Clinical and Translational Radiation Oncology (Mar 2025)
Evaluating the radiosensitivity of the oral microbiome to predict radiation-induced mucositis in head and neck cancer patients: A prospective trial
Abstract
Background: Predicting the occurrence and/or severity of oral mucositis (OM) before commencing radiotherapy (RT) remains very difficult. The aim of this prospective trial was to investigate whether the ex-vivo radiation sensitivity of oral keratinocytes from head and neck (H&N) cancer patients correlates with severe OM. Methods: Oral microbiopsies of healthy gingival mucosa were collected from 63H&N cancer patients undergoing (chemo)RT, of which 58 samples were useable. Keratinocytes from these microbiopsies underwent ex-vivo proliferation, irradiation, and subsequently the cell spreading assay. Tubes with the cell suspension were placed within the irradiation chamber of a 137Cs Gammacell 40 Exactor (Best Theratronics, Canada) and exposed to 0, 2, 4, 6, or 8 Gy at a dose rate of 0.63 Gy min−1. Cell suspension was then immediately pipetted into custom-made polydimethylsiloxane (PDMS) rings.The effect of demographic and clinical parameters on the cell spreading assay were also analyzed. Systematic clinical recording of OM was conducted twice a week by a specially trained examiner. Results: Most patients had node-positive disease and cancer of the oropharynx or oral cavity. The vast majority of patients received adjuvant RT and concurrent chemotherapy. Overall, 34 (58.6 %) participants developed grade 3 OM after a median dose of 32 Gy. No patient experienced a grade ≥ 4 event. There was a correlation between the cell spreading assay area and grade 3 OM (p 0.05 for all). Conclusions: It is necessary to establish reliable predictors of severe OM before treatment in H&N cancer to allow early management of treatment-related sequelae. This prospective trial illustrates that the intrinsic ex-vivo radiosensitivity of oral keratinocytes could be correlated with RT-induced OM in patients with H&N cancer. This novel predictor requires validation in larger prospective cohorts.
