aceBuilding Machine Learning Models Based on Oxidative Stress Index Score to Predict Survival of Locally Advanced Rectal Cancer Receiving Different Neoadjuvant Therapy Patterns
Abstract: Background Liver enzyme biomarkers are known to contribute to the onset and progression of colorectal cancer. Objective To develop a novel oxidative stress index score integrating γ-glutamyl transferase and total bilirubin, evaluate its prognostic value in locally advanced rectal cancer patients receiving neoadjuvant therapy, and construct and validate machine learning-based survival prediction models incorporating oxidative stress index score. Design A novel liver enzyme indicator - oxidative stress index score…
"Sun Yat-sen University no.5 Hospital" by Chintunglee is licensed under CC BY-SA 4.0. To view a copy of this license, visit https://creativecommons.org/licenses/by-sa/4.0/.
Researchers developed a novel oxidative stress index score from liver enzyme biomarkers and built machine learning models to predict disease-free survival and overall survival in 970 locally advanced rectal cancer patients treated at Sun Yat-Sen University Cancer Center. The training cohort received total neoadjuvant therapy and the validation cohort received neoadjuvant chemoradiotherapy, followed by surgery.
The work matters because it shows a routinely available biomarker combination can enhance AI-based prognostication beyond conventional ypStage, potentially informing treatment decisions and follow-up intensity. Uncertainty remains about generalizability beyond a single center and retrospective data, and about prospective clinical utility and integration into care pathways.
- Study enrolled 970 locally advanced rectal cancer patients from Sun Yat-Sen University Cancer Center between May 2007 and August 2018, split into training cohort n=521 receiving total neoadjuvant therapy and validation cohort n=449 receiving neoadjuvant chemoradiotherapy.
- Novel oxidative stress index score was established by integrating -glutamyl transferase and total bilirubin, and higher score was associated with worse prognostic outcomes.
- Among 5 machine learning models tested, random survival forests showed optimal efficiency and was used to stratify patients into low-risk and high-risk groups for disease-free survival and overall survival.
Random survival forests models incorporating oxidative stress index score improved prediction of disease-free and overall survival for locally advanced rectal cancer patients receiving neoadjuvant therapy, outperforming conventional ypStage and enabling risk stratification.
The rundown
Researchers collected data from Sun Yat-Sen University Cancer Center between May 2007 and August 2018 and enrolled 970 patients who underwent total mesorectal excision with or without adjuvant chemotherapy after either total neoadjuvant therapy or neoadjuvant chemoradiotherapy.
They integrated -glutamyl transferase and total bilirubin into an oxidative stress index score and built five machine learning models based on the score and clinicopathological characteristics, with random survival forests selected for final risk stratification and validation across both cohorts.
Sources
- Peer-reviewedDiseases of the Colon & Rectum2026-08-12
How should this claim be treated?
The debate