Well, I'm Frank Detterbeck. I'll just open this session. We're going to talk about stage 1, stage 2 lung cancer. Three talks. The first one is Erin Gillespie going to talk to us about updates in stage 1, stage 2 non-small cell lung cancer. Thank you. Good morning, everybody. Welcome to CHESS. I'm so delighted to be joining everybody today. I think we have a great program for you this morning. So we're going to talk about major updates. If this will change. There we go. Here we go. These are my disclosures. So I think we're living in this just really exciting era in early stage lung cancer. And I thought I'd break this talk down into kind of four major categories in talking about updates and the things that are going on for stage 1 and stage 2. Touching briefly on screening and then going into movement of drugs and medications. And then we're going to talk a little bit about what's going on in stage 1 and stage 2 lung cancer. And then we're going to talk a little bit about what's going on in stage 2 lung cancer. One of the biggest struggles that we've had for a long time in non-small cell lung cancer is the fact that the majority of our patients were diagnosed in a stage 4 setting. And really that's two reasons behind that. Many patients are presenting with a paucity of symptoms and for a very long time we didn't have any opportunities for screening. Fortunately, with the advent of the NLST and of course a wide array of new treatments available for our patients across disease stages, we've not only seen a shift of patients more towards stage 1, but improved five-year survival. We're seeing that mark move for the first time in a long time, which is really exciting. I think most people are pretty familiar with the NLST. It was published in 2011, randomized trial, randomizing an enriched patient group. It's really important to remember that these are high-risk patients, randomized to low-dose CT versus chest x-ray. And actually the study was terminated early when it reached its end point of a 20% reduction in lung cancer mortality, ushering them in this new era of lung cancer screening. Just last year, we see our first major update from the USPSTF guidelines, expanding our cohort of patients who now qualify for screening. So we've expanded our age group, dropping it from 55 to 50, current or former smokers, dropping the pack year requirement from 30 to 20, and of course these patients who are undergoing screening should have no symptoms of lung cancer. So hopefully continuing to shift us into more patients being diagnosed in that early stage on stage two setting. But I'm not satisfied with survival. None of us should be satisfied with survival. We still have a lot of work to do. I can tell you there's nothing more disappointing as a surgeon than taking a patient with a very small tumor through a surgery we think should be curative and then they recur a couple years later. And with that in mind, we've seen a lot of new treatments starting to make their way into the earlier disease setting. We're living in this unprecedented era of molecular medicine. We understand tumors better than we ever have before. We understand the oncogenes. We understand the changes in the tumor suppressor genes. And then of course we're living in the immunotherapy era. And so immunotherapy is being used in a lot of different cancer types and in the lung we've had a lot of great results in particular in the metastatic setting. And there's been a huge amount of pressure to continue expanding what people are considering these life-saving therapies into earlier and earlier stages of disease. Now I think it's important as we kind of launch into talking a little bit about the adjuvant to neoadjuvant setting to touch briefly on surrogate endpoints. So we've seen most of our trials move away from kind of our standard overall survival as our primary endpoint and really that's been more commonly relegated as a secondary endpoint. And we're using a lot of surrogate endpoints for our studies. PFS has shown to have great correlation to overall survival in the metastatic setting. And now in the earlier stages of disease for our primary endpoint we're seeing two things commonly used, disease-free survival and event-free survival. And now the incorporation of other kind of surrogate or biologic endpoints, things like CTDNA, pathologic complete response, and major pathologic response. So how do we come up with these surrogate endpoints? Well we're not just picking them at random. There's actually statistical modeling that's used to try and help us to select these. And the idea behind it is that we want to find and to study an effect on an endpoint with the idea that that endpoint is going to correlate well with the endpoint that we ultimately want, right? I'm not going to go through complex mathematics. It's far too early. It's a Sunday. That would be unnecessarily cruel. But what we can do is we can use statistics to calculate a correlation coefficient. So how well that effect is getting us to the endpoint that we want and how well that effect will get us to the ultimate endpoint with the idea that a correlation coefficient greater than 0.85 is going to be a high level of correlation. So we're going to get ultimately the endpoint that we want using this surrogate marker. And there's a lot of studies going on right now trying to validate some of these endpoints. So validate the circulating tumor DNA, validate the pathologic complete responses with the idea that we want these ultimately to correlate with our big goal, overall survival, in addition to, of course, high quality of life for our patients. Absolutely critical. All right. So let's jump into some of the major updates in adjuvant therapy. So I'm going to focus on two trials today that I think have really kind of changed the landscape of how we're thinking about adjuvant treatment for our patients. The first is Empower 010. This was a randomized trial taking completely resected patients, stage 1b to 3a. And they're randomized to adjuvant treatment, one of two arms. So chemotherapy versus chemotherapy plus a tesolizumab. I'm not going to go through the stratification factors because, frankly, we just don't have time to go through all of the details. But primary endpoint for this trial was DFS in our PD-L1 positive cohort for stage 2 and 3a. So we're not looking specifically at a small portion of the patients that we enrolled, the 1b group, OK? But stage 2. So this is earlier stage. If that was positive, then we look at the entire group. And if that's positive, we continue to follow all the way out to overall survival. On the left side here, you can see our disease-free survival. This is what came out as our first publication. You can see this beautiful early separation of our curves with the blue line representing our chemo plus a tesolizumab group, the bottom being our best supportive care chemotherapy alone. We crossed our statistical significance boundary. Our hazard ratio is 0.66, so statistically significant. Everybody's really excited. We've got enhanced disease-free survival for stage 2, stage 3a patients, PD-L1 positive. And so the FDA actually approves this regimen for use in this cohort of patients. Fast forward just a little bit, and we get our overall survival this year. We still see this separation of curves, but now we sort of start to lose it towards the end. Our hazard ratios are no longer statistically significant in our overall cohort with PD-L1 greater than 1%. So they did some subset analysis to try and figure out what subgroup of patients were really driving this survival benefit that we saw kind of in their disease-free survival and what's still driving it to a degree in our overall survival. And it turns out that it's our EGFR-ALK negative. That's an important thing to remember. Stage 2 and stage 3 patients that have a PD-L1 greater than 50%. But remember, this regimen is still approved for everybody greater than 1%, so important to know this data, to look at the overall survival data. We seem to have an ongoing signal of benefit, stage 2, stage 3a, PD-L1 greater than 50%. The ADORA trial was an incredible trial that came out. Really exciting data from this. Again, completely resected patients, stage 1b all the way to 3a. Good performance status. Primary endpoint again here is our disease-free survival. They looked at this primarily in the 2 and 3a cohort again, but then they actually also did a subset analysis for all the stages kind of independently, and I'll actually show you those curves. And after completing therapy, so this is going through surgery, plus or minus chemotherapy if it was recommended, patients were then randomized to receive either an osomertinib or a placebo. Here you can see our overall disease-free survival in patients with stage 2, stage 3a. The blue line at the top is the osomertinib line. At the bottom it is placebo. You can see a massive and early separation of the curves, hazard ratio of 0.17. Very statistically significant. If we look even at the independent by stage here, stage 3a, this kind of really mirrors that overall survival curve that we can see. But even stage 2 we get an early separation, and even for stage 1b we have a separation of curves. All of them have a statistically significant hazard ratio of 0.5, 0.17, and 0.12 respectively. Based on this kind of early signals of survival, the study was actually evaluated early, unblinded early, so that patients who are in the placebo arm could be offered therapy on the osomertinib arm. And actually FDA approved this regimen based on, obviously, these response curves. So it is approved for patients stage 1b and above who are EGFR mutant, non-small cell lung cancer. So really exciting kind of two major trials moving things into the adjuvant setting for our earlier stages of disease, 1b and above for osomertinib, and 2 and above PD-L1 high for artesializumab as part of an adjuvant regimen. Let's touch briefly on neoadjuvant therapy. I don't think you can open a journal without hearing somebody talking about Checkmate 816 these days. This is the very first neoadjuvant immunotherapy trial. These are patients 1b to 3a, so again, kind of the same patient population. As a reminder, these are patients greater than 4 centimeters, potentially node positive. And they were resectable, really, really important. Evaluated up front, resectable. And they were randomized to either nivolumab plus chemotherapy versus chemotherapy alone. Full disclosure, there was initially a third arm. This third arm was closed. Patients went on to have surgery and then followed. Adjuvant treatment was up to the discretion of the treating physician. Primary endpoints for this, remember we talked about our surrogate endpoints, pathologic complete response is one, event-free survival is the other, with a number of secondary endpoints, including overall survival. Important to keep this in as confirmatory. 358 patients were enrolled and randomized. We had a really good progression through with a high level of completion of treatment, which is really, really encouraging. Actually, the addition of nivolumab did not reduce the chances of a patient going on to have surgery or going on to have a successful surgery, which is really encouraging. So let's talk about the pathologic endpoint first. This was their primary major first endpoint. And you can see here the pathologic complete response occurred in 24% of patients with a nevo and chemo versus 2.2% in the chemo alone. I'm not going to ask you to look at this tiny graphic off to the side. I put that up mostly because you can see that dotted line in the middle. You can see that the majority of bars are over towards the right-hand side. That's our nevo plus chemo arm, showing that we had a really intense depth of pathologic response throughout all subgroups that were included in the study. So it was basically regardless of age, gender, stage, histologic subtype of tumor, the only one that you can see crossing our zero boundary here is our smoking status for our never smokers. So how did that portend? How did that end up panning out for our event-free survival? Well, here you can see our event-free survival curve that was actually just presented. Our top line here is our nevo plus chemo arm. Bottom line is our chemotherapy. We do see this nice separation of the curves. The hazard ratio is statistically significant at 0.63. When we look at the subgroup analysis, so this event-free survival is for the overall cohort. When we look at our event-free survival subset analysis, though, it's important to know first and foremost that we've switched it around on you. So now the nevo plus chemo arm is actually over on the left side instead of the right side, but really we don't see that same really dramatic push of all of our lines being completely across the other side. There's really only a couple of groups that we still continue to see this trend for a significant benefit, and it should be noted that this is 3A, so we didn't see that same benefit in the 1B and 2, and the PDL went high, so greater than 50%. We're still waiting for additional data to mature overall survival, but it'll be interesting to see how much this regimen continues to be used in those earlier settings, 1B and 2. It is approved. Checkmate 816 is approved for 2 and above. PDL1 doesn't matter. So it is available. It's going to be used. Need to continue following that. Lot more coming down the pipeline. There's a huge number of trials that are ongoing right now that are going to be reading out throughout 2023 and 2024 using all the different types of immunotherapy, so nevo, atizo, pembro, and then the neoadora, so moving that osomertinib into the neoadjuvant setting. So a lot of really exciting trials going on. Surgical trials, you're going to hear my colleagues talk about this. I'm going to just touch very briefly as an introduction. CALGB140503 was a really exciting trial that was just presented by Dr. Al-Turki. This is looking at really small tumors, so finally I get to talk about the stage 1As. This is tumors two centimeters or less, node negative, pathologically confirmed in the operating room, and they were randomized to lobectomy versus a sublobar resection, and I'll let my colleagues define for you what that means. You're going to see a lot of curves coming out in the literature. A lot of people really excited about this. This was a non-inferiority trial, so we're not looking for it to be superior. We're looking for it to be non-inferior. You can see that these mirror each other very, very nicely, and one of the things that I was really pleased with that they presented at the meeting recently, the ISLC 2020, was the disease recurrence patterns, because I think that's something that's really important for us to think about and to look at, and they were pretty similar amongst the groups. The other big thing that we talk about a lot, and I think that is imperative throughout all stages of disease, but I think especially important when we're starting to think about things like sublobar resections is a big push to enhance quality of surgery. I think this is something we need to talk about a heck of a lot more. We actually published this retrospective review just this year looking at guideline-concordant surgery and adjuvant chemotherapy among patients with early-stage lung cancer that were enrolled in the U.S. ALCHEMIST trial, and so it's embarrassing to say this, but only 53% of patients actually had what we'd consider a standard-of-care lymphadenectomy, and if we look at some of the big database studies, including centers from around the country, it gets as low as 25%. We have to do better. We have to do a better lymphadenectomy for our patients to make sure that we're staging them correctly, to make sure that they are going on to have the right treatments. To that end, the COC standard 5.8 came out just a couple of years ago, and it's created a very clear and a very standard guideline for how we should be doing a lymphadenectomy, and that includes the submission of three separate mediastinal lymph node stations, one separate N1 lymph node station. I encourage everybody to do more than that. To me, that's just kind of the bare minimum with any kind of pulmonary resection for non-small-cell lung cancer. There's going to be compliance visits to try to make sure, with auditing of charts to ensure, and path reports specifically, to make sure that we're doing a better job for our patients. In line with that, we also have trials going on, trying to look at the pathologic piece to this, making sure that we're looking through our specimens adequately to find all the lymph nodes. The last thing we want is to have resected a lymph node. It's there in the specimen. It's N1 positive. We don't know about it, and we're not offering a patient potentially adjuvant therapy that they would qualify for. So in conclusion, we're seeing a huge movement of immunotherapy, targeted therapies into earlier stage, significant use of surrogate endpoints. I think it's critical to keep correlating these with overall survival. I think it's critical to keep collecting surgical endpoints in all of these trials to make sure we're providing patients with safe and high-quality surgery, and less, maybe more, when it comes to very small lung cancers. So my conclusion statement for today is we've come a long way. We have a long way to go, and in between, we are somewhere. Thank you. All right, well, I'm impressed at how much you've covered in 15 minutes here. All right, so I'm Frank Detterbeck. So how do I start this? Here we go. All right, and I'm Frank Detterbeck, Yale University. I have no conflicts regarding this presentation. So I was asked to talk about when is lobectomy indicated, and I think that certainly things are changing. So just to kind of set the stage, I think stage three, lobectomy is often indicated, but, you know, the issues really are, you know, surgery versus RT and pre versus post-chemo. I think stage two, yes, and I'm really going to focus on stage one and really the issue of, you know, what do we do now, low versus segment versus wedge. And the way I look at this when we're, you know, dealing with patients, and I'm not sure if I have a pointer, I guess. Can you see that? All right, so I look at it as, you know, there are short-term outcomes, there are long-term outcomes, there are benefits, there are downsides, you know. There's, you know, patient values and preferences that sometimes shift this, you know, fulcrum, and that's kind of the way that I individualize things. But within that, I think that there are certain key outcomes, short-term, intermediate, long-term, that end up really being the drivers of that decision. There are areas where you say, well, there's not really much of a difference in, you know, a driver over here, but there's going to be a big difference in a driver over here, so I can kind of focus my thought process on a particular thing because otherwise we get lost in stuff. So I'm going to focus mostly on kind of healthy patients with a typical, solid, speculated lung cancer, and we have three randomized controlled trials. I'm going to go through these a little bit. Sorry for the stuffy group. We're actually not very stuffy. Lung Cancer Study Group, the JCOG study, the CLGB study that was just mentioned, I'm going to talk a little bit about some non-randomized comparisons, but ones that are adjusted for confounders, at least to some extent. You can ask is there value in doing that, but I think that it still allows us to explore some nuances and effect modifiers. And, you know, just to kind of cut to the chase here, to manage to get this in 15 minutes, in terms of short-term outcomes, we have clear data that there is not a difference between lobectomy, segmentectomy, and wedge. So the three randomized studies found no difference, and if you look at non-randomized comparisons, the differences that are there are, you know, really not clinically relevant. They're, you know, perioperative mortality at 1.56 versus 1.3 or something like that. So, you know, not really clinically meaningful. So once again, if you want to focus, you know, where are you going to drive the needle, it's not in these short-term morbidity mortality differences, which I think comes to most of us as a surprise. We always thought that we were saving some morbidity and mortality by doing limited resections, but the data does not support that. In terms of intermediate-term outcomes, and I think of functional capacity, that would be one that I'd love to have data for, but I don't. So I can use PFTs as kind of a surrogate for that. And the bottom line there, I'll show you just a little bit of data on that, there's not really a significant difference in PFTs in healthy patients between lobe and sublobar resection. Again, a surprise to most people. And quality of life data is a bit limited. It's kind of in non-randomized comparisons, but also suggests that there's not a difference. And I guess in some ways it makes sense. If you're going to do a surgical intervention, probably what you do on the inside doesn't matter as much as the fact that you did a surgical intervention and the impact of that surgical intervention. So the bottom line is that I think if we're really going to focus on where are we driving the, what's driving the decision-making here, it's going to be long-term outcomes. So I'll show you just a little bit of data on short-term outcomes and intermediate-term outcomes. So this is the data on FEV1. So these are a number of studies that have looked at FEV1 segment versus lobe and looked at it at least six months post-operatively, because early post-operatively you can't really assess where you're going to end up. And I've kind of split these up into series that had frequent multi-segment resections. For example, a lingular sparing left upper lobectomy. Common segmentectomy that's done. You could argue that lingular sparing left upper lobectomy is like taking out a right upper lobe. So lobe-like, perhaps, segmentectomy. If you look in these studies, the difference is about 4%. If you look at studies that had few multi-segment resections, so mostly single segment versus lobe, the difference is a little bit bigger. But if you're talking about patients with relatively normal PFTs, that probably is not really clinically relevant. And I'll just cut to the chase here briefly. The two randomized studies that just came out, the JCOG study, showed a 3% difference in FEV1 between segment and lobe. And the LTORCHI study, the CLGB study, showed a 2% difference between mostly wedge and lobe. 2% at six months. So multiple studies here that show that, you know, this is not what's going to drive your decision making. There's not really a big difference to be gained here. To most people's surprise. So these are the major randomized studies. And there are some differences between the studies, so we can say were they comparing lobe versus segment and wedge, like in the Lung Cancer Study Group study. And there are size of tumors, and there are tumors that are mostly ground glass, and there are tumors that are solid. So Lung Cancer Study Group study was solid tumors, larger tumors, so it was done 1982 to 1988. CT was not required. Some of those patients probably did have a CT, because CT came online in the early 1980s. Bottom line is survival was decreased with segment or wedge. So then you have the CLGB study, which is two centimeters or less, so smaller tumors than in the Lung Cancer Study Group. I would argue if it's chest x-ray discovered lung cancer, it probably was at least two centimeters to start with, maybe one and a half, but it didn't have a lot of one centimeter tumors in that. So CLGB, lobe versus wedge, primarily 58% wedge, predominantly solid. We don't really have good data on that. Pure ground glass was excluded. No morbidity mortality difference, and no survival difference. And then we have the JCOG study, which is also two centimeters or less, but it was really focused kind of on these more ground glass tumors. They did extend all the way up to quote unquote solid, but this is solid on lung windows, which is probably in most of those, I think that actually disappeared on mediastinal windows. JCOG also showed no short-term advantages and no difference in survival. So here's just some data from Lung Cancer Study Group, lobectomy versus limited resection. This is the overall survival data. You know, there was extensive node dissection that was required, accrued a long time ago. There was a difference, but I think you have to question, you know, how relevant is this in today's setting and the tumors that we see and, you know, PET scans and everything else that we do, is this really relevant? So then there's the JCOG study, randomized controlled tile, segment versus lobe, two centimeters or less, large number of patients. So these are primarily focused on ground glass capacities. The consolidation to tumor ratio was greater than 0.25. Now the paper says 0.5, but actually if you read through the fine print, they extended it down to 0.25. So I'm not quite sure why they keep writing 0.5, because that's not really what's included in the study. They involved many lobe-like segmentectomies, but that's not really well defined. There was a mandatory node dissection, mandatory, you know, needed to be N0, needed to have, you know, margins that were greater than two centimeters, and the Japanese are meticulous about their node dissection and their margins, very meticulous about it. Five percent of the segment group were crossovers due to N positive or inadequate margins, but analyzed with the segment group. So these are results here, left up some of the key things to kind of keep in mind at the top, but these are the results that we got. So this is overall survival. Segment actually came out a little bit better than lobectomy, three percent. Now you could argue three percent is that really, you know, move my needle, but, you know, it certainly was not inferior. Now it came out a little bit better actually because there were less unrelated deaths, less deaths from other cancers. Now why doing a segmentectomy would decrease your chance of dying from pancreatic cancer or something else doesn't really make a lot of sense. So I think we have to probably not take too much, you know, put too much into the segmentectomy being better than lobectomy because the driver of that is not really logical. Recurrence-free survival was exactly identical, 88 percent. Lung cancer deaths were identical. There was an increase in local recurrence. There was an increase in the ability to re-resect at the point of a local recurrence, which may actually have offset, you know, may have resulted in overall survival not really being affected much by this local recurrence. There was also a slight trend towards increased second primary lung cancers in the segment group and increased rate of resection of those second primaries. So that's the CLGP study, mostly ground glass. And if you look in the subset analysis, again, I don't expect you to read it, but you can see that, you know, all of these pretty much fall on the same side. So there's not really a signal in the subset analysis that favors a, you know, particular degree of ground glass or histology or age or, you know, particular lobe. We don't have any signals there. All right, so now let's move to the CLGB study. This was just reported at the World Lung Conference a couple of months ago. Publication is going to come out in the New England Journal, as my understanding, probably in about a month or two, I think. So less than two centimeters, fairly large number of patients, presumably mostly solid, but we don't really know that. So that was kind of the design of that study was to focus a bit more on this. They excluded pure ground glass capacities, but I don't really know how many of these had some degree of ground glass. Mandatory interoperative confirmation of N0. The protocol required a margin of greater than two centimeters. We don't have any information on what actually happened. I'm a little bit, how would I say, I'm very curious to see that data, you know, what was actually achieved, because certainly my experience in doing this for 30 years is that you end up with a lot less margin than what you thought. You know, you're there in the operating room, oh, this looks great, this is a great margin, and then you get the path report and you say, oh, damn, really? So I'm interested to see what happens there. Fifty-eight percent wedge. Note there were a lot of registered patients that were excluded, but most of them because they had benign disease or something else, but 25 percent were excluded because they were node positive, and so 25 percent of these that were excluded. So, again, if you're going to do this and you're not going to do careful interoperative, you know, proof of N0, you're not really following what the study did. So these are the outcomes, you know, these are the key figures. You already saw this slide. This is overall survival, sub-lobar and lobar, you know, completely the same. I think this is probably the cleanest outcome here, cumulative risk of recurrence or death from lung cancer, and there's no difference here between the dotted line and the solid line, which is sub-lobe and lobe, unrelated deaths, also no difference. You also saw this slide. Now, you know, a couple things to point out here. So overall recurrence, it's pretty damn high. You know, compared to the JCOG study, this is dramatically different, and also, you know, the CLGB study, you know, five-year survival was at around 60 percent, whereas in the JCOG study it was 90 percent. I think it just illustrates that we're talking about different tumors in the JCOG study. I think we're talking about mostly ground-class tumors, indolent tumors. I think here we're talking about more substantial kind of traditional lung cancers. At least something is different, you know, in terms of the tumors, and in either the lobe or the sub-lobar, it's pretty damn high. And if you look at distant disease, it's pretty high. Subgroup analysis, not really a clear signal here. You know, these are pretty down the middle here, kind of all the way along. You can argue a little bit performance status. You can argue a little bit tumor size, that maybe there's a slight suggestion that lobe is better for larger tumors or worse performance status, but that's a soft call, and that's really reading between the lines. You know, interestingly, you might expect that lobe is better for larger tumors. You wouldn't necessarily expect that it's better for worse performance status, but if anything, the signals seem to go that way. We don't really know about the margins achieved. We don't really know about the nature of the tumors. Certainly the overall outcomes are very different than JCOG. We don't really know how recurrence was differated from new lung primary. I have some questions about that. When I look at the curves, I'm not so sure that that was differentiated well. I don't have time to go into that. All right, I'm going to go a little bit into some non-randomized comparisons, and I don't want you to look at details, but just kind of look at the overall thing. So I've looked at non-randomized comparisons that adjusted for confounding. Now, you know, the degree of adjustment for confounding varies. You know, just because you say I've adjusted for, you know, three factors doesn't mean you've really adjusted well. If the three factors were the main drivers, then maybe that's important. If the three factors were kind of things that weren't really important, then you really haven't accomplished anything. So we did a very detailed study of non-randomized comparisons. I had criteria for it, and, you know, different domains of confounding, and green is you adjusted well, red is you adjusted poorly, and you can see that all of these studies have some issues with the adjustment. You know, overall, you can say some had high confidence in the treatment effect. You know, many had very low confidence that the difference represents the treatment effect. And if you look here in adjusted five-year overall survival or lung cancer-specific survival, segment versus lobe, green is where there was a statistically significant difference by multivariate analysis. So this is kind of how we looked at this. So bottom line is hazard ratio favors lobectomy in these. The differences are large enough to be clinically relevant. Only about a third of them were statistically significant, and, you know, if you look at these by tumor size, sorry, if you look at these by tumor size, no particular signal about the particular domains of confounding or tumor size. This is wedge versus lobe. You see a lot more green and a lot more differences here, and I'm going to show you this in a different way. So these are hazard ratios, segment versus lobe, lung cancer-specific survival, segment versus lobe. You can look at the differences that were there, how big they are. So there seems to be some signal there that lobe is better. If you look at wedge versus lobe, signal seems to be stronger. Hazard ratios seem to be larger. Differences seem to be larger. Having said that, you know, lots of residual confounding in this data. I'll segue just a little bit to margins. So these are studies that looked at margins and looked at millimeters of margins. So we classically sort of talk about you should have at least two centimeters, but if you look at it, to me, it looks like the dividing line is a little bit more around a centimeter. You know, if it's greater than a centimeter, you're talking about maybe 10 percent recurrence. If it's less than a centimeter, you're talking about maybe 20 percent chance of recurrence. So certainly in most of these studies, there was a difference with more limited margins, and this is about what the difference is. I think about 10 percent versus about 25 percent. Also, kind of in details here, increasing ground glass proportion of these tumors suggests that the margin difference probably makes less of a difference. Okay, margin to tumor ratio also, you know, makes a difference greater than one. So I think I'm going to rush through this a little bit more. If you look at older patients, you know, younger patients, really not much difference. If you look at older patients, this is a NCDB study. There's a difference, but it's really not as much as people would expect. And I'm going to skip through this. So putting it all together, this is sort of my assessment of where we are with this. I think there are hard arguments, you know. Tumor greater than two centimeters, I don't think we have data at this point. I think we should stick with saying lobe is probably what you should do. I also think you're going to have margin issues if a tumor is greater than two centimeters and you're trying to do sub-lobar resections. The data we have for sub-lobar resection is tumors in the outer third of the lung. So if it's not in the outer third, I'm not sure that's a good idea. If it's in an anatomic location, it's at a junction of segments, it's in an area where it's going to be hard to get a two centimeter margin, I think you should probably stick with a lobectomy. And I think that you need to do intra and pre-op, you know, EBIS, whatever, node assessment to confirm that it's N0 or else you can't apply the data that we have that, you know, we should be doing sub-lobar resections. I think softer arguments, reading between the lines, high pedigree would move me towards doing the lobe. If you have a patient where there's, you know, going to be a lot of, you know, air leak from the lung, maybe that moves you towards the lobe. And I would just remind people that, you know, the downside of doing the lobe really isn't there. You know, if you look at the morbidity mortality data in general, PFT data, it's not like there's a big downside. And so, you know, I think this is where lobe still fits, where we probably should still be doing lobe and, you know, don't feel like there's a, you know, strong push to not doing a lobe because of a benefit. All right, I'm going to stop there. All right. Thank you, Frank. Morning, I'm John Howington. I'm a native of Nashville, and that picture is if you drove out, it's 8th Avenue right out here. If you went out 8th Avenue, seven miles, and took a right at the light at Otter Creek Road, that's Radnor Lake that I hiked last Sunday. So, we could go today. So, if you have some extended time in Nashville, I suggest you go out and clear your mind and do some hiking around here. It's a beautiful time of the year to do it. So, my talk will be less data heavy and more practical. Can I get a show of hands, how many surgeons in the room? That's what I thought. How many pulmonary critical care? Yeah. How many fellows? Okay. So, because you've seen a lot of data, and you might come away with this and say, and I have nothing to disclose. I'm going to define sublobular more practically for you anatomically what a sublobular resection is. I'm going to go into the ideal patients for it, give you some cases where I, in the last quarter, picked patients for a segmentectomy, small amount of data on sublobular resection, a different study that the others haven't shown you, and then talk about mentally invasive. Because if we're going to talk about providing best outcomes in patients, it has to be tied to doing it mentally and basically. If you cut someone in half to do a wedge resection of an early stage lung cancer, you've lost all the benefit of doing a sublobular resection. I mean, that data is clear, and I'll show that to you. So, difference between a wedge resection and an anatomic segmentectomy, all right? So, a wedge, as it's shown here on the screen, you're just taking, like, a wedge out of a cake or a pie with a stick and a mechanical stapler. You're not thinking about what is the lymphatic drainage of that portion of the lung. And commonly, as Erin Gillespie pointed out, surgeons do the wedge and then get out. It's a brief operation, and they haven't taken any lymph nodes. It really doesn't benefit, other than tissue for doing biomarker testing, that doesn't have anything over doing stereotactic radiation. So, as surgeons, the way we can differentiate the product that we provide to the patient and our referring physicians is to actually do that lymph node dissection to accurately stage the patient as stage one and not an advanced stage. And you saw in the trial CHB study, they said you had to confirm with frozen section that they were in zero. So, you can't take it, if you looked at general, in the U.S., Medicare data on sublobular resections, it doesn't do nearly as well as lobectomy, and it's all about the quality of the product performed. So, when you do a wedge resection, it is a curative intent operation, and that's now part of the language for the COC. But the idea that you're intending for your margins to be clear, and you can't just assume they are, as Frank said, once you use an endomechanical stapler and compress the tissues, you really don't exactly know how great your margins are, you kind of feel, but the pathologist can tell you. So, they should be checking a margin at the time of the operation, because if it's close or positive, they should be taking additional lung parenchyma. Mucin lymph node dissection should be done, and if I've got time, I'll show you a little video clip of how, with modern minimally invasive approaches, you can do an elegant hyaluronid dissection at low risk. And we should be, since we're doing this minimally invasively, we should be putting the specimens into a protective bag so we don't get local recurrence from that. A segmentectomy, as I mentioned, is an anatomic resection. So, you're going to the hilum of the lung, you're identifying the bronchial and vascular supply to that segment, and isolating those. At the time of doing that, you're taking out lymph nodes, because that actually creates the spaces so you can divide those structures and give it a more accurate name. So, this is, would be an ideal patient. This is a screened patient with a nine millimeter, so stage 1A1 lung cancer. This has a five-year survival of over 90 percent, and that's with thousands of patients resected across the country. If you look at this eighth edition of the lung cancer staging guidelines, we now have that data that it's not 75 percent, it's over 90 percent for this 1A1 tumor. So, Frank's already mentioned this, ideal patients for sublobar resections, ground glass lesions. Many of those are adenocarcinoma in situ. They have a low risk of nodal involvement, low risk of dysmetastatic disease. Peripheral solid lesions, I'll give you some examples of that. The ideal segments, it's where you're taking a segment and you're leaving the majority of the lung parenchyma in that lobe, still with the patient. That's where you can see a difference impact in PFTs. If you take the basilar segment of the lower lobe and leave the supseg, as another example, you're not leaving a lot of parenchyma, you're not going to clinically safely benefit that patient. Location, location, location, and Frank kind of hinted to that. So, this lesion, you know, we can't tell always visceral involvement. It's a peripheral lesion, that's fine. Central tumors have, there's no role for a sublobar resection in a central tumor. You have no idea where the lymphatic drainage is for those. So, Frank mentioned the outer third. So, here's some examples. This is a patient I'm operating on Tuesday. He had a rollover accident in April. He had had a coronary bypass, excuse me, stenting and put on dual antiplatelet therapy. So, his pulmonologist followed him for six months. The lesion didn't change. The PET wasn't that impressive, but it didn't go away. So, you got a robotic navigational bronchoscopy and that's an adenocarcinoma. So, it's sitting right there. It's part solid, part ground glass, and it's in that apical posterior segment of the left upper lobe. He's going to get a segmentectomy because he's got a significant lifespan ahead and the potential for a second primary is there. And he's like, I really was not interested in having the whole lobe of my lung taken out. I said, well, I got good data that you don't need the whole lobe taken out. And so, he saw the radiation oncologist, saw myself. Thankfully for me, he saw the radiation oncologist first and the radiation oncologist said, your best shot is a resection, but if you choose not to, we can do stereotactic radiation. Superior segment of the lower lobe. This is the left lower lobe. This is an 80-year-old lady that comes in. This is a nice CT showing. Here is, let me get the mouse here. Here is the superior segment of bronchus and the artery to that segment, right there where the lesion is. That tells you pretty clearly that if you do that anatomic resection, that lymphatic dissection, you know where it is. If this lesion was here in the lower lobe, you have no idea where the lymphatic drain is on that. That's gonna be a bad choice, this location right here, to do a segmentectomy. You're gonna wanna do a lobectomy in that case. And again, that's just for the non-surgeons in the group. That's the portion, so really the patient most of the lower lobe of the lung. And here, this is where, Frank, and this is a patient where you do it for medical comorbidities or poor PFTs. DLCO of 38% are predicted. So I chose to do, based on a medical condition, to do less than a lobectomy. But as Frank mentions, we're coming through a lot of emphysema here, so this is Swiss cheese. Even if we reinforce the staple line, even if we put the only approved lung sealant on it, it will, they'll have prolonged air leaks. And that's where you get into the fact that you don't typically get a significant difference in short-term outcomes between lobectomy and sub-lobe arbor sections because of that potential for air leak. You're also still dissecting out the pulmonary vein, and so your potential for atrial fib is pretty close. Not the same, but pretty close. So again, and in this case, we would take a little bit of the backside of the apical segment, the right upper lobe, just coming through that parenchyma. For the surgeons in the room, if you're doing it robotically I typically, I don't always use Firefly, but I come through well-vascularized pink lung tissue. I don't come in the gray to pink margin. Y'all want all that where the staple line is in healthy tissue, no different than when you're doing an asthmatic resection. There is this study that was out of Belgium where they did, looked at segmentectomy or lobectomy for early stage, and so they again looked at it this way. They looked at all stage one patients, stage 1A patients, and then the subgroup of two somnometers or smaller, and they looked at, and Erin nicely pointed this out, the difference between overall survival, cancer-specific survival, and recurrence-free survival. The bottom line is, and we've already covered this, in stage one, it favors lobectomy. In stage 1A, not so much, and specifically, in stage 1A, less than two somnometers, that bottom, that fourth spot, you couldn't straddle the line any better than that, right? I mean, so there really is, for those sub-two somnometer tumors, if it's in the right anatomic location, your surgeon can offer that patient every bit as good a cure with a segmentectomy as a lobectomy. Mentally invasive surgery is supposed to be more of the rest of my talk about it, and this has kind of been the push of my career. I was an early adopter of VATS lobectomy. I now do robotic lobectomy, but in general, in the US now, with the STS, they have a quality metric, sorry, I'll go back, if you aren't doing most of your lobectomies in anatomic resections mentally and basically, you can't be a three-star program. I mean, it's a requirement for that, so it's a quality metric in our systems. Standard approach, again, I thought the room would be mostly not surgeons, so when we're doing mentally invasive, VATS typically, this is the most common, although some people are doing Singapore, kind of a triangulated approach. The big difference is, small incisions, no rib spreading. Who in the room would want to have the open thoracotomy, and who wants the mentally invasive, right? And this was the instance, my first few years of my career, why I was an early adopter, because when patients came in the room, they couldn't stand up straight, when I saw them three weeks after the operation, said, there's gotta be a better way than this, okay? So robotics, they're all in a line. Most commonly, this approach, there are some people doing Singapore robotic as well, but the vast majority are four ports, eight millimeter ports, along the ninth rib, and then an assistant port, and we use an aerosol system. We knew, more than a decade ago, that mentally invasive had lower complication rates, shorter chest tube duration, shorter length of stay, but what we didn't know about was, is it as good an operation, and how are the patients doing? This study, looking at the STS general thoracic database, that's gonna be a subset, and that's why I talked about high volume centers, because centers that don't do high volume aren't gonna pay them extra money to have the general thoracic database, as well as the cardiac database. And so, as you can see, large numbers of cases, 1,200 robotic lobectomies, 12,000 VATS lobectomies, so robust numbers. Between the two, median length of stay, four days, no difference. Here's the important thing, when you're talking to your patients about the risk of a lobectomy, in the modern era, it should be less than 1%. Here are 12,000 VATS lobectomies at a 0.3% intraoperative mortality in 30 day. It's 0.6, less than 1%. And that's where it should be, and we're not gonna get to zero, but with proper patient selection, and a focus on outcomes. Most of the patients go home, right? So your other patients, they're less worried about a prolonged air leak, but they don't wanna go to the nursing home, because they're always worried if they go to the nursing home, they may not get out. Okay, and the conclusion from this was, there's no difference, whether you're doing it robotic or VATS, as long as the operation that you do has that approach to hyaluronidase and lymph node dissection, and doing an anatomic lobectomy. This was the VIOLET trial coming out of the UK, so this was a prospective, randomized trial, looking at VATS versus open surgery, so they were randomized after they were determined resectable, they were randomized to either an open or minimally invasive lobectomy, and the primary endpoint, which was an interesting endpoint, they were looking at quality of life, functional capacity, at five weeks after surgery. Commonly, I do four weeks post-op, so this would be just a week later. And they did, you can see there's a difference, but the question would be, how clinically significant is that? The thing to note here is, all the things that come into play here about how do you manage pain, what are you doing in the conduct of the operation, how big is your access, and all those things that aren't really laid out, but what you do see is it took six months for the open cohort to get to where the patients were functionally at two weeks after a VATS lobectomy, and we know that in our practice, those of us that do thoracic surgery. The other thing to note is the adverse events in the hospital, 44% in the open versus 33% in the minimally invasive, and serious adverse events after discharge, 38% versus 31, and for the US, this is important, 29% readmission rate for the VATS route versus 36% for open surgery. Your administrators would not be happy with the 36% readmission rate, I can promise you. Having been the Vice Chair of Quality, I'd be having a conversation with you. Here's another one, they had a significant, and I talked to you about that, in the open surgery group in this UK cohort, they had five deaths out of 250, so that's a 2% mortality within 30 days. That draws into question patient selection, management, so I showed you a 12,000 cohort across the US, it had a .6% versus 2%, so when we're reading these studies, you gotta think about the larger picture, so patient selection's important in how you manage them, and it's not just an anatomic resection. So in summary, surgery for stage one is best practice. That data is unequivocal. Sub-lobar resection's a good option for tumors less than or equal to somers, but location matters. Minimally invasive surgery is best practice in stage one, preferred in stage two, but not data, that's not, the data's not as robust there. No matter the age of the patient, treatment over no treatment. In the US, the number of patients over the age of 80 that get denied treatment because of age is depressing. So, thank you. Thank you. Good morning. My name's Greg Vedetic. I feel Friday at five. My surgeon's calling me saying, I've got an 80-year-old I'd like you to see right now that I can't operate on, and I've got to do it in five minutes. So I appreciate your patience by letting me speak to you as rapidly as possible this morning about why stereotactic body radiotherapy has been an amazing advance in the world of caring for the patients that we look after, because as we've said, we want to cure their cancer, and we also don't want to hurt them at the same time. I'm a radiation oncologist at the Cleveland Clinic, where I've had the privilege of working with a great thoracic surgery group who've actually helped advance stereotactic, because they've actually been invested in patient care and want to make sure that their patients are well looked after, because this is basically who I'm seeing at Friday. And they come in, and they want to be well looked after, and the word on the street is unless you get surgery, you don't get cured. Well, the point that I want to make to them is you will get cured, because we're going to do the right therapy for you for your cancer as well as for you. I do want to remind us all that lung cancer is a bad cancer. It's not the most common, but it's the most lethal. And as we've already heard, part of the problem we know is we don't see patients early on. We're talking about early stage lung cancer, which is no negative disease. It's that little block at the end of that histogram. And the other terrible thing about even early stage lung cancer, even within stage one, that survival keeps on decreasing, because as we've heard, it doesn't like to behave exactly according to plan, and metastatic disease happens relatively early. So you've got to have this idea like balancing, am I going to look after what I see without paying attention to what may not be seen? So local control and doing the right thing locally is important. But as Dr. Gillespie mentioned, we've got to look at what's happening after retreat. So this debate between surgeons and radiation oncologists really is relevant only if we think about the future for these patients. So as we know, this is no negative early disease. Ultimately it's going to be picked up, usually incidentally, I've had tons of patients who've had car accidents who get an early stage lung cancer. And then with screening what's going to happen, we're going to find more early stage disease. So if you have 250,000 patients a year, maybe 10 to 15 or 20% of them, you've got a lot of people with early lung cancer, and there's no question the standard was set by surgery. They've done the trials, they've told us how to expect outcomes. Is Dr. Detterbeck at work? But you know, how gold is gold, and we know gold varies according to where you find it. So yes, I want the right thing for my patients based on what I know about their cancers. But don't forget, and we just heard this from Dr. Howington, if you walk in and tell somebody you're going to be worse from your operation than your cancer, it's very discouraging for most patients. So it's not that surgery or radiation is better or superior or inferior, it's what's the combination that works best for our patients. Because this is the guy I see on the weekend also who's being referred from pulmonary after a COPD bout has been settled, and he's found to have a new growing nodule. So nobody in the surgical world wants to look after him. As for my pulmonologist colleagues who support our patients, most patients have baggage, heart baggage, but mainly lung baggage with lung cancer. And this has made the issue of safety as important as cancer control. And that's what these other conversations were about, about the role of surgery, how much surgery, how harmful that surgery could be. And we have something also that's designed to be less harmful. It's really important to understand that this is not done in an absence of understanding cancer. I just don't have a hammer nail syndrome and somebody sends something and I zap it. I look to treat the right cancer, and I'm very lucky to work in an institution where so many components of the care, in particular the role of EBUS in terms of staging mediastinums, is part and parcel of how we make a decision that really high risk or inoperable patients will get the greatest care if they get looked after by a radiation oncologist. So I want to dissuade people from the notion that we burn people. These are old pictures from an old textbook. This sort of cloud that's around a tumor represents 1950s style, where a lot of the lung was being treated for the cancer. And then as things evolve with technology, not by the radiation working better, but by aiming and shaping better, you have this diamond shaped thing. But effectively what happened about 25 years ago is very smart people in Sweden who had developed brain radiosurgery. So this is a surgeon, a neurosurgeon who developed brain radiosurgery. This is what's called a gamma knife. People have heard of gamma knife. And basically these same Swedes transposed this to the chest. So essentially I end up focusing on the lung tumor after proper staging, like the brain surgeon does on the brain tumor. So I end up with this kind of thing, which is really different from what you saw before. I'm like a super sniper. That's how I talk to my patients, so that they have an idea of two things. I'm going to kill cancer, and I'm not going to kill your lung. So stereotactic therapy, SABR, SBRT is like 25 years old. So we're still young, vibrant, exciting. And one of the things that really developed early on was how well this worked at getting rid of the cancer that was seen. And it did this across a whole range of retrospective studies. So you're talking about control rates in the 90s that previously had never been seen before with radiotherapy. And the other thing that was going along with this, again, because matching benefit has to match patient outcomes, was that this is non-toxic. So again, we've talked a lot about how surgery can have an impact on outcomes, while radiation can be detrimental as well, as I showed you early on. And one of the interesting things was the very low rates of severe toxicities which were coming out of the use of SBRT. And the good thing is that, just to jump a little bit, this has been really established not just retrospectively, but it's come out of some prospective studies as well, that over the past 25 years, by doing prospective work, we've identified that the safety, and in other words, the lack of lung injury, permanent lung injury, or permanent detrimental effects to patients, is also correlated with their cancer outcomes. So again, this is not so super data-heavy, but it is that people still want to understand their therapy. And again, just in the same way that lobectomy versus sloboblobar resection has been tested, it's actually been tested that SBRT is actually outcome-driven, not just guesswork. And there's a very important study that came out a few years ago, randomized phase 3, that looked at stereotactic radiotherapy, comparing it to conventional radiotherapy, for early stage lung cancer. And not only was the control better, but the overall survival was better. So in other words, we get rid of cancer, and we get rid of cancer in a way that's more effective with distant failure as well, by probably controlling better early on. So we have a standard, and for inoperable patients, most of it was done retrospectively, because a lot of these patients are fragile. But ultimately, you have to do have prospective studies, and this is sort of the landmark study that was done by Bob Timmerman, who originally was in Indiana. And it was a prospective study that was trying to understand, does this really work if you do it in a multi-institutional setting? And it was a endpoint of control. And what he found in this small study, but really important one, because it was achievable to do this high-tech therapy in multiple institutions, was that the control rate was really good. So that's what you want. You want to just be like a surgeon. You want to get rid of what you see. The failure rates were just like surgeons. In other words, we saw failure rates, even recent trials, where it's like 16% to 20%. So we're seeing the same kind of distant failures you do with surgical series. So that's making you feel like you're as effective as a surgeon. And then ultimately, again, that it's as safe as you want it to be. So with some toxicity rates that are in a relatively acceptable zone of 10% to 15%. So what's happened is, SBRT has been really popular. A lot of people want it, even if they're surgical candidates, and I'm not here to promote this over surgery. This is not a debate of surgery versus radiotherapy for early-stage patients. But the point is, is that we had to set parameters which showed that we're as effective. And what's happening now in the stereotactic world is a couple of things. We're trying to understand what we do better. This is a patient I treated a number of years ago with a very large tumor. And you'd think, how in the hell did you do that? But I did do it, and I did it safely. And he had excellent local control out of this, and unfortunately, had distant failure. No surprise, unfortunately, because of his primary. So when we look at outcomes over the past 25 years, in the same way that we want to demonstrate that we're effective, we have to look at our local control, which has been excellent. We want to look at our cancer-specific survival, and the reason that is, is because we're dealing with those old patients who actually have a hard time living beyond their other comorbid issues, it gets very difficult to compare head-to-head with surgical series, so that even though I can say, yes, I get rid of your cancer, the hard part is a lot of these patients succumb to things like COPD exacerbations. And that's appropriate for that population, because ultimately, again, you want to cause no harm for the benefit of getting rid of their cancer. So we control it, the cancer is well-managed, we don't tend to fail locally, and the toxicity is relatively minimum for the therapy that we're offering. And we also have done quality-of-life studies that have shown that not just changing the cancer does good, but we actually don't change the outcomes, and these are two prospective studies, one of which I led, looking at early versus a year later, and there were no changes in any of the components of that measure functionality. So again, reassuring that we're not just thinking we're doing well in the long-term, but patients who are followed actually do quite well. This is my anecdote, this is a lady I put on a trial, she was inoperable as judged by our team, she had an early squeam of the left upper lobe, I treated her, and that's the 10-year-later picture. The curious thing is on the right, she actually developed a small cell that I treated as well. So you know, you cure people, but you don't stop following them, because unfortunately lung cancer brings risk of more cancer. So we have a standard for medically inoperable patients. The advances we're trying to do, besides our own technology, is exactly what Dr. Gillespie was mentioning on adjuvants, so there are studies with immunotherapy now, because most of these patients cannot tolerate anything like chemotherapy, but these are things that are in the works. So as it stands right now, if you have an inoperable patient, we're very happy to offer them cure. Because this topic was on general therapy, I didn't want to forget mentioning something that may be familiar to many, as well as the pulmonologists, and that's the use of ablative therapies. I just wanted to mention them because they're less frequently used, depending on your institution, and our institution elected to focus on stereotactic therapy, but there are different ways of skinning a cat, and this is using a physical probe that is inserted into the tumor and destroys it by either microwave or radio frequency, or in some cases, cryoablation. The only thing I want to mention is, as much as these are interesting therapies, they are physically different than something like stereotactic. Obviously there's a big needle that gets put into the chest, so that brings its morbidity issue, which has had some impact on maybe whether or not people want to offer this to high-risk or inoperable patients. And just to summarize, there have been comparisons between these two therapies' approaches for inoperable patients, and although generally speaking for small cancers, both of them seem to work really well, the complication rates with SBRT are significantly less, basically because they don't violate the patient in any way, and it appears that when you look at the cancer control over the long term, it's probably somewhat better because we treat every tumor, whereas smaller tumors for the RFN are the only ones that are really appropriate for that therapy. So the point that I want to make in this kind of overall view of why SBRT or SABR has become so popular is because it gets rid of cancer, and it does it very effectively, but it also does it with very minimal morbidity, and there's some fascinating work that's going on comparing surgery with SBRT in more, I'll call them less risky patients, whether they're high-risk patients or even in some settings with operable patients, and those are valid studies, they're often being led by surgeons because they actually want the same outcomes that we do, which is getting rid of the cancer and doing it really well. So with that, I'm very grateful to have been part of this platform and to be at this meeting today.

early-stage lung cancer

treatment updates

screening

low-dose CT screening

molecular medicine

immunotherapy

specific treatments

surgery

radiation therapy

patient outcomes