Artificial intelligence-based diagnosis of upper gastrointestinal subepithelial lesions on endoscopic ultrasonography images.

BACKGROUND: Endoscopic ultrasonography (EUS) is useful for the differential diagnosis of subepithelial lesions (SELs); however, not all of them are easy to distinguish. Gastrointestinal stromal tumors (GISTs) are the commonest SELs, are considered potentially malignant, and differentiating them from benign SELs is important. Artificial intelligence (AI) using deep learning has developed remarkably in the medical field. This study aimed to investigate the efficacy of an AI system for classifying SELs on EUS images. METHODS: EUS images of pathologically confirmed upper gastrointestinal SELs (GIST, leiomyoma, schwannoma, neuroendocrine tumor [NET], and ectopic pancreas) were collected from 12 hospitals. These images were divided into development and test datasets in the ratio of 4:1 using random sampling; the development dataset was divided into training and validation datasets. The same test dataset was diagnosed by two experts and two non-experts. RESULTS: A total of 16,110 images were collected from 631 cases for the development and test datasets. The accuracy of the AI system for the five-category classification (GIST, leiomyoma, schwannoma, NET, and ectopic pancreas) was 86.1%, which was significantly higher than that of all endoscopists. The sensitivity, specificity, and accuracy of the AI system for differentiating GISTs from non-GISTs were 98.8%, 67.6%, and 89.3%, respectively. Its sensitivity and accuracy were significantly higher than those of all the endoscopists. CONCLUSION: The AI system, classifying SELs, showed higher diagnostic performance than that of the experts and may assist in improving the diagnosis of SELs in clinical practice.

[A case of mucosa-associated lymphoid tissue lymphoma of the bile duct diagnosed using intraductal ultrasonography and transpapillary investigation of the bile duct].

A 78-year-old woman with a chief complaint of mammary tumor was referred to our hospital and was diagnosed with mucosa-associated lymphoid tissue (MALT) lymphoma using needle biopsies. She presented with MALT lymphoma lesions in the stomach, duodenum, and intraperitoneal lymph nodes. Abdominal ultrasonography revealed thickened internal layer of the bile duct. Cholangiocarcinoma, immunoglobulin G4-related sclerosing cholangitis, and MALT lymphoma were suspected. She did not report any abdominal pain, and did not have jaundice. Endoscopic retrograde cholangiopancreatography and intraductal ultrasonography indicated diffuse, irregular wall thickening with the formation of nodule lesions in the extrahepatic bile duct. Transpapillary cytology indicated CD20-positive atypical lymphocytes. MALT lymphoma of the bile duct was diagnosed. CVP with rituximab therapy was performed. After treatment, computed tomography, gastrointestinal endoscopy, endoscopic retrograde cholangiopancreatography, intraductal ultrasonography, and transpapillary cytology revealed that the tumor and the biliary wall thickening had disappeared. It is difficult to diagnose MALT lymphoma of the bile duct. We report a case in which intraductal ultrasound and transpapillary cytology were useful for establishing a diagnosis and evaluating the therapy for MALT lymphoma of the bile duct.

ATAD2 is associated with malignant characteristics of pancreatic cancer cells.

Pancreatic cancer is one of the most aggressive human cancers and is associated with a poor prognosis. To develop a novel strategy for pancreatic cancer treatment, it is essential to elucidate the molecular mechanisms underlying the invasion and proliferation of cancer cells. ATPase family AAA domain containing protein 2 (ATAD2) is a highly conserved protein with an AAA+ domain and a bromodomain. Accumulating studies have demonstrated that ATAD2 is associated with the progression of multiple cancers. The present study demonstrated that ATAD2 depletion suppressed cell invasion and migration. In addition, ATAD2 knockdown suppressed anchorage-independent growth of pancreatic cancer cells. Finally, ATAD2 depletion was demonstrated to sensitize pancreatic cancer cells to gemcitabine. The results of the present study indicate that ATAD2 is involved in the malignant characteristics of pancreatic cancer.

AMPD3 is associated with the malignant characteristics of gastrointestinal stromal tumors.

Gastrointestinal stromal tumors (GISTs) are mesenchymal tumors of the gastrointestinal tract. It is well known that activating mutations in the receptor tyrosine kinases KIT and platelet-derived growth factor receptor-α have essential roles in the pathogenesis of GISTs. The activation of these receptor protein kinases triggers multiple signaling pathways that promote cell proliferation and survival; however, the exact mechanism by which the activation of these kinases promotes the progression of GISTs remains uncertain. The aim of the present was to search for genes that are associated with the progression of GIST. The present study used reverse transcription-quantitative polymerase chain reaction to demonstrate that adenosine monophosphate deaminase 3 (AMPD3) was highly expressed in GISTs. Furthermore, transfection of GIST-T1 cells with KIT-specific small interfering RNA (siRNA) demonstrated that the expression of AMPD3 was dependent on KIT expression, while the depletion of AMPD3 in human GIST-T1 cells using AMPD3-specific siRNA resulted in the suppression of cell migration and invasion. In addition, AMPD3 depletion sensitized GIST-T1 cells to the tyrosine kinase inhibitor imatinib. The results of the present suggested that the combined inhibition of tyrosine kinases and AMPD3 may be effective for the treatment of GISTs.

Treatment with near-infrared radiation promotes apoptosis in pancreatic cancer cells.

Cancer remains one of the leading causes of human mortality worldwide. Radiation and chemotherapy are commonly used for cancer treatment; however, the combination of these therapies and surgery do not completely eradicate cancer cells. Near-infrared radiation (NIR) is a low-energy form of radiation that exerts multiple effects on mammalian cells. Previous studies have reported that NIR induces DNA double-strand breaks and apoptosis of cancer cells. In the present study, a 915-nm laser was used to examine the effects of NIR on pancreatic cancer cells. Irradiation of pancreatic cancer cells using a 915-nm laser significantly induced caspase-3 activation and apoptosis. In addition, the combination of gemcitabine treatment and a 915-nm laser synergistically increased the number of apoptotic cells. The results of the present study indicate the use of infrared irradiation and chemotherapy may be a possible therapy for the treatment of cancer.

Ring-enhancement pattern on contrast-enhanced CT predicts adenosquamous carcinoma of the pancreas: a matched case-control study.

OBJECTIVES: Adenosquamous carcinoma of the pancreas (ASC) is a rare malignant neoplasm of the pancreas, exhibiting both glandular and squamous differentiation. However, little is known about its imaging features. This study examined the imaging features of pancreatic ASC. METHODS: We evaluated images of contrast-enhanced computed tomography (CT) and endoscopic ultrasonography (EUS). As controls, solid pancreatic neoplasms matched in a 2:1 ratio to ASC cases for age, sex and tumor location were also evaluated. RESULTS: Twenty-three ASC cases were examined, and 46 solid pancreatic neoplasms (43 pancreatic ductal adenocarcinomas, two pancreatic neuroendocrine tumors and one acinar cell carcinoma) were matched as controls. Univariate analysis demonstrated significant differences in the outline and vascularity of tumors on contrast-enhanced CT in the ASC and control groups (P < 0.001 and P < 0.001, respectively). A smooth outline, cystic changes, and the ring-enhancement pattern on contrast-enhanced CT were seen to have significant predictive powers by stepwise forward logistic regression analysis (P = 0.044, P = 0.010, and P = 0.001, respectively). Of the three, the ring-enhancement pattern was the most useful, and its predictive diagnostic sensitivity, specificity, positive predictive value and negative predictive value for diagnosis of ASC were 65.2%, 89.6%, 75.0% and 84.3%, respectively. CONCLUSIONS: These results demonstrate that presence of the ring-enhancement pattern on contrast-enhanced CT is the most useful predictive factor for ASC.

Clinical characteristics of adenosquamous carcinoma of the pancreas: a matched case-control study.

OBJECTIVES: Adenosquamous carcinoma of the pancreas (ASC) is a variant of pancreatic ductal adenocarcinoma (PDAC), but the prognosis remains unclear. The purpose of this study was to clarify the prognosis of ASC using a matched case-control design. METHODS: We evaluated clinical characteristics of ASC treated between 2001 and 2011 in our institution. As controls, PDAC cases matched with ASC cases for sex, age, pretreatment Eastern Cooperative Oncology Group performance status, location, initial therapy and American Joint Committee on Cancer TNM staging for pancreatic cancer were also evaluated. RESULTS: Of the 914 cases of pancreatic neoplasm, 28 cases (3.06%) of ASC were identified, and 56 cases of PDAC were matched as controls. Median overall survival (OS) was significantly worse for ASC (8.38 months) than for PDAC (15.75 months; hazard ratio [HR], 1.94; 95% confidence interval, 1.07-3.51; P = 0.026). Of the 22 unresected cases, median OS was again significantly worse for ASC (4.67 months) than for PDAC (12.36 months; HR, 2.39; 95% confidence interval, 1.27-4.51; P = 0.007). CONCLUSION: These results demonstrate that ASC is more aggressive than PDAC.