A machine learning model for predicting the lymph node metastasis of early gastric cancer not meeting the endoscopic curability criteria.

BACKGROUND: We developed a machine learning (ML) model to predict the risk of lymph node metastasis (LNM) in patients with early gastric cancer (EGC) who did not meet the existing Japanese endoscopic curability criteria and compared its performance with that of the most common clinical risk scoring system, the eCura system. METHODS: We used data from 4,042 consecutive patients with EGC from 21 institutions who underwent endoscopic submucosal dissection (ESD) and/or surgery between 2010 and 2021. All resected EGCs were histologically confirmed not to satisfy the current Japanese endoscopic curability criteria. Of all patients, 3,506 constituted the training cohort to develop the neural network-based ML model, and 536 constituted the validation cohort. The performance of our ML model, as measured by the area under the receiver operating characteristic curve (AUC), was compared with that of the eCura system in the validation cohort. RESULTS: LNM rates were 14% (503/3,506) and 7% (39/536) in the training and validation cohorts, respectively. The ML model identified patients with LNM with an AUC of 0.83 (95% confidence interval, 0.76-0.89) in the validation cohort, while the eCura system identified patients with LNM with an AUC of 0.77 (95% confidence interval, 0.70-0.85) (P = 0.006, DeLong's test). CONCLUSIONS: Our ML model performed better than the eCura system for predicting LNM risk in patients with EGC who did not meet the existing Japanese endoscopic curability criteria. We developed a neural network-based machine learning model that predicts the risk of lymph node metastasis in patients with early gastric cancer who did not meet the endoscopic curability criteria.

[An anatomical case of obstructive jaundice due to the rapid enlargement of hepatic peribiliary cysts in end-stage alcoholic liver cirrhosis].

A man in his 60s had end-stage alcoholic cirrhosis. About six months before his death, hepatic peribiliary cysts (HPBC) rapidly increased, and he developed jaundice and liver failure. The pathological autopsy performed after his death revealed that his intrahepatic bile duct was pressured due to multiple cysts caused by HPBC, which resulted in liver failure. Some cases of HPBC have been associated with alcoholic cirrhosis;however, no other cases of increased HPBC in a short period of time have been reported. Although identifying the cause of increased HPBC in a short time is difficult in this case, it may be have been caused by continuous alcohol drinking after the onset of HPBC. Most patients with HPBC have liver cirrhosis and obstructive jaundice that may promote liver failure as in this case. Therefore, patients with HPBC should not only be instructed for abstinence but also promptly consider effective treatments in the event of obstructive jaundice to prevent liver dysfunction.

Prognostic Factors of Oral Intake After Endoscopic Gastroduodenal Stent Placement for Advanced Gastric Cancer.

Gastroduodenal stents are effective for gastric outlet obstruction (GOO) due to gastric cancer. However, some patients are unable to eat again soon. We retrospectively analyzed the cause of short-feeding periods. Between 2011 and 2015, we performed stent placement in 22 patients who could not eat solids. The effects of clinical characteristics on duration of oral intake were analyzed using Cox proportional hazards models. Univariate analyses revealed that the degree of GOO [no oral intake/liquids only; hazard ratio (HR), 10.9; 95% confidence interval (CI), 2.5-48.1; P=0.003], performance status score (2 or 3/0 or 1; HR, 5.7; 95% CI, 1.8-16.9; P=0.004), and poststenting chemotherapy (no/yes; HR, 5.7; 95% CI, 1.9-18.9; P=0.002) were significant factors for cessation of oral intake. Multivariate analysis showed that GOO and chemotherapy were significant factors. Gastroduodenal stents were less effective for patients with severe stenosis or without poststenting chemotherapy.