AGA Abstracts Mo1208 Pancreatic Cancer Risk in Peutz-Jeghers Patients; Results of a Large Dutch Cohort Study and Implications for Surveillance Femme Harinck, Susanne E. Korsse, Margot G. van Lier, Katharina Biermann, Johan Offerhaus, Nanda Krak, Caspar W. Looman, Wendy van Veelen, Ernst J. Kuipers, Anja Wagner, Evelien Dekker, Elisabeth M. Mathus-Vliegen, Paul Fockens, Monique van Leerdam, Marco J. Bruno Introduction: Although Peutz-Jeghers syndrome (PJS) is thought to be associated with an increased pancreatic cancer (PC) risk, estimates of this risk differ widely. This hampers counseling of PJS patients and the development of optimal screening strategies. We therefore aimed to determine the risk of developing PC in a large cohort of Dutch PJS patients. Methods: All patients included in this cohort study had a definite diagnosis of PJS as defined by diagnostic criteria recommended by the WHO, a proven LKB1 germline mutation, or both. Patients were followed prospectively between January 1995 and July 2011; clinical data from the period before 1995 were collected retrospectively. All patients with a diagnosis of PC were identified. As screening of the pancreas can also lead to the detection of other malignancies in the pancreatic region, we also selected patients diagnosed with distal cholangiocarcinoma or ampullary carcinoma. To ascertain the diagnosis, clinical, radiological and histological data was reviewed by experts of all relevant disciplines. In inconclusive cases, immunohistochemical staining was performed. Cumulative risks were calculated using Kaplan-Meier analysis and relative risks using Poisson regression analysis. Results: We included 144 PJS patients (49% males) from 61 families, accounting for 5640 person years of follow-up. Seven PJS patients (6 males) from 7 families developed PC at a median age of 54 years (IQR 36-62 years), 6 of whom had ductal adenocarcinoma and one acinar cell carcinoma. All 7 patients were symptomatic at the time of presentation with advanced stage disease. The median survival after diagnosis was 5 months (IQR 3-17 months). Additionally, 2 cases of distal cholangiocarcinoma and 2 cases of ampullary carcinoma were diagnosed (3 males) at a median age of 55 years (IQR 44-69). The cumulative PC risk was 23% (95% CI 4.1;41.3) by the age of 70 years (Figure). Compared to the general population the PC risk was significantly increased (HR 76.2, p<0.001). The cumulative risk for developing PC, distal cholangiocarcinoma or ampullary carcinoma was 29% (95% CI 10.4;47.6) by the age of 70 years (Figure). Conclusion: Patients with Peutz-Jeghers syndrome are at very high risk for developing cancer in the pancreatic region, including pancreatic cancer, with a cumulative risk of 29% by the age of 70 years. This very high risk and the notion that early detection of these malignancies or its precursors will hopefully lead to an improved survival, make PJS patients good candidates for screening. This screening should be performed with annual endoscopic ultrasonography (EUS) and/or MRI within well-defined research protocols. The impact of such an approach on survival of these patients remains to be demonstrated. Mo1209 Is Postprandial Chylomicronemia More Important Than Plasma Total Triglyceride Concentration in the Etiology of Recurrent Acute Pancreatitis in Lipoprotein Lipase Deficiency? Marco J. Bruno, Janneke de Wal, Jaap Twisk, Daniel Gaudet, Norman Miller Background: In various conditions associated with recurrent acute pancreatitis (AP) plasma triglycerides are implicated to play a key role in the initiation of the inflammatory event. Chylomicrons (CM), are the main carriers for triglycerides in circulation during the postpran- dial phase. We have studied patients with lipoprotein lipase deficiency (LPLD), also known as familial (hyper)chylomicronemia, and noticed a 350-fold increased risk of AP. Although non-chylomicronemic hypertriglyceridemic states are occasionally associated with AP, this risk increase is much smaller than that in LPLD. This and other observations made during the development of alipogene tiparvovec, a gene therapy for LPLD, prompted us to perform an extensive literature review and develop an adapted and novel hypothesis regarding the role of CM in the etiology of AP in LPLD. Results: We gathered evidence from the literature supporting the hypothesis that CM, particularly large nascent chylomicrons, in high concen- tration have a pro-inflammatory effect on the pancreas, mediated by free fatty acids released from the particles by the action of pancreatic lipase. This evidence has been provided by studies in: (a) patients with hypertriglyceridemia and/or chylomicronemia, (b) animals with naturally occurring mutations of the lipoprotein lipase (LPL) gene, (c) LPL gene knockout mice, (d) experimental pancreatic preparations In Vivo and In Vitro, and (e) isolated pancreatic acinar cells. The most frequent and likely chain of events involves CM at high concentration aggregating in small capillaries and becoming exposed to pancreatic lipase. Such aggregation is likely to involve other blood elements, and to be initiated by hypercoagulability factors relating to nascent CM. This hypothesis provides ample explanation for the observed extreme increase in AP risk in LPLD patients. It also explains the results of a historical case-note review study we undertook, in which a decreased risk of AP was seen in LPLD patients treated S-622 AGA Abstracts with alipogene tiparvovec, in whom we separately demonstrated a persistent improvement of postprandial handling of newly formed CM following ingestion of a standardized liquid low-fat meal (containing 13g of fat and 896 kcal, supplemented with [3H]-palmitate to label newly formed CM particles). Conclusion: Our experience supports the hypothesis that excess CM, and particularly nascent CM, are the main cause of AP in LPLD, not only by virtue of their high triglyceride content, but also due to other characteristics, which may lead to local intra-pancreas capillary obstruction and subsequent (pro)-inflammatory events mediated by lipase-released fatty acids. Mo1210 Incidence of Pancreatic Ductal Adenocarcinoma and Extrapancreatic Malignancy in Patients With Branch Duct Intraductal Papillary Mucinous Neoplasm of the Pancreas Hiroyuki Hisai, Etsu Miyazaki, Michihiro Ono, Shota Yamada, Gota Sudo Background: Previous studies have suggested that patients with branch duct intraductal papillary mucinous neoplasm (BD-IPMN) of the pancreas may be at increased risk of developing pancreatic ductal adenocarcinoma (PDA) and extrapancreatic malignancy (EPM). The aim of this study was to elucidate the incidence and clinical characteristics of synchronous and metachronous PDA and EPM in patients with BD-IPMN. Materials & Methods: Between January 1997 and November 2011, 341 patients with BD-IPMN (160 men and 171 women; mean age 73.9 years, range 47-98), confirmed by surgical resection and typical findings of various imaging modalities including CT, EUS, MRCP and/or ERCP were enrolled. Clinical features of PDA and EPM associated with BD-IPMN, occurring in the prediagnostic or postdiagnostic period, were investigated. Patients with malignant IPMN (noninvasive and invasive intraductal papillary mucinous carcinoma) were excluded. The diagnosis of PDA and EPM was histologically verified in all patients. Median follow-up period was 42 months (range 1-176). Results: Of 341 patients with BD-IPMN, concomitant PDA was detected in 14 (4.1%) patients. Thirteen of 14 PDAs were diagnosed at an advanced stage. Twelve PDAs were synchronously detected at the time of BD-IPMN diagnosis, whereas two were identified during the irregular follow-up. All PDAs occurred in separate from the BD-IPMN lesions. There was significantly higher incidence of diabetes mellitus in patients with PDA than those without PDA (P=0.0003). The diameters of the main pancreatic ducts were significantly larger in patients with PDA than those without PDA (P=0.009). In addition, a total of 107 EPMs were detected in 95 of 341 (27.9%) patients with BD-IPMN. Thirty-eight, 56 and 13 EPM had been diagnosed before, during, and after the diagnosis of BD-IPMN. Gastric adenocarcinoma (34 patients, 31.8%) was the most common EPM. Other EPM included liver cancer (n=12), colorectal cancer (n=11), lung cancer (n=11), bile duct cancer (n=9), prostatic cancer (n=7), and others (n=23), respectively. Conclusions: Particular attention of the entire pancreas should be paid to the development of PDA in patients with BD-IPMN. Furthermore, routine systemic surveillance is necessary for patients with BD-IPMN. Prospect- ive multicenter studies are required to evaluate the true incidence of PDA and EPM in patients with BD-IPMN. Mo1211 Impact of Tumor Markers on Compliance With Current Consensus Guidelines for the Resection of Cystic Pancreatic Tumors Leticia P. Luz, Chang Min Cho, John M. DeWitt, C. Max Schmidt, Stuart Sherman, Julia K. LeBlanc, Mohammad A. Al-Haddad Background: Cystic pancreatic tumors (CPTs) are increasingly diagnosed but their manage- ment remains challenging. International Consensus Guidelines published in 2006 defined several criteria for surgical resection, but did not include cytopathologic findings, biochemical or molecular tumor markers obtained during endoscopic ultrasound-fine needle aspiration (EUS-FNA). Objectives: To describe the impact of preoperative cyst tumor markers on surgical decision making as reflected by compliance with current consensus guidelines for resection of CPTs. Methods: Retrospective review of prospectively collected EUS data in a single academic institution between 7/2008 and 6/2011. Our cohort includes patients with CPTs where cyst fluid tumor markers were available prior to surgical resection. Results: 45 consecutive patients (31 [69%] women) who underwent pancreatic resection of CPTs were included. Cyst fluid DNA analysis was available in all patients and showed K-ras mutations in 16 (36%), 2 or more allelic losses in 5 (11%) and elevated DNA concentration (> 40 ng/ mL) in 2 (4%). Cyst fluid CEA was available in 36 (80%); mean CEA level was 8588.8 ng/ ml(SD 5026.2). The rationale for resection of CPT was: main duct involvement in 14, lesion size >3cm in 11, presence of solid component/mural nodule in 8, abnormal cytology in 5 (cytology positive for neuroendocrine tumors [NETs] in 3, suspicious for carcinoma in 2), pancreatic symptoms in 4, lesion growth by at least 50% upon follow-up (including one patient with strong family history of pancreatic cancer) in 2, and high risk DNA analysis in 1. The rationale for resection of CPT was in compliance with current guidelines, in addition to cases with diagnostic or suspicious cytology, in 42/45 CPTs (93%). Final pathology of all resected tumors demonstrated: intraductal papillary mucinous neoplasm (IPMN) with low grade dysplasia (LGD) in 16/45 (36%), IPMN with moderate dysplasia in 9/45 (20%), mucinous cystic neoplasm in 5/45 (11%), NET in 4/45 (9%), IPMN with high grade dysplasia (HGD) in 3/45 (7%), invasive adenocarcinoma in 3/45 (7%), serous cystadenoma (SCA) in 2/45 (4%), pseudocysts in 2/45 (4%) and retention cyst in 1/45 patient (2%). In the 3 non- compliant cases, rationale for resection was: CPT growth in 2 patients and DNA analysis (low clonality k-ras mutation with 4 allelic losses) in 1 patient. Of those, pathology showed 1 SCA and 2 IPMNs with LGD. Conclusions: Despite the wide-spread availability of tumor markers preoperatively, resection of CPTs remains in compliance with current consensus guidelines in the vast majority of cases (93%) in this cohort. This observation could be explained by the limited ability of such markers to identify lesions with HGD or malignancy, and their role appears to be limited to the early classification CPTs as mucinous or non-muc- inous.