A multilineage screen identifies actionable synthetic lethal interactions in human cancers.

Area of Science:

• Oncology
• Genetics
• Systems Biology

Background:

• Cancer arises from genetic alterations, creating dependencies exploitable for therapy.
• Many identified genetic dependencies lack consistency across different tumor types.
• A need exists for robust methods to discover actionable genetic interactions in cancer.

Purpose of the Study:

• To introduce SCHEMATIC, a novel strategy for identifying a core network of highly penetrant, actionable genetic interactions.
• To develop a context-aware, data-driven approach for matching genetic alterations to targeted cancer therapies.

Main Methods:

• Systematic combinatorial knockouts across tumor lineages to perturb fundamental cellular processes.
• Hierarchical pooling analysis of identified synthetic lethal interactions to assess reliability and lineage specificity.
• Convergence analysis of interactions onto multigene systems and identification of drug sensitivity biomarkers.

Main Results:

• Identified 1,805 novel synthetic lethal interactions, with 95% previously unreported.
• Found that 51% of interactions segregate by tissue type or biomarker status, and 34% are penetrant across lineages.
• Discovered 49 essential multigene systems, including MAPK signaling and BAF complexes, and 266 interactions correlating with drug sensitivity, such as KDM5C/6A alterations sensitizing to TIPARP inhibition.

Conclusions:

• SCHEMATIC provides a robust framework for discovering actionable genetic interactions in cancer.
• The identified interactions and biomarkers can guide the development of targeted therapies.
• This approach enhances the precision of matching genetic alterations to effective cancer treatments.