Passenger Mutations in More Than 2,500 Cancer Genomes: Overall Molecular Functional Impact and Consequences

Area of Science:

• Genomics
• Cancer Biology
• Computational Biology

Background:

• The traditional cancer model classifies mutations as either impactful "drivers" or inconsequential "passengers."
• This dichotomous view may oversimplify the complex mutational landscape of tumors.

Purpose of the Study:

• To investigate the existence of medium-impact mutations beyond the driver-passenger dichotomy.
• To explore the relationship between mutation impact, subclonal architecture, and mutational signatures.
• To assess the predictive power of aggregated passenger mutations for cancer phenotypes.

Main Methods:

• Utilized the comprehensive variant dataset from the ICGC/TCGA Pan-Cancer Analysis of Whole Genomes (PCAWG) project.
• Analyzed mutation impact, subclonal architecture (early vs. late mutations), and mutational signatures.
• Adapted an additive-effects model from complex-trait studies to evaluate the collective impact of mutations.

Main Results:

• Identified a third group of medium-impact putative passenger mutations, challenging the strict driver-passenger dichotomy.
• Demonstrated correlation between molecular impact, subclonal architecture, and distinct mutational signatures.
• Showed that aggregated passenger mutations, including potential weak drivers, provide significant additive variance (approximately 12%) in predicting cancerous phenotypes.
• Estimated the frequency of potential weak-driver mutations in samples lacking known drivers.

Conclusions:

• The mutational landscape of cancer is more nuanced than a simple driver-passenger model.
• Aggregated effects of seemingly low-impact mutations can substantially contribute to cancer development and progression.
• This framework enhances our ability to predict cancer phenotypes and identify potential therapeutic targets by considering the collective impact of all mutations.