A novel risk score system for prognostic evaluation in adenocarcinoma of the oesophagogastric junction: a large population study from the SEER database and our center

Construction of a novel nomogram for predicting overall survival in patients with Siewert type II AEG based on LODDS: a study based on the seer database and external validation

Background

In recent years, the incidence of adenocarcinoma of the esophagogastric junction (AEG) has been rapidly increasing globally. Despite advances in the diagnosis and treatment of AEG, the overall prognosis for AEG patients remains concerning. Therefore, analyzing prognostic factors for AEG patients of Siewert type II and constructing a prognostic model for AEG patients is important.

Methods

Data of primary Siewert type II AEG patients from the SEER database from 2004 to 2015 were obtained and randomly divided into training and internal validation cohort. Additionally, data of primary Siewert type II AEG patients from the China Medical University Dandong Central Hospital from 2012 to 2018 were collected for external validation. Each variable in the training set underwent univariate Cox analysis, and variables with statistical significance (p < 0.05) were added to the LASSO equation for feature selection. Multivariate Cox analysis was then conducted to determine the independent predictive factors. A nomogram for predicting overall survival (OS) was developed, and its performance was evaluated using ROC curves, calibration curves, and decision curves. NRI and IDI were calculated to assess the improvement of the new prediction model relative to TNM staging. Patients were stratified into high-risk and low-risk groups based on the risk scores from the nomogram.

Results

Age, Differentiation grade, T stage, M stage, and LODDS (Log Odds of Positive Lymph Nodes)were independent prognostic factors for OS. The AUC values of the ROC curves for the nomogram in the training set, internal validation set, and external validation set were all greater than 0.7 and higher than those of TNM staging alone. Calibration curves indicated consistency between the predicted and actual outcomes. Decision curve analysis showed moderate net benefit. The NRI and IDI values of the nomogram were greater than 0 in the training, internal validation, and external validation sets. Risk stratification based on the nomogram's risk score demonstrated significant differences in survival rates between the high-risk and low-risk groups.

Conclusion

We developed and validated a nomogram for predicting overall survival (OS) in patients with Siewert type II AEG, which assists clinicians in accurately predicting mortality risk and recommending personalized treatment strategies.

The development and validation of automated machine learning models for predicting lymph node metastasis in Siewert type II T1 adenocarcinoma of the esophagogastric junction

Background

Lymph node metastasis (LNM) is considered an essential prognosis factor for adenocarcinoma of the esophagogastric junction (AEG), which also affects the treatment strategies of AEG. We aimed to evaluate automated machine learning (AutoML) algorithms for predicting LNM in Siewert type II T1 AEG.

Methods

A total of 878 patients with Siewert type II T1 AEG were selected from the Surveillance, Epidemiology, and End Results (SEER) database to develop the LNM predictive models. The patients from two hospitals in Suzhou were collected as the test set. We applied five machine learning algorithms to develop the LNM prediction models. The performance of predictive models was assessed using various metrics including accuracy, sensitivity, specificity, the area under the curve (AUC), and receiver operating characteristic (ROC) curve.

Results

Patients with LNM exhibited a higher proportion of male individuals, a poor degree of differentiation, and submucosal infiltration, with statistical differences. The deep learning (DL) model demonstrated relatively good accuracy (0.713) and sensitivity (0.868) among the five models. Moreover, the DL model achieved the highest AUC (0.781) and sensitivity (1.000) in the test set.

Conclusion

The DL model showed good predictive performance among five AutoML models, indicating the advantage of AutoML in modeling LNM prediction in patients with Siewert type II T1 AEG.