Abstract
Adverse drug reactions (ADRs) are a major public health concern, and early detection is crucial for drug development and patient safety. Together with the increasing availability of large-scale literature data, machine learning has the potential to predict unknown ADRs from current knowledge. By the machine learning methods, we constructed a Tumor-Biomarker Knowledge Graph (TBKG) which contains four types of node: Tumor, Biomarker, Drug, and ADR using biomedical literatures. Based on this knowledge graph, we not only discovered potential ADRs of antitumor drugs but also provided explanations. Experiments on real-world data show that our model can achieve 0.81 accuracy of three cross-validation and the ADRs discovery of Osimertinib was chosen for the clinical validation. Calculated ADRs of Osimertinib by our model consisted of the known ADRs which were in line with the official manual and some unreported rare ADRs in clinical cases. Results also showed that our model outperformed traditional co-occurrence methods. Moreover, each calculated ADRs were attached with the corresponding paths of “tumor-biomarker-drug” in the knowledge graph which could help to obtain in-depth insights into the underlying mechanisms. In conclusion, the tumor-biomarker knowledge-graph based approach is an explainable method for potential ADRs discovery based on biomarkers and might be valuable to the community working on the emerging field of biomedical literature mining and provide impetus for the mechanism research of ADRs.
Highlights
Adverse drug reactions (ADRs) are a cause of significant morbidity and mortality in patients and a source of financial burden in the healthcare system (Patton and Borshoff, 2018)
We constructed the Tumor-Biomarker Knowledge Graph (TBKG), which is a weighted heterogeneous graph with four types of objects extracted from the MEDLINE corpus: tumor, biomarker, drug, and ADR
We use the biomarkers and ADRs related to the drug “osimertinib” as an example to show the TBKG results
Summary
Adverse drug reactions (ADRs) are a cause of significant morbidity and mortality in patients and a source of financial burden in the healthcare system (Patton and Borshoff, 2018). Pharmacokinetic parameters can be altered by the disease itself, or hepatic or renal impairment, or reduction of serum-binding proteins due to malnutrition. They experience a relatively high rate of ADRs from antitumor drugs and more experience rare and severe ADRs, which could seriously impact the quality of life (Shrestha et al, 2017). There are still two challenges to achieving good performance: (1) the unstructured biomedical literature contains many irrelevant words and contexts, and how to extract ADR-related entities and fully explore their relations (e.g., tumor-biomarker-drug) is difficult; and (2) some predicted unseen ADRs are unexpected and cause confusion, which means explainability and validation become critically important for automatic detection
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
