Predicting atezolizumab response in metastatic urothelial carcinoma patients using machine learning on integrated tumour gene expression and clinical data

Abstract Atezolizumab is a treatment for metastatic urothelial carcinoma (mUC), yet only 23% of mUC patients benefit from it. Worse yet, accurately predicting such responders remains challenging, despite existing biomarkers. Here we employed eight machine learning (ML) algorithms to predict mUC pati...

Full description

Saved in:
Bibliographic Details
Main Authors: Chayanit Piyawajanusorn, Ghita Ghislat, Pedro J. Ballester
Format: Article
Language:English
Published: Nature Portfolio 2025-06-01
Series:npj Precision Oncology
Online Access:https://doi.org/10.1038/s41698-025-00969-8
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850222299601436672
author Chayanit Piyawajanusorn
Ghita Ghislat
Pedro J. Ballester
author_facet Chayanit Piyawajanusorn
Ghita Ghislat
Pedro J. Ballester
author_sort Chayanit Piyawajanusorn
collection DOAJ
description Abstract Atezolizumab is a treatment for metastatic urothelial carcinoma (mUC), yet only 23% of mUC patients benefit from it. Worse yet, accurately predicting such responders remains challenging, despite existing biomarkers. Here we employed eight machine learning (ML) algorithms to predict mUC patient response to atezolizumab using tumours’ gene expression profiling and clinical data from two independent cohorts. The CART-OMC model developed on the discovery dataset achieved the highest performance, with a validation set Matthews correlation coefficient (MCC) of 0.437, using the expressions of just 29 ML-selected genes, including CXCL9 and IFNG. Univariate biomarkers like TMB, TNB, and PD-L1 were less predictive with MCCs of 0, 0.316, and 0, respectively. Upon merging these datasets, the best-performing model (LGBM-OMC; MCC of 0.252) also outperformed top modelling approaches such as EaSIeR (MCC ~ 0) and JADBio (MCC of 0.179). We make these promising ML models freely available to predict atezolizumab response in other mUC patients.
format Article
id doaj-art-049298ce6b294ae4bbc6a4511eb83df9
institution OA Journals
issn 2397-768X
language English
publishDate 2025-06-01
publisher Nature Portfolio
record_format Article
series npj Precision Oncology
spelling doaj-art-049298ce6b294ae4bbc6a4511eb83df92025-08-20T02:06:23ZengNature Portfolionpj Precision Oncology2397-768X2025-06-019111410.1038/s41698-025-00969-8Predicting atezolizumab response in metastatic urothelial carcinoma patients using machine learning on integrated tumour gene expression and clinical dataChayanit Piyawajanusorn0Ghita Ghislat1Pedro J. Ballester2Department of Bioengineering, Imperial College LondonDepartment of Bioengineering, Imperial College LondonDepartment of Bioengineering, Imperial College LondonAbstract Atezolizumab is a treatment for metastatic urothelial carcinoma (mUC), yet only 23% of mUC patients benefit from it. Worse yet, accurately predicting such responders remains challenging, despite existing biomarkers. Here we employed eight machine learning (ML) algorithms to predict mUC patient response to atezolizumab using tumours’ gene expression profiling and clinical data from two independent cohorts. The CART-OMC model developed on the discovery dataset achieved the highest performance, with a validation set Matthews correlation coefficient (MCC) of 0.437, using the expressions of just 29 ML-selected genes, including CXCL9 and IFNG. Univariate biomarkers like TMB, TNB, and PD-L1 were less predictive with MCCs of 0, 0.316, and 0, respectively. Upon merging these datasets, the best-performing model (LGBM-OMC; MCC of 0.252) also outperformed top modelling approaches such as EaSIeR (MCC ~ 0) and JADBio (MCC of 0.179). We make these promising ML models freely available to predict atezolizumab response in other mUC patients.https://doi.org/10.1038/s41698-025-00969-8
spellingShingle Chayanit Piyawajanusorn
Ghita Ghislat
Pedro J. Ballester
Predicting atezolizumab response in metastatic urothelial carcinoma patients using machine learning on integrated tumour gene expression and clinical data
npj Precision Oncology
title Predicting atezolizumab response in metastatic urothelial carcinoma patients using machine learning on integrated tumour gene expression and clinical data
title_full Predicting atezolizumab response in metastatic urothelial carcinoma patients using machine learning on integrated tumour gene expression and clinical data
title_fullStr Predicting atezolizumab response in metastatic urothelial carcinoma patients using machine learning on integrated tumour gene expression and clinical data
title_full_unstemmed Predicting atezolizumab response in metastatic urothelial carcinoma patients using machine learning on integrated tumour gene expression and clinical data
title_short Predicting atezolizumab response in metastatic urothelial carcinoma patients using machine learning on integrated tumour gene expression and clinical data
title_sort predicting atezolizumab response in metastatic urothelial carcinoma patients using machine learning on integrated tumour gene expression and clinical data
url https://doi.org/10.1038/s41698-025-00969-8
work_keys_str_mv AT chayanitpiyawajanusorn predictingatezolizumabresponseinmetastaticurothelialcarcinomapatientsusingmachinelearningonintegratedtumourgeneexpressionandclinicaldata
AT ghitaghislat predictingatezolizumabresponseinmetastaticurothelialcarcinomapatientsusingmachinelearningonintegratedtumourgeneexpressionandclinicaldata
AT pedrojballester predictingatezolizumabresponseinmetastaticurothelialcarcinomapatientsusingmachinelearningonintegratedtumourgeneexpressionandclinicaldata