Synthetic lethality-mediated precision oncology via the tumor transcriptome

Area of Science:

• Oncology
• Genomics
• Computational Biology

Background:

• Precision oncology currently focuses on targeting mutations in cancer driver genes.
• Exploring the tumor transcriptome offers new avenues for guiding cancer patient treatment.
• Genetic interactions within tumors are increasingly recognized as crucial for therapeutic response.

Purpose of the Study:

• To introduce SELECT (synthetic lethality and rescue-mediated precision oncology via the transcriptome), a novel framework for predicting cancer therapy response.
• To leverage tumor transcriptome data and genetic interactions for enhanced treatment selection.
• To validate the SELECT framework across diverse cancer types and clinical trials.

Main Methods:

• Development of the SELECT framework integrating genetic interactions and transcriptome data.
• Application of SELECT to analyze data from 35 published targeted and immunotherapy clinical trials.
• Testing SELECT's predictive performance across 10 different cancer types.
• Validation using data from the multi-arm WINTHER trial.

Main Results:

• SELECT demonstrated predictive capability in 80% of the analyzed clinical trials.
• The framework successfully predicted patient response in the WINTHER trial.
• Predictive signatures and analysis code were generated and made publicly available.

Conclusions:

• SELECT provides a robust method for predicting patient response to cancer therapies using tumor transcriptome data.
• The framework enhances precision oncology by incorporating genetic interactions.
• SELECT's public availability supports further research and prospective clinical studies in personalized cancer treatment.