In Silico Prediction of Mechanism of Action for Cancer Therapeutics

Authors


Abstract

Cancer is currently the second leading cause of death in the U.S. and is projected to become the principal cause in the near future. While radiation and surgery are common cancer treatment methods, chemotherapy remains a key treatment option, offering distinct advantages over other therapy options, especially in the management of metastasized tumors. Understanding the mechanism of action (MoA) of current and newly developed drugs is crucial to ongoing drug development research. Foreknowledge of how a candidate drug works can yield a wealth of information, including which cancers a drug may treat more effectively based on the susceptibility of the cancer to drugs with the same MoA. Previous studies concerning prediction of MoA have relied on costly experimental measurements as input for their predictions. We have developed an a priori quantitative structure-activity relationship (QSAR) for the in silico prediction of MoA without the need for experimental measurements. This model enables us to relate structural features of a chemical to its efficacy with a predictive accuracy of over 80 %, thus identifying the MoA of a candidate drug without costly, time-consuming experimental tests.

Ancillary