Pathway-driven analysis of synthetic lethal interactions in cancer using perturbation screens

Area of Science:

• Oncology
• Genomics
• Computational Biology

Background:

• Synthetic lethality (SL) is a promising cancer therapeutic strategy when direct gene targeting is difficult.
• Identifying SL interactions is crucial for developing novel cancer treatments.

Purpose of the Study:

• To identify potential SL interactions across pan-cancer and individual cancer types using a novel computational framework.
• To leverage biological function and signaling pathway information to improve the accuracy of SL interaction discovery.
• To propose FDA-approved drugs for specific cancer mutations based on screening data.

Main Methods:

• Comprehensive analysis of large-scale CRISPR, shRNA, and PRISM screens.
• Development of a new computational framework integrating gene function and signaling pathway data.
• Application of the framework to pan-cancer and 12 individual cancer types.

Main Results:

• Identified several putative SL interactions, including KRAS-MAP3K2 and APC-TCF7L2 in pan-cancer.
• Discovered cancer-type specific SL interactions such as CCND1-METTL1 (liver), TP53-FRS3 (blood), SMO-MDM2 (skin), and CCNE1-MTOR (gastric).
• Proposed potential drug therapies, including cabazitaxel for VHL-mutated kidney cancer and alectinib for KRAS/NRAS-mutated lung cancer.

Conclusions:

• Integrating signaling pathway information enhances the identification of more promising SL interactions.
• This approach improves concordance between different screening methods (CRISPR, shRNA).
• Findings provide clinically relevant insights, such as the potential efficacy of dasatinib for colorectal cancer with WNT pathway mutations.