Immune dysregulation in COVID-19: Mathematical modeling of the within-host dynamics

The COVID-19 pandemic has spurred extensive research into viral transmission and control, yet the mechanisms of the human immune response to SARS-CoV-2 remain incompletely understood, particularly the role of natural killer (NK) cells and cytokine regulation in disease severity. Mathematical modeling provides a powerful approach to bridge this gap by linking viral dynamics with immune interactions. In this work, we developed a mechanistic within-host model, formulated in a system of coupled ordinary and delayed differential equations, to investigate the contributions of NK cell activity, interferon signaling, and pro-inflammatory cytokines to viral clearance and disease outcome. Model parameters were estimated from experimental data, and computational simulations were used to explore how dysregulated NK responses and cytokine feedback loops may drive divergent clinical outcomes. Local sensitivity analysis identified the most influential parameters shaping host–pathogen dynamics, highlighting potential control points for intervention. In addition, knockdown simulations were performed to mimic potential therapeutic interventions, allowing us to evaluate their advantages and limitations in silico. These findings provided mechanistic insights into COVID-19 immune dynamics and offered a foundation for guiding the design of future treatment strategies.