Computational inference of cancer-specific vulnerabilities in clinical samples

Kiwon Jang,Jong Won Lee,Min Ji Park,Haeun Hwangbo,Suhwan Chang,Min Kyung Sung,Jaeyun Jung,Sinae Kim,Jung Kyoon Choi,Sei-Hyun Ahn,Jae Soon Park

doi:10.1186/s13059-020-02077-1

Abstract

BackgroundSystematic in vitro loss-of-function screens provide valuable resources that can facilitate the discovery of drugs targeting cancer vulnerabilities.ResultsWe develop a deep learning-based method to predict tumor-specific vulnerabilities in patient samples by leveraging a wealth of in vitro screening data. Acquired dependencies of tumors are inferred in cases in which one allele is disrupted by inactivating mutations or in association with oncogenic mutations. Nucleocytoplasmic transport by Ran GTPase is identified as a common vulnerability in Her2-positive breast cancers. Vulnerability to loss of Ku70/80 is predicted for tumors that are defective in homologous recombination and rely on nonhomologous end joining for DNA repair. Our experimental validation for Ran, Ku70/80, and a proteasome subunit using patient-derived cells shows that they can be targeted specifically in particular tumors that are predicted to be dependent on them.ConclusionThis approach can be applied to facilitate the development of precision therapeutic targets for different tumors.

Highlights

There have been substantial efforts to profile cancer dependency by loss-of-function screens in cell lines [1,2,3,4,5,6,7,8,9,10], providing valuable resources that can lay foundation for new ways to fight cancer
We merged two independent CRISPR-Cas9 screens of 28 and 25 breast cancer cell lines [9, 10], each based on a dependency score named CERES and BAGEL [9, 15], respectively
For each dependency and independency, we performed our in silico CRISPR/RNAi to generate the perturbed transcriptome

Summary

Introduction

There have been substantial efforts to profile cancer dependency by loss-of-function screens in cell lines [1,2,3,4,5,6,7,8,9,10], providing valuable resources that can lay foundation for new ways to fight cancer. The screening methods are applicable only to in vitro cell culture, limiting the discovery of therapeutic targets for clinical samples. In the absence of matched normal samples, it is difficult to identify truly cancer-specific dependencies and characterize them in association with somatic alterations. Systematic in vitro loss-of-function screens provide valuable resources that can facilitate the discovery of drugs targeting cancer vulnerabilities

Methods

Results

Conclusion