Transforming animal study toxicology reports into structured, harmonized data using large language models
Archives of Toxicology · 2026-09 · 100 (9) · 4105-4134
DOI: 10.1007/s00204-026-04414-y ↗Kurz gefasst
Fachartikel in der Zeitschrift Archives of Toxicology, online erschienen am 14.05.2026. Verfasst von Tatyana Y Doktorova und 15 weiteren Autor:innen. Beteiligte Einrichtung: Predictive Modelling, pRED, Roche, Basel, Switzerland. tatyana.doktorova@roche.com. Hauptthemen laut MEDLINE: Toxicology. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 11 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
Preclinical toxicology study reports contain the expert interpretations required to distinguish test article-related effects from incidental findings, yet these conclusions often remain embedded in unstructured text form that limit systematic reuse and integration with computational safety approaches. To address this gap, we developed a Large Language Model (LLM) - supported pipeline that converts toxicology reports into structured, machine-readable datasets harmonized with SEND terminology. The pipeline combines automated document preprocessing, section identification, schema-constrained information extraction, and semantic harmonization, complemented by targeted human curation. We evaluated the system performance using 200 Roche toxicology study reports, encompassing clinical pathology, histopathology, organ weights, exposure data, and study-level conclusions. Across domains, extraction performance was strong, characterized by consistently high sensitivity and precision for most parameters. Histopathology, organ weight, and NOAEL-related endpoints demonstrated the greatest robustness, with sensitivity typically above 95% and precision frequently exceeding 97%. Lower performance for parameters such as route of administration and substance identifiers reflected heterogeneous reporting practices rather than LLM-based method limitations. The structured datasets generated by this pipeline enable cross-study querying, identification of compounds with defined toxicological liabilities, integration with raw SEND data, and development of high-quality labels for predictive toxicology models. Practical utility has been demonstrated through representative use cases. These results demonstrate that LLM-assisted extraction can reliably capture expert toxicological interpretations at scale and provide a foundation for data-centric safety assessment, strategic decisions, reverse and forward translational toxicology research.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Predictive Modelling, pRED, Roche, Basel, Switzerland. tatyana.doktorova@roche.com.
- Predictive Modelling, pRED, Roche, Basel, Switzerland.
- Computational Sciences Center of Excellence, gRED, Roche, Basel, Switzerland.
- Computational Toxicology, Translational Safety, Genentech Inc., South San Francisco, CA, USA.
- PASS, pRED, Roche, Basel, Switzerland.
- TSAC, pRED, Roche, Basel, Switzerland.
ORCID-Profile: Tatyana Y Doktorova
Zitieren
APA
Doktorova, T. Y., Chernov, I. S., Formaggio, A., Zheng, X., Serra, A., Mohan, P. L. B., Duchateau-Nguyen, G., Draganov, D. I., Hayes, M., Hasselgren, C., Siramshetty, V. B., Gaudio, S., Tecilla, M., Maliver, P., Musvasva, E., & Anger, L. T. (2026). Transforming animal study toxicology reports into structured, harmonized data using large language models. Archives of Toxicology, 100(9), 4105-4134. https://doi.org/10.1007/s00204-026-04414-y
Vancouver
Doktorova TY, Chernov IS, Formaggio A, Zheng X, Serra A, Mohan PLB, et al. Transforming animal study toxicology reports into structured, harmonized data using large language models. Archives of Toxicology. 2026;100(9):4105-4134. doi:10.1007/s00204-026-04414-y
BibTeX
@article{Doktorova2026,
author = {Doktorova, Tatyana Y and Chernov, Ilya Schneider and Formaggio, Alberto and Zheng, Xiaochen and Serra, Alvaro and Mohan, Preethi Latha Bhavani and Duchateau-Nguyen, Guillemette and Draganov, Dragomir Ivanov and Hayes, Matt and Hasselgren, Catrin and Siramshetty, Vishal B and Gaudio, Stefano and Tecilla, Marco and Maliver, Pierre and Musvasva, Eunice and Anger, Lennart T},
title = {Transforming animal study toxicology reports into structured, harmonized data using large language models},
journal = {Archives of Toxicology},
year = {2026},
volume = {100},
number = {9},
pages = {4105--4134},
doi = {10.1007/s00204-026-04414-y}
}
EUVIMED