Cancer type-specific variation in patterns of driver alterations across 50,000 tumors

Area of Science:

• Genomics
• Oncology
• Cancer Biology

Background:

• The impact of somatic driver alterations on cancer development is influenced by the specific tissue environment.
• Understanding these context-specific properties necessitates large datasets of genomically analyzed tumors with clinical annotations.

Purpose of the Study:

• To characterize cancer type-specific patterns of driver alterations across a large cohort of tumors.
• To identify novel driver hotspots and analyze their context-specific features and clinical implications.

Main Methods:

• Analysis of genomic data from 54,331 tumors across 448 histological cancer subtypes.
• Identification of driver alterations, including gene fusions and hotspots.
• Evaluation of subclonality, emergence timing, and biological properties of drivers.
• Assessment of human leukocyte antigen (HLA) restricted neoantigens and HLA loss in relation to ancestry and resistance.

Main Results:

• Defined cancer type-specific driver alteration patterns, identifying 164 new hotspots.
• Found that one-third of drivers occurred in non-canonical contexts with distinct features (subclonality, later emergence).
• Observed associations between gene fusions, co-occurring drivers, and earlier disease onset.
• Identified ancestry-specific differences in HLA-restricted neoantigens and cancer-type-specific patterns of resistance via somatic HLA loss.

Conclusions:

• The functional role and oncogenic impact of driver alterations are highly dependent on the cancer type and clinical context.
• Findings provide insights into non-canonical driver behavior, early disease onset mechanisms, and immune evasion strategies.
• Results have implications for understanding cancer heterogeneity and developing targeted therapies, including T cell receptor therapy eligibility.