Mr. Pancreas Goes To Washington

The Best Of DDW 2024

Digestive Disease Week is clearly the premier annual gastroenterology meeting, with over 13000 participants, 4300 oral abstracts and poster presentations, and 400 lectures. Attendees are overwhelmed by the sights and sounds of GI science and technology. It's very easy to miss important things, which you only realize you missed once you got home!

This special issue of Sweetbreads will help. There were hundreds of pancreas-related abstracts and talks presented at DDW in 2024. I have compiled the best of these for your review.

Abstract Mo1411: PREDICTORS OF POST-ERCP PANCREATITIS: A COMPREHENSIVE SYSTEMATIC REVIEW AND META-ANALYSIS OF 229 STUDIES WITH OVER 3.3 MILLION PATIENTS

In this fascinating glimpse into the causes of post-ERCP pancreatitis, Beran et al. conducted a comprehensive metanalysis of over 3 million patients across 229 studies. They found 36 unique risk factors for developing post-ERCP pancreatitis. The highest patient-related risk was in patients with asymptomatic choledocholithiasis; prior post-ERCP pancreatitis; sphincter of Oddi dysfunction; age < 40; and female gender. The highest procedure-related risk was in patients who had pancreatic acinarization, difficult cannulation, any pancreatogram, and EPBD of an intact papilla. Fellow involvement was a risk (RR 1.5) but biliary sphincterotomy was not.

Abstract Tu1429 THE USE OF CANNABINOIDS IS ASSOCIATED WITH A LOWER RISK OF PANCREATIC CANCER

This abstract reviewed NIS data from 2016-2017. During that time, 41,180 patients were diagnosed with pancreatic cancer with a mean age of 68 and slight male predominance (52.1%). In that group, the highest risk factors for pancreatic cancer were chronic pancreatitis (OR 4.67); prior pancreatic cyst (OR 3.77); and pre-existing diabetes (OR 16.69). Interestingly, users of cannabis had a very low risk of pancreatic cancer (OR 0.63), and the prevalence of pancreatic cancer among users of cannabis was low at 0.1%. Interesting and obviously more study is needed!

Abstract Tu1461: EXOCRINE PANCREATIC INSUFFICIENCY IN PATIENTS WITH MALIGNANT PANCREATIC NEOPLASMS

In this interesting observational cross-sectional study, Gutierrez-Lozano et al. found that 83% of patients with pancreatic cancer had exocrine pancreatic insufficiency even though only 54% had diarrhea. If patients were jaundiced or had a history of alcohol consumption, the risk was even higher. Whipple surgery conferred the highest risk of EPI, up to an OR of 13.8 for severe EPI. This study concludes that EPI should be proactively screened for in this population, even if they have no symptoms.

Abstract Sa1451 GLUCAGON-LIKE PEPTIDE-1 RECEPTOR AGONISTS USE DOES NOT INCREASE THE RISK FOR ACUTE PANCREATITIS AND IS ASSOCIATED WITH LOWER COMPLICATIONS IN PATIENTS WITH TYPE 2 DIABETES WHO DEVELOP ACUTE PANCREATITIS: A MULTI-CENTER ANALYSIS

Nieto et al. tried to answer the often-asked question about whether the extremely popular weight loss drugs GLP-1 agonists really are safe. In this large retrospective cohort study, patients taking GLP-1 agonists did not develop acute pancreatitis at a higher rate than matched controls. Notably, if patients did develop pancreatitis and were on GLP-1 agonists, they had less complications such as shock, need for mechanical ventilation, or AKI. In addition, all-cause mortality was lower in the GLP-1 group who also had diabetes!

Abstract Tu1405: UPFRONT SURGICAL RESECTION IS SUPERIOR TO NEOADJUVANT THERAPY FOLLOWED BY SURGERY (NATS) FOR STAGE I PANCREATIC DUCTAL ADENOCARCINOMA (PDAC): SURVIVAL ANALYSIS OF THE NATIONAL CANCER DATABASE (NCDB) (2004 - 2020)

Is surgery or neoadjuvant therapy best in resectable stage I pancreatic cancer? Neoadjuvant therapy has become the standard of care in pancreas cancer. Perrotta et al. performed a survival analysis using the National Cancer Database from 2004-2020. The results, frankly, were quite surprising. 5966 patients were identified. There was worse overall survival in the group that received neoadjuvant therapy vs upfront surgery. A representative Kaplan-Meier curve from the study is below.

More to say on this important issue I'm sure!

Presentation Number 1010: IMPACT OF EUS-GUIDED CHOLEDOCHODUODENOSTOMY VERSUS TRANSPAPILLARY ENDOSCOPIC BILIARY DRAINAGE ON THE INTRA- AND POST-OPERATIVE OUTCOME OF PANCREATICODUODENECTOMY: A MULTICENTER PROPENSITY SCORE MATCHED STUDY

EUS-guided biliary drainage is the "new kid on the block" for either primary drainage of a bile duct obstructed by malignancy or after failed ERCP. However, because the intervention occurs within the surgical field, surgeons have been highly reluctant to opt for this means of biliary drainage since there is the possibility of making surgery either difficult or potentially even impossible.

Fritzsche et al. studied patients in the Dutch Pancreatic Cancer Audit between 2020 and 2022. During this time period, 641 patients underwent Whipple surgery for pancreatic cancer and, of that group, 34 (5.3%) had EUS-guided choledochoduodenostomy (EUS-CD). Major post-operative complications occurred in 29% of the ERCP group and in 18% of the EUS-CD group, and this was not statistically significant. In addition, the creation of the EUS-CD using a LAMS did not complicate the eventual performance of Whipple surgery, with only 28% reporting that the LAMS increased difficulty slightly.

Obviously more to come on this important topic (see below)!

Presentation Number Mo1438: PRIMARY EUS-GUIDED BILIARY DRAINAGE VERSUS ERCP FOR MALIGNANT BILIARY OBSTRUCTION: META-ANALYSIS OF RANDOMIZED CONTROLLED TRIALS

EUS-guided biliary drainage has finally come into its own as a procedure for draining malignant biliary obstruction, and in expert hands is extremely useful. However, does this mean that it can replace ERCP completely? Jaber et al. attempted to answer this question with this metanalysis of 6 RCTs with 528 patients. They found no significant difference between the rates of clinical or technical success using either ERCP or EUS-CD first. They also found no significant difference in adverse event rates between the groups.

Once again, depending on the local expertise of the endosonographer, it appears that in fact, yes, EUS can replace ERCP for the drainage of malignant biliary obstruction. Your mileage may vary however and ERCP is far from dead!

Abstract EP60: ENDOSCOPIC ULTRASOUND-GUIDED CELIAC PLEXUS BLOCK ONLY RARELY LEADS TO LONG LASTING REDUCTION OF ABDOMINAL PAIN AND OPIOID CONSUMPTION IN PATIENTS WITH CHRONIC PANCREATITIS

In older series, EUS-guided celiac plexus block has been found to be only moderately effective in providing long-term analgesia in patients with chronic pancreatitis. Does this maxim still hold? Ruus et al.'s study says yes. They reviewed the benefits of celiac plexus block and neurolysis on 24 patients with chronic pancreatitis over a period from 2018 to 2023. The authors found that only 9% of patients had a reduction in opiate consumption lasting more than 6 months. More importantly, 4% of the patients developed long-lasting complications from the procedure, leading the authors to conclude that EUS-guided celiac plexus block should rarely be used for patients with chronic pancreatitis.

Abstract Number Tu1449: NAVIGATING THE PANCREAS CYST SURVEILLANCE DILEMMA: IS LESS, POTENTIALLY MORE? LONG-TERM FOLLOW-UP OF LOW-RISK BRANCH-DUCT INTRADUCTAL PAPILLARY MUCINOUS NEOPLASMS (BD-IPMNS) IN AN ENDOSCOPIC ULTRASOUND COHORT

It's hard to believe that we are still struggling with how to manage pancreas cysts in 2024! But we are and this abstract (as well as the next one) sheds a little more light on the subject. In this study from Mayo Clinic, Pisipati et al. followed 131 patients with BD-IPMN for at least 5 years. At baseline EUS, patients were stratified into very low risk, low risk or high risk based on accepted criteria. None of the 20 patients with worrisome features on baseline EUS progressed to malignant transformation on follow-up imaging. Surgical intervention and disease-specific mortality was 0% at long-term follow-up. Cyst growth remained low at over 8 years of follow-up.

Abstract Tu1441: OUTCOMES AND COST OF PANCREAS CYST SURVEILLANCE IN PATIENTS 70 YEARS AND OLDER

Given the findings of the abstract above, is it really cost-effective and clinically beneficial to monitor patients older than 70 years old with pancreas cysts. Bosch et al. say yes! They reported on 451 patients in their pancreas cancer prevention clinic. 124 were > 80 years old and 190 were between 70 and 75 years old. They found that although it became more costly to monitor patients > 80 years old, the rate of resection and the findings of HGD/cancer were similar in that age group compared to younger patients. The authors concluded that age alone should not be the sole determinant for discontinuing surveillance.

Abstract Tu1480: NONAGENARIANS DO NOT HAVE INCREASED MORTALITY OR WORSE POST-OPERATIVE OUTCOMES WHEN UNDERGOING PANCREATIC SURGERY

This abstract demonstrated that even 90 year olds could do well after pancreatic surgery! Pedro et al. reported a retrospective cohort study (2016-2019) using the National Inpatient Sample. 62% of the group > 90 years old had surgery for malignancy in the head of the pancreas (524 patients). After adjusting for confounders, nonagenarians did not have increased odds of in-hospital mortality after pancreatic surgery [adjusted odds ratio [aOR] 1.47; 95% confidence interval [CI] 0.41-5.28)], requirement of mechanical ventilation [aOR 1.00], in-hospital shock [aOR 0.73; (0.08-6.07), post-operative shock [aOR 1.00], wound dehiscence [aOR 1.00) and post-operative infection [aOR 1.16; (0.15-8.92)].

Finally, for a peek into the future, we have to throw in some AI ;)

Presentation Number 90: MACHINE LEARNING ANALYSIS OF ENDOSCOPIC ULTRASONOGRAPHY TEXTURE AND CLINICAL INFORMATION PREDICTS PANCREATIC NEOPLASTIC PROGRESSION IN HIGH-RISK INDIVIDUALS WITHIN 18 MONTHS

Yan et al. used data from the CAPS at Johns Hopkins to help come up with a machine learning model using clinical and EUS texture features to help predict the development of pancreatic cancer within the next 18 months. Using 24 clinical features, the model had 50% accuracy in predicting progression to HGD or PDAC (see below). EUS texture analysis-based model using 726 images had 78% accuracy. The model using combined clinical and EUS features had 96% accuracy! Wow.

This is obviously only a taste of the wealth of pancreas information presented at Digestive Disease Week? (DDW) this year! If you want to dive deeper, look here.

See you in San Diego in 2025! American Society for Gastrointestinal Endoscopy (ASGE) American Gastroenterological Association (AGA) American Association for the Study of Liver Diseases (AASLD) SOCIETY FOR SURGERY OF THE ALIMENTARY TRACT

FOLLOW me and SUBSCRIBE to this NEWSLETTER for the most pancreas content