ORIGINAL RESEARCH article
Front. Gastroenterol.
Sec. The Pancreas
Volume 4 - 2025 | doi: 10.3389/fgstr.2025.1523827
This article is part of the Research TopicPancreatic Cancer Awareness Month 2024: Current Progress in Pancreatic Cancer Treatment and ManagementView all articles
Exploring the Survival Benefits of Surgical Treatment for Pancreatic Adenocarcinoma Using the DeepSurv Neural Network Model
Provisionally accepted
- Beijing Tiantan Hospital, Capital Medical University, Beijing, China
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Objective: To apply the DeepSurv algorithm to construct a deep learning model for survival analysis of pancreatic adenocarcinoma patients. The model predicts the survival rates of surgical and non-surgical treatments across different stages, including various T and N stages within stage IV, to evaluate the survival benefits of surgical treatment and provide a basis for clinical decision-making.Methods: Clinical data of patients with pathologically confirmed pancreatic adenocarcinoma were selected from the Surveillance, Epidemiology, and End Results (SEER) database of the National Cancer Institute from January 2000 to December 2020. Based on TNM staging, patients were randomly divided into an experimental group and a model group. The model group was used to construct the DeepSurv model, while the experimental group was divided into surgical and non-surgical groups. Simulated paired data were created by converting surgical cases to non-surgical and vice versa. The experimental and paired groups were input into the DeepSurv model to obtain predicted paired survival rates and perform paired Wilcoxon signed rank test. The same method was used to predict and test the survival rates of different T and N stages within stage IV.Results: A total of 16,068 patients were included in the analysis. The constructed DeepSurv model had a C-index of 0.85 and a Brier score of 1.01e-3. The later the stage, the lower the survival rate for surgical treatment, but it remained higher than that of non-surgical treatment, with P<0.05. For stage IV patients, surgical treatment had a higher survival rate than nonsurgical treatment in T1-3 and N0 stages, with P<0.05, but there was no statistical difference in survival rates for T4 and N1 stages.Conclusion: Surgical treatment reduces mortality rates for patients at various stages, but its effectiveness diminishes in later stages. For stage IV patients, except for T4 and N1 stages, surgical treatment is beneficial compared to non-surgical treatment.
Keywords: Pancreatic adenocarcinoma, Survival Prediction, DeepSurv, primary tumor resection, neural network model
Received: 06 Nov 2024; Accepted: 22 Apr 2025.
Copyright: © 2025 Wang, Yan, Dong, cheng, Yu and Zhang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence:
Xin Wang, Beijing Tiantan Hospital, Capital Medical University, Beijing, China
Hongyi Zhang, Beijing Tiantan Hospital, Capital Medical University, Beijing, China
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.