Back

Integrative Modeling of SARS-CoV-2 Infection Dynamics to Inform COVID-19 Vaccination Strategies

Waema, R.; Kaumbutha, C. M.; Orwa, T.

2025-08-24 epidemiology
10.1101/2025.08.19.25333964 medRxiv
Show abstract

Non-pharmaceutical interventions were very instrumental in the early phases of COVID-19 pandemic. Fortunately, the urgency to control the crisis prompted an accelerated vaccine development process, saving millions of lives globally. Despite these measures, cases of COVID-19 and other variants SARS-CoV-2 are still being reported in different countries and regions. We present a within-host mathematical model of SARS-CoV-2 infection that incorporates target cell dynamics, innate and adaptive immune responses, and vaccine interventions. Analytical results identify the basic reproduction number, R0, as the threshold for infection persistence. Simulations show that immune responses, both lytic and non-lytic, are critical in controlling viral replication, with vaccine-induced immunity further reducing viral load and protecting epithelial cells. Immune-boosting strategies and monoclonal antibody therapies targeting intracellular replication outperform entry-blocking interventions alone, while combination approaches yield the greatest reduction in peak viral load and fastest clearance. Timing is crucial: early vaccination or treatment maximizes benefits, whereas delays allow higher viral titers to persist. These results underscore the importance of early, multi-mechanism interventions, particularly for vulnerable populations such as the elderly and immunocompromised. The model offers a framework for evaluating treatment strategies and can be extended to incorporate pharmacokinetics/pharmacodynamics or patient-specific calibration for improved predictive accuracy.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.