A Multi-Scale Hybrid Model of CSC Emergence Through IL-6 Signaling: Integrating GRNs into Agent-Based Tumor Simulations

Mahule Roy, Subhas Roy

Abstract: Cancer stem cells (CSCs) represent a critical subpopulation within tumors that drive metastasis, therapeutic resistance, and post-treatment recurrence. While interleukin-6 (IL-6) signaling has been established as a key mediator in the conversion of non-stem cancer cells (NSCCs) to CSCs, existing computational models predominantly focus on sustained cytokine expression patterns. Emerging experimental evidence suggests that IL-6 secretion often occurs in transient, pulsatile bursts under stress conditions such as hypoxia, chemotherapy, or chronic inflammation. This study develops a multi-scale hybrid computational model to investigate how temporal dynamics of IL-6 signaling—specifically pulsed versus sustained patterns—influence CSC emergence through the LIN28–let-7 regulatory axis. We constructed a gene regulatory network (GRN) incorporating IL-6, NF-κB, LIN28, let-7, c-Myc, and OCT4, modeled via ordinary differential equations (ODEs) with shifted Hill functions to capture nonlinear regulatory interactions. This intracellular GRN was integrated into a multicellular agent-based framework using CompuCell3D, enabling simulation of tumor population dynamics under spatially heterogeneous cytokine fields. We systematically compared four signaling regimes: (1) pulsed IL-6 with sustained NF-κB, (2) sustained IL-6 with sustained NF-κB, (3) stochastic IL-6 with sustained NF-κB, and (4) pulsed IL-6 with pulsed NF-κB. Our simulations reveal that transient IL-6 pulses, particularly when synchronized with NF-κB activation, promote more robust and dynamic CSC emergence compared to sustained or stochastic signaling patterns. The coordinated pulsing enhanced bistable switch activation frequency, pushing more NSCCs across the stemness threshold defined by OCT4 expression. These findings suggest that cytokine temporal dynamics represent a critical determinant of tumor plasticity and propose pulsed microenvironments as potential accelerators of stem-like transitions. This work provides a novel computational framework for studying cytokine dynamics in tumor evolution and offers insights for therapeutic strategies targeting CSC emergence.