AGA Abstracts 403 Branch Duct Type-Intraductal Papillary Mucinous Neoplasm (BDT-IPMN) of the Pancreas: Does Immunosuppression (IS) Change the Natural History? Kanwar R. Gill, Mario Pelaez-Luna, Andrew Keaveny, Timothy A. Woodward, Michael B. Wallace, Suresh T. Chari, Thomas C. Smyrk, Naoki Takahashi, Jonathan E. Clain, Michael J. Levy, Randall K. Pearson, Bret T. Petersen, Mark D. Topazian, Santhi Swaroop Vege, Michael L. Kendrick, Michael B. Farnell, Massimo Raimondo Background: De novo cancer development is a significant risk of long-term IS after solid organ transplantation (SOT) including liver (LT), kidney (KT) and heart transplantation (HT). Patients undergoing SOT may be found to have incidental IPMNs. The natural history of IPMNs in the setting of IS after SOT in unknown. Aim: To determine if IPMNs have an increased potential for malignant transformation post-SOT in the setting of IS. Methods: Between 1996 and 2007 all patients diagnosed with BDT-IPMNs based on CT, MRI, and/ or EUS were retrospectively reviewed from our database. Patients who received SOT (LT, KT, HT) and maintained on IS were identified (T-BDT-IPMN group). Patients with follow- up imaging (CT, MRI and/or EUS) for BDT-IPMNs were included. Demographics, clinical, imaging and pathological data, including predictors of BDT-IPMN malignancy (PoM) (cyst related symptoms, cyst size >3 cm, mural nodules, positive cytology) were collected. The pre- and post-SOT courses were noted, including type and duration of IS therapy as well as type and timing of imaging follow-up studies. Similarly, a control group of BDT-IPMNs without IS (NT-BDT-IPMN) was identified for comparison. Univariate analysis was performed using student t test for continuous variables, X2 test for categorical variables. Results: A total of 26 IPMN patients with IS after SOT (22 LT, 3 KT, 1 HT) and 58 control patients were identified. Demographics, follow up time, cyst size, presence of PoM were similar in both T-BDT-IPMN and NT-BDT-IPMN groups (Table 1). All IPMNs were BDT with no coexistent main duct type. Nineteen patients had multifocal IPMNs. None of the IPMNs had mural nodules before SOT; only one IPMN was diagnosed with mural nodule during follow-up. No significant differences were found regarding PoMs between the two groups. Conclusions: Short term follow-up after SOT was not associated with any significant change in T-BDT-IPMN compared to NT-BDT-IPMN based on follow-up by EUS and MRI. Our results suggest that incidental BDT-IPMN <3 cm, even in the setting of IS after SOT can be followed conservatively as proposed by the published guidelines. Table 1. Demographic and cysts' characteristics 404 Acute Pancreatitis (AP) in Patients with Intraductal Papillary Mucinous Neoplasms of the Pancreas (IPMN): Frequency, Severity and Correlation with Tumors Characteristics Anne-Laure Pelletier, Pascal Hammel, Vinciane Rebours, Marie-Pierre Vullierme, Anne Couvelard, Alain Aubert, Frédérique Maire, Olivia Hentic, Dermot O'Toole, Alain Sauvanet, Philippe Lévy, Philippe B. Ruszniewski AP is frequent in patients (pts) with IPMN. Prevalence, severity and correlation of AP with IPMN histology have not been determined. Aims : 1 - To describe the characteristics of IPMN associated with AP. 2 - To compare clinical and histological features of IPMN in patients with (group 1) or without (group 2) AP. All pts were operated on and had pathological confirmation of IPMN. Methods : all consecutive pts operated on for an IPMN between 1995 and 2006 were included. Age, gender, symptoms (including AP, diarrhea, steatorrhea, weight loss, jaundice), IPMN morphology on CT scan, endoscopic ultrasound (EUS) and MRI after 2000, type of surgical resection and pathological data were recorded. Severity of AP was defined by the occurrence of organ failure and Balthazar score on CT scan. Patients with alcohol intake > 30 g/d or with another cause of AP were excluded. Results : among 185 pts included, 64 (34.6 %) had at least one AP (median and range : 2.1-10). AP revealed IPMN in 61/64 (95.3 %). Their characteristics are summarized in table 1. Median Balthazar score was 1 (0-6). Two pts had severe AP requiring intensive care, with favourable outcome. No recurrence of AP occurred after surgery (follow-up: 24 months, range 1-168). On imaging procedures, frequency of branch-duct type IPMN did not differ between groups 1 and 2 (25% and 23 %, respectively). A mass was more often found in group 2 (24% vs 12 %, p =0.02). Prevalence of calcifications in IPMN (16% and 8%), stenosis of the main pancreatic duct (16.5% and 18.8 %) did not differ between the 2 groups. Pathological data are shown in table 2. The degree of dysplasia was lower in group 1. Conclusion : One-third of IPMN pts had AP, which often revealed the disease. AP were more frequent in pts with branch- duct type IPMN and were associated with a lower grade of dysplasia. AP did not recur after IPMN resection. Table 1 - Clinical characteristics according to the occurrence of AP A-56 AGA Abstracts Table 2 - Correlation between AP and pathological analysis of IPMN 405 Characteristics of Intraductal Papillary Mucinous Neoplasm and Cyst of the Pancreas That Develop An Infiltrating Ductal Adenocarcinoma During Follow- Up Minoru Tada, Takao Kawabe, Saburo Matsubara, Yoko Yashima, Toshihiko Arizumi, Osamu Togawa, Yousuke Nakai, Naoki Sasahira, Kenji Hirano, Takeshi Tsujino, Hiroyuki Isayama, Masao Omata Background: We previously reported that patients with intraductal papillary mucinous neoplasm of the pancreas (IPMN) including small pancreatic cystic lesions were at an increased risk of pancreatic cancer (Clin Gastroenterol Hepatol 2006). In this study, we examined the baseline clinical features of patients who subsequently developed pancreatic cancer, with a focus on the infiltrating pancreatic cancer. Methods: 502 consecutive patients (average 67, range 25-93 years old, male: female=1:1), who were diagnosed as asymptomatic branch duct IPMN or pancreatic cyst that did not meet criteria for resection by the interna- tional guideline except size of the cystic lesion (average 17, range 2-80mm), were prospect- ively followed-up as outpatients up to 13 (average 4) years. Clinical examinations, blood tests and imaging diagnosis (US, EUS, CT or MRCP) were performed twice a year. Factors contributing to the development of pancreatic cancer were analyzed by Cox proportional hazard model. Results: During follow-up, pancreatic cancer was found in 13 patients after 4 years in average with an annual incidence rate of 0.66%. There were 2 cases consistent with intraductal papillary mucinous carcinoma, whose carcinoma appeared from the pre- existing cystic lesion. The remaining 11 were supposed to ordinary infiltrating cancer; characteristics of those were as follows; 72 (range 63-82) years old, 10 males, size of cystic lesions was 13 (range 8-20) mm, diameter of the main pancreatic duct was 2 (range 1-4) mm, and the number of cysts was multiple in 7. Older age, male, multiple cysts, and smaller diameter of the cystic lesion were selected as predictors for infiltrating pancreatic cancer by univariate analysis. Older age and male remained significant as the predictors after multivari- ate analysis (p<0.05). Conclusion: In male patients with multiple cystic lesions, the whole pancreas should be carefully examined by regular check-up especially, because of higher risk of development of ordinary pancreatic cancer. Infiltrating cancer was more frequently detected in patients with smaller diameter of cystic lesions. It is tempting to speculate that these cystic lesions might represent large PanIN rather than small IPMN which precedes infiltrating pancreatic cancer. 406 DNA Mutational Analysis Versus Cytology with and Without Fluid CEA Level in the Diagnosis of Mucinous Cystic Lesions of the Pancreas: A Multicenter Study. Eugene Zolotarevsky, Mayar Al Mohajer, Richard S. Kwon, Andres Gelrud, Nonthalee Pausawasdi, Nathan Schmulewitz, Grace H. Elta, Michelle A. Anderson Differentiating benign and mucinous pancreatic cystic lesions (PCL) remains difficult. DNA mutational analysis by PathFinderTG® (PFTG) of pancreatic cyst fluid has recently been used to help differentiate benign and mucinous PCL. The incremental benefit of the PFTG over cytology with and without CEA level is unknown. AIM: To investigate if PFTG is superior to cytology with and without cyst fluid CEA level in detecting mucinous PCL. METHODS: Patients undergoing EUS-FNA of PCL at 2 tertiary care centers had cyst fluid cytology, CEA, and PFTG analysis prospectively recorded and compared to known final diagnosis. PFTG results were reported as benign, or mucinous (indolent mucinous [IML], mucinous with potential for progression [MLPFP], and high grade mucinous [HGML]). Benign and IML were grouped as “indolent“ and MLPFP and HGML as “advanced”. RESULTS: Among 86 patients undergoing EUS-FNA, 22 reached final histologic (20) and cytologic (2) diagnosis. Final diagnosis included adenocarcinoma (4), mucinous lesion (12, IPMN [7] and MCN [5]) and benign (serous, pseudocyst, lymphoepithelial cyst [6]). PFTG was available for all patients, cytology for 21/22 patients, and CEA for 19/22 patients. In differentiating any type mucinous neoplasm and adenocarcinoma from benign cysts, PFTG was more sensitive (93.6%) than CEA > 192ng/mL (64.3%) or cytology positive for mucin (60.0%) (p<0.01), but provided only marginal benefit over cytology plus CEA level (p=NS)(Table 1). PFTG sensitivity for advanced mucinous lesions was superior to CEA or cytology although this did not reach statistical significance. Conversely, CEA level > 192ng/mL predicted mucinous lesion with the highest specificity (80%). When PFTG was combined with cytology it predicted a mucinous or more advanced lesion with 100% sensitivity. CONCLUSION: PFTG is more sensitive than CEA or cytology in predicting mucinous status of cystic pancreatic neoplasms and when combined with cytology, PFTG detected all mucinous lesions