Model

C-IMMSIM Model using TSOL18 & EG95

A simulation of the immune response to the selected epitopes was done using C-ImmSim software available online using default parameters (Rapin, Bernaschi, & Castiglione, 2010). The initial simulation used the fifteen selected epitope sequences, identified by IEDB as being immunogenic, as inputs for a vaccination’s antigen components for a vaccination. The antigen count halved at about 12 days from its initial concentration (Figure 1). Unfortunately, this simulation did not show any antibody titers in response to the simulated vaccination. This is likely due to a difference in the epitope prediction algorithm used in this software compared to IEDB, which resulted in many of the epitopes being either partially or fully unrecognized as immunogenic in the C-ImmSim simulation.

To overcome the issue of no antibody titers being generated, a second simulation was done using the whole protein sequences for the antigens TSOL18 and EG95 as inputs. This did result in a visible antibody titer being generated, including IgM and IgG antibody subtypes which are protective against the diseases (Figure 2). One peculiar dataset from the results was seen in the dendritic cell (DC) population, which demonstrated a downward trend of resting cells up until about seven days, around which internalized and type II presenting was at their highest (Figure 3). Another interesting result we found was the gradual increase of type I presenting cells until about day 25, suggesting this population of DCs is involved in a slower acting immune response.

Figure 1: A prediction of the expected antibody response upon injection of the TSOL18 and EG95 proteins. The black line depicts the level of antigen (the injected proteins) while the other lines indicate different types of antibodies. This graphic was obtained from a C-ImmSim simulation (Rapin et al. 2010).

Figure 2: A prediction of the expected B cell population upon injection of the TSOL18 and EG95 proteins. The black line indicates the total number of B cells (not visible on graph), while other lines indicate the different types of B cells involved in the immune response. This graphic was obtained from a C-ImmSim simulation (Rapin et al. 2010).

Figure 3: A prediction of the expected dendritic cell (DC) population upon injection of the TSOL18 and EG95 proteins. The black line indicates the total number of dendritic cells, while other lines indicate the different types of DCs involved in the immune response. This graphic was obtained from a C-ImmSim simulation (Rapin et al. 2010).

Next Steps

C-IMMSIM Model

The current simulation demonstrates that the selection of antigens for this project is appropriate for stimulating an immune response against the parasites causing Echinococcosis and Cysticercosis. Future work on modeling the immune response can focus on fitting the parameters such as MHC allele, antigen quantities, and protein sequences to be more representative of our project in order to make a more accurate prediction.

Additional next steps for an effective model of immune response would involve troubleshooting with selected epitope sequence, or even re-selecting the antigens, in order to generate the desired Th2 dominant immune response, determined to likely be protective against the parasitic diseases. We will search for software tools that include various types of adjuvants as input parameters, as this is also critical in determining the specific immune response, and the results of such a simulation would help inform us about our selection of adjuvant. Other crucial components for a vaccine efficacy model include an accurate prediction of the quantities of various cell types and antibodies present, and the duration of time over which the immune cells and antibody titers are elevated. A deeper analysis into these aspects would directly inform the dosage, delivery, and overall safety and efficacy of our vaccine.

Vaccine Distribution Model

Two key papers led to our understanding of various variables that affect the selection of an optimal strategy for vaccine distribution, which is an essential part of an effective vaccine campaign. Both papers address vaccine allocation with respect to the COVID-19 pandemic. The first of these papers, by Matrajt et al. (2021), describes a mathematical model that generates an allocation when provided a vaccine efficacy and a vaccine coverage (the fraction of the population that can be vaccinated) as inputs, both provided as a percentages. Further, the model has four different modes to minimize either the total deaths, symptomatic infections, ICU hospitalizations, or non-ICU hospitalizations.

The recommended vaccine allocation is communicated as a percentage of age groups from 0 years to 75+ to be vaccinated, and this allocation varies with the selected mode of the model as well as given inputs. Critically, the model considers older age groups to be high-risk groups, and younger age groups to be high-transmission groups. A key finding is that to minimize deaths, the optimal strategy when vaccine coverage is low (50% or less coverage) is to first allocate vaccines to high-risk groups. Given a high vaccine coverage and high vaccine efficacy (60% or greater), however, the optimal strategy to minimize deaths is instead to vaccinate high-transmission groups, which would significantly slow down the epidemic curve. Additional allocation strategies for other cases can be found in the paper. This helped us to realize that optimal vaccine distribution is complex, requires a dynamic model for accuracy, and also requires us to have a very good understanding of the efficacy of the vaccine and the course of parasitic disease within specific demographics.

The second paper, by Wen et al. (2023), presents another model that considers willingness for vaccination (i.e. demand for vaccination), vaccine inventory, costs, and budget as important factors for vaccine allocation. Considering this paper together with the previous one shows us that vaccine distribution must be informed by a multitude of social, economic, geographic and demographic factors. It is important for us to consider the optimal vaccine allocation because a vaccine will not be effective if it is not used appropriately to contain the spread of disease. Based on what we learned, we are determined to investigate further into vaccine allocation models and the necessary input parameters in order to create a more fully fledged vaccination project with real world potential.