Deep learning assists detection of esophageal cancer and precursor lesions in a prospective, randomized controlled study.

Endoscopy is the primary modality for detecting asymptomatic esophageal squamous cell carcinoma (ESCC) and precancerous lesions. Improving detection rate remains challenging. We developed a system based on deep convolutional neural networks (CNNs) for detecting esophageal cancer and precancerous lesions [high-risk esophageal lesions (HrELs)] and validated its efficacy in improving HrEL detection rate in clinical practice (trial registration ChiCTR2100044126 at www.chictr.org.cn). Between April 2021 and March 2022, 3117 patients ≥50 years old were consecutively recruited from Taizhou Hospital, Zhejiang Province, and randomly assigned 1:1 to an experimental group (CNN-assisted endoscopy) or a control group (unassisted endoscopy) based on block randomization. The primary endpoint was the HrEL detection rate. In the intention-to-treat population, the HrEL detection rate [28 of 1556 (1.8%)] was significantly higher in the experimental group than in the control group [14 of 1561 (0.9%), P = 0.029], and the experimental group detection rate was twice that of the control group. Similar findings were observed between the experimental and control groups [28 of 1524 (1.9%) versus 13 of 1534 (0.9%), respectively; P = 0.021]. The system's sensitivity, specificity, and accuracy for detecting HrELs were 89.7, 98.5, and 98.2%, respectively. No adverse events occurred. The proposed system thus improved HrEL detection rate during endoscopy and was safe. Deep learning assistance may enhance early diagnosis and treatment of esophageal cancer and may become a useful tool for esophageal cancer screening.

ARL9 is upregulated and serves as a biomarker for a poor prognosis in colon adenocarcinoma.

BACKGROUND: ARL9 is a newly identified member of the ARF family, and the clinical significance of ARL9 in colon adenocarcinoma is unknown. In this study, we aimed to explore the expression of ARL9 mRNA in colon adenocarcinoma, and its effect on the prognosis of patients with colon adenocarcinoma. METHODS: We investigated the differential expression of ARL9 between colon adenocarcinoma tissue and adjacent tissues through a bioinformatics analysis using The Cancer Genome Atlas (TCGA) database. The correlation between clinical characteristics and the mRNA expression level of ARL9 were analyzed. A survival analysis and a Cox regression analysis were used to determine the prognostic significance of ARL9. Finally, we conducted a gene set enrichment analysis (GSEA) to explore the ARL9 signaling pathways involved in the development of colon adenocarcinoma. The effect of the expression of ARL9 on the proliferation and migration of colon adenocarcinoma was analyzed by the CCK8 method and a cell scratch test, respectively. RESULTS: The mRNA expression of ARL9 in colon adenocarcinoma tissues was higher in comparison to the level in normal adjacent tissues (P < 0.05). The mRNA expression of ARL9 was not related to sex, tumor stage, T stage, N stage, M stage, but to age. The 5-year survival rate of colon adenocarcinoma patients with high ARL9 mRNA expression levels was significantly lower than that of patients with low ARL9 mRNA expression levels (P < 0.05). Age and the high mRNA expression of ARL9 were independent risk factors for a poor prognosis in patients with colon adenocarcinoma. The GSEA suggested that ARL9 may be able to upregulate cell adhesion, extracellular matrix receptor interactions, tumor-associated pathways, and downregulate the citrate cycle and tricarboxylic acid cycle pathway, which are involved in the development of colon adenocarcinoma. After knocking down ARL9, the proliferation and migration abilities of colon adenocarcinoma cells were decreased (P < 0.01). CONCLUSION: The mRNA expression of ARL9 is upregulated in colon adenocarcinoma, and higher mRNA expression levels are associated with a poor prognosis. Knocking down ARL9 can reduce the proliferation and migration of colon adenocarcinoma cells. ARL9 mRNA can be used as a prognostic biomarker in patients with colon adenocarcinoma.

Effectiveness and safety of endoscopic resection for duodenal gastrointestinal stromal tumors: A single center analysis.

BACKGROUND: Endoscopic resection for duodenal gastrointestinal stromal tumors (GISTs) is still considered a great challenge with a high risk of complications, including perforation, bleeding, tumor rupture, and residual tumor. AIM: To assess the effectiveness and safety of endoscopic resection for duodenal GISTs. METHODS: Between January 2010 and January 2022, 11 patients with duodenal GISTs were treated with endoscopic resection. Data were extracted for the incidence of complete resection, bleeding, perforation, postoperative infection, recurrence, and distant metastasis. RESULTS: The incidence of successful complete resection of duodenal GISTs was 100%. Three cases (27.3%) had suspected positive margins, and the other 8 cases (72.7%) had negative vertical and horizontal margins. Perforation occurred in all 11 patients. The success rate of perforation closure was 100%, while 1 patient (9.1%) had suspected delayed perforation. All bleeding during the procedure was managed by endoscopic methods. One case (9.1%) had delayed bleeding. Postoperative infection occurred in 6 patients (54.5%), including 1 who developed septic shock and 1 who developed a right iliac fossa abscess. All 11 patients recovered and were discharged. The mean hospital stay was 15.3 d. During the follow-up period (14-80 mo), duodenal stenosis occurred in 1 case (9.1%), and no local recurrence or distant metastasis were detected. CONCLUSION: Endoscopic resection for duodenal GISTs appears to be an effective and safe minimally invasive treatment when performed by an experienced endoscopist.