Welcome, Usman. Thank you. Good morning, everyone. First, I want to congratulate our course directors for organizing a fantastic event today and for the invitation to speak. So my goal with today's talk is to provide a little bit of clarity in how we address what's often a challenging clinical situation, and that's proving cancer in indeterminate biliary strictures. And I'm also going to highlight some aspects from recently published guidelines, including those from the ASGE. Here are my disclosures. Let's start with a little background information, though. You know, the current clinical definition of an indeterminate biliary stricture are biliary strictures of undetermined etiology from imaging and patient history or a non-diagnostic prior ERCP. Now, diagnosing cancer in this scenario can be difficult for a number of factors. First, there's the difficult to access bile duct location, the small caliber lumen. There's also limited ways that you can get tissue non-invasively. And also, you have overlapping features with benign diseases. So often, the burden of proof on proving cancer is on the endoscopist. It requires, in many cases, multiple procedures, and that can lead to delays in treatment and increase patient anxiety. When you have a biliary stricture, there is a long list of potential etiologies. The one malignancy that we always think about is cholangiocarcinoma, as you see in this first cholangiogram. But other malignancies in nearby organs can also cause biliary strictures. In the second cholangiogram, you see a pancreatic head cancer causing a distal bile duct stricture. But liver, gallbladder, and metastatic cancers can also cause strictures. Now, in some cases, we give good news to patients, like this one in the middle, where a patient was referred to me for a possible cholangiocarcinoma, but it was actually just a very large bile duct stone. Other benign diseases that cause biliary strictures include PSC and chronic pancreatitis. But those are tricky, because those patients can also develop malignancy over time. So it can be difficult proving that later on. Other benign causes are post-surgical strictures and autoimmune cholangiopathy. And just so we're all on the same page in terms of biliary anatomy and the terms that we describe when we make our recommendations, this is the distal bile duct, where you get strictures from pancreatic head cancers, and 30% of cholangiocarcinomas are located. More proximally, we have the higher region, where 50% of cholangiocarcinomas are located. And in a very select group of patients at expert centers, these patients may be eligible for liver transplant. And then this is the intrapathic region. So what are endoscopic tools that we can use to evaluate and sample these biliary strictures? I'll start with endoscopic ultrasound, or EUS. And this is an essential tool for two reasons. Number one, when the scope is positioned in the duodenum, you get really close-up views of the bile duct. And then you also have the ability to sample with fine needle aspiration, or FNA. EUS is great for picking up small masses, especially those that are less than 2 centimeters, which may not be picked up on cross-sectional imaging. What we often see is a small, you can see a small mass with a concentrically thickened wall. You can also pick up local lymph nodes or any liver metastases. And also, you can see a background of chronic pancreatitis. In terms of sampling, most of us follow what the ASGE guidelines recommend. If you have a distal biliary stricture or mass, then EUS with direct FNA of that biliary mass is recommended. Now, in the past, there has been data out of Mayo Clinic looking at their patients who went for liver transplant. And they showed that there was a small risk of tumor seeding from either percutaneous biopsy or FNA of higher masses. So if you have a patient that has any chance for going for a liver transplant, you do not want to directly FNA a higher mass. You would potentially sample any adjacent lymph nodes or liver mets, though. And if you look at the data and the ASGE guidelines, if you perform EUS with ERCP, then your sensitivity for picking up malignancy is going to be close to 90%. So for that reason, we definitely advocate doing both of those procedures together during your workup. But then ERCP is probably the test that's most often used, because it's more widely available than EUS. And it's usually the first procedure that's performed. Most endoscopists, if you do an ERCP for a biliary stricture are going to do a brush cytology. And the reason for that is simple. It's cheap, and it's easy. So endoscopists are pretty simple people. And even though it has a yield of only around 40%, this is probably the first test that is done. Now, passing a biopsy forceps freehand into the bile duct and using fluoroscopy to guide your biopsy of a stricture can increase your yield a little bit. And if you combine the brush cytology with the biopsies, then your yield can increase up to 66%. But that's still not great. And you're thinking, OK, how can I increase my diagnostic capabilities so that we can make this diagnosis? Well, don't worry. There is a light at the end of the bile duct tunnel. And that leads us to ERCP with cholangioscopy. Cholangioscopy has evolved over the past couple of decades. We started out with needing two people to control a mother-daughter scope. Then we had a single operator system with a fiber optic catheter that provided not such great imaging. And then since 2015, we've had a nice integrated catheter with digital imaging. And that's our current iteration. And the good thing with cholangioscopy is that not only are you getting great images inside the bile duct, but now you can directly visualize where you're going to biopsy. And you're not doing it blindly. We have, in the past, for the most part, been using this 10 French catheter that has four-way deflection, water irrigation, and a 1-millimeter accessory channel that allows us to do biopsies. But sometimes it can be difficult to maneuver that catheter, especially in distal, tight strictures. So more recently, there's been newer devices that have come out that are different caliber diameters. One thing to be aware of, there is a risk of bacteremia with cholangioscopy. So we always advise giving one dose of IV antibiotics. Since the new digital imaging system came about in 2015, there's been a number of studies trying to come up with endoscopic criteria and classification systems for diagnosing malignancy. And you can see that there's been significant variability in terms of accuracy and inter-observer agreement. Sometimes there's issues with patients who have had stents before. And there's stent-induced changes within the bile duct that kind of make for conflicting images. But I think most endoscopists would agree that there are a few features that we all think are indicative of malignancy when we see them. And that includes tumor vessels, papillary projections, a nodule or mass, and if you see an infiltrative or ulcerated lesion within the bile duct, those all should raise your suspicion for malignancy. And again, the data supports that if you had prior ERCP sampling that was negative, your visual impression using digital cholangioscopy can get you an accuracy of predicting malignancy close to 90%. But you know, seeing's not always believing for your oncologist and surgeon, so we still wanna try to get tissue. And that's where your directly visualized biopsies come into play. And again, adding the biopsy to your visual impression is gonna raise your sensitivity to 72%. Again, we're still not perfect. So this is where you're probably gonna have a conversation with your surgeon if your visual impression and your biopsies have some discrepancy, because that's gonna require a multidisciplinary approach. Now, this is a case where there was no discrepancy between what I saw in my biopsies. I had a patient who was incidentally found to have a right hepatic duct abnormality on PET-CT and MRCP, had an outside ERCP where they removed a stone, but there was a persistent stricture with negative brush cytology. So I did an ERCP with cholangioscopy, and you can very clearly see this abnormal mucosa of the bile duct, which is what I would call papillary projections. Some tumor vessels were already present, and I was thinking this is probably cancer, but I need to prove it. So then we did our direct visual biopsies. Again, these are pretty small forceps, and there's different techniques for how you do it. You can take one piece at a time, take multiple pieces, and then suction the bile into a trap, but usually you try to get a few pieces. And we did diagnose cholangiocarcinoma in this patient. They went for surgery pretty quickly, and they had resection and were found to have a localized 15-millimeter cancer and are essentially cured. So I think this case highlights what some recent international consensus guidelines also suggest, that perhaps if your suspicion's high, performing cholangioscopy, getting a visual impression in those direct biopsies during the first ERCP may decrease the need for additional procedures. So I've shown you ways of how we can get tissue samples, but can we do anything with those samples themselves to improve our diagnostic accuracy for malignancy? Oftentimes we sent off fish testing on our cytology or biopsy samples, and we've done that to PSC patients with dominant strictures. This is basically where you have fluorescent DNA probes that can pick up abnormalities in cells that are associated with pancreatic obiliary malignancies, and that has been shown to increase your diagnostic yield. You can also send off for next-generation sequencing on your samples. Again, in bile duct strictures, it can increase diagnostic accuracy for malignancy. In other cancers, we know that it can also help guide chemotherapy. I think in current practice, though, you know, these tests are send-outs, it takes a while to get the results back, there's increased costs, and it's not likely that these individual tests are gonna initiate treatment, but I think they're certainly part of the diagnostic discussion. And that leads me to AI. It's the hottest topic in GI, and it has made its way into the bile duct. There were a number of studies within the last year that looked at using AI in biliary strictures, both in previously collected images, but also using it real-time to not only detect malignancy, but help guide where you take a biopsy, and also use it as a tool to help less experienced endoscopists diagnose cancer. Again, we're gathering more data, so it's not commercially available yet, but I'm sure we're gonna hear more information on this topic. And, you know, they always say the less GI doctors know about managing a disease, we publish more and more guidelines on that topic. So within the last year, there's been a number of guidelines that have been published on how you diagnose, approach, and manage indeterminate biliary strictures. I think these are all useful to help all of us, and also serve as, you know, references in the future when we're managing our patients. So just a few take-home points, and in terms of approach to indeterminate biliary strictures, I would say two tests are always gonna be better than just one. If you're getting a referral for this patient, I would say definitely do the EUS with ERCP. The EUS is gonna help you pick up those small masses, lymph nodes, and liver mats. You can FNA the distal lesions, but don't FNA hyaluronic masses or strictures if there's any chance the patient will go for a transplant evaluation. If you're doing an ERCP without local expertise or equipment, then do brush cytology with fluoroscopically guided biopsies. And if you're gonna do ERCP with cholangioscopy, make note of your visual impression, and take the directed visual biopsies, especially when you're dealing with hyaluronic strictures, and consider doing it on the index ERCP to make your diagnosis earlier. Now, if you have negative sampling, but say you had a high suspicion based on your cholangioscopy images, then I would suggest doing a repeat attempt. You can also work up benign causes. And as always, you definitely want to have a multidisciplinary discussion with your surgeons and oncologists when you're dealing with these patients. All right, thank you.

cancer diagnosis

biliary strictures

endoscopic ultrasound

cholangioscopy

artificial intelligence