201 Cytoplasmic Vacuoles: An Under Recognized Morphologic Feature of PEN and SPN Vidhu Kaushik, MD 1 , Darshana Jhala, MD 1 , Michael Kochman, MD 2 , Nirag Jhala, MD 1 . 1 Department of Pathology and Laboratory Medicine, Hospital of University of Pennsylvania, Philadelphia, Pennsylvania; 2 Department of Medicine, Hospital of University of Pennsylvania, Philadelphia, Pennsylvania Introduction: Frequency of pancreatic neuroendocrine neoplasm (PEN) on fine needle aspirates (FNA) of the pancreas is uncommon and is noted in approximately 4% of the aspirated pancreatic tumors. These tumors pose diagnostic challenges on FNA for unsuspecting pathologists. One of their major morphologic mimics is solid pseuodpapillary neoplasm of the pancreas (SPN). Earlier report suggests that large cytoplasmic vacuoles are noted with SPN. Literature is sparse about noting cytoplasmic vacuoles in PEN. This study was undertaken with an objective to determine frequency with which cytoplasmic vacuoles are noted in PEN. Furthermore this study also was undertaken to determine if large cytoplasmic vacuoles can serve as a feature that can distinguish SPN from PEN. Materials and Methods: A computer generated search identified 30 consecutive cases (16 males and 14 females) of PEN and 5 cases of SPN. For each case either CT scan or EUS results were also reviewed to determine the size, location within the pancreas and spread of tumor. All slides were reviewed by two independent pathologists for the architectural and cellular morphology as has been reported earlier, with special emphasis on determining cytoplasmic vacuolization. Cytoplasmic vacuoles were categorized as either small or large and then correlated with final reported diagnosis. Cytoplasmic vacuoles were considered small when they represented maximum diameter of less than half the size of the tumor nuclei, whereas, they were considered large when they were either equal to or greater than the size of the tumor nuclei. All cases had confirmatory stains and/ or follow up. The results were compared and Fisher’s exact 2 tailed tests was utilized to determine statistical significance. Results: Patients with PEN ranged from 30 to 84 years (mean age: 60 years) and the tumor size ranged from 0.9 to 6.3 cm (mean: 3.1 cm). They were distributed in the head (nZ23), body/ tail (nZ6). In 3 cases they presented as cystic neoplasm on imaging studies. In comparison, patients with SPN ranged from 21 to 51 years (mean age: 34 years) and all five were located in the body/ tail. Cytoplasmic vacuoles were noted in Diff Quick stained smears but not in Papanicolaou stained smears. In 6 cases, Diff Quick stained slides were not performed. Small cytoplasmic vacuoles were noted in 18/24 (75%) cases and were frequently focally noted. Small vacuoles when present were often multiple in a single cell. Large cytoplasmic vacuoles were infrequently noted (8%). In comparison large cytoplasmic vacuoles were noted in 4/5 SPN cases. This difference was statistically significant (pZ0.003). Conclusions: Small cytoplasmic vacuoles can be observed with increased frequency in PEN especially on Diff Quick stained smears. This study lends support to prior study that large cytoplasmic vacuoles can distinguish between PEN and SPN. 202 Secondary Tumors of the Pancreas Diagnosed by Endoscopic Ultrasound-Guided Fine Needle Aspiration: A Ten-Year Experience Lindsay Waters, MD 1 , Marilyn Dawlett, CT(ASCP) 2 , Debora Smith, CT(ASCP) 1 , Gregory Staerkel, MD 2 , Nour Sneige, MD 1 . 1 Pathology, The Methodist Hospital, Houston, Texas; 2 Pathology, MD Anderson Cancer Center, Houston, Texas Introduction: Most pancreatic masses are primary neoplasms, specifically pancreatic adenocarcinomas, although metastases from other primary sites make up 4.5-11% of pancreatic masses. Determining whether a mass is a primary or secondary pancreatic neoplasm is necessary for appropriate patient management. We reviewed our experience using endoscopic ultrasound-guided fine needle aspiration (FNA) for diagnosis of the pancreatic tumors over 10 years to identify distinguishing cytologic and immunohistochemical characteristics of metastatic disease. Materials and Methods: Cytopathologic records from January 1, 2002, to April 1, 2012, at The University of Texas MD Anderson Cancer Center and The Methodist Hospital in Houston, Texas, were reviewed to identify all FNA specimens of tumors metastatic to the pancreas. Clinical history, radiologic findings, primary tumor type, FNA diagnosis, treatment, and clinical follow-up information were collected for each patient. Results: A total of 66 patients with disease metastatic to the pancreas were identified, 38 (58%) of whom were men and 28 (42%) of whom were women, with an average age of 63 years (range 40-89 years). Locations of tumors in the pancreas included 31 (47%) in the ampulla/head, 16 (24%) in the body, 16 (24%) in the tail, 2 (3%) in an unspecified site, and 1 (2%) with multifocal locations. Most metastases (98%) were single lesions. Sites of origin for these metastases were 27 (41%) renal, 9 (14%) lung, 7 (11%) skin and soft tissue, 6 (9%) breast, 5 (7%) colon, and 12 (18%) various other sites. No relationship was found between the primary tumor type and the location of the metastasis within the pancreas. Immunohistochemical staining of cell block material was performed to confirm the tumor type in 32 cases. CD10, Pax8, RCC, vimentin, and pancytokeratin immunohisto- chemical stains were most commonly used to determine whether the tumor was metastatic renal cell carcinoma, clear cell type or primary pancreatic adenocarcinoma. Follow-up information was available for 64 of 66 (97%) patients, and length of follow-up ranged from <1 year to 10 years (mean 2.3 years). Thirty-four patients (53%) were alive at the time of most recent follow-up. Of the 25 patients with metastatic renal cell carcinoma, clear cell type whose long-term clinical follow-up data were available (range <1-7 years, mean 2.4 years), 19 (76%) were alive at the time of most recent follow-up. Conclusions: Most secondary tumors of the pancreas were solitary (98%), with renal cell carcinoma, clear cell type being the dominant diagnosis (38%). More than half of the patients, including those with renal cell carcinoma, had a long survival potential suggesting that these patients may benefit from resection of the pancreatic metastases. Because metastases may mimic primary pancreatic adenocarcinoma both clinically and cytologi- cally, adjunct immunohistochemical staining of cell block material and review of clinical characteristics is essential to determine the correct diagnosis. 203 Clinical, Radiology and Cytomorphology Correlation of Endoscopic Ultrasound Guided Fine Needle Aspiration of Pancreatic Lesions - A Single Institutional Experience Ana Yuil-Valdes, MD, Arvind Rishi, MBBS, MD, Chiara Sugrue, MBA, MS, Nora Morgenstern, MD. Pathology and Laboratory Medicine, Hofstra- NSLIJ School of Medicine, New Hyde Park, New York Introduction: Endoscopic ultrasound guided fine needle aspiration cytology (EUS-FNAC) is an important diagnostic modality to assess preoperative extent and diagnosis of pancreatic neoplasms. Early diagnosis and knowledge of the extent of pancreatic malignancy plays a significant role in management. Advantages of EUS includes ability to target smaller lesions, additional sampling of lymph nodes and liver, palliative neurolysis, low complications, very rare needle track tumor seeding and cost effectiveness. This study was conducted with an objective to correlate radiological parameters with diagnostic categories on EUS-FNAC and clinical and surgical follow-up in atypical, indeterminate or suspicious category (AIS group). Materials and Methods: Clinical, radiological and cytomorphological features of all EUS-FNAC accessioned in cytology department at North Shore and Long Island Jewish Hospital, New York in year 2010 were analyzed. Radiological and cytomorphological correlation was done for the diagnostic categories and location of lesion within pancreas, size, and nature of lesion (cystic or solid). Results: Total of 197 cases of EUS-FNAC of pancreas were accessioned in year 2010. 54% (nZ106) lesions were cystic, 38% (nZ75) were solid and S110 Abstracts