Mo1329 DIAGNOSTIC YIELD AND PREDICTORS OF ADEQUATE CYTOLOGY SAMPLING FOR KI67 ON ENDOSCOPIC ULTRASOUND GUIDED FINE NEEDLE ASPIRATION FOR GRADING OF PANCREATIC NEUROENDOCRINE TUMORS

Pancreatic neuroendocrine tumors (pNETs) account for 1-2% pancreatic tumors and require differentiation from carcinomas since their prognosis and management differ significantly. Endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) is routinely employed in grading pNETs through estimation of Ki67 index on cytology samples. Therefore, obtaining sufficient cytology sample for accurate tumor grading is essential for prognostication and guiding management. We aimed to report our diagnostic yield for obtaining adequate sample on EUS-FNA for diagnosis and tumor grading based on Ki67 index for pNETs. We also aimed to assess the procedural predictors of adequate sampling. A single center retrospective study including patients with suspected pNETs based on imaging findings (computed tomography and magnetic resonance imaging) were enrolled between January 2008 and October 2017.Collected data included age, sex, adequacy of sample for Ki67 index estimation, FNA needle gauge, approach to FNA, number of passes, location and size of tumor were recorded. The primary outcome was the ability to diagnose and grade pNET using cytological samples obtained through EUS-FNA. Cytological diagnosis of pNET was confirmed if the EUS-FNA sample tested positive for neuroendocrine markers such as chromogranin, synaptophysin and CD56 on immunocytochemical studies. Categorical variables were compared using the chi-squared test or Fisher exact test. A two-tailed p-value <0.05 was used to denote statistical significance. Of the 57 patients included in the study, all were diagnosed with pNET based on cytology samples obtained by EUS-FNA. Furthermore, 77% patients (n=44) had sufficient cytology sample to allow tumor grading through estimation of Ki67 index. The mean age was 58 ± 14 years and 63% were men. 56% (n=32) patients had grade I tumors (n=32), 11% (n=2) had grade 2 tumors, 11% (n=2) had grade 3 tumors, while tumor grading could not be determined due to sub-optimal cytology sample in 23% of patients (n= 13). There was no difference in sample adequacy for Ki67 index between the trans-gastric (76%) and trans-duodenal (66%) approach (p=0.63), tumor size < 2 vs. ≥ 2 cm (p=0.74), tumor location (pancreatic head, body or tail) (p=0.54), needle size (22 vs. 25 gauge) (p=0.69), presence of liver metastasis (p=0.150) and number of passes < 3 vs. ≥ 3 passes (p= 0.31). Patients with and without pancreatic ductal dilation had comparable cytology samples to obtain a Ki67 index (75% Vs. 78%, p=1.00). Our study shows that EUS-FNA is an excellent diagnostic modality to obtain adequate samples for performing cytological diagnosis of pNETs with a yield of 77% to be able to obtain sufficient cytology samples for estimation of Ki67 index. The results suggest that EUS-FNA can be a reliable method for diagnosing and grading of pNETs.Tabled 1Baseline Characteristics of Study Populationn (%)Age (mean ± SD)58 ± 14 yearsMales32 (63%)BMI (mean ± SD)29 ± 5.3Tumor size ≥ 2 cm39 (69%)CT scan18 (32%)MRI5 (9%)Octreotide Scan34 (60%)PET scan10 (18%)EUS-FNA56 (98%)EUS-FNB1 (2%)Pancreatic head tumors11(19%)Pancreatic body tumors32 (56%)Pancreatic tail tumors14 (25%)22 gauge EUS-FNA needle38 (79%)25 gauge EUS-FNA needle10 (21%)Trans-gastric approach38 (86%)Trans-duodenal approach6 (14%)≥ 3 passes36 (64%)Grade 1 tumors on EUS-FNA73 (56%)Grade 2 tumors on EUS-FNA6 (11%)Grade 3 tumors on EUS-FNA6 (11%)Liver metastasis15 (26%)Lymph node involvement16 (28%)Main pancreatic duct dilation8 (14%)Pancreatic atrophy6 (11%)Pancreatic calcifications3 (5.3%)*Abbreviations: BMI- body mass index, EUS-FNA-endoscopic ultrasound guided fine needle aspiration EUS-FNB- endoscopic ultrasound guided fine needle biopsy CT scan- computed tomography scan, MRI scan- magnetic resonance imaging scan PET scan- positron emission tomography scan Open table in a new tab