, very impressive. I don't know quite what to do in the scenario of stage IB. It wasn't as clear a separation; it wasn't nearly as impressive. I did notice that in the New England Journal publication, they didn't really separate out that IB, II, and III, like what was done in the presentation, I believe, at ASCO. But what I'm hearing from you is that for the (stage) IBs, there was separation of the curves and you describe that as clinically meaningful. Can you just dive into that a little bit further?
Chaft: I think it depends on whether you consider disease-free survival independent of overall survival an adequate endpoint and, mechanistically, we worry more about these patients getting an adjuvant EGFR inhibitor relapsing after discontinuation compared to an adjuvant checkpoint inhibitor. But it's funny to me to hear that you didn't find the (stage) IBs compelling because the hazard ratio was far superior to what we saw in PEARLS or in IMpower and so in the (stage) IBs, I think the hazard ratio is somewhere around 0.56. It was tremendous. So, clinically significant to me, these curves with the adjuvant EGFR inhibitors are many fold greater an advantage than adjuvant immunotherapy. And we've seen widespread and rapid uptake of adjuvant immunotherapy. And I really think the difference is the concern about relapse and the lack of confidence that these drugs are going to cure. However, I would argue that many years without disease will translate into a cure. And this study was not powered to show cure, and it might not be the one to demonstrate it, but it might. And then perhaps we'll have more doctors believing.
Sands: It will be very interesting to see the overall survival differences. And it's not to say that I don't ever prescribe it, but that's more of a discussion point. For me, I think the big difference in my mind between the targeted therapies and the immunotherapies are actually the tail of the curve. When I discuss immunotherapies with patients, I highlight the fact that actually I don't think it really helps for the majority of patients, but for those who it does help, it is an unbelievable difference or it can be really an extraordinary difference. And actually because of that, I'll admit that the PACIFIC trial, so we're talking stage III — chemotherapy, radiation, and then getting durvalumab after that in the stage III setting (a little outside the scope of this discussion) — but the initial data on that that showed improvement in disease control, the progression-free survival was statistically significant, although the overall survival had not yet matured. And I expressed some skepticism at that time, just because of the responses we see in metastatic disease from immunotherapy, where I think we may have patients who are essentially cured in the metastatic setting of their incurable disease. We'll only know with another decade of data and watching what happens with patients. But seeing that happen, I wondered if maybe in the stage III, we were just seeing these extraordinary outcomes that we see in stage IV, but the maturing of that data has actually led to further support of the durability of these responses. And at 5 years (we see) around a third of patients not having recurrence. This is a different setting, but this is where I think that targeted therapy is a bit different. I'm also a bit skeptical to the durability of 3 years. And one of the things I wonder, as the data matures out, is 3 years going to hold? Are people going to consider potentially continuing the treatment? But then in patients, you have some who truly are cured of their disease. Sometimes it's a little harder to know exactly what to do in that setting. Also, if there's toxicities and patients stop the drug, I don't anticipate that there's going to be ongoing benefit after stopping, whereas the immunotherapy, if there are complications of some kind and you stop treatment, you may have really gotten the full benefit, even though you've had to stop early. And that's some of my thoughts. But please feel free to respond to that as well.
Chaft: That all goes back to mechanism. And I absolutely agree with you. Having designed one of the five adjuvant immunotherapy trials (including) yours, I regret making the duration of therapy a year. I had initially advocated for 6 months and wish I had pushed harder because I agree with you. I don't think patients need a year of adjuvant immunotherapy. The mechanism of the targeted therapies is different. With osimertinib, I'm seeing more pneumonitis in the adjuvant setting than I've ever seen in the palliative setting. I think that's important because you're right, once you stop the drug, the benefit is probably gone. However, for those who tolerate it, they tend to tolerate it really well. And I agree that we're likely going to be seeing patients stay on past 3 years. At this point, my discussion with patients in the clinic is if you are tolerating it and have no evidence of disease recurrence, you will be taking it for at least 3 years. And at 3 years, we will review the updated data and decide what to do. There are two ongoing studies: one is in the United States, and the other, I believe, is solely in Asia. But in the United States, they're studying patients as early as stage I. So we'll get a little more granularity there. There are a lot of purely ground-glass opacities that are stage I EGFR and they are cured by surgery. So I hope they pick the right patients. I haven't seen the subtleties of the trial design. And there's a study ongoing of 5 years and again, were 5 years enough? I don't know.
Sands: So, related to the trials, as you pointed out, you are the lead on the ANVIL trial, which has accrued but is still maturing. So of course I know that you don't yet know data as to what we will see because that's all tightly guarded. But can you speak a bit then to what you've done in ANVIL, as well as some of the other studies in the adjuvant space?
Chaft: ANVIL is the NCI-sponsored version of PEARLS and IMpower, where the study was adjuvant nivolumab vs best supportive care or observation after completion of standard therapy. And the study did allow for post-operative radiotherapy, which has fallen out of favor based on an underpowered European study. However, the ANVIL study, as you point out, fully enrolled very quickly. However, we're still awaiting data. We're waiting for an endpoint to be hit and I wish I had a crystal ball to tell you when that would be, because it's a little sad to put so much heart into it and then be behind the other two studies, but that's where we are. There are other ongoing adjuvant studies. There's a fully accrued international study of adjuvant alectinib vs chemotherapy. And I will admit that I'm a believer in chemotherapy, even in these oncogene-driven patient populations. So I question the study design. The other studies we've seen mostly out of Asia that compared an EGFR inhibitor to chemotherapy were not so compelling. So I'm not sure why the decision was to randomize to chemo vs alectinib. However, it will be really exciting to see the data. I'm sure alectinib will come out on top given the data we have in the metastatic setting. But I think the most exciting stuff we have is the neoadjuvant data, honestly.
Sands: Let's dive into that. So in the neoadjuvant setting, we do have now CheckMate 816 showing the inclusion of nivolumab with chemotherapy with three cycles of treatment prior to surgery. This really showed to me a surprising pathologic complete response. Of course, we're waiting for data to mature, but take us through an overview of that data and your thoughts around it.
Chaft: Patients with resectable EGFR and ALK-negative non–small cell lung cancer were randomized to chemo vs chemo plus nivolumab for three cycles. Endpoints were pathologic and clinical, and it showed a really remarkable 13- or 14-fold increase in pathologic complete response [path CR]. Chemotherapy alone did pathetically in that sense — just 2% path CRs. And then the event-free survival data was equally compelling. I think all of the clinical data trended favorably towards immunotherapy plus chemo. The caveats here, again, despite the authors pushing for this in all-comers, there was a benefit in PD-L1 high greater than that in PD-L1–negative tumors. So that's something to keep in mind. And I think the most important take-home message from preoperative chemo plus nivolumab is if this is something you are going to give in the clinic, make sure that ACTH cortisol and thyroid function are checked preoperatively because the last thing we would want is a endocrine crisis in the operating room upon induction of anesthesia. But beyond that, I'm giving chemo plus nivolumab and I'm excited about it. I can't wait to see the results.
Sands: I love that you just highlighted including ACTH cortisol and thyroid pre-op. This is a new paradigm and so that is really something for people to make sure that is being checked. I think I've seen quite a few surgeons really enthusiastically request chemo plus nivolumab as neoadjuvant treatment, but it's important for them to be aware of this testing that's now really indicated because of the potential toxicities related to immunotherapy. So just to underscore what you said, I think in patients who are going to get neoadjuvant treatment at all, very clearly should be getting immunotherapy, namely nivolumab, at this time with that chemotherapy. What about in patients that you're seeing? Can you take us through some of your thought process on who you might say, okay, given this scenario, go ahead with surgery and then let's consider adjuvant vs giving neoadjuvant, which of course you've said is generally your preference. Are there different scenarios that would lead you to recommend different treatments?
Chaft: Yes. If the tumor is medial and the patient is at risk of post-obstructive infection or if they're having hemoptysis, I will always recommend upfront surgery, or radiation if they're not an upfront surgical candidate. Sometimes it happens. If I'm referred a patient that is considered borderline resectable and I'm asked to shrink the tumor before the operating room, I refer them back and say, put your money down that you can take it out now. Otherwise, they should be getting chemoradiation. So I think we're not there yet about knowing who is absolutely going to respond, despite most folks responding. So patients need to be given the best chance at cure upfront. We shouldn't be dabbling.
Sands: Now, the one caveat as well just to address is that there was about 20% of patients that did not go ahead with the surgery. There was no difference. I think numerically (patients) actually did a little better in the nivolumab arm. So certainly, in giving neoadjuvant treatment, this is a risk and in some cases, particularly when talking about IIIA disease, there can be reasons to not proceed with surgery for bad disease. But what did you make of that data and any concerns about the possibility of not getting surgery?
Chaft: Not at all. Every neoadjuvant study ever done has reported that. And honestly, if you took 100 patients who meet a surgeon or plan to go to the OR, many of them won't go to the OR. A comorbidity will be uncovered or they will be found to be unresectable. So that data does not move me. If they're referred for induction therapy and otherwise appropriate for immunotherapy, then I'm all for the combo.
Sands: And to that point, those may also be the patients who, even if they were getting adjuvant treatment, the disease might be worse in that case. And we certainly don't know neoadjuvant vs adjuvant in these cases.
Chaft: And never will.
Sands: I can't imagine that study being done either, quite frankly, but all right. Well, we have covered a lot of ground. Before we wrap up, Dr Chaft, anything you would like to highlight further as far as what you see in the field either going forward or from the data that we've discussed?
Chaft: I would go back to where I started. I think comprehensive biomarker testing in the early stage is absolutely essential. Soon we'll have four studies resulted and read out with adjuvant immunotherapy. And as you pointed out, it's really a subset, a minority, of these patients who benefit. So I think we'll learn more by doing that and be able to better guide our patients by knowing as much as we can about their tumors prior to just non-discriminately prescribing.
Sands: And I'll just take a moment to highlight the ALCHEMIST trials for those who have gotten surgery and then are being seen for adjuvant treatment. There are the ALCHEMIST trials to certainly consider. And with that, I would like to thank my guest, Dr Chaft. Thank you for joining the discussion today.
Chaft: Thanks for having me.
Sands: Thank you for tuning in. If you have not done so already, take a moment to download the Medscape app to listen and subscribe to this podcast series on lung cancer. This is Dr Jacob Sands for InDiscussion.
Listen to additional seasons of this podcast.
Resources
Osimertinib in Resected EGFR-Mutated Non–Small-Cell Lung Cancer
American Joint Committee on Cancer
Durvalumab After Chemoradiotherapy in Stage III Non–Small-Cell Lung Cancer
Neoadjuvant Nivolumab Plus Chemotherapy in Resectable Lung Cancer
Follow Medscape on Facebook, Twitter, Instagram, and YouTube
Medscape © 2023 WebMD, LLC
Any views expressed above are the author's own and do not necessarily reflect the views of WebMD or Medscape.
Cite this: Standard of Care in Perioperative Management of Lung Cancer: The Past, Present, and Future - Medscape - Mar 16, 2023.
Comments