# A Mechanistic Digital Twin of uPAR-Driven Prostate Cancer Invasion Integrating ODE Signalling and Agent-Based Modelling

## TL;DR

This paper presents a digital twin framework that combines molecular and spatial models to study how uPAR signaling affects prostate cancer invasion.

## Contribution

A novel multiscale framework integrating ODE and ABM models to mechanistically link uPAR signaling to tumor invasion dynamics.

## Key findings

- uPAR inhibition significantly reduces tumor invasion and growth in simulations.
- Enhanced uPA signaling shows only modest and non-significant invasion trends.
- Molecular-level changes in uPAR signaling lead to observable spatial tumor behavior.

## Abstract

Background: Aberrant signalling through the urokinase-type plasminogen activator receptor (uPAR) is a key driver of tumour invasion and progression in prostate cancer, yet linking molecular-level perturbations to emergent spatial invasion phenotypes remains challenging. Methods: In this study, we developed a multiscale in silico framework combining molecular docking, mechanistic ordinary differential equation (ODE) modelling, and agent-based modelling (ABM) to investigate uPAR-driven invasion dynamics. Results: Molecular docking and MM-GBSA analyses were used to prioritise caffeic acid phenethyl ester (CAPE) as a candidate uPA/uPAR modulator, while uPAR inhibition was implemented mechanistically at the signalling level within the ODE model rather than through direct energetic parametrisation. Steady-state signalling outputs were mapped to effective proliferation and motility rates, which served as inputs to a spatial ABM of tumour invasion. The integrated simulations showed that uPAR inhibition results in statistically significant reductions in spatial invasion and tumour growth compared with baseline conditions, whereas enhanced uPA signalling produced only modest, non-significant trends. Conclusions: These findings demonstrate how subtle intracellular signalling perturbations can translate into pronounced population-level invasion phenotypes when embedded in a spatial context. Overall, the proposed digital-twin framework provides a coherent and extensible approach for connecting molecular prioritisation with quantitative predictions of tumour invasion behaviour in prostate cancer.