Improving Prognostic Accuracy in Non-Metastatic Gastric Cancer: A Comparative Analysis of Statistical and Machine Learning Survival Models in SEER Cohorts
Keywords:
Gastric cancer; Survival analysis; Machine learning; Random survival forest; SEER database..Abstract
Gastric cancer (GC) remains a leading cause of global cancer mortality. Accurate prognosis in non-metastatic disease remains challenging when treating with curative intent. This is due to the survival heterogeneity that exists within tumor-node-metastasis (TNM) stages. In this study, six survival models were compared in 20,132 patients using 10-fold cross-validation. Models including Cox proportional hazards, Weibull accelerated failure time, survival support vector machine, random survival forest (RSF), gradient boosting survival analysis (GBSA), and DeepSurv were used. Performance was evaluated using C-index, integrated Brier score (IBS), calibration, decision curve analysis (DCA), and net reclassification improvement (NRI). Over a median 41 month follow-up, 13,272 deaths occurred. Similar discrimination was found among top models; DeepSurv (C index 0.7542), GBSA (0.7540), and RSF (0.7532) higher than Cox (0.7417). In terms of clinical utility, RSF got the lowest prediction error (IBS 0.1516), followed by GBSA (0.1534). However, DeepSurv indicated poor calibration and limited net benefit at 12 months. NRI analysis indicated that ensemble models (RSF, GBSA) consistently improved reclassification over Cox, whereas DeepSurv exhibited a decline at 12 months. Age, histology, and nodal burden were the primary predictors. Ensemble models, particularly RSF, offer a balance of both accuracy and clinical utility for risk stratification in non-metastatic GC.
Downloads
Downloads
Published
Issue
Section
License
You are free to:
- Share — copy and redistribute the material in any medium or format for any purpose, even commercially.
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
- The licensor cannot revoke these freedoms as long as you follow the license terms.
Under the following terms:
- Attribution — You must give appropriate credit , provide a link to the license, and indicate if changes were made . You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
Notices:
You do not have to comply with the license for elements of the material in the public domain or where your use is permitted by an applicable exception or limitation .
No warranties are given. The license may not give you all of the permissions necessary for your intended use. For example, other rights such as publicity, privacy, or moral rights may limit how you use the material.
