Predicting Chemotherapy Response in Patients With Advanced or Metastatic Pancreatic Cancer Using Machine Learning
JCO Clinical Cancer Informatics · 2025
DOI: 10.1200/cci-25-00124 ↗Kurz gefasst
Fachartikel in der Zeitschrift JCO Clinical Cancer Informatics, 2025. Verfasst von Jamin Koo und 10 weiteren Autor:innen. Beteiligte Einrichtung: ImpriMed Korea, Inc, Seoul, Republic of Korea. Hauptthemen laut MEDLINE: Pancreatic Neoplasms, Machine Learning, Antineoplastic Combined Chemotherapy Protocols. Die Arbeit nennt 31 Literaturangaben.
Themen: Künstliche Intelligenz in der Pflege, Onkologische Pflege
Abstract (Originalsprache)
PURPOSE Selecting an optimal first-line chemotherapy regimen for advanced or metastatic pancreatic cancer is challenging because of varying efficacy and toxicity profiles of fluorouracil, leucovorin, irinotecan, and oxaliplatin (FOLFIRINOX) and gemcitabine/nab-paclitaxel (GnP). This study aimed to develop machine learning (ML) models that predict survival outcomes and guide treatment selection using routinely available clinical data. METHODS We retrospectively analyzed 191 patients who received systemic chemotherapy for advanced or metastatic pancreatic cancer at Gangneung Asan Hospital and the Asan Medical Center between 2014 and 2023. Seventeen demographic and clinical variables, along with survival outcomes, were collected. The data set was stratified and split into training and test sets (4:1). CatBoost-based ML models were trained to predict 12-month overall survival (OS) for each regimen. A minimal subset of variables was selected using 5-fold cross-validation to optimize receiver operating characteristic (ROC)-AUC. Patients were classified as high or low risk based on model-derived thresholds. RESULTS The median age of the cohort was 62 years, and 64% was male. The ML models achieved ROC-AUCs of 0.81 for FOLFIRINOX and 0.82 for GnP. Predictive accuracies on test data were 0.77 and 0.80, respectively. Median OS differed significantly between predicted high- and low-risk groups: 9 v 15 months for FOLFIRINOX (hazard ratio [HR], 2.8; P < .0001) and 9 v 18 months for GnP (HR, 2.5; P < .01). In addition, 27% of patients predicted to be high risk for FOLFIRINOX were classified as low risk for GnP, and 32% vice versa. CONCLUSION ML models trained on multicenter data can effectively predict early mortality risk and help personalize chemotherapy selection in advanced or metastatic pancreatic cancer, potentially improving clinical outcomes.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- ImpriMed Korea, Inc, Seoul, Republic of Korea.
- ImpriMed, Inc, Mountain View, CA.
- Department of Chemical Engineering, Hongik University, Seoul, Republic of Korea.
- Department of Oncology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea.
- Department of Hematology and Oncology, University of Ulsan College of Medicine, Gangneung Asan Hospital, Gangneung, Republic of Korea.
ORCID-Profile: Jamin Koo, Changhoon Yoo, George Courcoubetis, Baek-Yeol Ryoo, Sungwon Lim, Moonho Kim
Zitieren
APA
Koo, J., Choi, G., Cheon, J., Yoo, C., Courcoubetis, G., Ryoo, B. Y., Kim, K. P., Chang, H. M., Oh, H. S., Lim, S., & Kim, M. (2025). Predicting Chemotherapy Response in Patients With Advanced or Metastatic Pancreatic Cancer Using Machine Learning. JCO Clinical Cancer Informatics. https://doi.org/10.1200/cci-25-00124
Vancouver
Koo J, Choi G, Cheon J, Yoo C, Courcoubetis G, Ryoo BY, et al. Predicting Chemotherapy Response in Patients With Advanced or Metastatic Pancreatic Cancer Using Machine Learning. JCO Clinical Cancer Informatics. 2025. doi:10.1200/cci-25-00124
BibTeX
@article{Koo2025,
author = {Koo, Jamin and Choi, Gyucheol and Cheon, Jaekyung and Yoo, Changhoon and Courcoubetis, George and Ryoo, Baek-Yeol and Kim, Kyu-Pyo and Chang, Heung-Moon and Oh, Ho-Suk and Lim, Sungwon and Kim, Moonho},
title = {Predicting Chemotherapy Response in Patients With Advanced or Metastatic Pancreatic Cancer Using Machine Learning},
journal = {JCO Clinical Cancer Informatics},
year = {2025},
doi = {10.1200/cci-25-00124}
}
EUVIMED