Combining a whole blood infection assay with biomathematical modeling to classify innate immune function

Objectives
Sepsis is a heterogeneous clinical syndrome, resulting from highly diverse pathological conditions and showing variable disease kinetics in individual patients. So far, the treatment of sepsis has focused on hyper-inflammation and anti-inflammatory therapies and there is still no approved immunomodulatory approach for the treatment of sepsis. Furthermore, recent studies suggest that both hypo-and hyper-inflammatory responses are activated simultaneously in sepsis, further complicating immune therapy approaches. Our main objective is to provide novel methods for the individualized quantification of immune effector functions in sepsis patients by combining a human whole-blood assay of infection with biomathematical modeling.

Methods
Blood was obtained from healthy volunteers or cardiac surgery patients and infected with either Staphylococcus aureus or Candida albicans over a 4 h time course. Association of the two model pathogens with immune cells and the activation pattern on these immune cells was determined via flow cytometry. Based on time course data regarding pathogen distribution to different immune cells, a virtual infection model was generated and used to perform detailed and quantitative predictions on the dynamics of host-pathogen interaction.

Results
Previously-generated data using blood from healthy donors infected with either C. albicans or S. aureus showed clear differences in the immune response and the rates of immune cell reaction differ considerably for infection with both pathogens. Our next step was to perform the whole-blood assay with a more homogenous population than sepsis patients. Within a pilot study, blood from patients that underwent cardiac surgery with extracorporeal circulation was analyzed. Characterized by a clear time-defined and strong inflammatory stimulus this surgery offers the possibility to distinguish between inter-individual differences and effects of inflammation. Analysis of blood samples from patients at three different time points (before cardiac surgery, immediately after surgery and 1 day after admission to the intensive care unit) revealed a strong post-operative increase in white blood cell count. Furthermore, variations in immune cell reaction rates and immune cell activation were determined and found to be higher after surgery e.g. increased rates for phagocytosis. Interestingly, pathogen-specific patterns of immune response like differences in association kinetics were diminished after surgery. Once optimized, similar analyzes of blood samples of sepsis patients and patients who have survived sepsis will follow in future studies.

Conclusion
To make immunomodulatory therapy approaches for sepsis patients possible it is necessary to develop tools that enable the classification of sepsis patients by their immune status. The use of an ex vivo whole blood model of infection combined with advanced biomathematical quantification allows for the quantification of immune effector functions, thereby forming a basis for future patient stratification.

7 / Combining a whole blood infection assay with biomathematical modeling to classify innate immune function