Machine learning-based model for prediction and feature analysis of recurrence in pancreatic neuroendocrine tumors G1/G2.

BACKGROUND: Pancreatic neuroendocrine neoplasms (PanNENs) are a heterogeneous group of tumors. Although the prognosis of resected PanNENs is generally considered to be good, a relatively high recurrence rate has been reported. Given the scarcity of large-scale reports about PanNEN recurrence due to their rarity, we aimed to identify the predictors for recurrence in patients with resected PanNENs to improve prognosis. METHODS: We established a multicenter database of 573 patients with PanNENs, who underwent resection between January 1987 and July 2020 at 22 Japanese centers, mainly in the Kyushu region. We evaluated the clinical characteristics of 371 patients with localized non-functioning pancreatic neuroendocrine tumors (G1/G2). We also constructed a machine learning-based prediction model to analyze the important features to determine recurrence. RESULTS: Fifty-two patients experienced recurrence (14.0%) during the follow-up period, with the median time of recurrence being 33.7 months. The random survival forest (RSF) model showed better predictive performance than the Cox proportional hazards regression model in terms of the Harrell's C-index (0.841 vs. 0.820). The Ki-67 index, residual tumor, WHO grade, tumor size, and lymph node metastasis were the top five predictors in the RSF model; tumor size above 20 mm was the watershed with increased recurrence probability, whereas the 5-year disease-free survival rate decreased linearly as the Ki-67 index increased. CONCLUSIONS: Our study revealed the characteristics of resected PanNENs in real-world clinical practice. Machine learning techniques can be powerful analytical tools that provide new insights into the relationship between the Ki-67 index or tumor size and recurrence.

Pancreatic Juice Exosomal MicroRNAs as Biomarkers for Detection of Pancreatic Ductal Adenocarcinoma.

BACKGROUND: Pancreatic ductal adenocarcinoma (PDAC) is a lethal neoplasm because of difficulties in early detection. Several studies have recently suggested that exosomes may have potential as novel biomarkers. This study aimed to isolate exosomes from pancreatic juice and to investigate whether exosomal microRNAs (ex-miRs) could be used as biomarkers for PDAC. METHODS: Pancreatic juice was collected from patients with PDAC and chronic pancreatitis (CP) by endoscopic retrograde pancreatography. Exosomes were extracted by ultracentrifugation. The presence of exosomes was confirmed by electron microscopy and Western blotting using anti-CD63, -CD81, and -TSG101 antibodies. Relative levels of ex-miR-21 and ex-miR-155 were quantified and compared between PDAC and CP patients. RESULTS: A total of 35 pancreatic juice samples (27 PDAC and 8 CP) were collected. Relative levels of both ex-miR-21 and ex-miR-155 were significantly higher in PDAC patients compared with CP patients (p < 0.001 and p = 0.008, respectively). By contrast, no significant difference was apparent in relative levels of miR-21 and miR-155 in whole pancreatic juice from PDAC patients compared with CP patients (p = 0.08 and p = 0.61, respectively). Ex-miR-21 and ex-miR-155 levels discriminated PDAC patients from CP patients with area under the curve values of 0.90 and 0.89, respectively. The accuracies of ex-miR-21 levels, ex-miR-155 levels, and pancreatic juice cytology were 83%, 89%, and 74%, respectively. When combining the results of ex-miR profiling with pancreatic juice cytology, the accuracy was improved to 91%. CONCLUSIONS: We successfully extracted exosomes from pancreatic juice. Ex-miRs, including ex-miR-21 and ex-miR-155, in pancreatic juice may be developed as biomarkers for PDAC.

Clinicopathological characteristics of non-functioning cystic pancreatic neuroendocrine tumors.

BACKGROUND/OBJECTIVES: The biological features of cystic pancreatic neuroendocrine tumors (PNETs) remain unclear. The aim of this study was to clarify the clinicopathological characteristics of non-functioning PNETs (NF-PNETs) with a cystic component. METHODS: The medical records of 75 patients with NF-PNETs who had undergone resection in our institution were retrospectively reviewed. Clinicopathological factors were compared between PNETs with and without a cystic component. Expression of somatostatin 2 receptor (SSTR-2) was also analyzed. RESULTS: Cystic PNETs were diagnosed in 14 patients (19%). The proportion of men was significantly higher for cystic than solid PNETs (79% vs. 44%, P < 0.05) and cystic PNETs were significantly larger than solid PNETs (25 mm vs. 17 mm, P < 0.01). However, there were no significant differences in the prevalence of lymph node metastases (14% vs. 10%, P = 0.64), hepatic metastasis (7% vs. 3%, P = 0.54), or disease-free survival rate (both 86%, P = 0.29) between PNETs with and without a cystic component. SSTR-2 expression was more frequently observed in PNETs with a cystic component than in those without (100% vs. 70%, P < 0.01). CONCLUSIONS: Although cystic PNETs were larger upon diagnosis than solid PNETs in this study, prognosis after surgical resection did not differ significantly between these types of PNET. Somatostatin receptor scintigraphy and somatostatin analogues may be more useful for diagnosing and treating cystic PNETs, respectively.

Role of SpyGlass-DStm in the preoperative assessment of pancreatic intraductal papillary mucinous neoplasm involving the main pancreatic duct.

BACKGROUND/OBJECTIVES: It is often difficult to determine an adequate resection line during pancreatectomy for intraductal papillary mucinous neoplasm involving the main pancreatic duct during partial pancreatectomy. The aim of this study was to evaluate the usefulness of improved peroral pancreatoscopy using SpyGlass-DStm in the preoperative assessment of intraductal papillary mucinous neoplasm involving the main pancreatic duct. METHODS: We collected and retrospectively analyzed clinicopathological data from seven consecutive patients who underwent preoperative assessment of intraductal papillary mucinous neoplasm involving the main duct using SpyGlass-DStm. RESULTS: Good imaging quality of the intraductal protruding lesion was obtained in all seven patients, and only one adverse event was noted wherein a patient had mild pancreatitis. Six patients underwent pancreatectomy. In one patient, masked-type concomitant pancreatic ductal adenocarcinoma and low-length dysplastic lesion was found near the surgical margin, which was not detected by preoperative imaging modalities including SpyGlass-DStm. The sensitivity of targeting biopsy during SpyGlass-DStm to diagnose high-grade dysplasia was 0%. CONCLUSIONS: SpyGlass-DStm can be safely performed in patients with intraductal papillary mucinous neoplasm involving the main duct, and has excellent visualization of the target lesion. However, challenges include poor diagnostic ability of targeting biopsy, and, therefore, intraoperative frozen section is still needed to obtain negative surgical margins.

Elevated bile amylase level without pancreaticobiliary maljunction is a risk factor for gallbladder carcinoma.

BACKGROUND: Elevated bile amylase level in patients with pancreaticobiliary maljunction (PBM) or high confluence of pancreaticobiliary ducts (HCPBD) is well known as a risk factor for gallbladder carcinoma (GBC) development. However, the effects of occult pancreaticobiliary reflux (OPR), a condition characterized by high bile amylase level in the presence of an anatomically normal pancreaticobiliary junction, on GBC development remain unclear. The aim of this study was to assess the relationship between OPR and GBC. METHODS: Clinicopathological data of 52 patients who were preoperatively diagnosed with gallbladder (GB) tumor (22 malignant, 30 benign) were retrospectively reviewed. All of the patients underwent preoperative endoscopic retrograde cholangiopancreatography to evaluate pancreaticobiliary junction morphology and bile amylase level. The relationship between the histological diagnosis of GB lesions, and pancreaticobiliary junction morphology and bile amylase level were investigated. RESULTS: Pancreaticobiliary maljunction, HCPBD, and normal pancreaticobiliary junction (NPJ) were identified in 12, nine, and 31 patients, respectively. The rates of GBC in patients with PBM, HCPBD, and NPJ were 58% (7/12), 67% (6/9), and 29% (9/31), respectively. Of the 31 patients with NPJ, 22 had OPR and nine of these had GBC. None of the patients with NPJ and normal bile amylase level had GBC. Additionally, among patients with NPJ, bile amylase level was significantly higher in patients with GBC than in patients with benign tumors. CONCLUSIONS: Occult pancreaticobiliary reflux, like PBM and HCPBD, is a risk factor for GBC development.

GNAS and KRAS mutational analyses of intraductal papillary neoplasms of the pancreas and bile duct developing in the same individual: A case report.

Intraductal papillary mucinous neoplasm (IPMN) of the pancreas and intraductal papillary neoplasm of the bile duct (IPNB) are considered as counterparts of each other, and it is suggested that these two entities have similar molecular alteration pathways. However, the occurrence of IPMN of the pancreas and IPNB in the same patient is rare. We report a surgical case of a 69-year-old woman who developed invasive IPMN of the pancreas and underwent pancreatectomy, 6 months after hepatic resection of invasive IPNB. Molecular analysis revealed GNAS/KRAS mutation in both invasive IPMN of the pancreas and IPNB. This is believed to be the first case report investigating GNAS/KRAS mutational status in both IPMN of the pancreas and IPNB developing in the same patient, and these two entities may show similar molecular alternations.