Computational Approaches to Cancer Cell Dormancy: From Detection to Dynamic Modelling

Area of Science:

• Oncology
• Computational Biology
• Systems Biology

Background:

• Cancer cell dormancy is a critical yet poorly understood state characterized by reversible growth arrest and delayed recurrence.
• Advances in single-cell and multi-omic technologies aid in detecting dormant and persister cell populations, but their molecular underpinnings remain elusive.

Purpose of the Study:

• To review the application of computational methods in elucidating cancer cell dormancy.
• To explore how these methods infer dormant cell identity, heterogeneity, microenvironmental influences, state transitions, and reactivation dynamics.

Main Methods:

• Review of computational approaches applied to single-cell transcriptomics, lineage tracing, spatial profiling, and integrative multi-omic analyses.
• Examination of mathematical and statistical frameworks for modeling dormancy and reactivation.

Main Results:

• Computational analyses reveal significant context-dependent variability in dormant cell states, challenging the concept of a universal dormancy signature.
• Existing methods struggle with fragmented definitions, rare-state detection, and analyzing temporal processes from static data.

Conclusions:

• Progress in understanding cancer dormancy requires computational frameworks that treat it as a dynamic, multi-scale systems problem, not a static classification.
• Integrating diverse computational approaches is essential for resolving the complexities of dormancy and its clinical implications.