Improved prediction of PARP inhibitor response and identification of synergizing agents through use of a novel gene expression signature generation algorithm

Abstract

Personalized medicine: Signature-guided cancer therapy Personalized cancer therapy is one of the holy grails of oncology, as the ability to determine what treatment would best benefit a patient would serve not only to improve outcomes, but also mitigate side effects from less effective treatments. Here, we develop algorithms to predict what patients will respond to a given therapeutic modality, as well as ways to specifically target any observed phenotype, by integrating large scale data sets that profile cancer cell line gene expression and sensitivity to hundreds of drugs. Furthermore, we show how these gene expression signatures can be used to predict novel synergizing agents to further enhance the efficacy of these therapeutics. Taken together, this work stands to advance the era of personalized medicine by enabling precision medicine approaches in the clinic.