High-accuracy prediction of colorectal cancer chemotherapy efficacy using machine learning applied to gene expression data

Aug 4, 2026

High-accuracy prediction enables better selection between FOLFOX and FOLFIRI, two standard first-line chemotherapy options for colorectal cancer, yet treatment choice has not been fully clarified. A newly developed machine learning model is trained on public gene expression datasets from GEO to derive molecular signatures predictive of the efficacy of 5-FU based combination therapy. The approach uses 5-fold cross-validation with LASSO and VarSelRF feature selection and evaluates classifiers including random forest and support vector machines. Validation and test results exceed 90% accuracy with 85–95% specificity and sensitivity.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages23

Reading Time58 min

AccessFree to download

Introduction

Methods

Results and Discussion

Keywords

TYPE Original Research PUBLISHED 18 January 2024 DOI 10.3389/fphys.2023.1272206 OPEN ACCESS EDITED BY Rajesh kumar Tripathy, Birla Institute of Technology and Science, India REVIEWED BY Jennie L. Williams, Stony Brook University, United States Elife Zerrin Bagci, Namik Kemal University, Türkiye *CORRESPONDENCE Mohsin Saleet Jafri, [sjafri@gmu.edu](sjafri@gmu.edu) RECEIVED 03 August 2023 ACCEPTED 26 December 2023 PUBLISHED 18 January 2024 CITATION Amniouel S and Jafri MS (2024), High-accuracy prediction of colorectal cancer chemotherapy efﬁcacy using machine learning applied to gene expression data. Front. Physiol. 14:1272206 . doi: 10.3389/fphys.2023.1272206 COPYRIGHT © 2024 Amniouel and Jafri. This is an openaccess 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) and the copyright owner(s) 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. High-accuracy prediction of colorectal cancer chemotherapy efﬁcacy using machine learning applied to gene expression data Soukaina Amniouel 1 and Mohsin Saleet Jafri 1,2* 1School of Systems Biology, George Mason University, Fairfax, VA, United States, 2Center for Biomedical Engineering and Technology, University of Maryland School of Medicine, Baltimore, MD, United States Introduction: FOLFOX and FOLFIRI chemotherapy are considered standard ﬁrstline treatment options for colorectal cancer (CRC) . However, the criteria for selecting the appropriate treatments have not been thoroughly analyzed. Methods: A newly developed machine learning model was applied on several gene expression data from the public repository GEO database to identify molecular signatures predictive of efﬁcacy of 5-FU based combination chemotherapy (FOLFOX and FOLFIRI) in patients with CRC. The model was trained using 5-fold cross validation and multiple feature selection methods including LASSO and VarSelRF methods. Random Forest and support vector machine classiﬁers were applied to evaluate the performance of the models. Results and Discussion: For the CRC GEO dataset samples from patients who received either FOLFOX or FOLFIRI, validation and test sets were >90% correctly classiﬁed (accuracy), with speciﬁcity and sensitivity ranging between 85%-95% . In the datasets used from the GEO database, 28.6% of patients who failed the treatment therapy they received are predicted to beneﬁt from the alternative treatment. Analysis of the gene signature suggests the mechanistic difference between colorectal cancers that respond and those that do not respond to FOLFOX and FOLFIRI. Application of this machine learning approach could lead to improvements in treatment outcomes for patients with CRC and other cancers after additional appropriate clinical validation. KEYWORDS colorectal cancer, FOLFOX, FOLFIRI, chemoresistance, machine learning, gene expression, feature selection 1 Introduction Colorectal cancer (CRC) is the most frequent malignant disease of the gastrointestinal tract, the third most frequent cancer affecting both men and women and is oneofthe leading causes of cancer-related morbidity and mortality in spite of widespread, effective measures of preventive screening, and major advances in treatment options (Fouad et al., 2018; Sung et al., 2021). In recent decades, the overall long-term outcome of patients curatively resected has not signiﬁcantly changed. The 5-year survival rate for CRC is 63% but drops to 14% for metastatic CRC. More than half of colorectal adenocarcinomas are still diagnosed only when the disease involves regional or distant structures (Araghi et al., 2021). Thus, further investigation is still needed to develop effective approaches for medical in

FAQs

What problem does the study address in colorectal cancer treatment selection?

The study targets the lack of thoroughly analyzed criteria for choosing between FOLFOX and FOLFIRI for colorectal cancer patients.

How is the machine learning model built and evaluated?

The model is trained on GEO gene expression data using 5-fold cross-validation, LASSO and VarSelRF feature selection, and classifiers such as random forest and support vector machines.

What performance results are reported for predicting chemotherapy efficacy?

Validation and test sets correctly classify samples at greater than 90% accuracy, with specificity and sensitivity ranging from 85% to 95%.

Aug 4, 2026

Upload by Ava Thompson