Advancing drug safety science by integrating molecular knowledge with post‐marketing adverse event reports

Abstract

Promising drug development efforts may frequently fail due to unintended adverse reactions. Several methods have been developed to analyze such data, aiming to improve pharmacovigilance and drug safety. In this work, we provide a brief review of key directions to quantitatively analyzing adverse events and explore the potential of augmenting these methods using additional molecular data descriptors. We argue that molecular expansion of adverse event data may provide a path to improving the insights gained through more traditional pharmacovigilance approaches. Examples include the ability to assess statistical relevance with respect to underlying biomolecular mechanisms, the ability to generate plausible causative hypotheses and/or confirmation where possible, the ability to computationally study potential clinical trial designs and/or results, as well as the further provision of advanced features incorporated in innovative methods, such as machine learning. In summary, molecular data expansion provides an elegant way to extend mechanistic modeling, systems pharmacology, and patient-centered approaches for the assessment of drug safety. We anticipate that such advances in real-world data informatics and outcome analytics will help to better inform public health, via the improved ability to prospectively understand and predict various types of drug-induced molecular perturbations and adverse events. | 541 MOLECULAR KNOWLEDGE FOR ADVANCED DRUG SAFETY (inter-)national regulatory authorities worldwide. Major instances of such safety data collections include: • The Sentinel Initiative of the US Food and Drug Administration (FDA), 2,3 • The FDA Adverse Event Reporting System (FAERS; formerly AERS), maintained by the FDA, 4 • The Vaccine Adverse Event Reporting System (VAERS), maintained by the FDA, co-managed by the Centers for Disease Control and Prevention, 5 • The European database of suspected adverse drug reaction reports (EudraVigilance), maintained by the European Medicines Agency (EMA), 6 and • VigiBase -the global database of individual case safety reports (ICSRs), maintained by the World Health Organization's (WHO) Uppsala Monitoring Center (UMC). 7How to cite this article: Soldatos TG, Kim S, Schmidt S, Lesko LJ, Jackson DB. Advancing drug safety science by integrating molecular knowledge with post-marketing adverse event reports. CPT

Modeling tumor size dynamics based on real‐world electronic health records and image data in advanced melanoma patients receiving immunotherapy

“…Application of population modeling methods to RWD to quantify disease trajectory, treatment effects, and associated sources of variability in clinical use settings is an emerging area of pharmacometrics. Although recent reports have described applications to therapeutic drug monitoring data, 23 , 24 , 25 real‐world pharmacokinetic studies, 26 postmarketing surveillance, 27 and characterization of disease progression or drug effects in cardiovascular medicine, 28 women's health, 29 , 30 and nephrology, 31 applications in oncology remain an untapped opportunity. The analysis of tumor dynamics in cancer patient populations based on real‐world imaging data from EHRs, quantified trough lesions segmentation has not been described to our best knowledge.…”

Investigation of a Data Split Strategy Involving the Time Axis in Adverse Event Prediction Using Machine Learning

“…Adverse events often lead to the withdrawal of drugs during their development or after their marketing. , Accurate prediction of the potential risk of adverse effects makes it possible to avoid the withdrawal by identifying the potential risks in advance. , Therefore, various predictive evaluation methods, such as toxicogenomics represented by TG-GATEs and DrugMatrix and high-throughput assays represented by Tox21 and ToxCast, − have been developed. Among them, in silico approaches utilizing machine learning (ML) are widely used in adverse event prediction because they do not require drug synthesis or complex experiments. , …”

“…Among them, in silico approaches utilizing machine learning (ML) are widely used in adverse event prediction because they do not require drug synthesis or complex experiments. 3,4 For the evaluation of the prediction accuracy of ML models, most studies have adopted the random split cross-validation method. 8,9 The applicability domain (AD), which defines and restricts the applicable compounds for a model considering the chemical spaces of the training set, determines whether such a model can be applied to the test set.…”

Quantitative approaches to drug safety: The 2022 PSP special issue

“…The Drug Safety special issue contains a variety of original research articles that collectively demonstrate how several different quantitative strategies can be applied to understand, predict, and ultimately prevent a wide range of adverse events. These are complemented by a review article that provides useful “big picture” context, 4 and two brief Perspectives that describe somewhat specialized issues that are sometimes overlooked in classical toxicity studies. These Perspectives cover drugs potentially carried in breast milk by nursing mothers 5 and toxicity caused by snake bites 6 .…”

“…Finally, it is worthwhile to highlight a couple of publications that discuss and use methods that are not typically encountered in CPT:PSP . The review article by Soldatos et al 4 advocates for the construction of biological networks through the integration of molecular data (such as target lists or drug‐induced changes in gene expression) with adverse event reports, often obtained through sophisticated text mining strategies. Along similar lines, in the paper by Jeong et al, 17 the authors use mechanistic simulation results to build a classifier using a convolutional neural network.…”