Assessing machine learning approaches for predicting failures of investigational drug candidates during clinical trials

• Published in: Computers in biology and medicine 2023-02, Vol.153, p.106494, Article 106494
• Main Authors: John, Lijo, Mahanta, Hridoy Jyoti, Soujanya, Y., Sastry, G. Narahari
• Format: Article
• Language: English
• Subjects: Artificial intelligence; Chemical compounds; Classifiers; Clinical trials; Computer applications; Datasets; Drug development; Drug withdrawal; Drugs; Drugs, Investigational - therapeutic use; Failure analysis; Feature extraction; Feature selection; Fingerprints; Learning algorithms; Machine Learning; Molecular descriptors; Optimization; Performance measurement; Pharmaceutical industry; Pharmacology; Prediction models; Proteins; Toxicity; Toxicity testing; Toxicology
• ISSN: 0010-4825; 1879-0534
• DOI: 10.1016/j.compbiomed.2022.106494

One of the major challenges in drug development is having acceptable levels of efficacy and safety throughout all the phases of clinical trials followed by the successful launch in the market. While there are many factors such as molecular properties, toxicity parameters, mechanism of action at the target site, etc. that regulates the therapeutic action of a compound, a holistic approach directed towards data-driven studies will invariably strengthen the predictive toxicological sciences. Our quest for the current study is to find out various reasons as to why an investigational candidate would fail in the clinical trials after multiple iterations of refinement and optimization. We have compiled a dataset that comprises of approved and withdrawn drugs as well as toxic compounds and essentially have used time-split based approach to generate the training and validation set. Five highly robust and scalable machine learning binary classifiers were used to develop the predictive models that were trained with features like molecular descriptors and fingerprints and then validated rigorously to achieve acceptable performance in terms of a set of performance metrics. The mean AUC scores for all the five classifiers with the hold-out test set were obtained in the range of 0.66–0.71. The models were further used to predict the probability score for the clinical candidate dataset. The top compounds predicted to be toxic were analyzed to estimate different dimensions of toxicity. Apparently, through this study, we propose that with the appropriate use of feature extraction and machine learning methods, one can estimate the likelihood of success or failure of investigational drugs candidates thereby opening an avenue for future trends in computational toxicological studies. The models developed in the study can be accessed at https://github.com/gnsastry/predicting_clinical_trials.git. [Display omitted] •Estimate the likelihood of success or failure of investigational candidates.•Five highly robust and scalable machine learning algorithms have been used.•Dataset of 2,212 compounds consisting of approved and withdrawn drugs.•Time-split approach generates training and hold-out test set.