An automated radiation modulator (ARM) was developed to enable a high-throughput radiation sensitivity test using patient-derived organoids (PDOs) as an in vitro diagnostic device. Treatment strategies for head and neck cancer include surgery and radiotherapy. However, patient responses to radiotherapy vary widely. To overcome limitations in efficiency and scalability, the ARM was developed to provide controlled, reproducible, and high-throughput radiation delivery. Its feasibility was validated by comparing it with conventional radiation methods using two HNSCC cell lines, as well as by applying it to PDO-based radiation sensitivity tests. The ARM successfully classified head and neck cancer PDOs into radiation-sensitive and radiation-resistant groups. OncoSensi, a radiation sensitivity screening method utilizing the ARM, was cross-validated with clinical radiotherapy outcomes, including recurrence status, in 14 patients with head and neck cancer. The multi-parameter OncoSensi model achieved a sensitivity of 80% and a specificity of 75%, demonstrating superior predictive performance compared to the single-parameter model, which yielded a sensitivity of 70% and a specificity of 50%. Statistically significant difference in recurrence-free survival (RFS) was observed between the OncoSensi-sensitive and -resistant groups. Therefore, ARM-based radiation sensitivity screening can serve as a practical tool for implementing precision medicine in radiotherapy for patients with head and neck cancer, ultimately contributing to improved treatment efficacy and patient prognosis. • A novel Automated Radiation Modulator (ARM) was developed to enable high-throughput, reproducible radiation delivery for assessing radiotherapy sensitivity using patient-derived organoids (PDOs). • The ARM-based screening platform, OncoSensi, demonstrated strong predictive power for radiotherapy outcomes in HNSCC patients, with 80% sensitivity and 75% specificity. • OncoSensi effectively distinguished between radiation-sensitive and -resistant PDOs, showing significant correlation with patient recurrence-free survival and supporting its clinical utility in precision radiotherapy.

Particulate matter (PM) is a major component of air pollution and is associated with respiratory diseases and lung dysfunction. Physiologically relevant models are essential for studying PM-induced airway responses in vitro. Conventional two-dimensional (2D) monolayer models comprising only bronchial epithelial cells are limited in their ability to mimic the complex structure of airway tissue. In contrast, three-dimensional (3D) airway organoids contain basal, ciliated, and goblet cells, which better reflect airway architecture. However, the standardization of 3D cultures has been a major challenge. In the present study, to address this, we developed the 3D Airway Organoid-Pillar Dish (AO-PD) model, a platform that enables reproducible formation of standardized 3D airway. We compared this model with a 2D monolayer model under PM exposure. The 3D AO-PD model showed higher cell viability and significant inflammatory and epithelial-mesenchymal transition responses than that of the 2D model. These results indicate that the 3D AO-PD model more accurately reflects PM-induced airway damage and provides greater resistance to environmental insults owing to its pseudostratified epithelial structure and enhanced cell-cell interactions. Thus, the 3D AO-PD model serves as a robust in vitro platform for investigating PM-induced airway toxicity and inflammation-related responses.