A Predictive Model of Vaccine Reactogenicity Using Data from an In Vitro Human Innate Immunity Assay System.

A primary concern in vaccine development is safety, particularly avoiding an excessive immune reaction in an otherwise healthy individual. An accurate prediction of vaccine reactogenicity using in vitro assays and computational models would facilitate screening and prioritization of novel candidates early in the vaccine development process. Using the modular in vitro immune construct model of human innate immunity, PBMCs from 40 healthy donors were treated with 10 different vaccines of varying reactogenicity profiles and then cell culture supernatants were analyzed via flow cytometry and a multichemokine/cytokine assay. Differential response profiles of innate activity and cell viability were observed in the system. In parallel, an extensive adverse event (AE) dataset for the vaccines was assembled from clinical trial data. A novel reactogenicity scoring framework accounting for the frequency and severity of local and systemic AEs was applied to the clinical data, and a machine learning approach was employed to predict the incidence of clinical AEs from the in vitro assay data. Biomarker analysis suggested that the relative levels of IL-1B, IL-6, IL-10, and CCL4 have higher predictive importance for AE risk. Predictive models were developed for local reactogenicity, systemic reactogenicity, and specific individual AEs. A forward-validation study was performed with a vaccine not used in model development, Trumenba (meningococcal group B vaccine). The clinically observed Trumenba local and systemic reactogenicity fell on the 26th and 93rd percentiles of the ranges predicted by the respective models. Models predicting specific AEs were less accurate. Our study presents a useful framework for the further development of vaccine reactogenicity predictive models.

Bridging Computational Vaccinology and Vaccine Development Through Systematic Identification, Characterization, and Downselection of Conserved and Variable Circumsporozoite Protein CD4 T Cell Epitopes From Diverse Plasmodium falciparum Strains.

An effective malaria vaccine must prevent disease in a range of populations living in regions with vastly different transmission rates and protect against genetically-diverse Plasmodium falciparum (Pf) strains. The protective efficacy afforded by the currently licensed malaria vaccine, Mosquirix™, promotes strong humoral responses to Pf circumsporozoite protein (CSP) 3D7 but protection is limited in duration and by strain variation. Helper CD4 T cells are central to development of protective immune responses, playing roles in B cell activation and maturation processes, cytokine production, and stimulation of effector T cells. Therefore, we took advantage of recent in silico modeling advances to predict and analyze human leukocyte antigen (HLA)-restricted class II epitopes from PfCSP - across the entire PfCSP 3D7 sequence as well as in 539 PfCSP sequence variants - with the goal of improving PfCSP-based malaria vaccines. Specifically, we developed a systematic workflow to identify peptide sequences capable of binding HLA-DR in a context relevant to achieving broad human population coverage utilizing cognate T cell help and with limited T regulatory cell activation triggers. Through this workflow, we identified seven predicted class II epitope clusters in the N- and C-terminal regions of PfCSP 3D7 and an additional eight clusters through comparative analysis of 539 PfCSP sequence variants. A subset of these predicted class II epitope clusters was synthesized as peptides and assessed for HLA-DR binding in vitro. Further, we characterized the functional capacity of these peptides to prime and activate human peripheral blood mononuclear cells (PBMCs), by monitoring cytokine response profiles using MIMIC® technology (Modular IMmune In vitro Construct). Utilizing this decision framework, we found sufficient differential cellular activation and cytokine profiles among HLA-DR-matched PBMC donors to downselect class II epitope clusters for inclusion in a vaccine targeting PfCSP. Importantly, the downselected clusters are not highly conserved across PfCSP variants but rather, they overlap a hypervariable region (TH2R) in the C-terminus of the protein. We recommend assessing these class II epitope clusters within the context of a PfCSP vaccine, employing a test system capable of measuring immunogenicity across a broad set of HLA-DR alleles.

Identification, Selection and Immune Assessment of Liver Stage CD8 T Cell Epitopes From Plasmodium falciparum.

Immunization with radiation-attenuated sporozoites (RAS) has been shown to protect against malaria infection, primarily through CD8 T cell responses, but protection is limited based on parasite strain. Therefore, while CD8 T cells are an ideal effector population target for liver stage malaria vaccine development strategies, such strategies must incorporate conserved epitopes that cover a large range of class I human leukocyte antigen (HLA) supertypes to elicit cross-strain immunity across the target population. This approach requires identifying and characterizing a wide range of CD8 T cell epitopes for incorporation into a vaccine such that coverage across a large range of class I HLA alleles is attained. Accordingly, we devised an experimental framework to identify CD8 T cell epitopes from novel and minimally characterized antigens found at the pre-erythrocytic stage of parasite development. Through in silico analysis we selected conserved P. falciparum proteins, using P. vivax orthologues to establish stringent conservation parameters, predicted to have a high number of T cell epitopes across a set of six class I HLA alleles representative of major supertypes. Using the decision framework, five proteins were selected based on the density and number of predicted epitopes. Selected epitopes were synthesized as peptides and evaluated for binding to the class I HLA alleles in vitro to verify in silico binding predictions, and subsequently for stimulation of human T cells using the Modular IMmune In-vitro Construct (MIMIC®) technology to verify immunogenicity. By combining the in silico tools with the ex vivo high throughput MIMIC platform, we identified 15 novel CD8 T cell epitopes capable of stimulating an immune response in alleles across the class I HLA panel. We recommend these epitopes should be evaluated in appropriate in vivo humanized immune system models to determine their protective efficacy for potential inclusion in future vaccines.

Synthesis of insulin pump controllers from safety specifications using Bayesian model validation.

Insulin pump controllers seek to alleviate the chronic suffering caused by diabetes that affects over 6% of the world population. The design of control laws for insulin pump controllers has been well studied. However, the parameters involved in the control law are difficult to synthesize. Traditionally, ad hoc approaches using animal models and random sampling have been used to construct these parameters. We suggest a synthesis algorithm that uses Bayesian statistical model validation to reduce the number of simulations needed. We apply this algorithm to the problem of insulin pump controller synthesis using in silico simulation of the glucose-insulin metabolism model.

Optimization of a dendritic cell-based assay for the in vitro priming of naïve human CD4+ T cells.

Methods to prime human CD4(+) T cells in vitro would be of significant value for the pre-clinical evaluation of vaccine candidates and other immunotherapeutics. However, to date, there is no reliable method for the induction of primary human T cell responses in the laboratory. Here, we optimized a culture strategy incorporating highly purified lymphocytes and dendritic cells, in the absence of any exogenous growth factors, for the in vitro sensitization of naïve CD4(+) T cells against a variety of protein antigens. This fully autologous approach, which was superior to the more traditional PBMC assay for supporting the induction of primary human T helper cell responses in culture, elicited effector cells capable of producing a variety of Th cytokines, including IFNgamma, TNFalpha, IL-2, IL-5, IL-17 and IL-21, and memory cells that could be restimulated multiple times with a specific antigen. Through simple modifications to this culture method, we evaluated the role of dendritic cell maturation state and regulatory T cells on the sensitization of naïve T helper cells, which highlights its utility for addressing basic questions of human immunobiology. Finally, using the formulated yellow fever vaccine, YF-VAX (R), we provide a proof-of-concept demonstration of the utility of the system for evaluating the T cell immunogenicity of vaccine candidates in a pre-clinical setting.

In vitro antibody response to tetanus in the MIMIC system is a representative measure of vaccine immunogenicity.

In response to the recurrent failure of animal vaccine protection studies to accurately predict human trial results, we have developed a fully human modular immune in vitroconstruct (MIMIC) to serve as a preliminary screen for efficacy testing of potential vaccine formulations. To validate the potential of this approach, we monitored the in vitro-generated tetanus (TT)-specific antibody levels in a cohort of donors before and after receiving tetanus vaccination. Purified CD4_T cell and B cell populations were combined with autologous tetanus vaccine-pulsed dendritic cells to generate specific antibody. Enumeration of TT-specific IgG antibody-secreting cells by ELISPOT displayed a significant increase in the magnitude of this population after vaccination. The relative magnitudes of the in vitro-generated TT-specific antibody response before and after vaccination largely recapitulated the TT-specific IgG serum titer profiles measured in the same individuals. These findings provide evidence that the MIMIC system can be a rapid and representative in vitro method for measuring vaccine immunogenicity via induction of the memory B cell response.