A Case of Superficial Primary Malignant Melanoma of the Esophagus Detected and Treated at Stage 0.

The patient was a 79-year-old male. At three years and eight months after his initial presentation, upper gastrointestinal endoscopy revealed a black-flattened elevated lesion in the middle third of the esophagus, which was diagnosed as malignant melanoma on biopsy. No lymph node or distant metastasis was found. A diagnosis of cT1bN0M0 Stage I was thus made. We performed a robot-assisted, minimally invasive esophagectomy and D2 dissection. The postoperative diagnosis was pT1a (MM), N0, M0, vascular invasion+, stage 0. The patient was recurrence-free for 14 months after surgery. We presume that an aggressive biopsy diagnosis is important for the early detection of malignant melanoma.

Semiautomatic contour tracking method for biological object segmentation in thin-section electron microscope images with modified zero DC component-type Gabor wavelets.

In electron microscopic image processing, artificial intelligence (AI) is a powerful method for segmentation. Because creating training data remains time-consuming and burdensome, a simple and accurate segmentation tool, which is effective and does not rely on manual drawings, is necessary to create training data for AI and to support immediate image analysis. A Gabor wavelet-based contour tracking method has been devised as a step toward realizing such a tool. Although many papers on Gabor filter-based and Gabor filter bank-based texture segmentations have been published, previous studies did not apply the Gabor wavelet-based method to straightforwardly detect membrane-like ridges and step edges for segmentation because earlier works used a nonzero DC component-type Gabor wavelets. The DC component has a serious flaw in such detection. Although the DC component can be removed by a formula that satisfies the wavelet theory or by a log-Gabor function, this is not practical for the proposed scheme. Herein, we devised modified zero DC component-type Gabor wavelets. The proposed method can practically confine a wavelet within a small image area. This type of Gabor wavelet can appropriately track various contours of organelles appearing in thin-section transmission electron microscope images prepared by the freeze-substitution fixation method. The proposed method not only more accurately tracks ridge and step edge contours but also tracks pattern boundary contours consisting of slightly different image patterns. Simulations verified these results.

Postpolypectomy bleeding of colorectal polyps in patients with continuous warfarin and short-term interruption of direct oral anticoagulants.

BACKGROUND AND AIMS: Postpolypectomy bleeding (PPB) is the most common adverse event after colorectal polypectomy. Use of anticoagulants is an important risk factor for PPB. This study aimed to evaluate PPB in patients receiving treatment with warfarin and direct oral anticoagulants (DOACs). METHODS: Between August 2017 and July 2019, 5449 patients with 12,601 polyps who underwent endoscopic snare resection of colorectal polyps were enrolled. Endoscopic snare resection was performed in patients receiving continuous warfarin (C-warfarin) and in patients who experienced 1 day cessation of (O-) of DOACs in accordance with the Japanese Gastroenterological Endoscopy Society guidelines. RESULTS: The PPB rate in the group receiving anticoagulants was statistically higher than that in the group without anticoagulants (8.5% [33/387] vs 1.2% [63/5,062], respectively; P < .001). By multivariate logistic regression analysis, male gender (odds ratio [OR], 2.17; P = .007), warfarin (OR, 4.64; P < .001), DOACs (OR, 6.59; P < .001), and multipolyp removal (OR, 1.77; P = .007) were significant risk factors for PPB. PPB was observed in 9 and 21 patients in the C-warfarin and O-DOACs groups, respectively: C-warfarin (8.0% [9/113]), O-dabigatran (6.1% [2/33]), O-rivaroxaban (14.8% [9/61]), O-apixaban (9.8% [9/92]), and O-edoxaban (1.8% [1/56]). The PPB rate with the O-edoxaban group was significantly lower than that with the O-rivaroxaban group (P < .05). CONCLUSIONS: Use of anticoagulant therapy was an independent risk factor for PPB. The rates of PPB in patients receiving C-warfarin and O-DOACs were also higher than those in patients not receiving anticoagulants. Edoxaban may be safe through short-term withdrawal in patients undergoing endoscopic snare resection of colorectal polyps.

Elevation of Neutrophil-to-Lymphocyte Ratio before First-Line Chemotherapy Predicts a Poor Prognosis for Second-Line Chemotherapy in Gastric Cancer.

OBJECTIVES: The neutrophil-to-lymphocyte ratio (NLR) has been proposed as an indicator of cancer-related inflammation. The aim of our study was to examine the prognostic value of the NLR for patients with advanced gastric cancer receiving second-line chemotherapy. METHODS: The association of overall survival (OS) in second-line chemotherapy and the clinicopathological findings including NLR were analyzed retrospectively. The selection criteria were patients who received second-line chemotherapy between January 2010 and June 2015, had histologically confirmed gastric adenocarcinoma, and were followed up until death or for 180 days or longer. RESULTS: Eighty-six patients met the selection criteria. Multivariate analysis revealed that performance status 2, hemoglobin < 10 g/dL, and NLR before first-line chemotherapy ≥3 were adverse predictive markers. NLR before second-line chemotherapy was not associated with OS. A prognostic model was constructed dividing patients into three groups according to the number of adverse predictive factors: good (no factor), intermediate (one factor), and poor (more than two factors). The median OS for the good, intermediate, and poor groups was 14.3, 7.2, and 4.4 months, respectively (p < 0.001). CONCLUSIONS: Patients with advanced gastric cancer with performance status 2, hemoglobin < 10 g/dL, and NLR before first-line chemotherapy ≥3 are not likely to benefit from second-line chemotherapy.