BackgroundSelf-expandable metal stents (SEMSs) are widely used for malignant biliary obstructions. Nitinol-covered SEMSs have been developed to improve stent patency. Currently, SEMSs may be uncovered, partially covered, or fully covered; however, there is no consensus on the best stent type for the management of malignant distal biliary obstruction (MDBO).MethodsPatients with unresectable MDBO receiving SEMS (Wallflex™) were retrospectively analyzed. Time to recurrent biliary obstruction (TRBO) and survival time were compared among the three types of SEMSs. Univariate and multivariate analyses were performed to identify risk factors for stent dysfunction.ResultsIn total, 101 patients received SEMSs for unresectable MDBO (44 uncovered, 28 partially covered, and 29 fully covered SEMSs). Median survival time was 200, 168, and 276 days in the uncovered, partially covered, and fully covered SEMSs groups, respectively. There were no differences in survival among the three groups. Median TRBO was 199, 444, and 194 days in the uncovered, partially covered, and fully covered SEMSs groups, respectively. Partially covered SEMSs had longer TRBO than uncovered (p = 0.013) and fully covered (p = 0.010) SEMSs. Tumor ingrowth occurred only with uncovered SEMSs and stent migration occurred only with fully covered SEMSs. Multivariate analyses confirmed that partially covered SEMSs have lower risk of dysfunction.ConclusionsPartially covered SEMSs with a proximal uncovered flared end have longer patency than uncovered and fully covered SEMSs by preventing tumor ingrowth and stent migration.

From 52 cancer-related gene analysis, only TP53 mutation was associated with malignant IPMNs. TP53 mutation could also be detected in pure pancreatic juice, potentially making it a useful tool to diagnose malignant IPMNs preoperatively.

Objectives The aims of this study were to identify genetic characteristics of intraductal papillary mucinous neoplasm (IPMN)-associated pancreatic ductal carcinoma (PDC) and to detect these markers using pancreatic juice. Methods From 76 cases, 102 tissues were obtained: 29 cases were noninvasive IPMN, 18 were PDC derived from IPMN (D-PDC; noninvasive part, n = 16; invasive part, n = 18), and 29 were PDC concomitant with IPMN (C-PDC; IPMN part, n = 10; PDC part, n = 29). Moreover, pancreatic juice samples from 28 cases were obtained (noninvasive IPMN, n = 13; D-PDC, n = 7; C-PDC, n = 8). Fifty-one cancer-related genes were analyzed by next-generation sequencing. Results TP53 mutation rates in D-PDC, C-PDC, and noninvasive IPMN were 67%, 66%, and 10%, respectively. Moreover, KRAS mutational patterns between 2 simultaneous tumors differed in 1 (6.3%) of the 16 D-PDC cases and in 8 (80%) of the 10 C-PDC cases (P = 0.0006). TP53 or multiple KRAS mutations were detected using pancreatic juice more frequently in C-PDC cases than in noninvasive IPMN cases (75% and 23%, respectively, P = 0.03). Conclusions Multiple KRAS mutations along with TP53 mutation are genetic markers for C-PDC, which could be detected using pancreatic juice preoperatively.

No abstract

Although comprehensive gene analyses of pancreatic cancer provide new knowledge on molecular mechanisms, the usefulness and possibility of the analyses in routinely available clinical samples remain unclear. We assessed the possibility and utility of target sequencing of endoscopically obtained pancreatic cancer samples. Fifty‐eight pancreatic cancer patients who underwent EUS‐FNA or endoscopic biopsy were enrolled. The extracted DNA quantity was assessed and used for next‐generation sequencing (NGS) of 50 cancer‐related genes from which gene mutations, copy number alterations, and microsatellite instability (MSI) were extracted via secondary analysis. A median of 19.2 ng (3.8–228) of DNA was extracted from formalin‐fixed paraffin‐embedded samples. Gene alterations were detected in 55 of 58 samples (94.8%), including all samples with a DNA concentration below the detection limit (n = 11). Four frequently altered genes were KRAS (83%), TP53 (66%), SMAD4 (26%), and PTEN (17%), and molecular targetable genes were detected in 13 cases (22.4%). Five samples (8.6%) had many mutations and suspected MSI with impaired mismatch repair genes. A Cox regression analysis revealed that metastasis (p < 0.005, hazard ratio [HR] 10.1), serum CEA >5 ng/ml (p = 0.01, HR 2.86), ≤10 detected hotspot mutations (p = 0.03, HR 9.86), and intact Ras signaling (p < 0.005, HR 5.57) were associated with a poor pancreatic cancer prognosis. We performed small, targeted sequencing of pancreatic cancer using available samples from real clinical practice and determined the relationship between gene alterations and prognosis to help determine treatment choices.