AI Article Synopsis

  • * It analyzed data from 1,235 patients and identified key independent prognostic factors such as tumor size and specific tumor markers (CEA, CA125, CA19-9).
  • * The new score model demonstrated improved predictive accuracy (higher AUC) compared to existing systems, providing a reliable tool for evaluating patient outcomes post-surgery.

Article Abstract

Prediction of prognosis after radical resection of gastric cancer has not been well established. Therefore, we aimed to establish a prognostic model based on a new score system of patients with gastric cancer. A total of 1235 patients who underwent curative gastrectomy at our hospital from October 2015 to April 2017 were included in this study. Univariate and multivariate analyses were used to screen for prognostic risk factors. Construction of the nomogram was based on Cox proportional hazard regression models. The construction of the new score models was analyzed by the receiver operating characteristic curve (ROC curve), calibration curve, and decision curve. Multivariate analysis showed that tumor size, T, N, carcinoembryonic antigen, CA125, and CA19-9 were independent prognostic factors. The new score model had a greater AUC (The area under the ROC curve) than other systems, and the C-index of the nomogram was highly reliable for evaluating the survival of patients with gastric cancer. Based on the tumor markers and other clinical indicators, we developed a precise model to predict the prognosis of patients with gastric cancer after radical surgery. This score system can be helpful to both surgeons and patients.

Download full-text PDF

Source
http://dx.doi.org/10.1177/10815589231179927DOI Listing

Publication Analysis

Top Keywords

gastric cancer
16
patients gastric
12
tumor markers
8
score system
8
roc curve
8
patients
5
curve
5
prediction survival
4
survival radical
4
radical gastrectomy
4

Similar Publications

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!