Diagnostic model and assay for personalized vaccine

Seasonal influenza poses a major global public health challenge, causing 3-5 million cases of severe illness and 290,000-650,000 deaths annually. Influenza vaccines are crucial for preventing illness, reducing severity of infection, and limiting virus transmission; however, their efficacy varies across populations due to factors such as immune status and pre-existing comorbidities. Immunocompromised patients (e.g. rheumatoid arthritis, HIV or organ transplant recipients) are particularly vulnerable to influenza and consequent complications include hospitalization and death, as their immune systems often fail to mount adequate responses to standard vaccines. Leveraging data from our previous ERC-St grant-funded studies, we analyzed the serological and multi-omics profiles of various cohorts across 5 time points over 4 influenza seasons. This analysis, covering multiple virus strains, identified robust molecular biomarkers linked to vaccine responses, which were validated through wet-lab experiments. Machine learning models based on pre-vaccination biomarkers were able to predict vaccine response in independent samples. Building on these findings, we propose to develop a diagnostic assay to measure these biomarkers and apply our established prediction model to stratify patients based on their responsiveness to the influenza vaccine. Samples from various patient cohorts, provided by our collaboration partners, will be used for verification and validation. Once validated, these biomarkers will be integrated into an innovative, fast, and reliable diagnostic test to predict vaccine responsiveness, through collaboration of an experienced industrial partner. Additionally, our team will create a user-interface app to translate biomarker measurements into diagnostic outcome. By enabling personalized vaccine strategies particularly for patients, this project has the potential to significantly improve influenza vaccine efficacy, reduce disease burden and ultimately save lives.