Modeling metastatic progression from cross-sectional cancer genomics data

Area of Science:

• Cancer genomics
• Computational biology
• Evolutionary medicine

Background:

• Metastasis formation is a critical factor in cancer mortality.
• Early stages of cancer dissemination and spread are difficult to observe.
• Genomic data from primary tumors and metastases can reveal insights into metastasis dynamics.

Purpose of the Study:

• To develop a computational model (metMHN) for analyzing joint progression of primary tumors and metastases.
• To elucidate the relationships between genomic events, metastasis formation, and clinical emergence.
• To enable chronological reconstruction of mutational sequences and estimate metastatic seeding times.

Main Methods:

• Developed metMHN, a cancer progression model utilizing cross-sectional cancer genomics data.
• Applied the model to a dataset of nearly 5000 lung adenocarcinomas.
• Analyzed statistical dependencies among genomic events and metastasis formation.

Main Results:

• metMHN successfully deduced joint progression of primary tumors and metastases.
• Identified TP53 and EGFR as key mediators in lung adenocarcinoma metastasis.
• Revealed that copy number alterations predominantly influence post-seeding adaptation.

Conclusions:

• metMHN provides a powerful tool for understanding cancer metastasis using genomic data.
• The findings highlight specific genes and genomic alterations involved in metastasis.
• This approach aids in reconstructing tumor evolution and estimating metastasis timing.