Endoscopic Ultrasonography for Vascular Invasion in Pancreatic Cancer

Abstract

The prognosis in patients with pancreatic cancer is poor with a five-year survival rate of 8%-9%. 1,2 Curative resection is indispensable procedure for achieving a higher survival rate. 1,2 Surgical resectability of pancreatic cancer can be determined by evaluating vascular invasion, especially in the major arteries (superior mesenteric, celiac, and common hepatic arteries) and the veins (superior mesenteric and portal veins). 3,4 Although contrast-enhanced computed tomography (CECT) is the standard method for the diagnosis of pancreatic cancer, there are reports that endoscopic ultrasonography (EUS) has some advantages for detecting vascular invasion. 5,6 In 2006, a systematic review comparing EUS and CECT for pancreatic cancer demonstrated that EUS is the superior technique for tumor detection, tumor staging, and nodal staging, but it has not been concluded which of these is superior in evaluating resectability. 7 On the other hand, a meta-analysis showed that the sensitivity to diagnosing vascular invasion was higher in EUS (86%) than in CECT (58%), 5 which was confirmed by a systematic review (EUS 72% and CT 63%). 6 In a meta-analysis study, however, there was no significant difference between the sensitivity (CECT 71% vs. magnetic resonance imaging [MRI] 67%) and the specificity (CECT 92% vs. MRI 94%) for vascular invasion. 3 Although CECT has better inter-observer agreement, accuracy, and skill, it is now believed that EUS is superior to CECT when evaluating vascular invasion and resectability in pancreatic cancer.In the current issue, Fujii et al. studied the diagnostic ability of EUS for major vascular invasion in pancreatic cancer, and evaluated the relationship between EUS findings and pathological distance. 8 The sensitivity, specificity, and accuracy of EUS diagnosis for vascular invasion were 89%, 92%, and 91% in the veins and 83%, 94%, and 93% in the arteries, respectively. 8 The values showed narrow improvement from previous studies, despite the development of EUS. In accordance with the relationship between tumors and major vessels, Fujii et al. classified the EUS findings into four types: type 1, clear invasion, encasement of a vessel by a tumor; type 2, a tumor that makes contact with a vessel, with loss of the hyperechoic vessel layer; type 3, a tumor that makes contact with a vessel, without loss of the hyperechoic vessel layer; type 4, clear non-invasion, with space between a tumor and a vessel. 8 The vascular invasion defined by Fujii et al. is far simpler than in previous studies. 8 According to Buchs et al., the findings of vascular invasion in EUS are as follows: (1) loss of the hyperechoic vessel wall/tumor interface; (2) direct visualization of tumor within the vessel lumen; (3) vascular encasement or

Semantic Segmentation of Pancreatic Cancer in Endoscopic Ultrasound Images Using Deep Learning Approach

“…The accurate and precise segmentation of the cancer area is critical for the surgical therapy of pancreatic cancer. In addition, vascular invasion is an important factor to diagnose the stage of cancer and plan for the therapeutic strategy [ 33 , 34 ]. Thus, our future work will include an investigation of vascular invasion in pancreatic cancer using a deep learning approach.…”

The Role of Endoscopic Ultrasound and Ancillary Techniques in the Diagnosis of Autoimmune Pancreatitis: A Comprehensive Review

“…Stenosis of the splenic or superior mesenteric artery is also described in a few cases. EUS has a sensitivity and specificity of 85% and 91%, respectively, in diagnosing vascular invasion by PC [ 78 ], and peripancreatic major vein invasion is more accessible by EUS than by CT [ 79 ]. Of note, in the two largest retrospective series of patients with AIP, EUS detected peripancreatic vessel involvement in only 19.3% [ 48 ] and 7.4% [ 37 ] of cases, respectively.…”