Hunting for the perfect shot: Organoid- and AI-based Identification of Oncology Drug-Vaccine Interactions

Infectious diseases caused >2 million deaths in 2019 (pre-COVID-19). Cancer patients die more often from infections than healthy individuals and are less effectively protected by vaccines. Several studies link oncology drugs to altered vaccine responses. While designed to inhibit tumor progression, these drugs also affect immune functions and thus may interfere with vaccine responses – for better or worse. A systematic analysis of this problem has so far been hampered by a lack of suitable models and the low throughput of clinical trials. Currently, it is unknown how most oncology drugs, routinely administered to patients, affect vaccine responses. In OrAIOn, I will fill this knowledge gap and systematically survey how a wide range of oncology drugs impact vaccination outcomes. To this end, I will use two high-throughput (HT) approaches: 1) a human tonsil lymphoid organoid-based screen with a custom library of >400 oncology drugs; 2) a computational screen with >20,000 publicly available drug profiles. To predict the impact of multiple drug combinations on vaccination, I will exploit a recent AI model. Finally, I will use in vitro and in vivo models, patient records, and public human datasets to determine and validate the underlying mechanisms. OrAIOn’s groundbreaking outcome will be the first comprehensive database of validated drug-vaccine interactions with detailed mechanistic insights. I will go beyond the state of the art by pioneering tonsil-based HT screens and yielding >3000 single-cell RNA-seq profiles from perturbed lymphoid tissue cells. My insights will inform follow-up clinical trials, leading to better vaccine recommendations and tailored vaccines. They will also improve anti-cancer vaccine strategies. In the long term, OrAIOn will enable me to realize my vision to provide personalized vaccination strategies to all cancer patients.