S2289 S. KAREEM TRAP Score for EUS-FNB-Guided Malignant Tissue Output: Comparison of Pancreatic and Non-Pancreatic Solid Lesions

Introduction: The objective of this study was to compare a scoring system for predicting malignancy during EUS-FNB in pancreatic vs non-pancreatic biopsies. Methods: This retrospective single-study was conducted at Liaquat-National-Hospital in Karachi, Pakistan. Electronic Medical Records of consecutive pancreatic and non-pancreatic EUS-FNB were collected from January 2019 - July 2024, and excluded incomplete procedures. S.KAREEM-TRAP score was developed using 4 parameters: Tissue size (T: 1 if >4 mm, 0 if <4 mm), ROSE (R: 1 if positive, 0 if negative), Age (A: 1 if ≥60 years, 0 if <60 years), and Needle passes (P: 1 if >2 passes, 0 if ≤2 passes), with a scoring range of 0-4. SPSS version-25 was used for data analysis. Diagnostic metrics for pancreatic and non-pancreatic biopsies were calculated, including sensitivity, specificity, and predictive values with a threshold S. KAREEM TRAP score of ³3. ROC-curves and AUC-values to assess the S. KAREEM-TRAP score’s predictive performance for malignancy. Results: 154 EUS-FNB cases were analyzed (102/154, 66.2% malignant). Median age was 60 years (IQR: 48-67), man predominance (62.3%). Pancreatic biopsies constituted 58.4% of the cases. Overall, Malignancies were identified in 79.5% of patients aged ≥60 years, 77.8% of those with >2 needle passes, 88.8% of cases with a tissue core size >4 mm, and 94.1% of ROSE-positive cases. For pancreatic lesions, S.KAREEM-TRAP score of ≥3 yielded a sensitivity of 86.36% and specificity of 95.83%. For non-pancreatic lesions, the score demonstrated even sensitivity 94.44%, and specificity 100%. Conclusion: The S. KAREEM-TRAP score may be a valuable tool for predicting higher positive outcomes in EUS-FNB-TA for both pancreatic and non-pancreatic solid-lesions.