Brain Cancer Drug Discovery: Clinical Trials, Drug Classes, Targets, and Combinatorial Therapies.

Brain cancer is a formidable challenge for drug development, and drugs derived from many cutting-edge technologies are being tested in clinical trials. We manually characterized 981 clinical trials on brain tumors that were registered in ClinicalTrials.gov from 2010 to 2020. We identified 582 unique therapeutic entities targeting 581 unique drug targets and 557 unique treatment combinations involving drugs. We performed the classification of both the drugs and drug targets based on pharmacological and structural classifications. Our analysis demonstrates a large diversity of agents and targets. Currently, we identified 32 different pharmacological directions for therapies that are based on 42 structural classes of agents. Our analysis shows that kinase inhibitors, chemotherapeutic agents, and cancer vaccines are the three most common classes of agents identified in trials. Agents in clinical trials demonstrated uneven distribution in combination approaches; chemotherapy agents, proteasome inhibitors, and immune modulators frequently appeared in combinations, whereas kinase inhibitors, modified immune effector cells did not as was shown by combination networks and descriptive statistics. This analysis provides an extensive overview of the drug discovery field in brain cancer, shifts that have been happening in recent years, and challenges that are likely to come. SIGNIFICANCE STATEMENT: This review provides comprehensive quantitative analysis and discussion of the brain cancer drug discovery field, including classification of drug, targets, and therapies.

Feature Selection Pipelines with Classification for Non-targeted Metabolomics Combining the Neural Network and Genetic Algorithm.

Non-targeted metabolomics via high-resolution mass spectrometry methods, such as direct infusion Fourier transform-ion cyclotron resonance mass spectrometry (DI-FT-ICR MS), produces data sets with thousands of features. By contrast, the number of samples is in general substantially lower. This disparity presents challenges when analyzing non-targeted metabolomics data sets and often requires custom methods to uncover information not always accessible via classical statistical techniques. In this work, we present a pipeline that combines a convolutional neural network with traditional statistical approaches and an adaptation of a genetic algorithm. The developed method was applied to a lifestyle intervention cohort data set, where subjects at risk of type 2 diabetes underwent an oral glucose tolerance test. Feature selection is the final result of the pipeline, achieved through classification of the data set via a neural network, with a precision-recall score of over 0.9 on the test set. The features most relevant for the described classification were then chosen via a genetic algorithm. The output of the developed pipeline encompasses approximately 200 features with high predictive scores, providing a fingerprint of the metabolic changes in the prediabetic class on the data set. Our framework presents a new approach which allows to apply complex modeling based on convolutional neural networks for the analysis of high-resolution mass spectrometric data.