Good morning, everyone. We're going to get started. My name is Eric Folk. It's a pleasure to be here with colleagues and friends, and I want to introduce you, my two colleagues who will be speaking after me. And first off, Dr. Katherine Oberg. She's the director of interventional pulmonary at South Bay in UCLA, and the associate program director for the interventional pulmonary fellowship at UCLA. And second, Dr. Raymond Osorio-Gabon, who doesn't need introduction. He's a legend in his own right. I had the pleasure of having met him in Memphis when I was working here. Many years ago, I set footing in his tumor board, and he quickly picked up there was someone who didn't belong, and he's like, who are you, and what are you doing here? And I said, hi, I'm new in town, I'm in a private practice, and I came to learn. And that's how we became friends, and I learned a lot from him. So he's the chief scientist for Baptist Memorial Healthcare in Memphis, Tennessee. Both of them have illustrious career, and I'm honored to be sharing the podium with them. So without any further ado, the talk that they have asked me to direct today is Don't Stop Until You're Getting Off Tissue, talking about adequacy of lung cancer diagnosis in 2022 with small samples. These are my disclosures. I serve as a consultant for multiple device companies, including Boston Scientific, Medtronic, and Cook. I'm the global PI for Navigate and Precise, both of which are funded by industry, and I have an app for EVUS staging, which the proceeds go to the AAVIP. So this is the spectrum of interventional pulmonary in a mind map, and I always start with this because what we do has been expanding and expanding over time. So today I'm only going to be speaking about flexible bronchoscopy biopsies, as well as EVUS, navigation, and robotic, a little bit of robotic, because it is the latest topic in our momentary. So a good speaker once told me, you have the most attention in the first two minutes, so use them wisely. So here's my conclusion. Does bronchoscopy and EVUS providing off tissue for molecular testing? Absolutely yes. The question is, what alternatives do we have when we have situations that are difficult? Difficult access, difficult patient, lack of human or technological capital, or prohibitive risk. And that slide, this picture that you see on the right, is a cell block from discarded material. I had my cytopathologist step in the room, Paul Vanderland, many years ago, and I was throwing away the suction syringe, and he said, let me just quickly look at what you're throwing away. And this is the discarded material. So we have plenty of tissue. It's just how we use it and how we process it that matters. So I'm going to try to convince you in the next 18 minutes of that. So as all of you or most of you would know, we can do flexible bronchoscopy through moderate sedation, deep sedation, LMA, ET tube, et cetera, et cetera. We can get with endobronchial ultrasound to central lesions, whether they're in the lung parenchyma or the mediastinum. And we can obtain, as I said, enough and adequate tissue. So when we use this same tool in the airway and in the esophagus, we can provide better staging and more information for our colleagues, both the microbiologist and the infectious disease, the cytopathologists, and our colleagues in thoracic oncology. In this mind map, I want to show you the diagnostic methods and considerations that many of us take when we're asked to do a biopsy, not only the considerations of what is the probability of cancer, the risk of complication, the diagnostic yield, and my area of expertise or my personnel in the room, also the patient's comorbidities and their Carnovsky scale. But I take into account whether this is a central or a peripheral lesion, whether it's in the mediastinum, or whether it's a pleural effusion and the diagnostic yield may be good with pleural fluid. I will not be talking about the pleural or pleural fluid today, but this is what goes on in our minds as we approach this problem. And with that, I have to say, we all want 100% certainty, yes. But that is obtained by surgery. And surgery has the highest risk or the highest risk of complication. Now if we are risk adverse, we want the minimum. So we go to sputum. Well, sputum gives us very little diagnostic certainty, as Paul Vanderland says. So we have to find a tradeoff between risk and benefit. And when we do that, we obtain small biopsies. Small biopsies do not give you all the architecture that large biopsies give you. For example, they don't tell us details like adenomatous hyperplasia. They don't tell us when we have adenocarcinoma in situ. We have to conclude that from the clinical information. But a resected sample gives you more architecture and more about the surrounding areas. Having said that, a small biopsy still allows you to differentiate cancer versus no cancer, allows you to differentiate adenocarcinoma versus squamous, and of course adenosquamous, and other types of non-small cell lung cancer, and even small cell lung cancer. Having said that, what is the path that we follow? Well, when we first approach the patient, not all biopsy tools are created equal, even though we just say needle and we say forceps. There are different kinds of forceps, different kinds of needles. Recently, one of my favorite papers is this article by Chris Manley and his group in Fox Chase in which they did a randomized trial of 22 and 19-gauge needles for safety, diagnostic yield, and molecular profile. You can see here that whether they process them as a core biopsy or they process them as a cell block, the results were indistinguishable. Not only that, but the size of the needle didn't matter. Not for the diagnosis, not for the quality of the sample, not for separating blood, and not for moleculars. So maybe the tool is not as important, at least not in this randomized control trial. However, if we look at the GI literature, we will find that they use different needles. They use cutting needles, different than the regular needles we normally use. And some manufacturers have actually come up with needles that are frenzy needles or provide you actually biopsy rather than cytology. So now I'm getting on a murky thing where half of you are saying, I'm not even sure what a core biopsy is. So let me go there. But before that, different sizes of forceps will give you different size of tissue, not necessarily a higher yield, but different size. So keep that in mind. Now when we obtain a small biopsy, and this is where I get into the core concept, we can do a cytology, which would be a smear, whether it's air-dried or alcohol-fixed thin prep. And we cannot do immunohistochemistry or moleculars on this material on a commercial basis, on a regular basis. For that, we need to transform that sample into a cell block, cut it as if it was tissue, so spin it, cut it, and then treat it as tissue as a histologic sample in which we can do an H&E immunohistochemistry. And then the part that is called a quality assessment by the pathologist. And they determine the degree of cellularity, and they determine if it's a sample they want to spend the time and money into doing next-generation sequencing or PCR or immunohistochemistry. So with that concept, where does the core concept come from? In 1950, a scientist did a breast biopsy using a fine needle and just jiggling it in the breast and saw that he could obtain a diagnosis, a cytologic diagnosis of breast cancer. However, if he used a bigger needle, he obtained more architecture, and he called that a core. And that's it. There's no definition exactly by the size of the needle or by the type of the needle. A surgical biopsy is clear what it is, but a core needle is an elusive concept because of that origin that it has. Having said that, let's look at the landscape of alterations for non-small cell lung cancer that have targetable mutations that we can address with medications. And nobody better than my colleague Ray here to talk about this. Every year I have to update this slide. But just keep in mind that only 50 percent of people will have a targetable mutation at best, and some of these targetable mutations don't have the exciting outcomes that we know for EGFR, for example. So keep those in mind, and they keep changing every time we talk about this. Now this slide shows you the expected frequencies of the different genomic alterations in lung cancer. They vary by geographic area, so keep that in mind. EGFR is one of the most common at 14 percent, and KRAS in this series at least is 29 percent. And this is a combination of six series, six large studies. About half of the patients will not have any genomic alterations. So we did this study in 2013 in which we compare CT-guided needle biopsy versus, I'm sorry, this is a retrospective study, so it has its inherent biases, but we did CT-guided biopsy versus transbronchial biopsy. And we found that the quality of the tissue was not different in any of them from surgery except for CT-guided. CT-guided was inferior, but we found out that it was probably because of the processing. This was for single gene testing for EGFR, ALK, and KRAS at that time were the targets that had been approved for treatment. So maybe it's not the tissue, but how you process it. Yes, absolutely. There are best practices to process the tissue immediately as you obtained it, and I will not go into that today. But just to say, this meta-analysis shows in this forest plot, you see that the EGFR per region of the world matters. The average is about 95%. In small samples, you can have enough tissue to do EGFR in 95% of the cases, you will have adequate samples for EGFR. But that varies from sites of the world. So North America and Eastern Europe have a lower prevalence than Asia, and definitely lower than Australia. Same thing, this is per region, and same thing for ALK in a forest plot and the prevalence of this aberration in those populations. Having said that, what happens after we treat those patients? Well, we may have acquired resistance, being the most common T7-90M. How do our tests with small samples perform in these conditions? Actually very well. Not only can we identify, like I said, T7-90M, we can identify ROS1 translocation or a histologic change to small cell, which is really rare. And this will determine, in clinical trials, the sequencing of medications that we use. Because as we know, TKIs have been through several iterations, and we are now at third generation TKIs. Now our experience at Beth Israel regarding T7-90M in small biopsies is more than 50% of the biopsies are actually small biopsies, and the performance in these small biopsies is very reliable. So now we know it works on the initial biopsy, it works on re-biopsy and acquired mutations, and it works in PD-L1. This is what a pathologist does when we ask them for PD-L1. They actually check for a TPS, or the tumor proportion score, or the percentile of tumor cells that stain. And as you can see here, less than 1% looks like the top image, 1 to 49%, and then more than 50%. And this is the staining that they do and the scoring that they report to us. How do EVOS samples perform in this way? 2% are inadequate. The remaining of the samples are either 100 to 1,000 cells or more than 10,000 cells, which is what we hope to obtain. If we do immunohistochemistry for PD-L1, anti-PD-L1 antibody, you'll see that 4% are insufficient. So now we know it's good for driver oncogenes, it's good for PD-L1. How about when we compare it to surgical specimens? Well, it depends the point that you ask. In this study, they saw a difference if you requested more than 1% or more than 50%, depending on the assay of PD-L1 that you're actually challenging it to. And there is no statistical difference in this study between obtaining a sample with EVOS, EUS, CT-guided, surgical, or plural. Now this is a recent paper by Dr. Oberg, so any questions, I ask you to direct the questions to her. This is a combination of shape-sensing robotic bronchoscopy and cryobiopsy. Her group actually, in this image you can see in the insert, is A is a regular needle, in B is actually a forceps biopsy, and in C is a 1.1 cryobiopsy. And they also looked at the diagnostic yield based on the size of the nodule and the biopsy modality. And the incredible part of it, or the really exciting part of it, is they proved the concept that it's not only navigating to the lesion, but using the right tool. The diagnostic yield was 90%, with 18% of the nodules being diagnosed only by cryobiopsy. So we are now in the territory of 90%. This is a molecular testing sample obtained by robotic bronchoscopy by the MSK group that is coming out soon, and they just did 128 malignant samples, and they showed that there's adequate tissue, as reviewed by the pathologist, in 108, and you can see that the failures were very small in this type of biopsy. So my summary is, we're providing off-tissue, adequate, and that the bronchoscopist and endoscopist is the only one who can determine if the lesion can be amenable to a biopsy. Thank you very much. Thank you, Eric. Good morning, everyone. Thank you to the organizing committee for the honor of this invitation. So is tissue still the issue? We have alternatives, and are they ready for prime time? So unlike Eric, I'm not going to give you the punchline first. I will save it for the end. These are my disclosures, none of which I think is pertinent to my talk today. So my objectives are to give an overview of biomarker testing in lung cancer today, talk about challenges with tissue-based testing, talk about this new kid on the block, blood-based biomarker testing, pros and cons of each, and maybe how we can make this work at home, and where are we going with all of this. Now we will talk about where we came from, a brief history lesson. Now is the winter of our discontent made glorious summer by this son of York, to quote Gloucester and Richard III. So this was our winter of discontent. We were in medical oncology just about ready to go hang ourselves when this paper was published. Our first, we were excited. We had had platinum compounds for non-small cell lung cancer for decades, always paired up with a doublet, and then we had all these new drugs that came, and we wanted to pair them up with platinum. And we thought anecdotally that we would really hit a home run. And then, of course, this paper was published, and it showed no daylight anywhere. This indeed was our winter of discontent. And then subsequently, Giorgio Scaliati's paper looking at pemetrexate, which was yet another new kid on the block at the time, we were really super excited about it, combined with platinum, did a clinical trial. So the first, the top survival plot on the left, no daylight. But the interesting thing was there was some preexisting understanding that the mechanism of action inhibiting thymidylate synthase, thymidylate synthase tend to be a little bit more concentrated in squamous cell cancer. So there was this idea, this a priori idea of testing whether there was a differential activity on the basis of histology. And that proved to be true. As you see in the middle slide, at least the progression-free survival for patients with non-squamous non-small cell lung cancer, the pemetrexate doublet, which is the CP, seemed to be slightly superior to the non-pemetrexate doublet. And you saw the flip with squamous cell cancer. So for the first time, we had reason in lung cancer to make a distinction between subsets of non-small cell lung cancer. If you recall, we used to just lump them all together. There was nothing we could do to make a difference. This was the paper that made it where we began to have reason to torture our pathologists to tell us, is this non-small cell, is this an adenocarcinoma, a squamous cell cancer, and so on. And subsequent to that, of course, the tyrosine kinase inhibitors of EGFR came about. And that, too, was its own sad story. Lots of anecdotal evidence of high-power responses, did a clinical trial, some promise, and then poof, at least the first drug failed in its phase three trial. So subsequently, we learned that there were people, indeed, who responded well to these drugs. They seemed to be female, relatively light tobacco users or non-tobacco users, predominantly Asian. So this clinical trial, the I-PASS trial, was done that enriched for people like that and compared the use of chemotherapy to the use of an oral-targeted agent, gefitinib, at the time. And you see here, if you observe closely, an interesting phenomenon where the chemotherapy started out looking superior, and then there was a cross, and then the oral-targeted therapy began to look superior. When the trial started, we didn't really know what it was that made these people respond. But somewhere early on in the trial, the idea of activating mutations of EGFR came about, and then there was an effort to test people retrospectively. And when you sorted people out who were tested, who had an EGFR mutation or not, suddenly you found the story becoming very clear, that those who had the activating mutation of EGFR had a better response to the oral-targeted therapy, and those who did not have the biomarker had worse survival. So this was the trial that made Tony Mock become an international superstar. And then subsequently, the principle was proven in this Japanese trial that only enrolled people who had an activating mutation of EGFR and clearly demonstrated that, yes indeed, the biomarker predicted who was going to respond and who was not. And since then, of course, the times have changed. So here is William Powell's famous paper talking about chipping away at the lung cancer genome that demonstrated that at that time, when this paper was published in 2012, almost half of all the adenocarcinomas, advanced adenocarcinomas, we really didn't know what the gene abnormality in them was. And then a few years later, Lynette Scholl published this lung cancer mutation consortium paper that showed you how the no-oncogenic driver crowd had shrunk a little bit, not quite 50% anymore. And here is my back-of-the-envelope calculation in 2021. After the KRAS G12C mutation became druggable, after we knew that patients who have a high level of expression of PD-L1 tumor proportion score could be treated with first-line monotherapy with an immune checkpoint inhibitor, suddenly that unknown world is no longer about half. It is now about 11%. The times there not only are changing, they are changing fast. So you see the cadence here, it is accelerating. Biomarker testing makes a difference, not because it gives us cool toys to play with, but because it saves lives. The only entranceway into personalized oncology care is through biomarker testing. And this is an analysis of the Medicare data set that demonstrates patients with metastatic lung cancer, non-small cell lung cancer, who had a biomarker test, who had a genomic test done, had superior survival to those who did not. Of course, it is not the test that saves lives, it is the test leading to the treatment, which is what you see on the right-hand side. The superior survival plot had a test, had a genomic discovery, received targeted therapy. And you see, if you don't have those two things side by side, everybody else is a distant second, third, fourth, and so on and so forth. So had a test, didn't get a targeted therapy, you were a distant loser. Didn't get a test, but somehow somebody gave you targeted therapy, well, that's no good. So the point is, you got a test, and then you have to treat appropriately. So that's the world in which we are now. The spectrum of relevance is galloping forward apace now. So we started out with stage four. Using first-line therapy is now easy for us to do. Looking at patients who may benefit from first-line immune checkpoint monotherapy is now something that we do all the time. Either we size there are emerging uses of biomarker testing to monitor treatment response, to detect emerging resistance. But it's now extending all the way through to adjuvant therapy, and even to the realm of early stage, early lung cancer detection. It's probably going to find its way in screening and management of incidental lung nodules. In the adjuvant therapy setting, yes, doc, how did the surgery go? Oh, I got it all. Well, we know oftentimes time proves us wrong, okay? And we know that adjuvant therapy has had a role. It has been chemotherapy. But now you see with the ADARA trial that patients who have had curative intense surgery for lung cancer, whose cancer expresses an activating mutation of EGFR, have that survival benefit, at least in terms of disease-free survivorship, compared to those who did not get the pill. You could drive a truck between those two survival points, unlike the one I showed you with Pemetrexid at the beginning where you needed a microscope to see the gap. We're doing better. And not only are we doing better with targeted therapy, here is the Empower 010 trial that shows you patients resected lung cancer after receiving chemotherapy who have a PD-L1 TPS 1% or greater respond pretty well to adjuvant therapy with this particular checkpoint inhibitor, atezolizumab. Fast-changing times, we're excited. But there are challenges. Biomarker testing is still not everywhere across the land. We know that biomarker testing saves lives, allows us to give tailored treatment to people, but still the majority of people who are candidates for it don't get it, don't get the treatment that that leads to. We know that disparities are already emerging. We know there's horrible geographic heterogeneity in access to biomarker testing. The logistic delays in procuring tissue, frail patients may be difficult to get tissue from, some biopsy sites are relatively inaccessible, the skill sets to access them may not be there. And then there's this infernal problem of too long turnaround time. It takes too long to get the answer. And then in the new adjuvant setting we're beginning to see there may be less of a challenge here and certainly in the adjuvant setting this should not be a big deal at all. So we have gone from the religious paradigm, no meet, no treat, if you don't give me histologic proof I will not treat a lung cancer patient, to where we're comfortable with the idea as our treatments become more effective and less toxic. Stereotactic body radiation therapy for example has taught us, yeah, it's not everybody who has to have a biopsy, go zap it, the patient will thank you years later. Okay, so we're becoming more and more comfortable with less and less material to work with. So some of the problems with tissue-based testing though, invasive tests, you need equipment, you need the skill sets, not everybody's an Eric Falk, okay. There are complications. Hyper variability in quality of specimens. QNS, those are the three worst letters when put together in my clinic. Turnaround time, tumor heterogeneity, the place you stick your needle in and suck out material does not necessarily represent the whole. We're learning all about that. Challenge of repeated procedures, patient anxiety. Well, blood-based genomic testing obviously, if we can make it work, would be much more patient-friendly. A blood test is easier than anything you're gonna do poking at a patient's chest. And you can use it for multiple different things, for diagnostics, to do your predictive assays, even some of these prognostic things. There's emerging evidence that what you get is not exactly what's in the tissue. It's not a 100% overlap. It actually looks like it may be a complementary test. So it's not either or. It may be this and that. Charu Agarwal at Penn has done some work trying to demonstrate that to us. And there's also the idea that the cells that escape from the tumor and circulate the DNA that you get may be a marker of the worst of the worst. The jailbreak. It's not the innocent person who wound up in jail. It's that hardcore criminal who's most likely to escape. There is that aspect that's emerging as well. So the advantages of plasma genomic testing we're beginning to see. So here's a tiny little single institutional study. Just, you know, looked at a prospective cohort of about 55 patients. Match that up with an immediate retrospective cohort of the same number of patients, same single institution. This was Penn. And so cohort one is the prospective blood-based biomarker tested cohort at the time of a biopsy. And then cohort two, the conventional reflex tested tissue genomics. And you see that the spectrum of genomic abnormality between cohort one and two is essentially the same. But what you find is the timeliness of having your result at the time of the first oncology visit. There's a landslide. And in terms of the therapeutic decision-making that you can do at the time of the first visit, there is another landslide. And if you talk about how long does it take for us to go from yeah, you got a cancer, to well, here's your treatment. Much shorter with cohort one. So, you know, blood-based biomarker testing has great advantages. It is new, so it is more expensive, and we're having to figure out how to make this work. Initial coverage was when you can't do a tissue-based test. You already tried, failed, patient doesn't want to do it, patient's too frail. Here was an option. But you also have the possibility that, you know, if this test becomes less expensive and more readily available, maybe we can begin to do it concurrently with some of the other things that we do. So the pragmatic use of this, it started out when tissue was not available, but now a lot of us are doing this sequentially, where the blood-based test is actually the first we do. And if it gives you an answer, you can run with it. If it gives you no answer, then yes, you can go on and get your tissue. And then there's this idea that, well, they are complementary tests. Maybe it's not either or. Maybe it's this and that. So my take-home message is lung cancer is rapidly fragmenting into splinters. It's no longer a monolithic, non-small-cell lung cancer. It is becoming genomically and biomarker identified subsets of disease. Biomarkers are the jackhammer that's splintering lung cancer into tiny little bits. They identify patients who can go into very different treatment pathways. But with that good, that excitement, comes the challenge of access. Access. Whatever we can do to make this easier for people to gain access to, whether it's a blood test that's easy, safe, cheap, expensive, effective. And it's expanding across the spectrum. Biomarker relevance is extending itself across the spectrum. And now we see, of course, that most lung cancer clinical trials, you can't write a protocol and send you through CTEP without a biomarker correlative study. It'll be turned back to you. Biomarker testing is here. We need to embrace it. So is tissue still the issue? My answer is yes, but. Thank you. All right. So thank you so much for having me. I am going to go through the little, tiny, narrow subject of staging and a little bit about NGS in 20 minutes. So these are my disclosures. I am a scientific consultant for various companies. I'm in the site PI for a plural trial on empyema, and none of which I think is relevant today. So again, this narrow subject of staging, you know, in 20 minutes, obviously, you can't go over everything. So I decided to pick out a few things that I think are relevant and pertinent. Why do we do it and the methods? Who needs it? Which I think is one of the biggest points. And then a bit about the frequency of and quality of staging that's performed. And then just a tiny bit on NGS. You know, that's already been very well covered, but I'm going to talk about a few studies that look at practice patterns out there about NGS and how well we're doing at performing NGS. We're all probably very familiar with this, right? This is the TNM, the 8th edition for lung cancer. And we know that nodal staging and metastasis plays a huge part in this at dictating what your stage is going to be. And stage matters because we know that survival is linked to what your clinical stage is. And so 1A is very different from, you know, metastasis stage 4. And so this is why, of course, we need accurate staging so that people get accurate treatment so that they have the best shot at survival. When we're looking at staging a patient, we always want to look at the highest stage first. So if you have a liver that looks like this, if you have a pleural space, this is a thoracoscopy I did recently, a pleural space that looks like this, where the visceral and parietal pleura are studded with tumor, then you have your diagnosis and your staging all in one procedure. So if you can do something like this, perfect. It's one procedure. You get the information that you need and you move on. But commonly it isn't like this, which is great. That's what we want, right? We want it to never look like this. And the better that we're doing with screening and managing incidentals and these types of things, we're seeing less and less of this, which is good. And so that leads us to the mediastinum and the hilum when we're talking about staging. There are several ways to do this, to approach this, of course. So we have needle-based techniques like endobronchial ultrasound or E-bus or endoscopic ultrasound, EUS, and things like mediastinoscopy, of course. I put PET-CT up here really more to just say that PET-CT alone does not obviate the need for invasive mediastinal and hilar staging. And I think this slide is very representative as to why. So we know that PET-CT alone, when you're looking at staging, is just not that great. In this study that was done, this is a meta analysis of several different studies looking at PET-CT for staging, you can see that the median sensitivity for malignancy was only 62% in this meta analysis. There are a lot of limits to PET-CT. The limits of detection are in the range of 7 to 10 millimeters. We know the anatomic resolution is not great. We don't have standardized criteria about what is really a positive PET scan result, and we don't have ideal cutoff points for SUV, right? They differ from place to place to place. So there are a lot of limitations when it comes to PET at staging alone, which is why we do things like invasive staging with E-bus, EUS, or mediastinoscopy. You know, knowing your lymph node map is important when it comes to staging, certainly, whether you have upper or lower mediastinum, the hilum, etc. And there are modalities that can get you access to these areas better than others. Everything that's above this V are things that are accessible by things like E-bus and mediastinoscopy, although 7 sometimes can get even a little challenging with mediastinoscopy. When you're looking at the hilum, E-bus really reigns king here, and I will make a very quick argument that I think it's sort of king overall. And then when you have lymph node stations 8 and 9, you're talking about things like EUS. Dr. Folk already showed this picture. It's a beautiful figure that he and his colleagues published in the New England Journal about a case where they used E-bus, EUSB, which is doing EUS with the E-bus scope to diagnose a patient. And this is something that we're doing more and more frequently, right? Because why do two procedures when you can do one? And if you can get more complete mediastinal and hilar staging doing E-bus and EUSB, we should be doing that. This is, you know, I had to check my interventional pulmonary impulses to go down this rabbit hole of, you know, E-bus versus mediastinoscopy and all of that, and I'm not going to go into that a lot today. Really this slide, I think, is a decent summary, and there are many more papers out there about this. Looking at the sensitivity for staging of E-bus and E-bus EUS, which you can see here, which are around 90, compared to something like mediastinoscopy, which, at least in this study, was about 81%. And you can see above this, I don't have an arrow there, but for PET-CT is only 62%. So I didn't even include that. This was looking more at things that I think are valid ways of staging the mediastinum, which would be E-bus, EUS, and mediastinoscopy. I will make a quick reminder that even from our own clinical practice guidelines, these are the ACCP guidelines that were published in 2013, a needle-based technique of E-bus, EUS, or combined E-bus, EUS, is recommended over surgical staging as the best first test. So I think it's important to remember this isn't a new thing. This is something that's been around for a while, and it's our own clinical practice guidelines that recommend this. So who needs mediastinal and Heiler staging? I think this is probably the most important part, honestly, and I always flip this around when I talk about this to say, who doesn't need mediastinal and Heiler staging? Like, this is really what we should be thinking about, right? Is that it's really the exception of somebody who doesn't need to be staged. If you have a small lesion, small meaning less than two to three centimeters, and the cutoff is controversial, and peripheral, meaning the outer one-third of the lung, then fine. I think you can get away without doing staging. If it's small, but it's central, you need to stage it. If it's small, but you have clinical, you know, suspicion on CT or PET of N1, 2, or 3 disease, you need to do staging. You can get away without staging the mediastinum and Heiler when you have other disease sites. So just like I showed you with the liver eaten up, or the pleural space, or whatnot, and I think that's great. If we can limit procedures, that's what we should be doing. But if you don't have that, essentially everybody else needs mediastinal and Heiler staging. And I always get this question, which is why I started incorporating this, which is even with a radiographically negative mediastinum and Heiler, and I'm going to answer that in the next slide or two. When we talk about central versus peripheral, this is what we're talking about. So this white area, the inner two-thirds, is what I would call a central lesion compared to a peripheral lesion, which is the outer one-third of the lung. And it matters because of this right here, right? That central lesions have a higher likely of having regional lymph node metastasis. And there have been, of course, many studies that have looked at this. I've just cherry-picked a few that I think illustrate the point, but there are a lot of them. This is a study even going back to the mid-90s, looking at 337 patients. Every one of the lesions was less than three centimeters. The pre-op CT scan had 91% that looked N0, so it looked like there was no nodal involvement. And ultimately, once they went on to surgery, 26% of these had some type of lymph node metastasis. And a third of tumors that were between two and three centimeters had lymph node mets. A similar story in this, this is jumping forward 10 years, 143 patients with peripheral lesions now are all under two centimeters. The pre-op PET scan, 95% with N0, looks like no involvement of the mediastinum and hilum. And at the end of the day, 17% of tumors that were 1.62 centimeters had evidence of lymph node metastasis in N1 or N2. The story continues. Now we're talking about lesions that are less than three centimeters. This is a pre-op PET, a hundred percent, you know, zero evidence of N2 or N3 disease. And at the end of the day, 16% had N2 once they were actually staged. And here they found that a central location was a predictor of somebody who would have N2 disease. And finally, more recently, a study again looking at very early stage lesions. So this is less than two centimeters, they're T1A or B. The pre-op PET scan, a hundred percent N0. So this is no N1, two, or three. Nothing's lighting up on the PET scan, nothing's enlarged over a centimeter. And ultimately, they found that 12% had N1 disease, 13% had N2 disease, and that tumors that were one to two centimeters or central had higher rates of positive N1 and N2 disease. And again, these are just a few of the many, many, many studies that are out there showing this. So I go back to this slide again to say, who doesn't need mediastinal staging? Because again, if it's a small lesion but it's central, if it's small but there's an enlarged node or something lighting up on PET, they have to have staging. So even with a radiographically negative mediastinum in hilum, yes, absolutely, absolutely. And the data is there to support that. These are the guidelines from the European Society of Thoracic Surgery. They updated them. This is their most recent update from 2014. And they give an algorithm about how you should go about evaluating someone with lung cancer. And you can see here that if the mediastinal nodes are positive on PET or CT, they recommend tissue confirmation with E-BUS EUS. This is Surgical Society of Thoracic Surgeons and they recommend E-BUS EUS, which is awesome. If they have mediastinal nodes that are negative but it looks like something in the hilum is abnormal, if it's a central tumor or if it's over three centimeters, they recommend that you have some type of tissue confirmation. The NCCN guidelines, and this is from 2020, they address this also. In patients with stage 1a disease, so peripheral, N0 clinically, they still recommend that you consider pathologic mediastinal lymph node evaluation. And they mention specifically that there's a low likelihood of lymph node involvement if it's a solid tumor less than a centimeter or a pure ground glass, so totally non-solid less than three centimeters. That here, pre-resection evaluation is optional. But anything else, so this is talking about a cutoff of one centimeter for a solid peripheral lesion, should have staging. So how are we doing with this? This is a study, many of the studies that I have here are from Dr. Osari Yagobon. He's prolific in this area and this is one of the many that I wanted to include. And this is actually looking at this concept of how are we doing with staging. So they looked at patients with curative intent resections from 2009 to 2018 in 11 different hospitals and they found that of almost 3,000 patients, only 22% were staged. 63% of patients who had a tumor greater than three centimeters who had radiographic N1 to N3 disease did not have staging performed. They found that the higher likelihood of staging had to do with certain variables like the operator, so if somebody was cardiothoracic boarded or if they had a thoracic focus or a higher volume of lung cancer patients or at a teaching institution they were more likely to stage appropriately. And they showed that this matters because then they look to see of the patients who had invasive staging and those who didn't what the survival was and you can see there's a significant difference here. I think the optimistic part to come out of this study is that they did look at the rates from 2009 to 2018 about minimally invasive or mediastinoscopy or whatnot staging and you can see thankfully that things are going up and up and up so we are getting better. So the take-home point from this is not a dismal one but I think it's a really important point to say that even now we're still not doing as good as we could do. This is another study from that group just published a few months ago in CHEST looking at non-metastatic non-small cell patients in Memphis and of 882 patients, 52% had some type of invasive staging and only 52% better than the other study, much better than the other study, but still not not perfect. 74% of these were guideline concordance staging and this is something I would have loved to go into but again at the time crunch but things like the patients who need staging are actually getting it. All of those patients who need staging all of them are getting it. That we're doing it in the correct way meaning that we're doing it in a systematic way. We're not doing it just targeted staging based off of something we saw on Pat. Anything greater than five millimeters or more we're sampling and that what you sample actually has lymph node tissue. So all of these things are really important when it talks about how you should actually stage someone. They found that guideline discordant staging is decreasing which is great and they found that guideline concordant staging those patients were more likely to have guideline concordant treatment which is perfect because that's really what we care about at the end of the day. And so overall 61% of the patients had both guideline concordant staging and treatment and they went the next step to kind of hammer it home to say that this really matters. So when you look at overall survival in these groups whether they had both you know guideline concordant staging and treatment versus just treatment versus just staging versus neither you can see the survival benefit here is clear. I'm not going to go into NGS a lot because that's already been so well covered. We know why it matters. We know here this is a study that was recently published that looked at over 3,400 patients and who got immunotherapy versus targeted therapy versus just chemo or best supportive care and you can see the survival curves as we've talked about over and over again are vastly different. You can do this in small biopsy specimens. We learned that with Dr. Folk and so I'll just sort of end with how are we doing because I think it's really important to look and see are we following our own guidelines. This is a study that was just published this year looking at surveys of US oncologists. So this was an ASCO task force that put this out looking at biomarker ordering and treatment practices and you can see the results that they found here. The respondents about half of them saw a low number of specifically lung cancer patients and about half saw a high number. There were 59% academic, 41% community and kind of a variety of practice experience. So you know 20% or so that were less than five years, 18% that had been out 26 years or longer and everywhere in between and I think this is is exciting news. 97% of the people in this survey and granted this is subject to bias that you get from surveys right of the people who answered the survey but 97% were ordering EGFR and ALG and ROS1 and BRAF which is awesome. They did find that other mutations like MET and RET and NTRK were higher in oncologists who were in an academic setting or who were thoracic specialists. Importantly they looked at wait time. So 58% of people are waiting at least two weeks to get these results back and over a third of people are waiting three or four weeks to get their results back. So just like we've already talked about this is a hugely important thing right about wait times and if you're waiting on this to give someone therapy. Interestingly most commonly these were coming from EBS TBNA specimens versus any other specimen. And then finally how are we as pulmonologists doing in this world? So this was another survey this was sent out to pulmonologists general and IP. About three quarters were general about a quarter was IP and then more community than academic. And you can see about half of people were only doing one to four new cases of lung cancer a month and then the other half were somewhere above that. And what they found is that awareness of EBS TBNA guidelines were not perfect. They did find that if you do more than you know 70 bus a month so if you're doing higher volume of EBS, if you're an interventional pulmonologist, if you're in an academic setting or if there is an institutional policy about who you have to stage and not, that you did better with following the staging guidelines. Similar for biomarker testing, IP training academics and an institutional policy in place where you just anyone that gets an adenocarcinoma or squamous gets sent off for NGS had a higher likelihood of getting tested. So in conclusion I would say staging is crucial. We have to be doing this. You have to stage anyone who has a tumor that's greater than three centimeters but you can make an argument for two centimeters. You can make an argument for one centimeter if it's solid and this is based off of clinical practice guidelines which are based off of data. Anyone who has nodal disease on CT or PET CT you have to. A central lesion even if it's small you have to and obviously any distant disease that you're going to go after. There are multiple staging methods though EBS with or without EUS is the preferred staging method. NGS you have to perform routinely. I have we've gone over this so many times but it's just so crucial to say it can absolutely be done on small specimens and we have work to do in this area on really all things. So with that I'll conclude and I think we'll probably take any questions.

lung cancer diagnosis

accurate staging

invasive staging procedures

EBUS

EUS

biopsy specimens

blood-based biomarker testing

targeted therapies

NGS testing