Endoscopic Ultrasound-Guided Radiofrequency Ablation for Pancreatic Cystic Neoplasms and Pancreatic Neuroendocrine Tumors

Overview

The primary aim of this study is to evaluate the technical feasibility and effectiveness of endoscopic ultrasound-guided radiofrequency ablation (EUS-RFA) in the treatment of pancreatic cystic neoplasms (PCNs) and pancreatic neuroendocrine tumors (pNETs) in patients who are not suitable candidates for surgical treatment or who refuse surgery. EUS-RFA, as a minimally invasive therapeutic option, may potentially reduce the need for intensive imaging surveillance, prevent further progression of these lesions, and serve as a bridging therapy in selected patients for whom surgical intervention may be considered in the future. At the same time, EUS-RFA could contribute to reducing overtreatment, thereby lowering the risk of complications associated with surgical procedures and leading to an overall reduction in healthcare costs. The secondary objectives of the study are to evaluate the safety profile of EUS-RFA, including the incidence and severity of adverse events both in the early postoperative period and during the first year of follow-up, as well as to assess the long-term efficacy of EUS-RFA, including disease progression.

Pancreatic lesions, including pancreatic cystic neoplasms (PCNs) and pancreatic neuroendocrine tumors (pNETs), represent a heterogeneous group of diseases with variable biological behavior, ranging from benign and indolent entities to lesions with significant malignant potential. Their detection has markedly increased over the past two decades, primarily due to the widespread use of high-resolution imaging modalities such as computed tomography (CT) and magnetic resonance imaging (MRI). Incidental pancreatic cystic lesions are currently identified in up to 20% of abdominal imaging studies performed for unrelated indications, with prevalence rising with age. Consequently, the clinical burden associated with these lesions is expected to increase substantially in aging populations. The management of pancreatic lesions remains challenging and requires a careful balance between the risk of malignant transformation and the morbidity associated with intervention. PCNs encompass both non-neoplastic lesions (e.g., pseudocysts) and neoplastic entities, including intraductal papillary mucinous neoplasms (IPMNs), mucinous cystic neoplasms (MCNs), serous cystadenomas (SCNs), and solid pseudopapillary neoplasms (SPNs). Among these, IPMNs and MCNs are associated with a well-documented risk of malignant progression. IPMNs, in particular, are classified into main duct, branch duct, and mixed types, with varying degrees of risk depending on morphological features and histopathological grade. However, accurate prediction of malignant potential in individual lesions remains difficult in routine clinical practice, as histological confirmation is often not feasible without surgical resection. Similarly, pNETs constitute a rare but increasingly recognized group of pancreatic tumors, accounting for approximately 3% of all pancreatic malignancies. These tumors exhibit diverse biological behavior and may be functional (hormone-secreting) or non-functional. Functional tumors, such as insulinomas or gastrinomas, often present with clinical syndromes related to hormone excess, while non-functional tumors are frequently detected incidentally. Tumor grading, based on proliferative indices such as Ki-67, and tumor size are key determinants of prognosis and management. Current guidelines generally recommend surgical resection for functional tumors and for non-functional tumors ≥2 cm in size. However, the optimal management of small (≤2 cm), asymptomatic non-functional pNETs remains controversial, as many of these lesions demonstrate indolent behavior and low risk of progression. Surgical resection, including procedures such as pancreaticoduodenectomy or distal pancreatectomy, remains the standard of care for many pancreatic lesions with suspected or confirmed malignancy. However, these procedures are technically demanding and associated with considerable morbidity (20-30%) and mortality (1-3%) even in specialized centers. Postoperative complications may include pancreatic fistula, delayed gastric emptying, hemorrhage, and long-term endocrine and exocrine insufficiency. These risks are particularly relevant in elderly patients and those with significant comorbidities. Furthermore, accumulating evidence suggests that a substantial proportion of patients undergoing surgery for PCNs ultimately have low-grade or benign disease, raising concerns about overtreatment. On the other hand, conservative management strategies based on long-term imaging surveillance are associated with their own limitations, including cumulative healthcare costs, radiation exposure (in the case of CT), and significant psychological burden for patients living with an untreated pancreatic lesion of uncertain malignant potential. Therefore, there is a clear unmet clinical need for minimally invasive treatment options that can effectively manage selected pancreatic lesions while minimizing procedural risk. Endoscopic ultrasonography (EUS) has become a cornerstone in the evaluation of pancreatic diseases, providing high-resolution imaging and enabling tissue acquisition via fine-needle aspiration or biopsy. More recently, EUS has evolved into a therapeutic platform, allowing for the delivery of targeted interventions directly into pancreatic lesions under real-time imaging guidance. Among these, EUS-guided radiofrequency ablation (EUS-RFA) has emerged as a promising minimally invasive technique. Radiofrequency ablation (RFA) induces localized tissue destruction through the application of high-frequency alternating current (typically 400-500 kHz), resulting in ionic agitation, frictional heating, and subsequent coagulative necrosis. Temperatures exceeding 60°C lead to protein denaturation, cellular dehydration, and irreversible cell death. When applied under EUS guidance, RFA allows precise targeting of pancreatic lesions while minimizing damage to surrounding critical structures, such as the pancreatic duct, bile duct, and major blood vessels. Preliminary clinical evidence supporting EUS-RFA has been reported in several pilot and multicenter studies. In an early feasibility study by Pai et al., EUS-guided RFA was performed in patients with pancreatic lesions, including cystic neoplasms and pNETs, demonstrating technical feasibility and absence of major periprocedural complications. In a multicenter prospective study conducted by Barthet et al., which included patients with pancreatic cystic lesions (predominantly IPMNs), a significant response rate of approximately 71% was observed at 12 months, including complete disappearance of cysts and resolution of mural nodules. Importantly, the authors reported a reduction in complication rates after protocol optimization. A meta-analysis evaluating EUS-RFA across multiple centers demonstrated high technical success rates approaching 100%, with clinical success (defined as reduction in lesion size) exceeding 90% and an overall adverse event rate of approximately 14.7%. These findings further support the feasibility and potential efficacy of this technique. In the context of pancreatic neuroendocrine tumors, Armellini et al. first described successful EUS-guided RFA of a small non-functional pNET in a patient unfit for surgery, with complete radiological response at short-term follow-up. Rossi et al. subsequently reported outcomes in a cohort of patients with pNETs treated using various RFA approaches, including EUS guidance, demonstrating effective tumor control and symptom resolution in functional tumors, albeit with some procedure-related complications such as pancreatitis. Further evidence was provided by Lakhtakia et al., who described rapid and sustained clinical improvement in patients with insulinomas treated with EUS-RFA, with normalization of glycemic parameters observed within days of the procedure. Similarly, Pai et al. reported reduction in vascularity and induction of necrosis in treated pNETs without significant complications. A large retrospective multicenter study from France, led by Barthet et al., evaluated over 100 patients undergoing EUS-RFA for various pancreatic lesions, including pNETs and cystic neoplasms. The study reported a complete response rate of approximately 60%, with higher success observed in pNETs smaller than 2 cm. Tumor size and indication (pNET vs. other lesions) were identified as significant predictors of treatment success, while proximity to the main pancreatic duct was associated with increased risk of complications. In a prospective multicenter study by Okasha et al., EUS-RFA was compared with EUS-guided ethanol ablation in patients with pNETs. The study demonstrated superior clinical remission rates in the RFA group (91%) compared to ethanol ablation (57%), suggesting a potential advantage of RFA as a local ablative therapy. Despite these encouraging results, the current body of evidence is limited by heterogeneity in study design, small patient cohorts, variability in RFA devices and settings, and relatively short follow-up periods. Therefore, prospective studies with standardized protocols are needed to validate these findings and better define the role of EUS-RFA in clinical practice. The present study aims to evaluate the feasibility, safety, and efficacy of EUS-guided radiofrequency ablation in patients with pancreatic cystic neoplasms and pancreatic neuroendocrine tumors. The study will focus on patients who are not optimal candidates for surgical resection due to comorbidities, advanced age, or patient preference for a minimally invasive approach. All procedures will be performed using a standardized EUS-RFA protocol to ensure consistency in technique and energy delivery. Pre-procedural evaluation will include detailed imaging and, where appropriate, histological confirmation using EUS-guided tissue acquisition. Patients will be followed for a minimum of 12 months after the procedure, with scheduled clinical assessments and imaging studies (CT, MRI, and/or EUS) to evaluate treatment response and detect potential complications or recurrence. Longer-term follow-up may be conducted where feasible to assess durability of response. The primary endpoints of the study will include: * Technical success, defined as successful delivery of RFA to the target lesion, * Clinical efficacy, assessed by imaging-based response. Treatment response will be categorized as: * Complete resolution: complete disappearance of the lesion or radiological evidence of complete necrosis, * Significant response: reduction in lesion size greater than 80% or complete resolution, * Treatment failure: reduction in lesion size less than 50% or no observable change. Secondary endpoints will include: * Safety profile, including the incidence and severity of procedure-related adverse events (e.g., pancreatitis, bleeding, infection, perforation, ductal injury), * Symptom resolution in patients with functional pNETs, * Need for repeat ablation sessions or additional interventions, * Progression-free survival and overall disease control, * Impact on patient quality of life. The study is not expected to demonstrate differences based on sex or gender; however, data will be analyzed accordingly to ensure comprehensive evaluation. By generating high-quality prospective data, this study aims to clarify the clinical utility of EUS-RFA as a minimally invasive treatment option for pancreatic lesions. The findings may contribute to refining patient selection criteria, optimizing procedural techniques, and establishing standardized treatment protocols. Ultimately, this approach has the potential to reduce the need for high-risk surgical procedures, decrease healthcare costs associated with prolonged surveillance and hospitalization, and improve patient outcomes and quality of life. Furthermore, the results of this study may inform future randomized trials comparing EUS-RFA with conventional surgical and conservative management strategies, thereby supporting evidence-based, personalized treatment of patients with pancreatic lesions.