A Retrospective Study to Establish a Prediction Model for Radiation-induced Brain Damage in Nasopharyngeal Carcinoma Patients Based on Pretreatment Imaging and Radiotherapy Data

Radiation therapy has become the preferred treatment for nasopharyngeal cancer due to the sensitivity of nasopharyngeal carcinoma to radiation. However, even with the use of intensity-modulated radiation therapy (IMRT), radiation-induced temporal lobe injury (RTLI) can be a severe complication. Patients with RTLI may experience long-term memory loss, personality changes, physical dysfunctions, and other symptoms, which seriously impair their quality of life and long-term prognosis. Currently, the diagnosis of RTLI primarily relies on clinical symptoms and imaging examinations such as computed tomography (CT) and conventional MRI. However, these methods only enable the diagnosis of RTLI at a late stage when it is irreversible and cannot be effectively treated. Therefore, the early identification or individualized prediction of RTLI after IMRT holds exceptional importance for improving the quality of life in nasopharyngeal carcinoma patients.

The exact mechanism of RTLI remains unclear. Many clinical covariates have been proven to be associated with RTLI in NPC patients, including stage, age, and dosimetric parameters. In addition, it was reported that each patient's temporal lobe exhibits unique genetic susceptibility to radiation exposure. In this study, we aim to predict the occurrence of RTLI by analyzing clinical factors and heterogeneity of temporal lobe tissue prior to irradiation. Finally, we want to construct and validate a prediction model for RLTI, which can support clinician decision-making in developing individualized treatment plans and providing preventive measures.