e16261 Background: Endoscopic ultrasound-guided ablation therapy (EUS-A) is a minimally invasive technique increasingly used for the management of surgically unresectable pancreatic cancer. The data on the efficacy and clinical outcome of EUS-A for unresectable pancreatic cancer is lacking. We conducted this systematic review and meta-analysis to evaluate the safety, technical and clinical success of EUS-A therapies specifically for unresectable pancreatic cancer. Methods: Studies were identified with a comprehensive search for EUS-A and pancreatic cancer on PubMed, Google Scholar, Web of Science and Embase data search as of October 2020. The technical and clinical success rates of EUS-A were primary outcomes while adverse events (AEs) rate was secondary outcome. We defined technical success as successful placement of probe within tumor and able to perform ablation regardless of tumor outcome. Clinical success was defined as symptomatic improvement and/or any reduction in tumor size or evidence of necrosis on radiological imaging after EUS-A. A compute pooled analysis was performed using the ‘meta’ package by Schwarzer et al. in the R programming language (version 4.0.2) using random effect model. Results: Nineteen studies including 192 patients (54% females) were included. Common pancreatic tumors were nonfunctional neuroendocrine tumor (NNET) 43.5% (97), followed by locally advanced pancreatic ductal adenocarcinoma (LAPDAC) 27.3% (61), insulinoma 17.9% (40), cystic neoplastic lesions 8.5% (19). The pancreatic head was the most common site of tumor 49.8% (111), followed by body, neck and tail 44.8% (100). EUS-RFA was the most commonly used ablative therapy 63% (12/19 studies), followed by EUS-EA (ethanol ablation) 26% (5/19 studies) while EUS-A using laser and lauromacrogol injection was used in one study each. The mean number of ablation sessions per patient was 1.4 based on the total 243 sessions in 176 patients. The pooled technical success rate of EUS-A was 99.5% [95% CI = 0.90-0.97, I2 = 0%]. The pooled clinical success rate calculated out of total number of pancreatic lesions was 91.1% [95% CI = 0.79-0.92, I2 = 16%]. Clinical improvement in symptoms was reported in 11 studies and a complete resolution or decrease in tumor size was reported in 16 studies. The pooled AEs rate was 22.9% [95% CI = 0.17-0.37, I2 = 43%]. Common AEs were abdominal pain 7.8% (15), pancreatitis 5.2% (10) and pancreatic pseudocyst 2.1% (4). The median follow-up was 9.5 months. Conclusions: EUS-A is a safe and promising modality in the management of unresectable pancreatic cancer with a high technical and clinical success rate. Large prospective studies and clinical trials are required for comparison of clinical outcome of different EUS-A therapies and its widespread application for unresectable pancreatic cancer.
Read full abstract