Building Hybrid Pharmacometric‐Machine Learning Models in Oncology Drug Development: Current State and Recommendations
CPT: Pharmacometrics & Systems Pharmacology · 2026
DOI: 10.1002/psp4.70113 ↗Kurz gefasst
Übersichtsarbeit in der Zeitschrift CPT: Pharmacometrics & Systems Pharmacology, 2026. Verfasst von Anna Fochesato und 6 weiteren Autor:innen. Beteiligte Einrichtung: Translational PKPD and Clinical Pharmacology, Roche Pharma Research and Early Development, Roche Innovation Center Basel, Basel, Switzerland. Hauptthemen laut MEDLINE: Machine Learning, Drug Development, Antineoplastic Agents, Neoplasms. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 86 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: F. Hoffmann-La Roche Ltd, Roche.
Abstract (Originalsprache)
Classic and hybrid pharmacometric‐machine learning models (hPMxML) are gaining momentum for applications in clinical drug development and precision medicine, especially within the oncology therapeutic area. However, standardized workflows are needed to ensure transparency, rigor, and effective communication for broader adoption. In this tutorial, we review pharmacometric (PMx) and machine learning (ML) reporting standards and evaluate them against hPMxML works in oncology contexts as a motivational example to identify current deficiencies and propose mitigation strategies for future efforts. The revealed gaps include insufficient benchmarking, absence of error propagation, feature stability assessments, and ablation studies, limited focus on external validation and final parametrization, and discrepancies between the performance metrics chosen and the original clinical questions. To address these, we propose a checklist for hPMxML model development and reporting, consisting of steps for estimand definition, data curation, covariate selection, hyperparameter tuning, convergence assessment, model explainability, diagnostics, uncertainty quantification, validation and verification with sensitivity analyses. This standardized approach is expected to enhance the reliability and reproducibility of hPMxML outputs, enabling their confident application in oncology clinical drug development, while fostering trust among all stakeholders.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Translational PKPD and Clinical Pharmacology, Roche Pharma Research and Early Development, Roche Innovation Center Basel, Basel, Switzerland.
- Clinical Pharmacology, Genentech, Inc., South San Francisco, California, USA.
- Predictive Modeling, Roche Pharma Research and Early Development, Roche Innovation Center Basel, Basel, Switzerland.
- Genentech Research and Early Development, Roche Innovation Center Basel, Basel, Switzerland.
ORCID-Profile: James Lu, Francois Mercier
Förderung
- F. Hoffmann-La Roche Ltd
- Roche
Zitieren
APA
Fochesato, A., Brooks, L., Bazgir, O., Pierrillas, P. B., Jamois, C., Lu, J., & Mercier, F. (2026). Building Hybrid Pharmacometric‐Machine Learning Models in Oncology Drug Development: Current State and Recommendations. CPT: Pharmacometrics & Systems Pharmacology. https://doi.org/10.1002/psp4.70113
Vancouver
Fochesato A, Brooks L, Bazgir O, Pierrillas PB, Jamois C, Lu J, et al. Building Hybrid Pharmacometric‐Machine Learning Models in Oncology Drug Development: Current State and Recommendations. CPT: Pharmacometrics & Systems Pharmacology. 2026. doi:10.1002/psp4.70113
BibTeX
@article{Fochesato2026,
author = {Fochesato, Anna and Brooks, Logan and Bazgir, Omid and Pierrillas, Philippe B and Jamois, Candice and Lu, James and Mercier, Francois},
title = {Building Hybrid Pharmacometric‐Machine Learning Models in Oncology Drug Development: Current State and Recommendations},
journal = {CPT: Pharmacometrics & Systems Pharmacology},
year = {2026},
doi = {10.1002/psp4.70113}
}
EUVIMED