An interpretable machine learning model for predicting response to anti-GD2 combined with umbilical cord blood-derived natural killer cell immunotherapy in neuroblastoma
6 affiliations
- 1School of Electronics and Information Engineering, Wuyi University
- 2State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Collaborative Innovation Center for Cancer Medicine, Department of Pediatric Oncology, Sun Yat-sen University Cancer Center
- 3State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Collaborative Innovation Center for Cancer Medicine, Department of Anesthesiology, Sun Yat-sen University Cancer Center
- 4Institute of Intelligent Manufacturing, Guangdong Academy of Sciences
- 5State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Collaborative Innovation Center for Cancer Medicine, Department of Experimental Research, Sun Yat-sen University Cancer Center
- 6NMPA Center for Innovation and Research in Regulatory Science
Abstract
Background: Neuroblastoma remains a major therapeutic challenge for children with high-risk or relapsed/refractory disease. Anti-GD2 therapy combined with umbilical cord blood-derived natural killer (UCB-NK) cell therapy represents an emerging strategy, but responses vary substantially. Imaging and bone marrow assessments require specialized resources, creating a need for complementary tools based on routinely available data. However, whether serial routine peripheral blood measurements can support response prediction in this treatment setting remains unclear.
Methods: Serial peripheral blood measurements from 42 patients were used to derive dynamic coupling features that reflected longitudinal trends and coordinated changes between biomarkers. These features were integrated with static variables to develop machine learning models for response prediction. Seven machine learning models were compared using the same repeated stratified cross-validation framework. Model selection considered discrimination, probability-estimation quality, and interpretability. SHapley Additive exPlanations (SHAP) was used to interpret the final model.
Results: Considering cross-validated performance, the quality of probability estimates, and clinical interpretability, L2-regularized logistic regression was selected as the final model for response prediction. The model leveraged dynamic peripheral blood features to predict response to anti-GD2 combined with UCB-NK cell immunotherapy, achieving a cross-validated AUC of 0.873 (95% bootstrap CI, 0.735–0.976) and a Brier score of 0.131 (95% bootstrap CI, 0.087–0.182). SHAP analysis characterized the direction and magnitude of feature contributions to model predictions.
Conclusion: This study developed an interpretable response-prediction model using serial routine peripheral blood tests obtained in clinical practice. The model may complement treatment-response assessment using routinely available data. However, validation in larger, independent cohorts is required before clinical use.
Related
- Predicting student academic performance using TabPFN and SHAP: An interpretable machine learning approach
- Immunotherapy with a posttranscriptionally modified DNA vaccine induces complete protection against metastatic neuroblastoma
- Interpretable metaheuristic-optimized machine learning for predicting biodiversity indices in the Hyrcanian forest
- Interpretable Machine Learning Links Group-Specific Plasma Proteomics to Regional Structural Neuroimaging in Alzheimer's Disease
- Learning interpretable kinetic models for biomolecular interaction networks
BibTeX
@article{dong2026interpretable,
title = {{An interpretable machine learning model for predicting response to anti-GD2 combined with umbilical cord blood-derived natural killer cell immunotherapy in neuroblastoma}},
author = {Qiqing Dong and Mengjia Song and Yumiao Mai and Juan Wang and Jia Zhu and Suying Lu and Feifei Sun and Junting Huang and Zijun Zhen and Xiya Tan and Yu Zhang and Yi Que and Yingxia Lan and Yongjun Cao and Fuqin Deng and Hao Qin and Yizhuo Zhang and Tong Xiang},
journal = {Frontiers in Immunology},
year = {2026},
month = oct,
doi = {10.3389/fimmu.2026.1940534},
url = {https://www.frontiersin.org/articles/10.3389/fimmu.2026.1940534/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - An interpretable machine learning model for predicting response to anti-GD2 combined with umbilical cord blood-derived natural killer cell immunotherapy in neuroblastoma
AU - Qiqing Dong
AU - Mengjia Song
AU - Yumiao Mai
AU - Juan Wang
AU - Jia Zhu
AU - Suying Lu
AU - Feifei Sun
AU - Junting Huang
AU - Zijun Zhen
AU - Xiya Tan
AU - Yu Zhang
AU - Yi Que
AU - Yingxia Lan
AU - Yongjun Cao
AU - Fuqin Deng
AU - Hao Qin
AU - Yizhuo Zhang
AU - Tong Xiang
PY - 2026
DA - 2026/10/29
JO - Frontiers in Immunology
DO - 10.3389/fimmu.2026.1940534
UR - https://www.frontiersin.org/articles/10.3389/fimmu.2026.1940534/full
AB - Background: Neuroblastoma remains a major therapeutic challenge for children with high-risk or relapsed/refractory disease. Anti-GD2 therapy combined with umbilical cord blood-derived natural killer (UCB-NK) cell therapy represents an emerging strategy, but responses vary substantially. Imaging and bone marrow assessments require specialized resources, creating a need for complementary tools based on routinely available data. However, whether serial routine peripheral blood measurements can support response prediction in this treatment setting remains unclear. Methods: Serial peripheral blood measurements from 42 patients were used to derive dynamic coupling features that reflected longitudinal trends and coordinated changes between biomarkers. These features were integrated with static variables to develop machine learning models for response prediction. Seven machine learning models were compared using the same repeated stratified cross-validation framework. Model selection considered discrimination, probability-estimation quality, and interpretability. SHapley Additive exPlanations (SHAP) was used to interpret the final model. Results: Considering cross-validated performance, the quality of probability estimates, and clinical interpretability, L2-regularized logistic regression was selected as the final model for response prediction. The model leveraged dynamic peripheral blood features to predict response to anti-GD2 combined with UCB-NK cell immunotherapy, achieving a cross-validated AUC of 0.873 (95% bootstrap CI, 0.735–0.976) and a Brier score of 0.131 (95% bootstrap CI, 0.087–0.182). SHAP analysis characterized the direction and magnitude of feature contributions to model predictions. Conclusion: This study developed an interpretable response-prediction model using serial routine peripheral blood tests obtained in clinical practice. The model may complement treatment-response assessment using routinely available data. However, validation in larger, independent cohorts is required before clinical use.
SN - 1664-3224
ER -