Localized Pancreatic Cancer x Dr. Michael Pishvaian 2026 UPDATE Part 1

[Music] Welcome back everyone, this is Sam. And this is Karine and we are two octux. This week's episode, as well as next week's episode, we'll be talking about pancreatic cancer, with a very special guest, a mentor and a personal friend of mine, Dr. Michael Pishfain. He's the associate professor in GI oncology at the Johns Hopkins Kimmel Cancer Center, and Mike is a fellowship trained GI oncologist who specializes in pancreatic, as well as refractory colorectal cancers. You're committed to precision medicine and providing your patients with the most up-to-date, awesome care. Mike, you also created tumor board Tuesday on X, which is very, very popular, and there's a molecular angle and you guys go over different cases in every disease type, every Tuesday at 8 p.m. And so this week's episode, we're going to be focusing on risk factors, diagnosis, treatment for localized pancreatic cancer, and next week we'll be covering metastatic disease. And so to start us off, Mike, what are our risk factors for pancreatic cancer? So that's a question that, as you can imagine, I get off them. And by the way, before I forget, thank you so much for allowing me to be part of this. You know, Sam, you were obviously one of the best fellows we had, and I'm so glad that you really have done so well for yourself, and it's nice to meet you too. And I think you guys have built a great program here that I think is helping a lot of trainees and fellows and everybody else. So great job, and I'm really proud to be part of this today for these sessions. So to talk about pancreatic cancer, you know, a lot of people will, I have the very active discussion with patients. You know, why did, why were you struck with this cancer? And the, and the reality is that 90 plus percent of patients, there's really no rhyme or reason why they were struck with this cancer. It is a cancer that affects people who are older, although ironically, the fastest small growing subgroup of patients are younger women for reasons that we absolutely don't understand. And about 7 to 10 percent have underlying genetic risk factors that I know we'll talk about. But of the other 90 is percent. There really aren't these very powerful risk factors. It's not like smoking and lung cancer, at least how that used to be. There is a slightly increased risk with smoking. There is a slightly increased with obesity. There is a slightly increased risk with diabetes, which is confounded by the fact that 10 percent of pancreatic cancer patients will get diabetes. All of those are legitimate risk factors, but they're very small or sort of minor risk factors. Most of the time we really don't understand why patients were struck with this disease. Definitely. And so like you alluded to, sometimes as a genetic risk. And so what are those genetic risk factors that we should all be aware of as clinicians? So forgive me if I'm going to get on my soapbox a little bit about this genetic risk factors, because I think this is something that we really need to do our best to educate our community about. Not just patients and their advocates and their caregivers, but really the whole physician community because this is something that still isn't firmly understood. So about 7 to 10 percent of all pancreatic cancer patients are born and underlying genetic occasion that they were born with that over the decades of their lives led to these cancers. And actually that list has been growing steadily when I joined Hopkins in 2020. The germline testing panel that we sent at that time was 32 genes, which is up from two or three genes 10 years earlier than that. Now actually we routinely test for 86 genes, and that number does seem to be growing every few years. Of course, a lot of people do know about BRCA1, RACO1, RACO2, as probably the predominant genetic risk factors, but there's a whole slew of other genetic mutations that have been either directly or at least loosely associated with pancreatic cancer. Many of them fall into that BRCA homologous recombination DNA repair pathway. So PALB2, GRIP1, BAC1, all these mutations that are associated with homologous recombination repair. And then there are other risk factors too, CTN2A, which can be associated with familial melanoma and pancreatic cancer. There are certainly lynched syndrome patients with microsatellite instability and MMR deficiency that can lead to pancreatic cancer and then some other rare causes as well. Probably the most powerful risk factor for pancreatic cancer are the STK11 germline mutations in Pugyegar syndrome, and that can lead to pancreatic cancer and that has a massive increase in the risk of pancreatic cancer. But it's just very, very, very rare. So we just don't see it all that often. And similarly, lethal, mania and TPP53 mutations. Did you probably see that more in sarcomas than I ever have seen in pancreatic cancer? I think the important thing about this discussion now is that there are definitely genetic risk factors. And we as a community have decided now about seven or eight years ago that 100% of pancreatic cancer patients should be tested, should have germline clustering to look for an underlying pathogenic mutation that may have led to this cancer. And importantly, that's irrespective of their stage of disease, that's irrespective of any family history. We should be testing every patient with pancreatic cancer when they walk on the door. That's a message that has not gotten after the community well enough. And I definitely, it's something that I'm very focused on making sure that all of our patients get tested. Everything you said is so incredibly important and high yield. So, you know, if there is a main takeaway from this episode, it's every patient with pancreatic cancer regardless of stage should have germline testing. So, so important. And so, what about somatic mutations in pancreatic cancer? Yeah, so somatic mutations, for years I was, you know, kind of giving talks about the fact that about 17 to 25% of patients with pancreatic cancer do have what we would call so-called actionable mutations. And that was important and important message to get out because even that message really wasn't getting out very well. I had holidays in mind, even good friends of mine who are committed pancreatic cancer specialists would sort of say to me tongue in cheek, why should we test everybody because only a few patients, only 2% of patients actually have mutations. But actually, again, 7 to 25% of patients have an underlying genetic mutation that could be therapy determinative, could relate to some treatment that we ultimately may want to give to the patients. Of that, 7 to 25% most of them fall into the category of DNA damage response from repair, DVR mutations. So, again, BRCA1, BRCA2, PALB2. And for those patients, just plain old platnums are very valuable. But now, as we are doing more testing, we are unearthing those very small slivers of the pie of patients who have very highly actionable genetic mutations. So, for example, about 1% of patients with pancreatic cancer have MSI high disease, about 0.7% of patients have intracut fusions. And actually, the newest testing, the newest therapy is for NRG1 infusion positive pancreatic cancers, which only exists in about 0.2% or 2 out of every thousand pancreatic cancer patients. And there could be a legitimate discussion as to whether we should be testing all of these patients. But the reality is that those patients have really disproportionate benefit when they get the appropriate, appropriately targeted therapy. These targeted therapies for the most part have been better than what we would expect with chemotherapy. And so, I feel we absolutely should be testing all patients. And in fact, the guidelines support that we should be testing every pancreatic cancer patient with advanced disease to try and identify these underlying genetic mutations. However, a lot of this has changed even or are minted in just the past few years because while we would talk traditionally about 17 to 25% of patients having mutations, now all of a sudden, after 30 years of trying, we actually have effective K-RAS inhibitors. And K-RAS mutations exist in about 90 to 95% of all pancreatic cancer. So all of a sudden, this disease has become a disease for which there are virtually 100% of patients have an actual alteration. Most of those are going to be K-RAS, but virtually all of them have an actual alteration and we should also be testing everybody. That's really exciting. So just to clarify as far as somatic testing, are you recommending for both localized and advanced disease? So the current NCCN guidelines are the way that it's worded very carefully. The way that it's worded is that all patients with advanced disease, including locally advanced disease, for whom therapeutic options are being considered should be tested. When you look at those numbers, that's probably at least 80 to 90% of patients. The sad reality is that most patients with earlier stage disease ultimately have recurrent disease, so I think those patients should be tested as well. There is a big push, although it hasn't needed that far yet. There's a big push that we should just change our wording and say 100% of patients with pancreatic cancer still have somatic tumor testing. At least then, Dylan, you know, I, in my circummabation, I do it up front so that I have the results in case I need future. Absolutely. That's so important. And so next, what is the usual clinical presentation of pancreatic cancer? So pancreatic cancer is when you, when you see enough pancreatic cancer patients, you start to see some of the old nations dispeled. And one of the old nations is that, you know, at least I learned in medical school, which is now a long time ago, is that pancreatic cancer patients walk in the door incredibly sick. And that is it. is still true for a decent percentage of patients. But I see plenty of patients who walk in the door feeling completely healthy. They went and they had a CAT scan because their doctor suspected a kidney stone or something else and lo and behold they had metastatic radicants. So I think the traditional textbook presentation is weight loss, loss of appetite, fatigue, and of course abdominal pain. But there are definitely palliations who don't have those symptoms and as I'm sure you remember from medical school about 10% of patients with acetic cancer have the classic presentation of painless genres where they turn yellow but they actually don't have pain and don't have any other outward overt symptoms of acetic cancer. What we have also learned and really become very clear just in the last five or 10 years is that new onset diabetes is a clear sign that a patient could have acetic cancer. This gets really hard because I think there's something like 30 million diabetics in this country in the United States. So obviously thank god 30 million people don't have acetic cancer. But the clues we look for are people who are older and never had diabetes all of a sudden getting diabetes or even people who had diabetes and had excellent control of their diabetes all of a sudden that diabetes getting much more difficult to control. So new onset or a change in diabetes for people who are over 50 can also potentially be a sign of underlying pain-grade symptoms. So important to remember that one absolutely. And so if a pain-grade cancer is suspected what is our workup that we need to be starting it and getting rolling? So one of the things that I really try and coach internal medicine docs, primary care doctors, gastro-concral just is that when you have a patient that's walking in the door with some of the more constitutional symptoms that a pain-grade cancer including weight loss and lost appetite, you know that doing an endoscopy and a colonoscopy, well yes, importance in and of itself isn't going to survey all of the areas that we're concerned about. So ultimately cross-sectional imaging is going to be critically important. Ultrasounds are not very helpful in pain-grade cancer. Of course if there are clear liver lesions, the ultrasound may pick it up, but a pancreatic mass can be easily missed on a pancreatic, can easily miss a pancreatic cancer. Even CT scans can miss pancreatic cancer if the contrast is not administered the right way. So a pancreatic protocol CT scan is really kind of the gold standard. MRIs, they aren't necessarily any better than a pancreatic protocol CT scan. It's just that a lot of times the CT scans are either not done with contrast or they're done with the contrast in a later beam of phase that may miss the pancreatic mass that ends up being caught on the MRI as well, but some kind of high-level, high-quality, proxxial imaging will be really critical. Of course to visualize the pancreatic mass, the true gold standard also is an endoscopic ultrasound, but of course that's an invasive procedure and it should be certainly done of course on any of the, on every patient and certainly not a screening test either. Right and I know one labs that we often grab a tumor marker, CA199, it can be elevated obviously in pancreatic cancer, but what else could elevate CA199? Is there any red herrings? Yeah definitely. CA199 can be elevated for a variety of reasons, including just sort of underlying, non-specific liver dysfunction. CA199 is released by the biliary cells, so the tumor irrigates the bile duct cells and causes the releases of CA199. And so you use the right term that CA199 can absolutely be about herring. In fact, we strongly recommend that CA199 is never a screening tool, so you know, a person who's seeing their, their concierge primary care docs should never get an annual CA199. Thankfully I haven't taken care of many, but I have taken care of maybe two patients in 20 years who were referred to me because of the elevated CA199 only, and we have done the million dollar work up and thankfully for those patients never found a pancreatic cancer, but really CA199 should not be a screening test. They're useful to track, CA199 can be useful to track the response to therapy for patients with pancreatic cancer, but not as a screening tool. And the other important thing about CA199 is when a pancreatic cancer is diagnosed, it is a useful tracking tool, but about 30% of patients do not have, do not generate CA199 with their pancreatic cancer. Of that 30%, maybe a third of them, so about 10% overall will make CEA instead. So often I will check CEA and CA199 up front, and then I'll follow the marker that's elevated. In about 20% of patients don't have any secretion of a relevant biomarker to the adult track overtime. That's really important. And actually higher numbers would have negative to my markers than I would have expected. So important clinically to be thinking about checking CEA as well. And next, how should our trainees be thinking about staging for pancreatic cancer? This is a running joke in my clinic because I am not infrequently caught off guard when a patient asks me what is your stage of pancreatic cancer and I sort of have to think about if they're not at a stage 4, I kind of have to think about it. Stage 1 of these two, I can't remember, because really what it comes down to is we have three relevant staging categories in pancreatic cancer. And that's patients who have clearly operable, clearly recyclable disease from the get-go at diagnosis. And that's only about 10% of patients. Patients who have clearly metastatic disease at diagnosis, so the opposite end of the spectrum and that's 55% of patients. And then about 30% to 35% of patients in the middle who have the localized disease. We used to separate localized disease into borderline recyclable and locally this unresectable. Part of the reason that we got away from that definition is because we used to think that locally this unresectable could never be rendered recyclable, but we now know that some of the patients can be rendered recyclable. So we try to lump it into locally, locally, but not necessarily unresectable disease. So those are the three categories that really ultimately matter partly because those are the three categories that will guide how we treat the patients. For patients who are clearly recyclable, you know, it's going to be surgery first or in certain circumstances, chemotherapy first. For patients who have localized disease, definitely chemotherapy first, but still possibly surgery. And then of course, the patient with advanced disease, it's going to be chemotherapy or systemic therapy only. And you know, the same is true in prostate cancer. I often get caught off guard when patients ask me for their stage in localized disease. So I can definitely relate. And so as far as the definition for patients in the localized setting that might be unresectable, what are some of the features that we can look for? Well, one of the things that's really important in patients who have localized disease, it's been studied and we actually know that patient outcome is better when patients are evaluated in a multidisciplinary setting. And that's because there's value to getting the input of the surgeons, the radiation oncologist, the medical oncologist, and the radiologist who's reviewing the scan with you. And while we have our NCCN definitions of recyclable, borderline recyclable and locally advanced, unfortunately, the tumors don't always read the textbook and don't exactly follow the guidelines exactly as we would love to classify them. So there is some variation in the eye of the, ultimately, the surgeon who would be doing surgery to define whether he or she thinks that this is a borderline recyclable or locally advanced recyclable. I try and hold my surgeons' feet to the fire and try and keep them within guidelines as much as possible. But at the same time, they are definitely, there is some subjectivity to defining a disease as recyclable, borderline recyclable, or localness unresexual. So you answer your question more specifically though, what we use to gauge whether or not these tumors are recyclable or unsectable, it really has to do with the interaction of the tumor with the local vasculature and also with the other organs as well. So the arteries are probably the most important part of this because the arteries are, it's harder to sacrifice. We used to say we can't sacrifice the arteries, but now there are surgical techniques that do allow the surgeons to reconstruct the arteries. But classically, our truly recyclable tumor is one that doesn't involve the local arteries at all. A borderline recyclable cancer is one that abuts, that sits next to the arteries, but does not incase any relevant artery more than 180 degrees. And then locally advanced diseases are those that surround the critical arteries, the siliac plexus, the siliac artery that's superior mesencharect artery by more than 180 degrees. And what it really comes down to is if the surgeon were to go in there and do surgery, we generally don't think that surgeons can quote unquote peel the tumors off the artery. That doesn't work. That leaves behind disease, that leaves behind margin-positive disease and it doesn't help the patient. So it's really about can they sacrifice that portion of the blood vessel? If it's a short section of the vein, they can actually cut out that section, remove the tumor and reconstruct the vein. As I mentioned, more and more techniques are evolving where they can cut out certain sections of some of the arteries as well. But if the tumor is positioned in a way that does not allow for resection of the artery, then unfortunately, that's not an operable cancer. I think that's a really great way to think about it as far as it being more prohibitive to have a but met on the artery than on the vein when fellows are looking at these cases. And so if a patient has what is deemed to be resectable upfront by the trained surgeons, what is your approach for those patients? That's a very controversial topic. So in a good way and a good way, I think that that's basis evolving. So for 25 plus years, it was surgery directly. And a lot of that was because we didn't have particularly good chemotherapies. So patients would go and they would have surgery and we would give them some genocide of being postoperatively and help a little bit. But it really wasn't powerful enough to to improve surgical outcomes. There was really no reason to give the chemotherapy up front. Really with the advent of full phyrinox and gem-sytidine nampacletaxyl, not that they're that they're great regimens, but they are definitely better than single-agine gem-sytidine, single-agine five-a-few. And we are now seeing tumor-at-actory responding and it increased the justification for giving preoperative therapy. And so for patients who had who don't have clearly resectable disease, meaning like borderline resectable disease, all of a sudden we realized that we really should be giving them chemotherapy up front in parts to try and kill some of the microscopic metastasis that the patients inevitably have, and also in part for some of those patients to render their tumor surgically alcohol by extracting it from the local blood vessel. So taking that, the efficacy of those regimens in the resectable patient population, a lot of us, and I include myself in this category, 100%, said, "Well, handcratic cancer is a systemic disease, you know, a half a centimeter tumor, a node negative, can still have a fairly high risk of eventually developing nettostexia, even with successful surgery." So clearly there are micrometastases at the time of diagnosis. We absolutely should be giving those patients chemotherapy first and then taking them to surgery. And so for a number of years, many of us were really focused on giving the odds of therapy first, even though we hadn't necessarily proven that it was better. The risk of doing chemotherapy first prior to surgery is that for various reasons, most of them tumor biology related, meaning that the cancer grew, about 30% of patients, 20 to 30% of patients who have a clearly optimal cancer who get on pre-operative systemic therapy, they actually never make it to the operating room, usually because their cancers grow and spread while they're actively on chemotherapy. They obviously had very poor biology disease that was going to be very aggressive, but at the same time, the concern is that they may have missed their opportunity for a potentially curative surgery. So all that background is say there are two large, ongoing randomized trials, both of which haven't actually completed a cruel one of the United States that has run to the Alliance and won through the Dutch. They're very, very similar studies and we hope that they will give us a definitive answer as to whether or not we should use preoperative therapy. So the study designs basically were patients were randomized, patients with clearly optimal disease were randomized to receive four months of preoperative fulfillinox, followed by surgery, and then two more months of postoperative chemotherapy, or they were taken directly to surgery and then upon recovery were given six months of post operative fulfillinox. Those trials are going to be really important to tell us definitively should we be giving preoperative chemotherapy or not. Right now, today, again, it's a case by case discussion that should be made in a multidisciplinary setting. There are some factors that guide it like patients with a C-A199 of over 200 should almost certainly get preoperative chemotherapy because their risk of developing this disease is definitely higher. If there's anything started off in the surgeon that what might be by the textbook, a resectable cancer that they feel that there may be some concern to intraoperatively getting all the cancer out, then that patient should receive preoperative chemotherapy. But to give every patient with resectable disease preoperative chemotherapy, we don't technically know that that's what we should be doing. Definitely. So I obviously trained under you and I remember a lot of the discussions of we feel we should give new adjuvant. I remember the test answer at the time and I think you tell me if you think this is still the test answer as if resectable surgery first, but obviously like you alluded to, there's research and trials, you know, hopefully they will answer who and how much neoadjuvant chemo is at all six months up front? Is it a sandwich approach? Three months, surgery, three months, those types of things. But would you agree Mike that the board, quote unquote, right answer is surgery first if resectable? Yeah, I do have to agree right now. The board answer is clearly operable resection first. Right. And so next, what are the types of surgery used in the treatment of pancreatic cancer? I can really introduce this as an on surgeon. Obviously, obviously we have the Whipple. And the Whipple is the pancreatic doodinous, the doodinectomy. So basically they take the head of the pancreas and some of the body of the pancreas as well as the the doodinum. And that's really in a big obligatory resection because of the shared blood supply. When they do that, they reconnect the intestine lower down to the back to the pancreas. I'm sorry, I said that wrong. Back to the stomach. And even if it's a even if it's a pylorus sparing Whipple, which most of them are these days in, you know, 2026, most of them are now pylorus sparing Whipples. They still can't reconstruct the patient such that the food content is going from the stomach directly into the intestine. Through the pylorus, it's now going through a connection that's made in the sort of inferior surface of the stomach. And that's a lot of the issue with the Whipple is that sort of now unnatural way for food to flow from the stomach down the inferior port, not through a sphincter, but directly into the intestine. They connect the pancreas, the remaining pancreas, and the but the biliary tree back up to the intestine, but it's actually reconnected upstream of where the food is actually entering. And so you don't get this sort of perfect mixing of digestive enzymes with the food as it's coming through. And that can contribute to a lot of the sort of post-Whipple digestive issues that patients find out. But about two-thirds of patients with pancreatic cancer that's resectable are appropriate for a Whipple surgery, and then about one-third of patients will get, we'll have a distal pancreas tumor. And so they're appropriate for a distal pancreate technique. And again, because of the oblique various shared blood supply, it's a distal pancreate technique with a splenectomy. Those surgeries tend to be easier to do. They tend to be easier to recover from and tend to cause less digestive complications long-term because that kind of surgery really shouldn't impact the stomach and the dyadim and the pancreatic duct and the biliary duct. Once in a while there are patients for whom an appropriate surgery is a total pancreateectomy, but that's a really rough surgery on a patient because they're losing their whole pancreas and they, by definition, become brittle diabetics that need a lot of blood sugar to control, basically, for the rest of their lives. And then the one surgery that's not appropriate for pancreatic adenocrystamins, I don't think many people do it, but you know, just wanted to reinforce. There's really no role for an enucleation of a pancreatic mass. In other words, where they just go in and scoop out the pancreatic tumor. That could be appropriate for certain times of pancreatic tumors, particularly well-selected pancreatic neuroendocrine tumors, but never for pancreatic adenocrystamins. And I think one board pearls, if someone has a distal pancreatic to me with a splenectomy, don't forget those vaccines. So if the splenect comes out, vaccines, vaccines, vaccines. And so I know you alluded a lot to new adjuvant and adjuvant chemotherapy regimens. How do we pick one and what are they? Um, pancreatic cancer, chemo therapy regimens are actually relatively straightforward. We partly because we don't have enough days to re-evenance of benefits. So the chemotherapy, adjuvantly post-operatively, have evolved since the 1990s. We first had gem-sidedine proving benefit as a single agent in the post-operative setting. And it did have a very modest but real improvement in survival rates. Shortly thereafter, we had single agent type FU, similarly proved benefit in the post-operative setting. There was a big trial, the ESCH-3 trial, that compared a single agent type FU to single agent gem-sidedine. The way that the Europeans gave the single agent type FU was actually a little bit harder to tolerate the gem-sidedine. And the outcomes were exactly the same between those two arms. And so that, so single agent gem-sidedine became the de facto standard of care in the adjuvant setting for pancreatic cancer. Fast forward to the 2010s when full phyrinox clearly showed benefit over single agent gem-sidedine in the metastatic setting, the French then replicated that in the adjuvant setting in 2018, we had clear unequivocal data that full phyrinox was superior to single agent gem-sidedine in the adjuvant setting. So that really is the gold standard. There was a regiment a couple of years before that in 2016. gem cytobene and capes cytobene together, given in the adjuvant setting. And it was an effective regimen, clearly better than single agent gem cytobene. We haven't to date ever had a head to head, ever had a head to head to compare some of adjuvant polpherinox to adjuvant gem capes. But because of the results with polpherinox, we're so, so much more powerful that polpherinox really is the gold standard. It's only a patient who clearly can't tolerate polpherinox for whom we should be offering gem cytobene, kid cytobene. And of course, then you're asking, you have any sort of familiarity with this, you're asking, well, what about gem cytobene and napackletaxle? Well, surprisingly, gem cytobene napackletaxle in the adjuvant setting was not proven to be better than single agent gem cytobene. It was called the A-Pack trial, and it came out in the, I think, early 2020s. And the way they designed the study, it actually did not prove to be gem cytobene napackletaxle in the adjuvant setting did not prove the superior to gem cytobene. And so really that is not in a regimen that is appropriate for use in the strictly adjuvant setting. In fact, I've had insurance rejected a couple of times when I tried to use it in the adjuvant setting. And they insurance companies, in that case, were not actually technically wrong. However-- The one time. The one time, yeah. However, in the preoperative setting, for those situations where you do choose preoperative chemotherapy, then you can definitely justify if you use full furionauts or gem cytobene napackletaxle. The one study that we have for that to compare the two was the SWAG 1515 study, compared to both of those regimens in the neoadjuvant setting and they both seem very appropriate. So when we choose to use preoperative therapy, we pick full furionauts or gem napackletaxle based on sort of the feature of the patient and what we can take and handle in their performance status. But either would be very appropriate. So switching gears from systemic therapy and surgery, when would you think about using postoperative radiation in localized pancreatic cancer? This has become a very controversial topic. And it's frustrating because we want radiation to be an important part of the treatment pancreatic cancer. But really, we have to be aware of the fact that radiation has never been proven to improve survival in any randomized trial for pancreatic cancer. And this became very apparent when we had the results of the RTOG study come out just about a year and a half ago. That was a large study that was run over almost 10 years to give a sense of how old that study was. When it started, it was using adjuvantium cytoding. It was later adopted to allow patients to receive adjuvantful furionauts. But basically the study was for patients who'd undergone a successful surgery for their pancreatic cancer, they were then randomized to. They were start on chemotherapy and then randomized to receive radiation therapy or not. And maybe to everybody's surprise, maybe not. There was pretty convincing evidence that there was zero benefit of adding radiation therapy in the adjuvant post-opera setting. And in fact, again, I think the gut feeling that many of us would have was that, "Oh, okay, there's a cause of margin." And we should definitely use radiation to quote unquote clean up or sterilize the fields. Actually, it was that group of patients that had absolutely no benefit in radiation therapy at all. If you really kind of tease the data out, the small group of patients who maybe had some benefit in radiation, but it was not a pre-plan analysis, were those who were no negative. So ironically, actually the lower risk groups group of patients had a slightly greater benefit in radiation therapy. But I think pretty clearly the data showed us pretty clearly that the positive margin patients do not benefit in adjuvant radiation. And this has become a point of controversy. I've had some headbutting with some of my radiation oncology colleagues about this data, but I no longer recommend adjuvant radiation therapy after surgery for pancreatic cancer. Thank you for sharing. And finally, when is pre-operative radiation indicated? So similarly, still don't have any fern data that we should be getting pre-operative radiation therapy to anybody. But this is an area where actually many of us still continue to do so. So a couple of studies. So for patients with resectable, well, let me start with borderlines. For patients with borderline resectable disease, there was a study done by the Alliance that tested the benefits of radiation therapy. I won't go into the details of the study, but basically that study showed the giving stereotactic radiation, or SBRT, clearly was not beneficial for patients with borderline resectable disease. So many of us feel for at least that population of patients, there's no clear role for radiation therapy. And similarly, for the clearly optimal patients, there's no clear reason to give radiation therapy. The one area where I think many of us are guilty, even without true proof, there were still guilty of using a lot of radiation for the patients who were locally advanced. There's been several large studies that have looked at the benefit of local radiation therapy for local events, pancreatic cancer, including a couple of these three trials, and they did not show any survival benefits. The radiation colleges will share a lot of criticism as to where the way those studies were done. This includes, for example, the lap of seven trial. And there have been publications out looking at the use of high dose of blade of radiation therapy in the local event setting, demonstrating an extension of overall survival. But that was never an arid in the Ice Faceory study. Many of us will use radiation therapy if there's any excuse from a palliative sense to use radiation therapy to pay for pain or other symptoms, then we will for local events disease. And there is just recently launched a national cooperative group study that's going to be assessing the value of high-disability radiation therapy in the TASR setting. But right now, we don't really have proof that we should even know many of us will. As far as using it to render somebody surgically operable, again, many of us will use it, typically in a multidisciplinary, especially after multidisciplinary discussion with surgery and with radiation therapy. And even though we do it often, we still don't have any clear proof that it helps. Great. Well, you had so many pearls on local pancreatic cancer in this week's episode. And so next week, we're going to be covering metastatic disease. So to our listeners, thank you for listening. Good luck with studying. And please feel free to reach out to us with corrections or comments on our Instagram or Twitter to op docs. [BLANK_AUDIO]