A data-driven approach to predicting successes and failures of clinical trials

By Kaitlyn M. Gayvert, Neel S. Madhukar, Olivier Elemento. 2016

300 citations recorded by the source corpus.

Over the past decade, the rate of drug attrition due to clinical trial failures has risen substantially. Unfortunately it is difficult to identify compounds that have unfavorable toxicity properties before conducting clinical trials. Inspired by the effective use of Sabermetrics in predicting successful baseball players, we sought to use a similar “moneyball” approach that analyzes overlooked features to predict clinical toxicity. We introduce a new data-driven approach (PrOCTOR) that directly predicts the likelihood of toxicity in clinical trials. PrOCTOR integrates properties of a compound’s targets and its structure to provide a new measure, the PrOCTOR score. Drug target network connectivity and expression levels, along with molecular weight, were identified as important indicators of adverse clinical events. Altogether, our method provides a data-driven broadly applicable strategy to identify drugs likely to possess manageable toxicity in clinical trials and will help drive the design of therapeutic agents with less toxicity.

Subfields: Computational Theory and Mathematics, Pharmacology, Statistics and Probability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Pharmacology

In Statistics and Probability

Citations in the corpus

Artificial intelligence to deep learning: machine intelligence approach for drug discovery (2021, methodology): “DeepTox (http:// bioinf. jku. at/ resea rch/ DeepT ox/ tox21. html) [334] and PrOCTOR (https:// github. com/ kgayv ert/ PrOCT OR) [335], are used for prediction of toxicity of new compounds and prediction of the toxicity probability in clinical trials, respectively.”
Artificial intelligence in drug discovery and development (2020, unknown): “It also recognized FDA-approved drugs that later reported adverse drug events [53].”
Uni-Mol: A Universal 3D Molecular Representation Learning Framework (2023, background): “• ClinTox This dataset contains the toxicity of the drug in clinical trials and the status of the drug for FDA approval[82].”
Integrated deep learned transcriptomic and structure-based predictor of clinical trials outcomes (2016, methodology): “One recently developed approach called PrOCTOR (Gayvert, Madhukar, and Elemento 2016) takes into account not only bioavailability related properties, but also target-related properties, such as tissue selectivity and established informative chemical features of the drugs with target-based features t”
Pre-training Molecular Graph Representation with 3D Geometry (2021, unknown): “[28] Kaitlyn M Gayvert, Neel S Madhukar, and Olivier Elemento.”
A Deep Learning Approach to Antibiotic Discovery. (2020, unknown): “coli, we prioritized these based on clinical phase of investigation, structural similarity to molecules in the primary training dataset, and predicted toxicity using a deep neural network model trained on the ClinTox database (Gayvert et al., 2016; Wu et al., 2017) (Table S2B).”
A deep learning approach to antibiotic discovery (2020, methodology): “…against E. coli , we prioritized these based on clinical phase of investigation, structural similarity to molecules in the primary training dataset, and predicted toxicity using a deep neural network model trained on the ClinTox database (Gayvert et al., 2016; Wu et al., 2017) (Table S2B).”
Artificial Intelligence in Cancer Research and Precision Medicine (2021, background): “Gayvert and colleagues published a random forest model that used distinct preclinical data types to predict drug toxicity and adverse events (79).”
Geometry-enhanced molecular representation learning for property prediction (2021, methodology): “We conduct experiments on multiple molecular benchmarks from 224 the MoleculeNet [53], including both classification and regression tasks [47, 33, 38, 16, 25, 28].”
Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development (2021, background): “It contains 1,484 drugs from clinical trials records (Gayvert et al. 2016).”