Machine Learning Model Integrating CT Radiomics of the Lung to Predict Checkpoint Inhibitor Pneumonitis in Patients with Advanced Cancer

Objective Checkpoint inhibitor pneumonitis (CIP) is a potentially life-threatening immune-related adverse event. Efficient strategies to select patients at risk are still required. The aim of our study was to assess the utility of a machine learning model, integrating pre-treatment CT lung radiomics...

Full description

Saved in:
Bibliographic Details
Main Authors: François Cousin MD, PhD, Thomas Louis MS, PhD, Pierre Frères MD, PhD, Julien Guiot MD, PhD, Mariaelena Occhipinti MD, Fabio Bottari MS, Wim Vos MS, Roland Hustinx MD, PhD
Format: Article
Language:English
Published: SAGE Publishing 2025-05-01
Series:Technology in Cancer Research & Treatment
Online Access:https://doi.org/10.1177/15330338251344004
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850152140599721984
author François Cousin MD, PhD
Thomas Louis MS, PhD
Pierre Frères MD, PhD
Julien Guiot MD, PhD
Mariaelena Occhipinti MD
Fabio Bottari MS
Wim Vos MS
Roland Hustinx MD, PhD
author_facet François Cousin MD, PhD
Thomas Louis MS, PhD
Pierre Frères MD, PhD
Julien Guiot MD, PhD
Mariaelena Occhipinti MD
Fabio Bottari MS
Wim Vos MS
Roland Hustinx MD, PhD
author_sort François Cousin MD, PhD
collection DOAJ
description Objective Checkpoint inhibitor pneumonitis (CIP) is a potentially life-threatening immune-related adverse event. Efficient strategies to select patients at risk are still required. The aim of our study was to assess the utility of a machine learning model, integrating pre-treatment CT lung radiomics features with clinical data, to predict patients at risk of developing CIP. Methods In this retrospective study, 116 patients with varied malignancies treated with immune checkpoint inhibitors (ICIs) were included. In this cohort, 35 patients presented with CIP and 81 patients did not. Each lung and its lobes were segmented on pre-treatment CT scans to perform a handcrafted radiomic analysis. Radiomic features were associated with clinical parameters to build generalized linear (GLM) and random forest (RF) models, to predict occurrence of CIP. The models were fine-tuned, validated and tested using a nested 5-fold cross-validation method. Results The RF models combining radiomic and clinical features showed the best performances with an area under the ROC curve (AUC) of 0.75 (95%CI:0.62-0.88) on the test set. The most accurate clinical model was a RF model and achieved an AUC of 0.72 (95%CI:0.51-0.92). The best radiomic model was a GLM model and achieved an AUC of 0.71 (95%CI:0.58-0.84). Conclusions Our CT-based lung radiomic models showed moderate to good performance at predicting CIP. We demonstrated the potential role of machine learning models associating clinical parameters and lung CT radiomic features to better identify patients treated with ICIs at risk of developing CIP. Advances in knowledge: Radiomics analysis of the lung parenchyma could be used as a non-invasive tool to select patients at risk of developing immune-checkpoint pneumonitis.
format Article
id doaj-art-7976a38d4f964545848c7ed3ce22d039
institution OA Journals
issn 1533-0338
language English
publishDate 2025-05-01
publisher SAGE Publishing
record_format Article
series Technology in Cancer Research & Treatment
spelling doaj-art-7976a38d4f964545848c7ed3ce22d0392025-08-20T02:26:03ZengSAGE PublishingTechnology in Cancer Research & Treatment1533-03382025-05-012410.1177/15330338251344004Machine Learning Model Integrating CT Radiomics of the Lung to Predict Checkpoint Inhibitor Pneumonitis in Patients with Advanced CancerFrançois Cousin MD, PhD0Thomas Louis MS, PhD1Pierre Frères MD, PhD2Julien Guiot MD, PhD3Mariaelena Occhipinti MD4Fabio Bottari MS5Wim Vos MS6Roland Hustinx MD, PhD7 Department of Nuclear Medicine and Oncological Imaging, University Hospital (CHU) of Liège, Liège, Belgium Radiomics (Oncoradiomics SA), Liège, Belgium Department of Medical Oncology, University Hospital (CHU) of Liège, Liège, Belgium Department of Respiratory Medicine, University Hospital (CHU) of Liège, Liège, Belgium Radiomics (Oncoradiomics SA), Liège, Belgium Radiomics (Oncoradiomics SA), Liège, Belgium Radiomics (Oncoradiomics SA), Liège, Belgium GIGA-CRC in vivo imaging, University of Liège, Liège, BelgiumObjective Checkpoint inhibitor pneumonitis (CIP) is a potentially life-threatening immune-related adverse event. Efficient strategies to select patients at risk are still required. The aim of our study was to assess the utility of a machine learning model, integrating pre-treatment CT lung radiomics features with clinical data, to predict patients at risk of developing CIP. Methods In this retrospective study, 116 patients with varied malignancies treated with immune checkpoint inhibitors (ICIs) were included. In this cohort, 35 patients presented with CIP and 81 patients did not. Each lung and its lobes were segmented on pre-treatment CT scans to perform a handcrafted radiomic analysis. Radiomic features were associated with clinical parameters to build generalized linear (GLM) and random forest (RF) models, to predict occurrence of CIP. The models were fine-tuned, validated and tested using a nested 5-fold cross-validation method. Results The RF models combining radiomic and clinical features showed the best performances with an area under the ROC curve (AUC) of 0.75 (95%CI:0.62-0.88) on the test set. The most accurate clinical model was a RF model and achieved an AUC of 0.72 (95%CI:0.51-0.92). The best radiomic model was a GLM model and achieved an AUC of 0.71 (95%CI:0.58-0.84). Conclusions Our CT-based lung radiomic models showed moderate to good performance at predicting CIP. We demonstrated the potential role of machine learning models associating clinical parameters and lung CT radiomic features to better identify patients treated with ICIs at risk of developing CIP. Advances in knowledge: Radiomics analysis of the lung parenchyma could be used as a non-invasive tool to select patients at risk of developing immune-checkpoint pneumonitis.https://doi.org/10.1177/15330338251344004
spellingShingle François Cousin MD, PhD
Thomas Louis MS, PhD
Pierre Frères MD, PhD
Julien Guiot MD, PhD
Mariaelena Occhipinti MD
Fabio Bottari MS
Wim Vos MS
Roland Hustinx MD, PhD
Machine Learning Model Integrating CT Radiomics of the Lung to Predict Checkpoint Inhibitor Pneumonitis in Patients with Advanced Cancer
Technology in Cancer Research & Treatment
title Machine Learning Model Integrating CT Radiomics of the Lung to Predict Checkpoint Inhibitor Pneumonitis in Patients with Advanced Cancer
title_full Machine Learning Model Integrating CT Radiomics of the Lung to Predict Checkpoint Inhibitor Pneumonitis in Patients with Advanced Cancer
title_fullStr Machine Learning Model Integrating CT Radiomics of the Lung to Predict Checkpoint Inhibitor Pneumonitis in Patients with Advanced Cancer
title_full_unstemmed Machine Learning Model Integrating CT Radiomics of the Lung to Predict Checkpoint Inhibitor Pneumonitis in Patients with Advanced Cancer
title_short Machine Learning Model Integrating CT Radiomics of the Lung to Predict Checkpoint Inhibitor Pneumonitis in Patients with Advanced Cancer
title_sort machine learning model integrating ct radiomics of the lung to predict checkpoint inhibitor pneumonitis in patients with advanced cancer
url https://doi.org/10.1177/15330338251344004
work_keys_str_mv AT francoiscousinmdphd machinelearningmodelintegratingctradiomicsofthelungtopredictcheckpointinhibitorpneumonitisinpatientswithadvancedcancer
AT thomaslouismsphd machinelearningmodelintegratingctradiomicsofthelungtopredictcheckpointinhibitorpneumonitisinpatientswithadvancedcancer
AT pierrefreresmdphd machinelearningmodelintegratingctradiomicsofthelungtopredictcheckpointinhibitorpneumonitisinpatientswithadvancedcancer
AT julienguiotmdphd machinelearningmodelintegratingctradiomicsofthelungtopredictcheckpointinhibitorpneumonitisinpatientswithadvancedcancer
AT mariaelenaocchipintimd machinelearningmodelintegratingctradiomicsofthelungtopredictcheckpointinhibitorpneumonitisinpatientswithadvancedcancer
AT fabiobottarims machinelearningmodelintegratingctradiomicsofthelungtopredictcheckpointinhibitorpneumonitisinpatientswithadvancedcancer
AT wimvosms machinelearningmodelintegratingctradiomicsofthelungtopredictcheckpointinhibitorpneumonitisinpatientswithadvancedcancer
AT rolandhustinxmdphd machinelearningmodelintegratingctradiomicsofthelungtopredictcheckpointinhibitorpneumonitisinpatientswithadvancedcancer