PT - JOURNAL ARTICLE AU - Margaret A. Myers AU - Amanda P. Smith AU - Lindey C. Lane AU - David J. Moquin AU - J. Robert Michael AU - Peter Vogel AU - Stacie Woolard AU - Amber M. Smith TI - The Nonlinear Relations that Predict Influenza Viral Dynamics, CD8<sup>+</sup> T cell-Mediated Clearance, Lung Pathology, and Disease Severity AID - 10.1101/555276 DP - 2019 Jan 01 TA - bioRxiv PG - 555276 4099 - http://biorxiv.org/content/early/2019/02/19/555276.short 4100 - http://biorxiv.org/content/early/2019/02/19/555276.full AB - Influenza A viruses cause a significant amount of morbidity and mortality. Understanding how the infection is controlled by host immune responses and how different factors influence severity are critical to combat the infection. During infection, viral loads increase exponentially, peak, then decline until resolution. The viral decline is often biphasic, which we previously determined is a consequence of density-dependent infected cell clearance. The second, rapid clearance phase corresponds with the infiltration of CD8+ T cells, but how the rate changes with infected cell density and T cell density is unclear. Further, the kinetics of virus, infected cells, and CD8+ T cells all contribute to disease severity but do not seem to be directly correlated. Thus, we investigated the relations between viral loads, infected cells, CD8+ T cells, lung pathology, and disease severity/symptoms by infecting mice with influenza A/PR8, simultaneously measuring virus and CD8+ T cells, and developing and calibrating a kinetic model. The model predicted that infection resolution is sensitive to CD8+ T cell expansion, that there is a critical T cell magnitude below which the infection is significantly prolonged, and that the efficiency of T cell-mediated clearance is dependent on infected cell density. To further examine the latter finding and validate the model’s predicted dynamics, we quantified infected cell kinetics using lung histomorphometry. These data showed that the area of lung infected matches the predicted cumulative infected cell dynamics, and that the area of resolved infection parallels the relative CD8+ T cell magnitude. Our analysis further revealed a nonlinear relationship between disease severity (i.e., weight loss) and the percent of the lung damaged. Establishing the predictive capabilities of the model and the critical connections that map the kinetics of virus, infected cells, CD8+ T cells, lung pathology, and disease severity during influenza virus infection aids our ability to forecast the course of infection, disease progression, and potential complications, thereby providing insight for clinical decisions.Author Summary Influenza A viruses infect millions of people each year. An understanding of how virus growth and host responses impact disease progression is critical to identify disease-specific markers that help predict hospitalization, complications, and therapeutic efficacy. To establish these relations, we developed and validated a mathematical model that accurately forecasts the kinetics of virus, infected cells, and CD8+ T cells. We discovered that the rate of infected cell removal by CD8+ T cells increases as infected cells decline, that there is a critical number of CD8+ T cells below which recovery is prolonged, and that recovery time depends on the number of CD8+ T cells rather than their efficiency. Further, examining lung pathology showed that the area of the lung infected and the area of the lung resolved parallel the model’s predicted cumulative infected cell and CD8+ T cell dynamics, respectively. Our analysis also revealed a nonlinear relation between the lung pathology and disease severity. These connections illustrated the predictive capabilities of our model and established how the spread and clearance of influenza virus within the lung contributes to disease progression. This information aids our ability to predict the infection course and potential complications, and make robust clinical decisions.