Ensemble machine learning for predicting breast cancer recurrence and mortality using clinical and hemogram data
8 affiliations
- 1Instituto de Tecnologia e Liderança, São Paulo, SP Brazil
- 2Center for Translational Research in Oncology (LIM24), Instituto do Câncer do Estado de São Paulo, Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo, São Paulo, Brazil
- 3Comprehensive Center for Precision Oncology, Universidade de São Paulo, São Paulo, Brazil
- 4Serviço de Hematologia, Hemoterapia e Terapia Celular and Laboratório de Investigação Médica (LIM) 31, Hospital das Clínicas HCFMUSP, Faculdade de Medicina, Universidade de São Paulo, São Paulo, Brazil
- 5Instituto do Câncer do Estado de São Paulo, São Paulo, SP Brazil
- 6Instituto D’Or de Pesquisa e Ensino (IDOR), São Paulo, SP Brazil
- 7Faculdade de Informática e Administração Paulista, São Paulo, SP Brazil
- 8Escola de Artes, Ciências e Humanidades, Universidade de São Paulo, São Paulo, Brazil
Abstract
Accurate prediction of breast cancer recurrence remains difficult because prognosis varies significantly across molecular subtypes. This underscores the need for scalable, affordable, and interpretable prognostic tools. We developed machine learning models integrating hematological indices with clinicopathologic data to predict 2- and 10-year recurrence or all-case death. We retrospectively analyzed 4277 women with primary breast cancer (2008–2022). The cohort included hormone receptor-positive (HR + ), HER2-positive, and triple-negative (TNBC) subtypes. We trained multiple classifiers and integrated them into a stacked ensemble using logistic regression as the final learner. Class imbalance was addressed with SMOTE applied to training sets. The ensemble achieved strong discrimination: general cohort AUC 0.859 (2-year) and 0.811 (10-year), with specificity 84–82% and sensitivity 68–61%. Subtype-specific performance remained robust across time horizons: HR + AUC 0.862/0.804, HER2 + AUC 0.877/0.831, and TNBC AUC 0.826/0.826 (2-year/10-year respectively). SHAP analysis identified advanced tumor stage, elevated inflammatory ratios (NLR, PLR, MLR), elevated red cell distribution width, and age as key adverse predictors with stronger effects on early recurrence. This interpretable tool uses routine blood tests and clinico-pathological data at diagnosis. It flaggs high-risk patients for intensified therapy, useful where genomic testing is unavailable.
Related
- Transformer-Based Deep Learning Integrating Multi-Regional MRI and Clinical Features for Predicting Biochemical Recurrence After Radical Prostatectomy: A Multicenter Study.
- Ensemble SHAP Aggregation and Attribution Variability in Clinical Machine Learning: A COVID-19 Mortality Study
- Machine learning in predicting recurrence of self-inflicted neck trauma in young adults
- Multi-Scale MRI, Radiology Reports, and Blood Biomarker–Guided Multimodal Deep Learning for Predicting Postoperative Recurrence in Cervical Cancer: A Multicenter Study
- Explainable machine learning and deep learning models for predicting in-hospital mortality in burn patients
BibTeX
@article{moreira2026ensemble,
title = {{Ensemble machine learning for predicting breast cancer recurrence and mortality using clinical and hemogram data}},
author = {Patricia Honorato Moreira and Arthur Shuzo Owtake Cardoso and Rafael de Oliveira and Joaquim Gasparini dos Santos and Renata Colombo Bonadio and Wesley Antonio Lopes de Lima and Thomas de Almeida Reichmann and Bruna Salani Mota and Alexandre Ferreira Ramos and Flavia Santoro and Roger Chammas and Luciana Rodrigues Carvalho Barros},
journal = {NPJ Breast Cancer},
year = {2026},
month = jul,
doi = {10.1038/s41523-026-01005-z},
url = {https://doi.org/10.1038/s41523-026-01005-z},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - Ensemble machine learning for predicting breast cancer recurrence and mortality using clinical and hemogram data
AU - Patricia Honorato Moreira
AU - Arthur Shuzo Owtake Cardoso
AU - Rafael de Oliveira
AU - Joaquim Gasparini dos Santos
AU - Renata Colombo Bonadio
AU - Wesley Antonio Lopes de Lima
AU - Thomas de Almeida Reichmann
AU - Bruna Salani Mota
AU - Alexandre Ferreira Ramos
AU - Flavia Santoro
AU - Roger Chammas
AU - Luciana Rodrigues Carvalho Barros
PY - 2026
DA - 2026/07/11
JO - NPJ Breast Cancer
DO - 10.1038/s41523-026-01005-z
UR - https://doi.org/10.1038/s41523-026-01005-z
AB - Accurate prediction of breast cancer recurrence remains difficult because prognosis varies significantly across molecular subtypes. This underscores the need for scalable, affordable, and interpretable prognostic tools. We developed machine learning models integrating hematological indices with clinicopathologic data to predict 2- and 10-year recurrence or all-case death. We retrospectively analyzed 4277 women with primary breast cancer (2008–2022). The cohort included hormone receptor-positive (HR + ), HER2-positive, and triple-negative (TNBC) subtypes. We trained multiple classifiers and integrated them into a stacked ensemble using logistic regression as the final learner. Class imbalance was addressed with SMOTE applied to training sets. The ensemble achieved strong discrimination: general cohort AUC 0.859 (2-year) and 0.811 (10-year), with specificity 84–82% and sensitivity 68–61%. Subtype-specific performance remained robust across time horizons: HR + AUC 0.862/0.804, HER2 + AUC 0.877/0.831, and TNBC AUC 0.826/0.826 (2-year/10-year respectively). SHAP analysis identified advanced tumor stage, elevated inflammatory ratios (NLR, PLR, MLR), elevated red cell distribution width, and age as key adverse predictors with stronger effects on early recurrence. This interpretable tool uses routine blood tests and clinico-pathological data at diagnosis. It flaggs high-risk patients for intensified therapy, useful where genomic testing is unavailable.
SN - 2374-4677
ER -