ORIGINAL RESEARCH article
Front. Oncol.
Sec. Cancer Metabolism
This article is part of the Research TopicMetabolic Reprogramming in Gastrointestinal Cancers: Crosstalk Between Tumor Plasticity and Therapy ResistanceView all articles
A Serum Metabolite-Based Machine Learning Model Predicts Response to Neoadjuvant Immunotherapy in Mismatch Repair-Deficient Colorectal Cancer
Provisionally accepted
Tao Ma1
Weili Zhang2
Yuxi Pan1Guojie Long1Xiuwei Mi1Junfeng Jiang1Fan Bai1Hao Zhang1
Tuo Hu1Ziyang Zeng2
Weidong Pan1*- 1Sun Yat-sen University Sixth Affiliated Hospital, Guangzhou, China
- 2Sun Yat-sen University Cancer Center, Guangzhou, 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
Background: Colorectal cancer (CRC) with microsatellite instability-high (MSI-H) or mismatch repair-deficient (dMMR) shows significant sensitivity to immune checkpoint inhibitors (ICIs). However, a considerable proportion of patients still exhibit primary or acquired resistance to ICIs. Until now, efficient and non-invasive biomarkers for accurately predicting immunotherapy efficacy remain unavailable. Methods: In this multicentre study, we employed liquid chromatography–mass spectrometry (LC–MS) and enzyme-linked immunosorbent assay (ELISA) to identify and validate serum metabolites associated with response to immunotherapy. Using machine learning algorithms, we constructed a random forest predictive model based on a panel of five metabolites. This model, termed the 5-Metabolite Predictive Model (5-MPM), incorporates prostaglandin E2 (PGE2), tryptophan, arginine, citrulline, and histidine. Results: The 5-MPM model demonstrated robust predictive performance in both training cohort and external validation cohort, with AUC values of 0.85 and 0.88, respectively. The SHAP analysis elucidated the contribution of each metabolite to model predictions. Integrating above five metabolites with metastasis stage did not further improve the predictive performance of this model. Discussion: This study provides the first systematic characterization of metabolic reprogramming in dMMR colorectal cancer with different response to immunotherapy, and establishes a non-invasive, high-precision predictive tool that offers a new basis for individualized therapeutic decision-making.
Keywords: colorectal cancer, immune checkpoint inhibitors, machine learning, predictive model, serum metabolomics
Received: 22 Oct 2025; Accepted: 06 Feb 2026.
Copyright: © 2026 Ma, Zhang, Pan, Long, Mi, Jiang, Bai, Zhang, Hu, Zeng and Pan. 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: Weidong Pan
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.