Management of localized non-small cell lung cancer (NSCLC)

DR LOVE: Welcome to Oncology Today: Post-Conference Lung Cancer edition, a program focused on recently presented data and new approvals in lung cancer. This is medical oncologist Dr Neil Love. For this program, I met with Dr Stephen Liu from Georgetown University Hospital in Washington, DC. To begin, Dr Liu commented on recent advances in localized non-small cell lung cancer.

DR LIU: I think one of the areas where we've seen probably the most excitement has been that early-stage space. And this is really relevant for our patients for a few reasons. One, as we get better at lung cancer screening, we're going to be seeing lung cancer at an earlier stage. So we're going to see a stage shift. We're going to have more patients in earlier stage. And even though we get excited about the advances we've seen in the metastatic setting, we're going to have a bigger impact in the earlier stage. We're going to cure more patients. And so the use of immunotherapy in this setting has come through very quickly. I've been impressed with how quickly we've generated large Phase III data sets. I think the first really was the NADIM trial run by a company in Spain, the GEP, and Dr Mariano Provencio. And what we saw there was the addition of nivolumab to neoadjuvant chemotherapy followed by nivolumab afterwards with improved outcomes.

Here in the US, we have CheckMate 816, and this was using 3 cycles of neoadjuvant nivolumab plus chemotherapy, and we are impressed by how much better the outcomes were. Both pathologic complete response rate, meaning no residual tumor left, but also the translation to event-free survival. And while I think in other tumor types we’re comfortable with pathologic CR as a surrogate, we didn't know what that meant in lung cancer. But with CheckMate 816, we see that PCR is a valid surrogate. Here 3 cycles of neoadjuvant nivolumab. This is not what we usually do in the metastatic setting. Usually we're giving a lot more immunotherapy. And so the question is, are those 3 cycles enough? Should we give more afterwards? And so we saw KEYNOTE-671, pembrolizumab given 4 cycles before and given after immunotherapy to complete a total of 1 year. And what we saw with that data, KEYNOTE-671, is, yes, we increase the likelihood of pathologic CR, yes, we improve event-free survival, but the big difference here, and what's most impactful, is that led to an improvement in overall survival. That was really the first study that led to people living longer.

Shortly on the heels of KEYNOTE-671, we saw the AEGEAN study with durvalumab and CheckMate 77T with nivolumab. And we see updates for all of these trials at pretty much every meeting. And there are so many different data sets, but I think all of these are adding more pieces to the puzzle. And one of the data points that we saw from Dr Tina Cascone was really the benefit seems to be even greater in people with more advanced disease. Some were wondering, well should we really consider this approach in someone with multi-station N2, someone that's a little bit more advanced, a little higher likelihood of recurrence? Or is this really for people with single-station N2, earlier stage? And we saw actually the benefit was even more pronounced in those that were more advanced. So should we start pushing the boundaries of surgery? But clearly the big question we have, do we need the continued immunotherapy afterwards? Because we're continuing more checkpoint inhibitors, which is increasing exposure, which is keeping people in our infusion units and really increasing the financial burden of this approach. Is it worth it? Are we seeing more benefit? And that's really a question that remains unanswered.

DR LOVE: So, and you mentioned the NeoCOAST study looking at durvalumab that was also updated at World Lung. When you look at the trials looking at these 3 IOs first of all, any way to differentiate them?

DR LIU: So if we're talking about pembrolizumab, durvalumab and nivolumab across these 3 randomized Phase III studies, they all show pretty similar outcomes so far and increased pathologic complete response rate and improvement in event-free survival and all to similar degrees. Right now, as of the time that we're talking, the only one that has a statistically significant improvement in overall survival is pembrolizumab in KEYNOTE-671. Will AEGEAN with durvalumab get there? Will CheckMate 77T with nivolumab get there? I suspect that it will. I don't see any real reason why these studies are different. And if there are differences in the outcomes, I think they are more a function of the trial design than the actual drugs. Now 1 minor difference, and I don't know if this really is going to make a big impact, one minor difference is there is a little bit less pembrolizumab given in KEYNOTE-671 than durvalumab and nivolumab. Because there you continue 1 year of treatment after surgery, but with pembrolizumab it was 1 year counting the neoadjuvant. So it's a slightly shorter course but to me, it's overall survival. So if any of the other studies don't hit overall survival, to me they're kind of off the table. Even though I don't think fundamentally there's a difference, if you've got a regimen that's got an OS benefit, I think you've got to go with it.

DR LOVE: And this question of whether or not it's actually even necessary to use postop IO, is that being addressed in ongoing trials?

DR LIU: Yeah, so we attempted to look at that. Dr Patrick Forde presented these data at WCLC in San Diego. And what was nice about this dataset is it just so happens we have a neoadjuvant only Phase III trial, and we have a perioperative Phase III trial both with the same PD-1 inhibitor nivolumab. So you could make a comparison at the patient level of CheckMate 816 to CheckMate 77T, both getting nivolumab before surgery. And if you're starting at the point of surgery, does giving more treatment provide more benefit? Now statistically, Neil, this is a little tricky. And there are some people that didn't like this kind of analysis because you do have to make a lot of adjustments here. The data set here in this comparison of 816 and 77T is not the entire population. They only include people that get to surgery and in the CheckMate 77T they only include people that get to surgery and had 1 dose of nivolumab afterwards. So you're cutting out a little less than half of the patients. And so now when you start looking at smaller subsets, you get a little bit into the statistics of small numbers and it's a little hard to make broad conclusions. Overall, the trends that they saw were that there definitely seemed to be greater benefit with more immunotherapy with the adjuvant nivolumab in PD-L1 negative and in people that did not get a pathologic complete response rate. And the people that had a path CR and the people that had PD-L1 positive cancers, the difference didn't seem to be so striking, but this is not a prospective comparative trial. I think that's what we need.

DR LOVE: Are there trials being done?

DR LIU: There are plans to determine what the contribution of the adjuvant portion is and whether we should escalate treatment if we don't achieve a pathologic CR, but they're still in the planning stages right now.

DR LOVE: I'm trying to think back to the preoperative breast trials, and I think they just all gave it before and after. I don't think any of them did, I think that's the problem. They did that trial and once they did that trial but in breast, I don't think they ever did that in breast, they always did postop adjuvant. But anyhow, really interesting and obviously a lot of clinical implications to be worked out there.

Let's talk a little bit about another study that was presented there, the NeoCOAST-2, also at the World Lung meeting looking at durvalumab combined with a bunch of other agents oleclumab, monalizumab, volrustomig, dato-DXd and durvalumab. And some of these, I think, are being looked at in the post chemoradiation setting.

Anyhow what about what they're reporting here at least in terms of I guess path CR?

DR LIU: So you're right. These are being studied post radiation that's in the COAST trial and NeoCOAST-2 trial. We've got our different immunomodulatory agents and an ADC thrown in there. And we've got so many new molecules and so many interesting targets that make a lot of sense on paper. We need to figure out which ones should we focus on and which ones should we invest in. And that's really what NeoCOAST-2 is. It's sort of a signal finding study where we have different arms but there's not really a comparative, there's not really a control arm. And so it's really just sort of seeing what are the signals here? And we didn't see the data from a TIGIT molecule. What we saw was oleclumab, which is anti-CD73, so you're talking about the adenosine pathway there. We saw monalizumab, which is NKG2A, a different checkpoint and we saw datopotamab DXd, a TROP2 antibody-drug conjugate. And these were given in the neoadjuvant setting. And what we saw here was really pathologic complete response and major pathologic response rates. And the takeaway point from Dr Tina Cascone, when she presented these data at WCLC, were that the pathologic CR rate with datopotamab/deruxtecan was extremely impressive. Here we see a path CR rate of 34%, 1 out of 3, with a pathologic complete response rate, two-thirds, 66%, with a major pathologic response. And so these were very impressive data. If we're trying to build on these neoadjuvant settings and push more people to path CR, this is really an appealing strategy.

DR LOVE: And the idea of, this is ADC plus chemo plus IO?

DR LIU: Correct. And what we saw here in terms of safety, as you expect, you're going to see more toxicity with more drugs. In that arm, with the datopotamab/deruxtecan, the datopotamab was not given in the adjuvant setting, only the durvalumab. In the other immunotherapy combinations, you continue both agents but there, once you went to surgery, afterwards it was just durva with datopotamab just part of the induction pushing people to path CR. And is that going to translate to improvements in event-free survival to improvements in OS? We need controlled trials, but this is pretty promising data.

DR LOVE: Just trying to process whether I've seen trials that combine an ADC, chemo and IO?

DR LIU: We've seen them in the Stage IV setting. They have not been randomized trials yet. And so the data that have been presented have been pretty small safety finding signals looking at datopotamab with pembro, datopotamab with durva. We've seen sacituzumab as well. We've seen these combinations in early studies but, you're correct, we have not seen randomized controlled data in that front-line setting yet.

First-line treatment of advanced NSCLC

DR LOVE: Really interesting. Alright, well let's talk a little bit more about first-line therapy now in terms of immunotherapy. And again at the World Lung we saw data on cemiplimab monotherapy in patients with PD-1 greater than 1. The EMPOWER-Lung 1 study. Any comments?

DR LIU: These data continue to be pretty impressive. And it is very difficult to compare across trials, but the long-term survival numbers with cemiplimab really do hold up. The Phase III EMPOWER-Lung 1 study randomized patients to front-line cemiplimab versus chemo and at 5 years we see the OS benefit persists, hazard ratio of 0.585. Actually if you're comparing across trials, which we're trained not to do, it's a little better than the pembro data at 5 years. But I would say these data are comparable. We're seeing long-term survivors with cemiplimab. And if there was any question as to whether cemiplimab would hold up over time, I think these data put that to rest. This is a solid checkpoint inhibitor. We are seeing long-term survivors whether it's given by itself or whether it's given in combination with chemotherapy. And so this is another very good option. I think that what we're hoping to do though is build on these single-agent regimens.

DR LOVE: I always use the example of the PACIFIC trial in terms of which IO to use. Because particularly when I talk outside of lung, I talk about the fact that pembro is really, was the main drug that came out in terms of metastatic disease and they looked at it in locally advanced nonresectable, but it didn't look as good as durva so now everybody uses durva even though they sort of think they're the same thing. And I guess that would apply here in terms of cemiplimab. And I'm assuming you agree with that. Of course, the thing we always ask people and get interesting answers is, does that mean if there was a biosimilar that was half the cost or whatever, if the drug itself was half the cost, would you be okay using it?

DR LIU: I'd be okay with anything that has the data. But again, the data here are long-term survival. It's not response rate, it's not PFS. We're getting long-term survivors. These drugs are different. They have different epitopes. They have different binding capacities, binding constants. So there are some differences. It's just that I don't think at this point, you know if you or I are in clinic, I don't think we can really leverage those differences in any meaningful way to our patients. And so I do feel comfortable with any and all — if there were a biosimilar, I would still want to see the data that achieves that. But if I have a patient that I prescribe pembrolizumab and their copay or the insurance denies it, but they allow cemiplimab, I have absolutely no problem switching from one to the other.

DR LOVE: I'm hearing that story a lot with a lot of different agents nowadays in oncology. Okay, HARMONi-2. I want to get your take. You tell me.

DR LIU: HARMONi-2 we've been waiting to see these data. This is, I think, the biggest story in immunotherapy for lung cancer of the year. This is a Phase III study looking at ivonescimab (AK112) versus the king, pembrolizumab, in the front-line setting for PD-L1 positive lung cancer. And these were presented by the IASLC president, Caicun Zhou. And what we saw in this HARMONi-2 study, we knew from press release it was a positive study, and the wording was very enticing, what we saw was, I think, better than any of us had imagined. They showed a PFS and the PFS with ivonescimab in that PD-L1 positive group, 11.1 months. That's phenomenal. Compared to 5.8 months with pembro that is really consistent with the original pembro data. And so the pembro arm did not underperform in the PD-L1 positive group. We had a doubling of the PFS with ivonescimab, a hazard ratio of 0.51. There were more toxicities, and it has to do with what ivonescimab is. Ivonescimab is a different molecule, a next-gen molecule. This is not a PD-L1 inhibitor. This is a bispecific antibody. It binds to PD-1, but it also binds to VEGF. And so you will see VEGF toxicity and that's what we saw. We saw proteinuria. You can see hypertension. You can see the traditional VEGF toxicities. It didn't lead to more discontinuations. In fact, the discontinuation rate from treatment-related AEs was lower with ivonescimab. But what it did do was significantly improve efficacy and this was, I think, greater than any of us thought. This is a molecule that we think is greater than the sum of its parts. It's not like adding bevacizumab or ramucirumab to a checkpoint inhibitor. Because of the bispecific nature here, you're looking at cooperative binding. And so when the molecule binds to VEGF, it's more likely to bind to PD-1 and vice versa and hold on tighter. And so it really is fundamentally different and these data I think were phenomenal.

DR LOVE: So I guess a lot of questions about this. John Heymach was the perfect person to do the discussion, and I guess one simple question is, that you've sort of already kind of addressed it, but do we really know if there was a randomized trial between this agent and pembro and bev that really would be positive trial?

DR LIU: That's a good question. We've seen some benefit with angiogenesis inhibitors targeting VEGF like bevacizumab, like ramucirumab. There have been some signals. And what those signals have shown was really some trend towards improvement in PFS, but it seemed to be transient. And the question is, would this be different? Here we saw tremendous improvement in PFS, but some questions remain. What is going to be the impact on overall survival? And, I think just as importantly, can we replicate this? These data were sponsored by a Chinese company. These data will be replicated in the US. And so in a western population and in a different regulatory system where trial conduct is a little different, will this hold up? And so these data do need to be replicated to get access to it. The data are extremely encouraging if these numbers hold up in another randomized study in a different population. And if it translates to some improvement in OS, you're really looking at a new king, you're looking at a new paradigm. And then the question is, everywhere that we're using not just pembro, but atezo, durva, should we be using this drug instead? And so this might be the whole new chapter. And there are other similar molecules that are also in development, so my hope is that we're on the cusp of a whole new era with much better outcomes.

DR LOVE: So I think that trial's going to set the Olympic record for accrual. It'll probably totally accrue in about 2 days. But when do you think it'll be open?

DR LIU: Some of the studies are already open.

DR LOVE: Really?

DR LIU: And they're looking at some different combinations, so they've moved pretty quickly, yeah.

DR LOVE: That's awesome. That's going to be a really exciting thing to look at. Alright, so we have talked a lot about sacituzumab, particularly in breast cancer, and we heard about this other TROP2 ADC sacituzumab/tirumotecan, that's about as good as I can do that one, but we did see data on that at ASCO. Any thoughts about that?

DR LIU: So they're calling this one sac-TMT for short.

DR LOVE: Okay, I like that. That's better. Much better.

DR LIU: These are all so hard to say.

DR LOVE: I like it too. Sac — I like it though. It's got a little punch to it.

DR LIU: Got a little punch to it. So sac-TMT — so they compared this in the OptiTROP-Lung study. And this was a randomized study but not a randomized controlled study. This was looking at sac-TMT, which is a novel TROP2 ADC, and they're combining it with a different checkpoint inhibitor at some different doses. And what we saw was pretty impressive. Given with that PD-L1 inhibitor, the response rate, when we looked at the higher dose, was 78% with progression-free survival not yet reached. We're seeing extremely high response rates with sac-TMT and TROP PD-1. Now the problem is there is not a control arm of IO alone. And so we do need randomized data, but in terms of finding what the right dose is, clear difference here between the 2 doses of KL-A167, which is the PD-L1 inhibitor. So there is going to be a randomized trial looking at pembro alone or pembro plus sac-TMT that's being called the TroFuse-007 study and so we'll see what those results show. But I think the question is, is giving an antibody-drug conjugate targeting TROP2, is that going to provide more benefit? Is that going to be worth the toxicity? And is this going to be an additive benefit, which is okay, or a synergistic benefit, which is much more preferred?

DR LOVE: What about tolerability? Sacituzumab, we hear about myelosuppression, GI, diarrhea, et cetera, what about here?

DR LIU: Interestingly enough even though the name is similar, more similar to sacituzumab/govitecan, the toxicities are much more similar to datopotamab/deruxtecan.

DR LOVE: Really?

DR LIU: And so we're seeing more stomatitis. We do see myelosuppression. And so there are real toxicities. The ILD rate was pretty low, not a huge study, so that'll be one to watch. But we are seeing stomatitis with this compound. And so that is a unique toxicity for us in the lung world. I know they're a little more familiar with it in the breast cancer world and some other tumor types. But stomatitis was a not infrequent toxicity we saw with sac-TMT.

DR LOVE: Yeah, the breast people say that data responds the same to mucositis as other stuff, everolimus, they do the steroid mouthwash, et cetera. How problematic has it been for you using dato?

DR LIU: It's been so far pretty low-grade in all of the reports, and I think the key there, as you mentioned, is to get out in front of it. Ideally, you're being proactive not reactive, but those strategies like dexamethasone mouth rinse I think are going to be the key, early recognition really important and questions, will ice chips help reduce effusion? Can we do sort of inexpensive things to try to improve tolerability?

DR LOVE: So speaking of dato, what about the TROPION-Lung01 study presented at World Lung also comparing dato to docetaxel?

DR LIU: So this was an important update but one that we knew was coming. We know this study missed its overall survival. TROPION-Lung01 was a randomized Phase III study comparing datopotamab deruxtecan versus docetaxel. This was in patients who already had platinum doublet chemotherapy and immunotherapy in non-small cell lung cancer, and we saw true enough hazard ratio right around 1, KM curves that overlap. But we already knew that the benefit for this drug was in non-squamous lung cancer, and we saw that when Dr Aaron Lisberg presented these data at ESMO 2023. That in non-squam there's a signal and in squam there wasn't. And we saw that held true with OS as well. In non-squam, hazard ratio of 0.84, median 14.6 versus 12.3 and in squam the hazard ratio is on the wrong side of 1, 1.32, the chemotherapy arm much better. This drug has no business in squamous lung cancer. Really should only be used in non-squamous lung cancer. And the question that we have today is, is this enough to lead to approval of the drug?

DR LOVE: What do you think?

DR LIU: I think, yes. It met one of its dual primary endpoints and what we don't talk about, I think enough, is that datopotamab had a much higher response rate and response is how you get patients to feel better. In the second-line setting when patients are progressing, if they're symptomatic from their tumor, the only way you can improve someone who is symptomatic is with a response. And so I think the response rate is a valid surrogate. Docetaxel as we know it's a tough drug. We've been trying to get rid of it for years. It causes a lot of toxicity. Patients do not feel well. Here if we can get higher response rates with a better progression-free survival, I see value in that. The survival difference is not statistically significant here, but I do think that it's enough that I want this drug available as an option to our patients. And my hope is that it will be, but time will tell.

DR LOVE: Was the squamous/non-squamous a prespecified randomization?

DR LIU: So we didn't expect the differences to be this strong in squam versus nonsquam. And so this was not an anticipated difference here and to be honest, Neil, we don't exactly know why. When we look at expression there seemed to be some differences here, but this is not something that we expected. And also what's interesting is we do not see this signal at all with sacituzumab/govitecan. And so some thought, well maybe it's TROP2 expression of the squam, but that doesn't explain any EVOKE data we saw with sacituzumab where there was no difference in squam and non-squam. So we still don't have a great reason as to why we're seeing such a difference here. The squam population is smaller so a lot of the hypotheses we have in terms of biomarkers and other correlated studies will be a little hard to study because we don't have quite as many patients in that squam group, but this was not really something we anticipated.

DR LOVE: Yeah, it seems like, it wouldn't be surprising if they did require another randomized trial just for nonsquamous, but I think maybe what's hanging over here is what you said, how much people hate docetaxel. What do you think about tolerability of the 2 options?

DR LIU: That's a great question. And you know tolerability is really in the eye of the beholder. It's not a nontoxic drug. It is still a drug with chemotherapy like toxicities. You see myelosuppression and you see stomatitis. There's a risk of ILD. And so it's not without its own toxicities. I think overall I feel it is going to be a better tolerated drug. But the nice thing about having it as an option, if it were to be approved, is if it is not well tolerated you could switch to the other one and vice versa. And so I don't object to having it here. I don't think it's worse than docetaxel in any way. I do think it will be better tolerated, but it won't be the same for every patient. And when you think of a toxicity like mucositis, I think there's a lot of range in terms of how patients can tolerate that. And for some it's really not an issue and for others it can be a real detriment to quality of life. And so I think that watching this closely, being proactive and pivoting and making adjustments is really important for this.

DR LOVE: Yeah, that's a great point. And speaking of TROP2, we also saw a really interesting paper presented at World Lung. I even told the breast people about this also and they know about this, membrane ratio thing. I mean, they've been so frustrated that you can't seem to correlate TROP2 levels with efficacy, which seems like it should be the case. Anyhow, how are they doing this new way of looking at TROP2?

DR LIU: It's really clever. And I'm glad this was one of the presidential — this was presented by Dr Marina Garassino in the presidential plenary. And some people wondered why is this a presidential platform and I really think this is a landmark study in a whole new direction. First, it gives a lot of insight to the fact that these companies have been working on biomarkers really diligently. When you look at TROP2 IHC we don't see those differences. A lot of people thought, well they just stopped there. But clearly that's not the case. They have been hard at work looking for some way to isolate who is benefiting from this group. And this is a very different paradigm. We don't have anything like this in lung cancer. What you do is you take a light microscopy slide, an H&E slide, you do immunohistochemistry with this TROP2 assay but you're not just reading 2 plus 3 plus there. Here now you are making a copy, a digital image of that slide, uploading it to a central place where a deep learning machine, AI has sort of helped identify how to best read this study, and you're looking at where the expression is. You're looking at the optical density based on antibody uptake in the membrane versus the cytoplasm. And what this should measure is really how much of that receptor is being internalized. Because having receptor on the outside to bind to the drug is not enough. What you need is binding and internalization. And so once it's internalized, because this is a cleavable linker, then the payload is released, not before. For noncleavable linkers, for PH-induced linkers, internalization is nice, but it's not necessary. And so this type of biomarker will not apply to all ADCs.

For example, I would not expect this approach to be useful for sacituzumab/govitecan, whereas for datopotamab, because it really requires internalization, this assay should be a good look at internalization. But when you get that biomarker readout, you actually want a low number. The number they're looking at is the membrane optical density over membrane plus cytoplasm and you want a number that's much less than 0 showing that there's a lot more receptor in the cytoplasm than just on the receptor, just on the surface. And if you're internalizing the receptor, you're going to internalize the drug. If you're internalizing the drug, you're going to release your payload and you're going to have efficacy. And what they saw was exactly what you want to see. In the biomarker positive group, docetaxel compared to datopotamab/deruxtecan, dato-DXd did much better, 7.2-month PFS versus 4.1, hazard ratio 0.52, KM curve that stands way out. But if I was thinking of what I would want to see here, I would want to see that this biomarker had nothing to do at all with docetaxel, had no value for docetaxel, because docetaxel doesn’t care if TROP2 is being internalized or not. And what we saw with docetaxel, if you were biomarker high or low, biomarker positive or negative, the PFS was the same, 4 months with or without. No difference. That’s exactly what you want to see. You want to see this biomarker only matters for datopotamab. And if you are biomarker positive, PFS 7.2. Biomarker negative, PFS was 4.0. That’s in those with no actionable genomic alterations. And actually, in the overall population, it was an even more pronounced difference.

DR LOVE: So one of my, on my waiting list of things to watch, I saw that there’s a presidential presentation on ESMO about digital imaging and they had 6 billion images, and they were using AI. Is that kind of like what they’re doing here? Is this sort of a digital AI kind of thing?

DR LIU: It is. So it is being uploaded and they are using AI to help sort of process these. And you bring up a very hot topic, a sensitive point, an interesting point here because I think you’re picking up, this is a little different. This is not IHC that you and I can do in the lab, that we can have a lab-derived test or an LDT. We can’t do this. This goes into this black box with this AI algorithm that spits out this number. And we can’t validate it. You and I, we can’t make an LDT. We have to use this central one. And the question that I asked actually to some of the former FDA employees, how does the FDA view sort of these AI algorithms? And I was a little surprised to hear there are already many AI-based biomarkers that are approved by the FDA. And so they’re actually out in front of this. I think they might use slightly different terminology, but it’s already being done.

However, this is not something we’ll be able to replicate. This is centrally validated. So this is something the FDA has to pay close attention to. I do think this is probably where we’re going. And biomarkers just positive or negative for IHC are probably a little too simplistic. Going forward, it probably is going to be more AI-based algorithms, using digital slides. But one could predict a world where you’re looking not just at expression on tumor cells, but you’re really factoring in the microenvironment, the spatial kinetics, spatial relationships between stroma cells and tumor cells, looking at infiltrating lymphocytes. And we can generate a lot more power. So this is probably the future, and I think we’re going to need to get used to it.

Targeted therapy for patients with NSCLC and actionable genomic alterations

DR LOVE: Yeah, really interesting and exciting as well. Let’s talk a little bit about targeted therapies. Just to note that osimertinib was approved following the ASCO meeting, obviously, where the data was presented after locally advanced disease. And I’m just kind of curious. What other actionable mutations will you, if you could access the drug, substitute for durva nowadays?

DR LIU: Yeah. So I can always count on you to ask the tough questions, and this is one that comes up a lot. We know that durvalumab has been a real difference maker, but we also know that immunotherapy tends to be more effective for smoking-related cancers. And for something like an EGFR mutant lung cancer, we wouldn’t expect to see as much benefit. There’s been a little bit of back and forth in terms of the retrospective data and how much benefit is derived. But the data for LAURA were off the charts. And so osimertinib in that setting is our standard.

What else can we use there? Would we do the same approach for ALK, for RET, for MET, for BRAF? Well when I think of what makes osimertinib so impactful here, well-tolerated drug that you can take for a very long time, highly potent and effective and CNS penetrant. And so what else checks those boxes? To me, ALK, absolutely. And so would I use alectinib in this setting? To be honest, Neil, I’ve been doing it for years. And so yes, alectinib in this setting, I think is absolutely appropriate. Would I use it for RET? Absolutely. I think if anything, ALK and RET check those boxes more than osimertinib. I think those drugs are better tolerated, better in the CNS and they work for all of them. So ALK, RET, no problem.

ROS1, this is tricky because you have drugs that are CNS active. We have a population where the benefit from IO is not quite as much. However, you have drugs that right now are probably not quite as tolerable as a drug like alectinib or osimertinib or selpercatinib or pralsetinib. And so I think that for ROS1, I’m not quite as happy with the tolerability, especially if we’re giving these drugs for long periods of time. For BRAF, for MET exon 14, those are populations where we can get benefit from immunotherapy. And again, drugs that are not always the easiest to take, especially for long periods of time. So those ones, to me, don’t make the cut until I see some data. But we will see data.

DR LOVE: So a medical oncologist with locally advanced disease in ALK comes to you and says, how about give me lorlatinib.

DR LIU: Worth the consideration, yes. But while lorlatinib is clearly, I think, the most effective ALK inhibitor that’s approved right now, I do not think it’s the best tolerated. And in the setting where we have very low burden of disease and where we’re going to be taking a drug for a long time, I think the long-term tolerability of alectinib is probably a lot more appealing than lorlatinib. It’s not the wrong answer. And neither of them are approved. And so I’m standing on shaky ground to start with. But I would probably reach for alectinib right now before lorlatinib in that setting.

DR LOVE: Yeah, I’m imagining kind of a sick patient who is still recovering from chemoradiation and then they get those CNS personality things on top of that. That does not sound like a good time there.

Alright. Let’s talk a little bit more about other targeted therapies that have been presented recently. So at ASCO, we saw some data on repo, repotrectinib in patients with ROS1. You were just talking about ROS1. Any thoughts?

DR LIU: Phenomenal drug. The response rates here, about 80% in that front-line setting. And waterfall plots, these are dream waterfall plots. Everything is south of 0. We’re seeing high response, very CNS active. The intracranial progression-free survival rate at a year, 91%. These are phenomenal numbers. Really, this is a triumph of medicinal chemistry. We develop these drugs to overcome resistance mutations and then to prevent resistance. This is what happens in this day and age. We see very high response rates. Overall, 79%. But in that front-line setting, really north of 80%. To have that kind of confidence where someone has a ROS1 fusion lung cancer, you give a drug, you know you’re going to have a RECIST response over 80% of the time. I think it’s phenomenal.

So these are very active drugs. The durability is very impressive. But there is toxicity. The toxicity here is a shame because it’s not really from targeting ROS1. We can target ROS1 without too much consequence. It’s the fact that we’re also targeting NTRK. And repotrectinib targets ROS1 but just like entrectinib, it’s also active in NTRK. And while that’s good if you have an NTRK fueled tumor. If you don’t, you’re targeting NTRK with no benefit. And targeting NTRK means you’re getting NTRK toxicities. And #1 of which is ataxia, dizziness, gait instability and impact on proprioception. We also see weight gain. You can see the withdrawal pain syndrome. You can see pretty significant effects with the drug. And so with this in mind, the way this drug is given I think is very smart. It starts at 160 mg once a day, you wait and then if well-tolerated, you can then escalate to twice a day. And so you don’t start at the full dose. You ramp up. Similar to kind of brigatinib.

DR LOVE: So what’s your preferred first-line agent for ROS1?

DR LIU: Today, it’s this. It’s always a clinical trial. And I think there are very exciting ROS1 inhibitors like the NVL-520 agent that was presented also at ESMO. But right now, off study, repotrectinib is my number 1.

DR LOVE: Okay. KRAS G12C. We have more data that came out at ASCO looking at adagrasib, the KRYSTAL-12 study. Any thoughts?

DR LIU: So adagrasib, another KRAS G12C inhibitor. What we saw in KRYSTAL-12, that Phase III study, if we look at progression-free survival, we saw an improvement, 5.5 months versus 3.8 months. Hazard ratio of 0.58. And I’ve got to say the response rates were much better with adagrasib, 32% versus 9%. Going into this study, really, we had sotorasib as the first. We were very impressed with it. Then when we saw the randomized CodeBreaK data, it didn’t seem quite as impressive as we like. And it lost a little bit of its luster. And then all of a sudden, adagrasib looked like maybe this was a better drug. The design to work in the brain, we’re going to see more potency. And we saw a lot of people shift towards adagrasib.

And what I think we saw with KRYSTAL-12 is that these drugs are probably much more alike than different. And the outcomes here were very reminiscent, to me, of the sotorasib data. Coming in, I thought maybe adagrasib is quite a bit better. Here, they look very similar. You do see higher response rates. You do see a better progression-free survival. But you’re not looking at those wide curves that you’re seeing really with ALK or EGFR inhibitors. You’re seeing a modest improvement from a first-generation drug. I think both of them in my mind are preferable to docetaxel. But they may not be the homeruns that we had hoped that they were.

DR LOVE: Do they have enough CNS activity that a patient who was asymptomatic, had multiple small brain mets, you would try it rather than radiation?

DR LIU: It depends on the size and location, so it’s all in the details. If someone had asymptomatic subcentimeter, yes. The answer is yes. But I’m watching closely. And I’m not taking it for granted that it will work. And while you’re looking at intracranial response rates, I think the more valid there in someone that’s asymptomatic is intracranial progression-free survival, intracranial control rates. And those can be quite long with an agent like this. If someone had large symptomatic brain mets, something in the brainstem, something that was in a dangerous location, there I’m probably relying on radiation up front.

Recent FDA approvals for NSCLC

DR LOVE: So a couple of press releases I’m curious your thoughts about. First, in August, we saw the approval of lazertinib/amivantamab first-line. Now we have 3 great options. Any thoughts? And how are you sorting through those?

DR LIU: Yeah, this is a tough spot. I think we’ll be arguing about the rank and file for these 3 regimens for the next few years. The approval of lazertinib and amivantamab is no surprise. MARIPOSA was a positive study, just as positive as FLAURA-2. And so we now have a third option. This brings lazertinib into the market. That’s a third-gen TKI that’s already been approved in South Korea. Now it’s approved in the US. But now we have 3 regimens.

Ami/lazer clearly superior to osimertinib in terms of progression-free survival, yes. Longer PFS. Performs well in these poor prognostic subgroups. But it comes with the cost of more toxicity and infusions. And so this is the challenge. Who really needs this extra benefit? When we look at a drug like osimertinib, we appreciate that the median progression-free survival of 18.9 months in FLAURA is not good enough. And a year and a half is not very long. However, it’s a very well-tolerated drug. And there are some patients that stay on that drug for a very long time with a very high quality of life. If we give chemotherapy or amivantamab up-front with that TKI, the PFS will probably be longer, yes. But the quality of life during that time is going to be different. We’re introducing more toxicity. We’re anchoring that patient to our cancer center because now, you’re getting an infusion every 2 or 3 weeks. And that gets in the way of just sort of normal life. And so these are really important considerations, which is why there’s not a clear right answer.

It’s a setting where, you know how it is, Neil, you ask 5 oncologists, you get 10 opinions about what to do, and so no clear path. Lots of right answers. This is shared decision making and we’ve really got to appreciate what the patient’s values are. There are some patients that say the increase in progression-free survival to me is not worth just coming into an infusion every 2 or 3 weeks. I’d rather be on a pill, see you 3 or 4 times a year. And in fact, in China, Yilong Wu, one of the surgeons there involved in ADAURA, ran a poll, a questionnaire. The vast majority, even with the better PFS, preferred TKI alone.

DR LOVE: How do you think subQ amivantamab would affect how people view that?

DR LIU: I think it makes things a little bit easier. The subQ, as you know from the PALOMA-3 study, which Natasha Leighl presented at ASCO, the subQ version, you don’t get the infusion reactions. There was less venous thrombolic events. There actually seemed to be a signal of better survival with subQ hinting that there might be some immunologic benefit with giving amivantamab subcutaneously. But you’re still getting the injections and you’re still having most of the same toxicities in terms of the dermatologic toxicity, which is something you can’t hide, which does affect your quality of life. SubQ is better. I think in the future though what’s going to make this even better is when we start delivering subQ to the patients and they’re not needing to come into the cancer center. And I think that’s going to make the convenience a lot better.

DR LOVE: So another approval we just saw was amivantamab plus carbo/pem in EGFR mutant active mutations. Any comments?

DR LIU: To me, this is the current standard of care second line. So postosimertinib, we looked at chemotherapy versus chemotherapy plus amivantamab. That’s the MARIPOSA-2 regimen. This was approved by the US FDA September 19^th, 2024. Adding amivantamab to chemotherapy significantly improved the response rate, almost doubling the response rate north of 60% with a profound improvement in PFS and a significant, well not statistically significant, but a major trend towards a survival benefit. I think this will hit survival.

And to me, this is the most active regimen we have in the second-line. But it comes with toxicity. I actually think amivantamab is a little easier to give with the chemo because you’re already getting an infusion, so that aspect doesn’t change things. And with some of the premeds we give with chemotherapy like the dexamethasone, I think that does counter some of the toxicity with amivantamab. So I’ve been pretty happy with tolerability in this setting, especially once you get to the every-3-week dosing of amivantamab. And I think this will be even better in the subQ setting. But there’s tox for sure. And it’s the dermatologic toxicity that’s most important. Maybe not something that’s dangerous medically to the patients, but something that’s very bothersome and affects quality of life. So we’ve got to be proactive and get out in front.

But to me, this is the standard of care second line. And this, I think, is the greatest argument and the greatest threat to using this drug in the front-line setting. Because the real question I have is not osimertinib versus ami/lazer. It’s ami/lazer then chemo versus osimertinib then MARIPOSA-2 or ami/chemo. And so it really is sequencing these not just in acumen. And so because this target is so important, I think, in the second-line setting, I feel a little more comfortable leaving it out of the front-line setting.

DR LOVE: That’s really interesting. Do you think that patritumab is going to be approved in these patients second-line EGFR mutant? And if it is, how do you think it’s going to compare to this?

DR LIU: This is a great question. And I think it’s good to have options. I think it will be approved. And what I like about patritumab is really the versatility. You really see this work across multiple mechanisms of resistance. However, to, again, tell both sides of the story, I think that a lot of people were a little disappointed by that progression-free survival. While we saw pretty high response rates and we saw efficacy across multiple types of resistance, a PFS below 6 months isn’t really the homerun that I think people were hoping that it is. So I think this will be approved. I think it will have a role. To me, the response rate with patritumab deruxtecan is not nearly as high as the response rate of ami/chemo. And so I would still probably put it behind ami/chemo, but it is a good option. And there are some settings where I might prefer it over ami/chemo.

DR LOVE: I have memories of being at a Society of Neuro-Oncology meeting, I think it was at least 10 years ago, when they first started to talk about tumor treating fields, which, of course, eventually got approved as part of treatment for GBM, but everybody was like, what? And then here we are. Approved in lung cancer as of 2 days ago. Wow.

DR LIU: Yes. And so this was based on the Phase III LUNAR trial presented by Dr Ticiana Leal. And the tumor treating fields were not given alone. They were given with standard treatment. So patients randomized to standard treatment or with the tumor treating fields. And what we saw is overall, the study was positive. But the signal was really in patients who were getting immunotherapy with the tumor treating fields. And the biologic hypothesis is it really does something to potentiate the immune response. Patients getting docetaxel didn’t seem to get nearly as much benefit. And so it was really people getting standard immunotherapy, but this was second-line immunotherapy. And this is a device, and so the approval process for a device is not quite as stringent as that for medicine.

And there are a lot of sort of openings in terms of the study design that leave some questions with some of our colleagues. For example, there was no placebo, no sham device in this. And so is it simply that people were getting more attention, being seen more frequently for the upkeep of these fields? The problem is I don’t really know if I could ask patients to enroll in placebo or sham device study because it’s a big commitment to ask someone to do that and I don’t really know if I would want to send patients to a sham device randomization. And I certainly know patients would not want to be randomized to a sham device.

And so the other point to this study is that it really reflects a population that in most cases doesn’t exist anymore because we’re not giving immunotherapy second-line, we’re giving it first-line. We’re giving it first-line alone or with chemotherapy. So single-agent immunotherapy in the second-line setting is something we really shouldn’t be doing anymore. And if this is where that had the benefit, would this translate to other settings where we use immunotherapy? Should we use it in the second-line setting? Should we use it with maintenance? I think there are some unanswered questions. But I do think that this is a very promising field. And this device clearly does something. There’s a biologic activity. And when we look to our neuro-oncology colleagues, this really has made an impact in glioblastoma multiforme. And so I do think that this is a very promising type of device, but I also think there are a lot of biologic questions that are not answered that we really need to understand more. There are a lot of ongoing studies, to the sponsor’s credit, that are looking to really develop the biology of this and identify where should we use this today in a world where we’re not giving immunotherapy second line? And even though it is approved, one question that’s still unanswered, will insurance companies pay for it?

DR LOVE: I guess another question is, will oncologists recommend it? Will you?

DR LIU: So I think that in settings where giving single-agent immunotherapy in the second-line setting, it is kind of appealing but I kind of need to see more to extrapolate the other fields. So right now, there is a lot involved in this. This is a device that you wear most of the day every day. It is a little bit limiting and it is, I think, a lot to ask. Note though that there is also data with mesothelioma. And so this is something that does have some precedence in a thoracic space. So I’m not opposed to its use at all. I don’t think it’s the wrong thing to do. I don’t know if it’s going to find a place in my regular clinic because in settings where patients have progressed on immunotherapy and are looking for second-line treatment, I really think those patients are probably better served with a clinical trial. And if I have an extremely motivated patient, I think there are a lot of other promising avenues I’d like to explore more. But I’m keeping an open mind. And so I want to see what happens.

Recent advances in small cell lung cancer

DR LOVE: So let’s finish out with some papers on what’s becoming one of my favorite topics nowadays. Now we’ve got some good stuff to talk about in small cell. It’s been a while. I got tired of talking about lurbi all the time and IO/chemo, et cetera. Now we’ve got some good stuff going on. Of course, starting out with the trial looking at durva in limited-stage disease.

The ADRIATIC study that was presented as part of the huge lung plenary session at ASCO this year. Any thoughts? Any comments?

DR LIU: Yeah, I thought this was really phenomenal data. What we saw is when you add durvalumab after chemoradiation, we have an improvement in progression-free survival and a significant improvement in overall survival. Now if you look at the hazard ratio, the hazard ratio for survival is 0.73. That’s not so different from what we saw with IMpower133 and CASPIAN in the extensive-stage setting. But if you look at the median survivals, in the extensive-stage setting, those led to about 2-month improvement in the median. Here, you’re looking at about a 2-year improvement in the median. Again, the impact is always greater when we identify these cancers earlier. And so as we find small cell in the limited stage, we can make more of a difference. We can have a bigger impact. But to me, walking out of that plenary session at ASCO, this was the standard of care, and everybody should have been getting it.

What was very interesting though is later in the year at the ASTRO meeting, Dr Kristin Higgins presented the NRG-LU005 data. That looked at atezolizumab for limited-stage small cell lung cancer. But the big difference there is atezo was given with the chemoradiation up-front. And when you give atezolizumab with chemotherapy and radiation, there was no benefit compared to placebo. And so you saw no benefit if you fold in the immunotherapy early. This is very similar to what we saw with PACIFIC-2. With PACIFIC-2 at ELCC 2024 when Jeff Bradley presented those data, giving the durvalumab with the chemoradiation negated any benefit from durvalumab even if you gave it after. It was no better than placebo. So when we give immunotherapy with chemoradiation, not only are we not seeing that benefit, but it’s actually detrimental to giving it after the radiation. And there are some theories as to why. The most likely here and the most endorsed is that when we give a checkpoint inhibitor, we are stimulating turning on all of our T-cells. If we’re turning on those T-cells and then killing them with lymphopenic chemotherapy and radiation, then we’re preventing those memory T-cells from ever forming and we’re negating any benefit. So giving the checkpoint inhibitor, the PD-1 or PD-L1 inhibitor, with chemoradiation is not a winning strategy.

DR LOVE: This is one of my favorites, ADC in small cell. I think it’s been tried before, but didn’t work. Now, it looks like maybe we’re getting somewhere with ifinatamab. Does that have a short name? I-DXd maybe, right?

DR LIU: I-DXd, yes.

This is the antibody-drug conjugate using that same cleavable linker, topoisomerase-1 payload targeting B7-H3. And it’s highly expressed on small cell lung cancer. And what we saw in this study, the Ideate-Lung01 study was really comparing 2 different doses. And what we saw was actually the higher dose was much better. And so this dose finding stipulation by the FDA in Project Optimus is really paying off in a lot of these studies, especially in small cell. We saw it with tarlatamab in the DeLLphi study and we’re seeing it with I-DXd as well, that the higher dose here, really the opposite we saw with tarlatamab, the higher dose better with a response rate of 55% compared to only 26% at the lower dose. And so the response rate was higher. The disease control rate with this drug in the previously treated setting was over 90%. These are phenomenal numbers. And so the I-DXd data, looking very promising. We need to start seeing the randomized data versus standard treatment. But I think this drug holds a lot of promise though it does have some important toxicities too.

DR LOVE: Yeah, let’s talk about that. Because anything with the last name of DXd always gets people’s attention. So what’s the deal in terms of ILD here?

DR LIU: Exactly. So you hit the nail on the head. It really is the interstitial lung disease, 9% in the low dose, 12% in the high dose. Mostly low-grade, but there was a fatal case in that low dose. And there was a Grade 3 case as well in the high dose. And so in this relatively small study, there was about 10% risk of ILD overall. And so that is something that we’re going to need to watch. It does cause other toxicities we associate with that same payload in terms of myelosuppression, but really pneumonitis is the one that caught everyone’s eye.

DR LOVE: And what about intracranial activity?

DR LIU: The pleasant surprise is that we’ve seen this consistently with ADCs, these are big molecules, and we were always taught that sort of these big molecules don’t cross the blood-brain barrier. But I think we’re learning the biology is not as simple as we thought, and the barrier is very different. And here, we are seeing response rates. In patients with target lesions in the brain, the response rate of I-DXd was over 50%. And so this is very important for small cell where CNS involvement is very common. So having an ADC that works so well in the body and in the brain is going to be a real win. And I’ve got to tell you there are many other ADCs in development for small cell. So this is going to become a crowded space in a hurry.

Lurbinectidin in combination with atezolizumab as front-line maintenance therapy for patients with early-stage SCLC

DR LOVE: Let me ask you one other thing. I think I just saw a press release about adding lurbi to IO as maintenance in small cell.

DR LIU: Correct.

DR LOVE: Can you talk about that? And what do you think it’s going to mean?

DR LIU: So the other advance we’ve seen in small cell lung cancer really is in that maintenance setting. And as you know, Neil, to date, we’ve tried maintenance therapy a lot. And every study has been negative. Maintenance is appealing in small cell because we can get that initial response pretty easily. But then when the cancer comes back, it doesn’t respond to anything. And the attrition rate in small cell is so high that even under controlled circumstances, getting any second-line treatment is about a 50/50 proposition. So if we can introduce a treatment before progression, could that be of benefit?

But multiple randomized maintenance studies have been negative including nivolumab in the CheckMate 451 study. So the IMforte study builds on the front-line setting of chemo/IO. So all patients get carboplatin/etoposide and atezolizumab and they’re randomized to standard atezolizumab maintenance or atezolizumab with lurbinectedin. And the premise here is 2-fold. Lurbinectedin, approved in the second-line setting, FDA accelerated approval. Active agent. Let’s give it early. But also, there is some synergy potentially between lurbinectedin and atezolizumab. It has to do with the effect of lurbinectedin on tumor-associated macrophages, which are particularly important for small cell lung cancer. And there was a study out of Spain from Antonio Calles called the LUPER study that combined lurbinectedin with pembrolizumab with really impressive results. So is folding in lurbinectedin going to be of benefit?

And while we haven’t seen the data, the press release does show that adding lurbinectedin to maintenance atezolizumab improved both progression-free survival and, importantly, overall survival. So we do have a survival benefit with lurbinectedin in that maintenance setting. And we need to see the data to see the magnitude, but it is, by press release, statistically significant.