AUTHOR=She Shengnan , Chen Hengwei , Ji Wei , Sun Mengqiu , Cheng Jiaxi , Rui Mengjie , Feng Chunlai 

TITLE=Deep learning-based multi-drug synergy prediction model for individually tailored anti-cancer therapies

JOURNAL=Frontiers in Pharmacology

VOLUME=Volume 13 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/pharmacology/articles/10.3389/fphar.2022.1032875

DOI=10.3389/fphar.2022.1032875

ISSN=1663-9812

ABSTRACT=While synergistic pairwise drug combinations are frequently employed to treat cancer, complicated high-order combinations are more effective at fighting tumors that have a complex pathophysiology, prefer compensating mechanisms, and eventually become resistant to treatment. Even though certain computational methods have been used to identify synergistic drug combinations in lieu of traditional in vitro and in vivo screening tests, the majority of previously published work has focused on predicting synergistic drug pairs for specific types of cancer. These sophisticated high-order combinations receive little attention. The main objective of this study is to develop a deep learning-based approach that integrated multi-omics data to predict novel synergistic multi-drug combinations in a given cell line. To develop this approach, we firstly created a dataset comprising of gene expression profiles of cancer cell lines, target information of anti-cancer drugs, and drug response against a large variety of cancer cell lines. Based on the principle of a fully connected feedforward Deep Neural Network, the proposed model was trained using this dataset. Furthermore, for model performance, we utilized three breast cancer cell subtypes (MCF-7, MDA-MD-468 and MDA-MB-231) and one lung cancer cell line A549 to validate the predicted results of our model. Our model achieved MSE of 2.50 and RMSE of 1.58 in the regression task, and gave the best classification accuracy of 0.94, AUC of 0.97, sensitivity of 0.95, and specificity of 0.93. The predictive performance of our model has been validated by extensive in vitro cellular experiments, showing that the predicted top-ranked multidrug combinations had superior antitumor effects to other combinations, particularly those that were widely used in clinical treatment. Our model has the potential to increase the practicality of expanding the drug combinatorial space and to leverage its capacity to prioritize the most effective multi-drug combinational therapy for precision oncology applications.