State-dependent evolutionary models reveal modes of solid tumour growth

Area of Science:

• Oncology
• Evolutionary Biology
• Computational Biology

Background:

• Spatial tumor growth influences cancer progression, treatment resistance, and metastasis.
• Quantifying spatial variations in tumor cell division rates within clinical tumors is challenging.

Purpose of the Study:

• To develop a method for inferring differential cell division rates based on spatial position within tumors.
• To analyze genetic patterns indicative of faster peripheral tumor cell division.

Main Methods:

• Reconstruction of phylogenetic trees from spatially sampled tumor cells.
• Development of a Bayesian state-dependent evolutionary phylodynamic model (SDevo).
• Application of SDevo to simulated and clinical hepatocellular carcinoma (HCC) data.

Main Results:

• Faster-dividing peripheral tumor lineages exhibit more branching and mutations in phylogenetic trees.
• SDevo accurately infers spatially varying birth rates in simulated tumors.
• Clinical HCC data reveal a 3-6 times higher division rate on the tumor edge compared to the center.

Conclusions:

• Spatial position significantly impacts tumor cell division rates, with faster growth at the periphery.
• The SDevo model provides a robust tool for analyzing spatial tumor evolution using multi-region sequencing data.
• This approach can enhance understanding of spatial growth restrictions and inform future cancer research.