Predicting immunotherapy response in patients with non-small cell lung cancer: superior performance of a positron emission tomography radiomics-based stacking ensemble model.
- 1Department of Nuclear Medicine, The First Affiliated Hospital of Zhejiang Chinese Medical University (Zhejiang Provincial Hospital of Traditional Chinese Medicine), Hangzhou, China.
- 2School of Medical Technology and Information Engineering, Zhejiang Chinese Medical University, Hangzhou, China.
Abstract
The efficacy of immunotherapy among patients with non-small cell lung cancer (NSCLC) varies, and thus reliable predictive biomarkers of response are needed. This study aimed to develop and validate a stacking ensemble learning model based on pretreatment positron emission tomography (PET) radiomics to predict response to immunotherapy in patients with NSCLC. In this retrospective study, 45 patients with NSCLC treated with immunotherapy were enrolled. Pretreatment Digital Imaging and Communications in Medicine-format PET images were preprocessed and normalized. Tumor volumes were segmented using three methods: manual delineation, fixed standardized uptake value (SUV) thresholding with a standardized uptake value of 2.5 (SUV2.5), and adaptive thresholding (40% of the maximum SUV). Radiomics features were extracted, followed by a multistep feature selection process involving mutual information, Spearman correlation, and least absolute shrinkage and selection operator (LASSO) regression. Predictive models constructed included a stacking ensemble model [with k-nearest neighbors (KNN), logistic regression (LR), and random forest (RF) as base learners], a clinical feature-based model, and a deep learning model. Model performance was evaluated on an independent test set according to the area under the curve (AUC), accuracy, precision, recall, and F1 score. On the test set, the stacking ensemble model achieved the highest predictive performance using features from the SUV2.5 segmentation dataset, with an AUC of 0.933 [95% confidence interval (CI): 0.846-0.994]. This significantly outperformed the best single model (LR; AUC=0.896) and a deep learning model (AUC =0.925). The clinical feature-based model showed moderate predictive capability, with extreme gradient boosting (XGBoost) yielding the highest AUC of 0.877 (95% CI: 0.644-1.000). The proposed PET radiomics stacking ensemble model, which leverages features from multiple segmentation strategies, demonstrated superior predictive performance for immunotherapy response in NSCLC compared to single models and other benchmarks. This approach offers an effective strategy for enhancing robustness in small-sample radiomics studies and holds promise as a noninvasive decision-support tool that can aid in personalized immunotherapy stratification.
Related
- Development and validation of a habitat-based radiomics fusion model for predicting immunotherapy response in advanced NSCLC: a multicenter retrospective study.
- Volumetric analysis and MRI radiomics for response assessment in canine glioma immunotherapy: an exploratory feasibility study
- An explainable machine learning model combining intratumoral and peritumoral computed tomography radiomics with maximum standard uptake value for predicting International Association for the Study of Lung Cancer grading and prognosis in stage I pure-solid lung adenocarcinoma.
- Predicting drainage success of peritonsillar abscesses: a radiomics-based machine learning approach using contrast-enhanced computed tomography
- A T-cell exhaustion-related signature for predicting response to immunotherapy among patients with lung adenocarcinoma
BibTeX
@article{zhang2026predicting,
title = {{Predicting immunotherapy response in patients with non-small cell lung cancer: superior performance of a positron emission tomography radiomics-based stacking ensemble model.}},
author = {Lixia Zhang and Zicheng Zeng and Peilin Sheng and Jing Peng and Xiyan Zhang and Caiyun Xu and Jianqing Wang},
journal = {Quantitative imaging in medicine and surgery},
year = {2026},
month = sep,
doi = {10.21037/qims-2026-0758},
url = {https://europepmc.org/article/MED/42997730},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Predicting immunotherapy response in patients with non-small cell lung cancer: superior performance of a positron emission tomography radiomics-based stacking ensemble model.
AU - Lixia Zhang
AU - Zicheng Zeng
AU - Peilin Sheng
AU - Jing Peng
AU - Xiyan Zhang
AU - Caiyun Xu
AU - Jianqing Wang
PY - 2026
DA - 2026/09/29
JO - Quantitative imaging in medicine and surgery
DO - 10.21037/qims-2026-0758
UR - https://europepmc.org/article/MED/42997730
AB - The efficacy of immunotherapy among patients with non-small cell lung cancer (NSCLC) varies, and thus reliable predictive biomarkers of response are needed. This study aimed to develop and validate a stacking ensemble learning model based on pretreatment positron emission tomography (PET) radiomics to predict response to immunotherapy in patients with NSCLC. In this retrospective study, 45 patients with NSCLC treated with immunotherapy were enrolled. Pretreatment Digital Imaging and Communications in Medicine-format PET images were preprocessed and normalized. Tumor volumes were segmented using three methods: manual delineation, fixed standardized uptake value (SUV) thresholding with a standardized uptake value of 2.5 (SUV2.5), and adaptive thresholding (40% of the maximum SUV). Radiomics features were extracted, followed by a multistep feature selection process involving mutual information, Spearman correlation, and least absolute shrinkage and selection operator (LASSO) regression. Predictive models constructed included a stacking ensemble model [with k-nearest neighbors (KNN), logistic regression (LR), and random forest (RF) as base learners], a clinical feature-based model, and a deep learning model. Model performance was evaluated on an independent test set according to the area under the curve (AUC), accuracy, precision, recall, and F1 score. On the test set, the stacking ensemble model achieved the highest predictive performance using features from the SUV2.5 segmentation dataset, with an AUC of 0.933 [95% confidence interval (CI): 0.846-0.994]. This significantly outperformed the best single model (LR; AUC=0.896) and a deep learning model (AUC =0.925). The clinical feature-based model showed moderate predictive capability, with extreme gradient boosting (XGBoost) yielding the highest AUC of 0.877 (95% CI: 0.644-1.000). The proposed PET radiomics stacking ensemble model, which leverages features from multiple segmentation strategies, demonstrated superior predictive performance for immunotherapy response in NSCLC compared to single models and other benchmarks. This approach offers an effective strategy for enhancing robustness in small-sample radiomics studies and holds promise as a noninvasive decision-support tool that can aid in personalized immunotherapy stratification.
SN - 2223-4292
SN - 2223-4306
ER -