An immuno-evolutionary model for optimized design of personalized cancer vaccines.

• Applicant: Dr. Matthijs Meijers
• Subject Area: Immunology ; Biophysics
• Term: since 2026
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 575815076

In the proposed work, I will study the immune pressure on cancer tumors and will build evolutionary models to quantify the effect of the immune pressure on tumor growth. It is the aim to learn and predict the immunogenicity of cancer peptides and the dynamics of the activation and expansion of T-cell clones that recognize cancer neoantigens. Integrating biophysical and statistical models for the immune response with evolutionary models for tumor growth, I will develop a quantitative theory to optimize the vaccine composition of personalized cancer vaccines. This theoretic work will be informed and validated by controlled experiments using mouse models, as well as by data from recent clinical trials with personalized cancer vaccines. The project can be broken down into two main research objectives: 1) Prediction of the T-cell immune response to immune stimulation. I will develop a biophysical and statistical model for the T-cell response to cancer neoantigens. In this model, I will integrate model components for the presentation probability of cancer peptides by the MHC, the likelihood of recognition of neoantigens by the T-cell repertoire, prediction of the TCR – pMHC binding affinity, and the clonal dynamics of T-cells upon activation. The aim is to predict the magnitude and the clonal composition of the responding T cell repertoire. This work will be informed by new experiments on mouse models with tumor injection and immune stimulation, as well as by available published data on validated immunogenic peptides. 2) An evolutionary model of the tumor response informs vaccine composition. I will use fitness modelling to predict tumor growth in the presence of immune pressure. The fitness of a cancer clone will depend on the recognition probability of its presented neoantigens by the responding T-cell repertoire, using the computational framework as established in aim 1. Combining the model for the immune response with evolutionary modeling of tumor growth, I will compute a quantitative score for the vaccine composition design that relies on the expected tumor growth. This quantitative framework for vaccine composition design will be validated using experimental data on mouse models, in which the tumor growth is tracked after immune stimulation. Additionally, the evolutionary model will be validated using clinical trials with personalized anti-tumor vaccines. I aim to predict the immunogenicity of the vaccine and its effect on tumor growth as a function of the peptide composition of the vaccine, the clonal composition of the tumor, and the patient’s HLA type. A predictive model of immune response to the given vaccine composition will allow for proposing mathematically objective criteria for vaccine design.

• DFG Programme: Fellowship
• International Connection: USA
• Host: Professorin Dr. Marta Luksza