Novel approaches to predict strength and breadth of influenza vaccine response

About This Grant

PROJECT SUMMARY Influenza vaccines remain the most cost-effective tool for reducing infection and disease burden. However, despite decades of research and development, the protection they provide is often suboptimal. A key challenge is the wide variability in vaccine induced immune responses among individuals, which is poorly understood. Addressing this knowledge gap is crucial for improving vaccine efficacy and vaccination strategies. Predictive modeling, with its demonstrated success in advancing medical research and public health, holds significant promise in this regard. By harnessing predictive models, we can tailor vaccination strategies to individuals, optimizing protection and improving outcomes. Current predictive models, however, suffer from critical limitations. Most rely solely on static, pre- vaccination data and focus narrowly on antibody responses to a single vaccine component (in current vaccines, these are H3N2, H1N1 and one or two of the B lineages). Existing models do not account for the dynamic nature of the immune system or the role of heterologous, non-vaccine-specific antibody responses, reducing their accuracy and practical utility. These shortcomings hinder the development of more comprehensive, adaptable models for predicting vaccine efficacy across diverse populations and viral strains. This project aims to overcome these limitations by developing and validating advanced predictive models for influenza vaccine responses. Our approach integrates systematically collected, longitudinal data with state-of-the-art statistical and machine-learning methods. Specifically, we will: 1. Develop predictive models that incorporate long-term antibody response trajectories and heterolo- gous strain data to improve predictions for vaccine strain responses. 2. Expand the scope of these models to predict heterologous breadth and overall antibody responses, offering a more complete understanding of vaccine-elicited immunity. 3. Generate and analyze high-throughput antibody landscape data to further refine and enhance our predictive models. By combining these innovations, we aim to establish a new framework for individualized vaccine re- sponse prediction. This framework will significantly improve upon existing models, enabling tailored vac- cination strategies designed to optimize protection for each individual. Ultimately, this work will provide a robust scientific foundation for guiding future influenza vaccination efforts, contributing to better public health outcomes worldwide.