So here are my disclosures again, and just again, courtesy to Dr. Vanessa Shammy, she and I were fortunate to help the ASG out with the EUS diagnostic course that we've done in the past, and some of the slides are taken from that slide deck as well. So again, Vivek asked me to put a top 10, and I think Nuzat had mentioned this before, for primary mucosal staging of upper GI neoplasms, I think a radial makes life much easier. Yes, you can do it with a linear, but why make your life much trickier, especially when sometimes these things can be hard to delineate? We know there's no role for EUS cancer staging with metastatic disease, unless you need tissue from a lymph node or something for some other reason. We talk about tangential imaging a lot, and that's really important. That's where with the radial scope, if you push too hard against the transducer, against the tissue, you can make things look like what I put, T3. So you want to be very perpendicular to the images, and Nuzat very nicely showed the axis of imaging, and that's really important. When you're doing a rectal ultrasound, and we do a lot of rectal ultrasound at our institution for rectal cancer, for large rectal polyps to see if it's deeply invasive, and I also do some rectal ultrasound for incontinence work. I think I'm one of the few left endosynographers that does EUS, or rectal ultrasound for fecal incontinence. So when you're performing rectal ultrasound, you've got to be really careful because the sphincter muscles can melt, especially if you have a low-lying rectal lesion. The last thing you want to do is mistake a T1 lesion and call it a T3 lesion and have a patient go for an APR when perhaps that patient could get a transanal or a minimally invasive surgery, and that was a focus of one of our research studies that I'll share with all of you. I think Nuzat shared this very nicely. I tend to go with an endoscope first for small, subtle lesions that are mucosal-based. For upper GI cancers, you'll be surprised even when the CT doesn't show liver mats, especially if you look in the left lobe right as you're coming up by the G-junction, the left lobe, you can see the heart beating just in that corner there. If you look, you'll often find a metastatic lesion just in that little triangle or that angle that is so common. You always want to assess the celiac, even in pancreatic cancer. We've had probably a series of 10 HCCs, hepatocellular cancers, that have had metastatic celiac lymph nodes and certainly in cholangiocarcinoma. Nuzat mentioned this earlier. I will reemphasize, not only do I say that antibiotics are an absolute must if you must an FNA of a mediastinal cyst or rectal type of cyst, but I would say that unless you have backing from your surgeon and they really want you to perform that FNA, do not perform that in a cystic mediastinal or rectal lesion because we as well have had some very severe cases early on. We don't do these FNAs of cystic rectal lesions or mediastinal lesions because of really bad abscesses and other bad problems. I'll share with you our own study where despite antibiotics with rectal lesions, the risk of infection is much higher with rectal despite giving antibiotics. So you really want to be careful if you're FNAing a rectal or peri-rectal lesion. The other thing is with EUS, just inspect the vasculature. You can often see a tumor thrombus and also primary involvement of that tumor into the vessel by EUS. Those are some of my top 10 tips here. You always want to make sure that there's no metastatic disease in the setting of esophageal cancer before you rush to an EUS. I think we know that, but it's just good to keep reminding, especially sometimes referring colleagues such as oncology or primary care. I think I shared with you, we're very accessible for EUS. Often patients will get sent for an EUS before they've even had serial or cross-sectional imaging, which is a great problem to have, but you've got to be careful about that as well. I'm just going to zip through this slide quickly because NUSAT has covered this with the histologic layers. You know the layers, and that's where for T-staging, EUS can really help. Anything that's into the muscular is appropriate, but not through it. Things that are through it are T3 lesions and the corresponding hypo- and hyper-echoic layers for the wall. This has already gone through in detail, so I'm not going to bother you. Now, a couple of words about lymph nodes. These criteria for lymph nodes were applied in the setting of esophageal cancer, and what I find is a lot of times folks will try to apply these same criteria for all malignancies, and that shouldn't be done. However, if you see a lymph node in the setting of esophageal cancer, you can see this tumor is breaking through the muscular is appropriate. We call that sort of pseudopodia, and then here's a lymph node. This is larger than a centimeter. It's round. It's hypo-echoic. It's got sharp borders, and when all these four criteria are met, you know, Dr. Shamian and Mark Catalano back in the day showed that you had a pretty good chance that this is going to be a malignant type of node. We still like to biopsy these nodes if we can, as long as there's no tumor in the way. So here's for the esophageal cancer. What's the rationale for EUS in the setting of esophageal cancer? Well, we know that if you identify a very early sort of superficial lesion by EUS, then perhaps they don't need an invasive surgery, and you can offer them an EMR and ESD. However, anytime there's lymph nodes involved, and I think EUS is quite good for lymph node involvement, then this is where, you know, I think in these scenarios right here, the first three, EUS is helpful, and we know based on other studies that if you combine new adjuvant therapy plus surgery versus surgery alone, the outcomes are better. So that's the rationale to try to get a proper stage. Well, we know that EUS is really good for primary staging. This is a meta-analysis where if you look at all T staging, pretty good anywhere from sort of 80% to 90% accuracy for T staging, and it's very good for distinguishing T4 and T3 lesions, and also for nodal disease, especially if you combine F and A. So EUS is excellent for primary staging, for T and N staging for esophageal cancer. And I think Vivek had alluded to this before. This is the NCCN guidelines, our National Cancer Center, where they incorporate EUS for the workup. I mean, you can go through all those other modalities, but EUS is right up there in the primary evaluation and workup for esophageal and GE junctions, and really even gastric cancer now. So here's how we do it. You get the probe. In the old days, we would really try to dilate this malignant stricture. We know that if you have a malignant stricture, it's going to be 80% of the time a T3 lesion. So I think nowadays, if you can't get through there, then don't risk perforation, because if you put the scope tip right above the tumor, you can get pretty much most of the information that you need. You won't be, if you can't get through here, you may not be able to access the celiac lymph nodes, which is important if you can. But I always do a radial for staging for esophageal cancer. And then Nuzat had mentioned changing the frequency. You can go from five to seven and a half to really look at the wall layers. Often what I'll do is I'll switch my screen from a full screen down to a half screen, so you get better penetration. And really, if you want to hone in, and you've got to be careful about that tangential imaging, because you can make something look T3 when it's not. And the consequences are quite significant. We use water, the balloon, whereas I think we didn't talk about this. I don't think anyone partly uses a balloon on the linear scope. You still need to use a balloon on the radial scope. And then for submucosal lesions, we'll put water in the lumen. Can't really do that too much in the esophagus. But all these techniques, and it's in your slides, that can be helpful for esophageal cancer. So here's a case. I think that all of us who do EUS will show you this and the value of EUS. Here's the endoscopic image. Let me just get my laser pointer here. So here's the endoscopic image. And by EUS, if you look right here, this is a hypocholic lesion. And these slides are courtesy of Dr. Vanessa Shammy. And doing a, because it looks mucosal and not deep, and I think you can appreciate that the muscularis propria is not at all involved. And the submucosa looks pretty nice and clean. So this would be a perfect case for an EMR. And that's what she did. And I've done cases like this, as well as my partners here. And the final path. Okay, so this, you can see the histology. This lesion was confined to the mucosa and taxa mucosa. And this was staged as a T1S lesion. So again, very powerful EUS in helping to determine whether you should undergo, or this patient should undergo an EMR or not. Now, I sort of touted the merits of EUS for primary staging. If you look at the data for restaging EUS, you know, Charlie Lightdale, this is a big review. But if you look at all the different studies, and I know Greg Zuccaro had one of the first studies that was published in the Red Journal many years ago. But look at the accuracy. It really falls down. And that's because once patients have been treated, then that thickness and on all the radiation chemo effects, you know, no longer can rely on the primary wall layers for T staging and N staging. So it's just not as accurate. You know, I trained at MUSC in Charleston back in the mid 90s. And Dr. Carolyn Reed, who unfortunately has passed quite ago from pancreatic cancer, she was an amazing cardiothoracic surgeon. We did a lot of studies together at that time, looking at the validity or the need to do to interrogate the celiac lymph nodes. But what they looked at, the cardiothoracic surgeons, is what is the utility of restaging EUS after new adjuvant therapy for esophageal cancer. And I've already shown you that restaging, if you look at T staging compared to surgical, the T staging falls significantly, the N staging. So, you know, in conclusion, the restaging EUS did not accurately predict pathologic stage in patients who received new adjuvant treatment. So does that mean that there's no role for EUS at all? I don't think so. And I'll show you where it can still be useful. And here, this is a case that I just did last week. This is an 80-year-old gentleman who came down to escape the cold from Upper Peninsula, Michigan, to be down in North Carolina with his daughter. And so the patient was treated, and the question was, is the tumor gone entirely or not? And you can kind of see the mucosal sort of endoscopic pictures. I think all of you can appreciate the buried esophagus associated with this. But I did endoscopic biopsies. I didn't see anything. However, I was doing a radial on this. And you can see, I measured the lymph nodes. It's 1.5 by 1.4. And if you look at this lymph node right here, it has all the features. It's round. It's hypochoic. It's over a centimeter. It's well demarcated. So just looking at it, that's probably going to be malignant. However, the proof is in the biopsy. So we did a biopsy. And sure enough, the power here of the EUS is that the endoscopic biopsies were all negative for tumor. However, the FNA was positive. So again, the patient's 81. Is this patient going to undergo further treatment? That remains to be seen. But this is where I think EUS is helpful. There's also some other studies that show that even though the primary T-staging for esophageal cancer is not useful, however, you can look at the amount of decrease in the wall thickness by EUS as a surrogate predictor of response. But I think the take-home message is EUS is great for primary staging, but not as great for restaging due to the caveats that we just discussed. So in our own group here, we actually, it was an NIH-sponsored study, how good is PET scan for predicting response? And what we showed here, I'm sorry, I'll just go back one slide here, is that if you had a really high SUV uptake to start out with, greater than 15, you had a much better likelihood, almost 80% of getting a complete pathologic and complete response. So the PET scanning is useful if your SUV, it's the amount of the SUV that decreases after your new adjuvant therapy that can predict. But there's, you know, because about 25% of patients will have complete response after new adjuvant therapy and which patients are those. So we'll shift gears now, because I think predominantly you're going to see esophageal cancer, but you will see some gastric cancer. And what's the rationale for EUS with gastric cancer? You always want to perform a good exam first. Where is it located? You want to mention that in extension to the GE junction, is that involved? And the reason it's important is there's a growing trend for new adjuvant therapy or perioperative chemotherapy. And we know this because if you combine this perioperative chemotherapy versus surgery alone, the outcomes are better, at least for a T2 disease. And this is some other data looking at other combined chemotherapy modalities. So here's a case, this lesion you can tell looks bad. There's already ascites on your endoscopic ultrasound. There's peritumoral lymph nodes. And you can actually see right here that in this individual, I think Vivek had pointed out that metastatic liver lesions can be hypo or hyper or isochoic. But this is an area in the left lobe of the liver where, excuse me, the CT did not show any meds to the liver, but clearly there's liver meds. So you always want to try to perform a very thorough and complete exam. So we'll shift gears now to the ampulla. And I think there's a really nice role for EUS in this setting of ampullar lesions because I've found that no cross-sectional imaging is great. I think they're all complementary, whether it's a CT, MR, MRCP, ERCP, but I think don't underestimate or undervalue the power of EUS for ampullar lesions. And this was work done by Dr. Artifon out of Brazil years ago. If you look at just the sensitivity for ampullar lesions compared to CT scan, 85% for T2 and T3 lesions versus very low. I think all of us would agree that a CT scan just isn't good, is an MRCP. MRCP is definitely better, but I think the role of EUS is also quite powerful. And we actually looked at this, one of our former fellows. So kind of a spin on this is if you combine worrisome features. So we have 500 patients with suspected ampullar lesions and then 69 that were confirmed and some that were excluded. But what I want to point your attention to these features. So by EUS, the size was less than two and a half centimeters. If the invasion was less than four millimeters, if the pancreatic duct was dilated or not, and if there are lymph nodes involved. And so if you apply these criteria and you had zero of these features, then it's, you know, you're not, whereas if you had more number of these, then you're much more likely to have advanced disease. So I think if you look at this endoscopic view right here and in EUS, I mean, this shows a pretty nice one centimeter ampullar lesion with really no great invasion further into the pancreas or other ducts and things. And so perhaps this lesion might be amenable to an endoscopic ampullectomy. So I think if you're going to try to do an ampullectomy, I think an EUS can give you some powerful information. You know, the whole role for staging in pancreatic cancer, you know, that's a talk in terms of the caveats. I think where we are right now is that everything is complementary. But again, the power of EUS here is, so this is a patient that was sent to us with obstructive jaundice. The outside ERCP was equivocal. The brushings were equivocal. CT did not show a mass. And the patient was very reticent to undergo a Whipple surgery, elderly. But I think you can appreciate right here, you can see the stent here. I don't think I need a laser pointer to show you, but you can see the stent. And then we did an FNA. And this came back as an adeno, a very small lesion. So the power is now the patient did undergo a successful Whipple, an elderly patient, and the margins were quite good. So this is again from Vanessa and her group at UVA, where they compared EUS versus MRI, and you could put CT or high quality CT, whatever modality. And the point is that the agreement between the two is not always that great. I mean, if you look at the agreement, it's quite low, actually, 27% by 27 patients. So in essence, I mean, what they've shown is that EUS and MRI had very marginal correlation in staging. So I think where we are now is that you can't say that EUS is the best staging modality because the quality CTs and MRs, we don't get the 360 degree radial involvement of the vasculature. But there's still power in doing the EUS because you can perform FNA and get a look. So I'll finally finish with some rectal lesions and cancers. And so what is the rationale? 1990 NIH conference recommended this in terms of looking for advanced local regional lymph nodes. And the T staging is very similar to what we discussed with esophageal cancer. And the goal really is to identify which patients should undergo new adjuvant therapy. And this already has been discussed. So, you know, the T and N staging, but it's very similar for rectal cancers as well. I think the big take home message that I'll leave you is where you might see nodes in other settings, you should not see rectal lymph nodes unless they're worrisome or concerning. So, you know, the endoscopic appearance is very important with rectal cancer lesions. Is it ulcerated? How far is it from the anal verge? What's the relationship to the valves of Houston? And I always perform a flex safe first, because again, most of our cases are coming from the outside. I like to get a good lay of the land. And before I do the ultrasound in here is a corresponding radial image of this rectal lesion. Just remember that the left and right are going to be different depending on how the patient's laying down in the left lateral decubitus position. So here's a case of a lesion that you can see up in the upper right, a very angry looking ulcerated lesion. And you can see by the EUS, here's the prostate. So the plane is very well preserved. And here's the lesion right there. I haven't shown you a really good image to show that it's a T3 lesion, but there's a lymph node involvement here. And this lymph node was positive. So what is the stage? This was a, sorry, a T3 lesion. This invaded into the muscular dyspropria, into the serosa. So this is a T3 lesion. And the lymph nodes are important because you shouldn't have lymph nodes in rectal cancer. So peri-rectal nodes do not occur in healthy individuals and independent of size. And Gavin Harewood showed this, I think when he was at Mayo and for Gleason has shown this again at Mayo that these nodes independent of size should be sampled if you can sample them. And so this is a study that one of our residents did many years ago. And this is looking at RUSFNA with peri-rectal or presacral or pelvic lesions. And just a couple of summary key points that I want to highlight on this study is that if you look here, so this is, we had 27 patients, the FNA was diagnostic in 90% cases. And in about half of the patients, we sort of obviated the need for surgery. So it's quite powerful. However, we had a 15% complication rate with four complications. This is despite antibiotics, despite, you know, very well-trained endosynographers, we had two peri-rectal, two presacral abscesses. So, you know, I think I'm not saying don't biopsy presacral or, you know, rectal lesions, but you really have to have a good indication just because you see a lesion, you know, you don't want to biopsy a cystic lesion, even a GIST tumor. You want to have the backing of your colorectal surgeons. And often it is our colorectal surgeons that will ask us to do an FNA on these, just so they know, especially for recurrence of rectal cancers. But very important point that, you know, rectal lesions and FNA, despite antibiotics, in our experience have a little bit higher risk of complications. So how good is EUS looking at T-stage? You know, quite good in the 80 to 90% range, maybe not as high up as esophageal cancer. And these, again, are pooled meta-analysis. So I think Dr. Call had brought this up, is EUS obsolete for rectal cancer? You know, MRIs are very, very powerful because you have the same accuracy, but you get much more useful information, especially if you have a stenotic lesion in the rectum. So, you know, here's a, you can see by a rectal, you know, beautiful MR imaging that shows the sort of pelvic structures much better. You can get distant pelvic nodes that we can't see, the perineal reflection distant from the mesorectal fascia. And it's not operator dependent. You know, I think we're showing you great data for rectal ultrasound, but unless you do a lot of this, you know, you can make mistakes. And, you know, we're all going to make mistakes, but you don't want to make too many of them, especially if it's going to send someone for an APR that doesn't need it. So I had alluded to this earlier in the morning. This was an oral presentation by one of our fellows. So the question here is how useful is a rectal ultrasound in identifying lesions that are suitable for TAE or minimally invasive type of surgery, TAE or TEM? So we had over 624 rectal ultrasounds that we did. And we, at the end, we have data on 118 patients, which is a fairly high number. So I'm optimistic that this paper will be published and accepted. But what I want to highlight here is that we had great sensitivity to determine if the lesion is T0 or T1. So none of the patients that underwent an APR had a T0 or T1 on rectal ultrasound. So where our colorectal surgeons feel rectal ultrasound outperforms MR are in the very low-lying rectal lesions where right near the sphincter muscles, we can delineate whether that lesion is going deep or not better than an MR in their opinion. So that's how we use it. So in summary, I think I'm good on time here. For luminal cancers, for all luminal cancers, you just want to make sure that there's no metastatic disease in cross-sectional imaging. For esophageal cancer, you want to look at the proximal and distal extent. You want to assess for T3 or nodal disease. Preoperative therapy matters. You always want to look for systemic metastases. Is the liver involved or not? And then finally, for gastric cancer, similarly, is it T2 or below? Because maybe that patient with no lymph nodes, we had a case just last week, an elderly patient where they don't need surgery. Sorry, they can undergo straight to surgery without perioperative therapy. Looking for meds, looking for ascites. Ampullary lesions, I think EUS has a vital role, and especially in identifying those lesions that are amenable to an endoscopic ampullectomy. For pancreatic cancer, I think EUS is complementary. I think the days of saying that EUS is the best, most sensitive test for pancreatic cancer, I think those days are gone. I know Dr. Ahmad had many studies, but that was back in 2003, 2004. As Amivek mentioned, she's a giant and has been doing this for many years. Back then, I think EUS outperformed, but I think even in their center, they'd probably say that they're complementary now. And perhaps the value is in performing genomic analysis with getting core biopsies. And then I mentioned with rectal cancer, T3 disease, preoperative delineating the sphincter muscles is quite useful, and then seeing which patients are best amenable for TAE or TAM surgery. So thank you for attention. I think Vivek, I've kept you on time, I believe.

endoscopic ultrasound

cancer staging

lymph node involvement

imaging modalities

complications

rectal cancer