Our next presentation is by the course co-director Anne-Marie Lennon whom you all have met today. She is the chief of GI at Johns Hopkins and a professor of medicine, and she will be discussing the evaluation and management of pancreatic cystic neoplasms. Thank you very much. It's an absolute pleasure to be here. Can I have people raise their hands? Has anybody in the audience ever got a referral for a pancreatic cyst? Excellent. Okay. Well, so in the next 20 minutes, we are going to do a few things. First and foremost, we're going to talk about what are the tests that you have and how do you interpret them. Number two, we're going to talk about what are the common cysts, and number three, we're going to talk on how do you manage them. And then finally, we're going to talk about what's new over the last year or two and where is the future. So these are my disclosures. Well, you've had an absolutely wonderful time at ACG. You go home, and on Monday week, you're in your clinic, and this patient comes in. So she's a 52-year-old lady who went to the emergency room with lower abdominal pain, was found to have a renal calculus. And when she had her scan, she had an incidental finding of a 2.7-centimeter cyst in the head of the pancreas. The CT report says it appears to communicate with the main pancreatic duct, which measures 3 millimeters. She's entirely asymptomatic, and the radiologist says they think it's a serious cyst. She went on to have an endoscopic ultrasound, and the report from your colleague said that the cyst was thin-walled. There was no solid component. And they biopsied it, and these are the results that are sent to you. The CEA is 354. The cytology shows no marked atypia, and there's a thing called a GNAS mutation. So the question is, how do we manage this patient? So this is, if you remember one thing from this talk, it's this slide. And the question, when you see a patient with a pancreatic cyst, question number one, what type of cyst do they have? That will allow you to determine what is the risk of malignant transformation, and that will allow you to determine what you do with your patient. So the question is, do you have a patient with a benign cyst, such as a serous cyst or a pseudocyst, which has no risk of malignant transformation, and you should discharge your patient? Do you have a patient with pancreatic adenocarcinoma, in which case clearly they need to have surgery, if that's feasible? Or finally, do you have one of the two precursors to pancreatic cancer, intraductal papillary mucinous neoplasms, or IPMNs, or mucinous cystic neoplasms, which we call MCNs? Most of those will never get pancreatic cancer, but a small number will, so most of them can be followed, and a very small number will need to be referred to surgery. So how do we determine what are the tools that we have? And these are what we have. Well, we have imaging. The first imaging we have is MRI. And sometimes, on the left-hand side, you have a great image of a microcystic serous cyst. It's classic. It's a done deal. You don't need to do anything else. However, on the right-hand side, we often see this as well, and that's where it's very challenging for our radiology colleagues to tell us exactly what it is. And if you look at a systematic review of meta-analysis, what you can see is that MRI is not perfect for telling you either what type of cyst you have or whether or not you have high-grade dysplasia. So what other tests do you have? Well, we have, of course, as gastroenterologists, have endoscopic ultrasound. We learned earlier on about the importance of informed consent, so I would tell my patient it is more invasive than the other tests. But what are the benefits of it? Well, it's the most sensitive test that we have for identifying a solid component, which is shown here. And if you see this, then you can tell your patients that they have an odds ratio of 9.3 for having high-grade dysplasia or invasive cancer, which is a very important thing to hear about. You can also sample the cyst. And here we have the cyst with a needle in it. And this is, I think, a fairly low-risk thing to do. And this is a large systematic review and meta-analysis of almost 5,000 patients. And these are the adverse events. So overall, you can tell your patients you've got an adverse event risk of 2.7%. Pancreatitis of around just under 1%. These are retrospective studies. So if you look at prospective studies, they rate the risk of acute pancreatitis around 1.5% to 3%. But you can see that the risk of bleeding, infection is low, and a small risk of perforation. So how do you interpret the results that you send? Sorry. We'll get back to that. So firstly, you have cyst fluid CEA. If it's high, greater than 192, you're dealing with an IPMN or an MCN. If it's very low, less than 5, you're likely dealing with another type of cyst, such as a cirrhosis or a pseudocyst. And if it's anywhere in between, it's really not very helpful to you. What about your cytology? Well, if you see mucin, you're likely dealing with an IPMN or an MCN. And the other piece of information that you're going to get from cytology is, is there high grade dysplasia present or invasive cancer? Either way, if you saw that, you would refer the patient to surgery. What about molecular markers? Well, this is my cheat sheet for how you interpret them. If you have a GNAS mutation, you have an IPMN. If it comes back and says you have a KRAS mutation, you have an IPMN or an MCN. And if you have a VHL mutation with no other mutation, you have a cirrhocyst. So let's go back to our patient and our potential list. So what type of cyst does she have? So let's start with a pseudocyst. I think we're all very comfortable with these cysts. You can occur in acute or chronic pancreatitis. They can be single or multiple. You can find them anywhere within the pancreas. If you look at your cyst fluid, you're going to have a low CEA. And on cytology, you may often have macrophages. And they are, of course, benign. A clinical pearl that I find useful is if your patient has no history of pancreatitis, then it's very unlikely that you are dealing with a pseudocyst. And so if we look at our patient, I think that the likelihood that this is a pseudocyst is very unlikely. So let's look at the second commonest type of benign cyst, a cirrhocyst. So these are more common in women. They typically present in their 50s, but of course, they can present at any stage. They tend to be single. There is classically no connection or communication between the cyst and the main pancreatic duct. The typical beautiful cirrhocyst that you'll get is one called a microcystic cirrhocyst, shown here, and also here, which is made up of multiple tiny cysts. And around 45% of cirrhocysts will present like this. Unfortunately, 50% will be what we call a macrocystic cyst, which is a single cyst, and 5% will be solid. Your cyst fluid CEA is going to be low. You typically should have no mutation, and if you're lucky, you'll have your VHL mutation. There are four cases ever of cirrhocyst adenocarcinoma reported in the literature. These are a benign disease, and if you can confirm the diagnosis, they need no follow-up. A clinical pearl that I find very useful, in a large study in 860 patients, all of whom underwent surgical resection, if you have a VHL mutation and no other mutation, there is 100% specificity for a cirrhocyst, so you can feel very, very comfortable about discharging your patients. If you recall, our patient had a US. This is her number, CEA 354. Remember our cirrhocysts are going to have a very low CEA of less than 5, so I think we can exclude it. So let's move to the right-hand side of our diagram. Could this be pancreatic cancer? And some cancers can present as a cyst, and that's because if they have rapidly growing, you will have cystic degeneration of the center. And here you can see a cyst, and as you move through it, you can see the large, solid component. And both pancreatic adenocarcinomas and pancreatic neuroendocrine tumors can present as cysts, and if you find this, you should refer them to a high-volume center. Now, the diagnosis is not difficult. If you recall, in our patient, there was no solid component. The cytology did not show any high-grade dysplasia, therefore, I think we can exclude this. So let's move to our last two cysts, mucinous cystic neoplasm and introductal papillary mucinous neoplasm. So what is an MCN? Who gets it, and does she have it? So MCNs are very simple. They almost always occur in women. They almost always occur in the body and the tail, and they are almost always single. So if you have a man or you have multiple cysts or your cyst is in the head of the pancreas, this is not an MCN. The cyst fluids, as we said, you're going to have a high CEA, you're going to have a KRAS mutation. They are precancerous. So in 4 to 12% of patients, they will ultimately develop invasive adenocarcinoma, but if you look at the reverse side, the vast majority of patients with MCNs will not get pancreatic cancer. And in a large systematic review and meta-analysis, there were no cases of invasive adenocarcinoma in an MCN less than three centimeters without concerning features. Now in the past, MCNs, people have said maybe you resect them, but one of the problems are that it's very difficult to differentiate a single IPMN in the body and tail from an MCN, and most of the more recent guidelines have recommended that they are followed in the same way as you would for an IPMN. So our final, so if you recall, she had a cyst in the head of the pancreas, and we said these occur almost exclusively in the body and tail, so we're going to exclude it, and we come down to our last case. These are by far the most common type of cysts that we see. There are three different types of IPMNs. The first type is called main duct type IPMN, and that is when your main pancreatic duct is enlarged, greater than five millimeters, and you have no pancreatic cysts as seen here. The second type of IPMN is a mixed type IPMN. That is where your pancreatic duct, again, is dilated, greater than five millimeters, and you have pancreatic cysts. Why do we care about this? The reason we care about it is that if you have main or mixed type IPMN, you have a significantly higher risk of your patient developing high-grade dysplasia or invasive cancer, and those patients are the type of patient you need to refer to either a pancreatic surgeon or a multidisciplinary group for evaluation of, do you watch them very carefully, or do you resect them? And I'd be happy to discuss that later. Finally, the third type of IPMN is a side branch or branch duct type IPMN. These occur equally in men and women. They typically present in their 70s, but it can present at any age. The youngest patient we've ever had was three months old. 40% of patients will have multiple cysts, and they can occur anywhere in the pancreas. You classically will have a connection or a communication between your cyst and your pancreatic duct, and that is how radiologists classically will differentiate an IPMN from an MCN, but again, it is imperfect. And here is your cyst fluid analysis. So again, branch duct IPMNs, they have, like MCNs, malignant potential. However, the vast majority of patients with IPMNs will not develop pancreatic cancer and can be followed. Okay, the great question, how do you follow them? And here we have the guidelines. Not one, not two, not three, not four, but five guidelines. And which one should you follow? And again, I'd be happy to discuss this later about what are the differences. And the bottom line is, I think Alan beautifully presented early on this beautiful high quality data. And if you look at this, the whole problem with these guidelines, whether they're grade or non-grade, is that the quality of the data on which they are based is low or extremely low. So no matter how much you want to have evidence-based, these are not evidence-based guidelines, and they are basically expert opinions. Most of the guidelines are really very similar, and the one that stands out is the AGA guideline that's been different to the others. So I'm going to give you the, I think, what most people will follow, and I'm going to share with you the ACG guidelines. So you have your IPMN or your patient with your pancreatic cyst. So she has her cyst in the head of the pancreas. So the first question is, what size is your cyst? If it's less than one centimeter, you're going to bring them back in two years. If it's one to two centimeters, you're going to bring them back in a year. If it's two to three centimeters, you're going to bring them back in six months. You'll see here that in gray, if you have a small cyst, you're going to follow them with MRI. If the cyst is bigger, two centimeters or bigger, you can follow them with either MRI or endoscopic ultrasound. You'll also notice that if the cyst is stable for many years, you can increase your surveillance interval. When should you think about referring your patient for consideration of watching very carefully versus surgery? Well, the following. If you have symptoms or signs of jaundice, highly concerning. Jaundice secondary cyst, very concerning. Acute pancreatitis due to the cyst, elevated CA19-9. Imaging showing a mural nodule or soil component, a dilated pancreatic duct, or cyst size greater than three centimeters. And again, I'd be very happy to discuss that point later. And finally, cytology that shows high grade dysplasia or invasive cancer. What's a level below that? So something maybe you don't need to send them, but you should have a little red flag you're worried about it. So if you have new onset of diabetes or a rapid increase in the cyst size, then that's something you should be slightly more concerned about. It is, in some studies, associated with a higher risk of high grade dysplasia or invasive cancer. And in these patients, we recommend a short interval MRI or EUS. And what about stopping surveillance? And the only guidelines that provided guidance was the ACG, and they recommended talk to your patients age 75 and assess, is surveillance really appropriate? For me, that may be a little bit ageist because I have some patients who are age 50 who aren't feasible for it. And I think I personally look at comorbidities and I say, what's the likelihood of this patient? Do they have multiple comorbidities? Is it actually appropriate to follow them? Now finally, if you send your patients to surgery, do you need to follow them? And the answer is, with the exception, if you have a pseudocyst, a cirrhocyst, there's no follow up. If you have an MCN without invasive cancer, you need no follow up. The only patients that you need to follow up are IPMNs, and that's because it affects the entire pancreas, and therefore you do need to follow the remnant. Now I was asked by my co-director to talk about patients with a family history of pancreatic cancer. So I'm going to talk about this briefly. Do you need to follow them? Your patient has, she comes in and says, you know, my great grandfather had pancreatic cancer. So this is the guidelines. So if you have one first degree relative with pancreatic cancer, they do not need to be followed. If you have two relatives with pancreatic cancer, one of whom is a first degree relative of your patient, then the current guidelines recommend surveillance starting at age 50 or 10 years younger than the youngest relative who developed cancer. What about if you have a germline genetic mutation? So this is in your handout. You don't need to read it. But if you just remember a few things, we see patients with Putzjeger. They are at a significantly increased risk of pancreatic cancer, and those patients should be screened, irrespective of whether they have a family history. What about HMPCC? All of us see patients with HMPCC occasionally. Those patients are at an increased risk of pancreatic cancer, and if they have an infected first degree relative, they should start surveillance at age 50. And finally, we often see patients with a BRCA mutation. And the current, the CAHPS guidelines, which were published in Gut in 2020, recommend that if you have one first degree relative, then you should start surveillance at 50. This is quite, this is a hot topic for debate, and there is a new ASGE guideline coming out in 2022, and we look forward to seeing what it says. So how do you follow these patients? Very simple. Less than two centimeters, you follow them every year. Two to three centimeters, you follow them every six months. And unlike your sporadic, you alternate MRI and EUS. So the last section, I just wanted to talk about what's new, and what's happened in the last, you know, two to three years. So one is we have the microbiopsy forceps. These are biopsy forceps that you can put through your scope. There have been quite a lot of papers published. The most recent one is a systematic review and meta-analysis, which looked at just over 400 patients. And what they found was that the microbiopsy forceps are superior to FNA in terms of the diagnostic yield. You have a high sensitivity and specificity of 90 and 94% respectively for identifying IPMNs and MCNs from different types of cysts. The adverse event in this large systematic review and meta-analysis was 7%, and if you recall earlier on, I normally quote a rate adjuster of around 1%. You also need a 19-gauge needle to put them into. And what about the surgical concordance? So how often, how good are they when you get the results back and the patients go to surgery? And you can see here it's fairly good. So 85% to 93% concordance with your cyst type and 75% to 100% concordance with grade. What about needle confocal laser endomicroscopy, also known as NCLE? There are now a large number of prospective trials, seven in total. The most recent one is by Dr. Krishna, it was the index study which was published last year looking at 144 patients, which showed excellent sensitivity and specificity for differentiating IPMNs and MCNs from other types of cysts. They've been shown to be more accurate than cyst fluid CEA or cytology with a 3% adverse event. Again, you need a 19-gauge needle, and if you look at the studies, almost all the cysts that they're used in are large cysts, typically over two to three centimeters. What else do we have? So this is a very nice study published by the group in Pittsburgh, Atra Singhi, in GUT in 2017, and he looked at molecular markers. When I look at where we are at the moment, I think we have the tools to tell most patients what type of cyst you have, so I think that's done. The next question is, if you have an IPMN, how do you tell which patients have got high grade dysplasia or early invasive cancer who truly will benefit from a Whipple or distal pancreatectomy? This is a nice study done by Atra where he showed that the presence of P53 or PIK3CA mutation was associated with almost 80% sensitivity and 96% specificity for high grade dysplasia or invasive cancer. So what about the future? So where are we now? I've given you the current what should you do when you go back and see the patient in a week or two's time, and where we say, okay, what's the size of the largest cyst? But this is really where we want to go, and I hope this is not where we'll be in the future. So what I hope we're going to be in the future is that we're going to use not only the imaging and the different modalities that we have, but they're going to be combined, and again for IPMNs with other types of things, looking at where are your mutations, how do you identify your patients with high grade dysplasia, and I think the really exciting work is the work that's been done looking at can you use, you know, what's the role of inflammation for causing progression? Could we give patients anti-inflammatories? Could we stop progression by giving them a vaccine for KRAS? And I hope that my talk, I hope when I come, if you invite me back to give a talk in three to five years' time, that it will be very different. So thank you so much for your time.

Dr. Anne-Marie Lennon

pancreatic cystic neoplasms

evaluation and management

cyst types

diagnostic tools

surveillance intervals