Enhancing Severe Neutropenia Prediction: PKPD ‐Informed Labeling for Machine Learning Models Trained on Real‐World Data
Clinical Pharmacology & Therapeutics · 2026
DOI: 10.1002/cpt.70116 ↗Kurz gefasst
Fachartikel in der Zeitschrift Clinical Pharmacology & Therapeutics, 2026. Verfasst von Conor J O'Hanlon und 9 weiteren Autor:innen. Beteiligte Einrichtung: Roche Informatics, F. Hoffmann-La Roche AG, Kaiseraugst, Switzerland. Hauptthemen laut MEDLINE: Machine Learning, Neutropenia, Docetaxel, Antineoplastic Agents. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 20 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
Accurately labeling outcomes in real‐world data for machine learning is challenging due to data sparsity and imbalances. This study developed and evaluated a pharmacokinetic–pharmacodynamic (PKPD)‐informed labeling strategy to enhance the risk prediction of docetaxel‐induced neutropenia. Machine learning models were trained on real‐world data from 4,248 patients using two approaches for comparison. The “naive” labeling method used only neutrophil observations, while the “PKPD‐informed” method used simulations from a semi‐mechanistic model to determine the neutrophil nadir for each treatment cycle. Three machine learning models (logistic regression, XGBoost, TabPFN) were trained with baseline laboratory data to predict severe neutropenia (neutrophil count
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Roche Informatics, F. Hoffmann-La Roche AG, Kaiseraugst, Switzerland.
- Roche Pharmaceutical Research and Early Development, Roche Innovation Center, Basel, Switzerland.
- iCAN Digital Precision Cancer Medicine Flagship, University of Helsinki and Helsinki University Hospital Comprehensive Cancer Center, Helsinki, Finland.
- Hematoscope Lab, Comprehensive Cancer Center, Helsinki University Hospital, Helsinki, Finland.
- Department of Oncology, University of Helsinki, Helsinki, Finland.
- Department of Clinical Chemistry, HUS Diagnostic Center, Helsinki University Hospital, Helsinki, Finland.
ORCID-Profile: Conor J O'Hanlon, Jonas Denck, Elif Ozkirimli, Candice Jamois, Oscar Brück, Ken Wang
Zitieren
APA
O'Hanlon, C. J., Denck, J., Ozkirimli, E., Bendels, S., Jamois, C., Chavanne, C., Fey, D., Porkka, K., Brück, O., & Wang, K. (2026). Enhancing Severe Neutropenia Prediction: PKPD ‐Informed Labeling for Machine Learning Models Trained on Real‐World Data. Clinical Pharmacology & Therapeutics. https://doi.org/10.1002/cpt.70116
Vancouver
O'Hanlon CJ, Denck J, Ozkirimli E, Bendels S, Jamois C, Chavanne C, et al. Enhancing Severe Neutropenia Prediction: PKPD ‐Informed Labeling for Machine Learning Models Trained on Real‐World Data. Clinical Pharmacology & Therapeutics. 2026. doi:10.1002/cpt.70116
BibTeX
@article{OHanlon2026,
author = {O'Hanlon, Conor J and Denck, Jonas and Ozkirimli, Elif and Bendels, Stefanie and Jamois, Candice and Chavanne, Clarisse and Fey, Dirk and Porkka, Kimmo and Brück, Oscar and Wang, Ken},
title = {Enhancing Severe Neutropenia Prediction: PKPD ‐Informed Labeling for Machine Learning Models Trained on Real‐World Data},
journal = {Clinical Pharmacology & Therapeutics},
year = {2026},
doi = {10.1002/cpt.70116}
}
EUVIMED