Advances in endoscopic ultrasound for autoimmune pancreatitis: Tissue acquisition, imaging innovations, artificial intelligence, and multidisciplinary integrations

Abstract

Autoimmune pancreatitis (AIP) is an immune-mediated inflammatory condition of the pancreas that can be difficult to differentiate from pancreatic malignancy and therefore remains challenging to diagnose. Type 1 AIP is the pancreatic manifestation of immunoglobulin G subclass 4-related disease and can be associated with other systemic fibro-inflammatory conditions. Type 2 AIP is characterised histologically by granulocytic epithelial lesions and is frequently associated with inflammatory bowel disease. Over the previous years, endoscopic ultrasound (EUS) has become an important part of the diagnostic work-up of patients presenting with suspected AIP, particularly when imaging studies demonstrate focal pancreatic enlargement, or a mass-like lesion. Recent advances in EUS imaging, including contrast-enhanced harmonic EUS and EUS elastography, may improve characterisation of the pancreatic lesion and help distinguish an inflammatory pancreatic process from a pancreatic malignancy. Moreover, EUS-guided tissue acquisition plays an important role in this context by allowing histological assessment of the lesion and hence helping exclude a malignancy. EUS-guided fine-needle biopsy is currently the preferred tissue acquisition technique, as it allows the preservation of tissue architecture, which helps facilitate the diagnosis of AIP. A multidisciplinary approach involving a panel of specialists in hepatopancreatobiliary endoscopy, hepatology, rheumatology, radiology, cardiology, nephrology, and histopathology is valuable for integrating the different components of the diagnostic work-up and guiding subsequent management. Finally, emerging artificial intelligence algorithms for EUS may support the diagnostic process and facilitate recognition of AIP. This review discusses current and emerging EUS techniques, tissue acquisition options, safety considerations, multidisciplinary integration and the potential role of artificial intelligence in the diagnosis and management of AIP.

Keywords: Autoimmune pancreatitis type 1; Autoimmune pancreatitis type 2; Endoscopic ultrasound; Immunoglobulin G subclass 4-related disease; Artificial intelligence

Core Tip: Endoscopic ultrasound (EUS) is central to the diagnostic work-up of autoimmune pancreatitis, particularly in patients with focal pancreatic enlargement or mass-like lesions. Advances in EUS imaging and EUS-guided fine-needle biopsy using end-cutting needles may improve lesion characterisation and diagnostic accuracy by preserving tissue architecture. Artificial intelligence has the potential to further enhance EUS interpretation and reduce interobserver variability. Given the diagnostic complexity and need to exclude malignancy, multidisciplinary collaboration is valuable for integrating imaging, histology, serology and clinical findings and for guiding subsequent correct diagnosis and management.