Variable-interval temporal feathering to optimize organs-at-risk repair for head and neck adaptive radiotherapy

Background and Purpose: Temporally feathered radiation therapy (TFRT) for head-and-neck cancer (HNC) combines variable-dose intensity modulated radiation therapy (IMRT) subplans to increase organs-at-risk's (OAR) recovery. This study evaluated whether incorporating OAR-specific radiosensitivity into variable rest-time planning can reduce side effects. Materials and Methods: Using radiosensitivity profiles of five OARs (larynx, esophagus, parotid, spinal cord, brainstem), we developed an optimized TFRT (oTFRT) model to maximize OAR recovery by optimizing rest times, guided by the linear-quadratic and normal tissue complication probability models. High-dose sparing factors were generated using Monte Carlo simulation. OAR recovery under IMRT and standard TFRT (sTFRT) was compared with oTFRT in two experiments: one with total oTFRT dose matched to IMRT, and one without dose restrictions. Results: Under dose-matched experiment, the oTFRT yielded improved recovery for larynx (88.1% versus 75.9%), parotid (67.5% versus 58.5%), and esophagus (93.6% versus 89.1%), while brainstem recovery (83.9% versus 82.4%) and spinal cord recovery (67.5% versus 65.0%) were slightly higher than IMRT. Under dose-unrestricted experiment, oTFRT improved recovery for brainstem and spinal cord (88.5% and 77.9%, respectively) versus the second-best plan (83.9% and 66.9%), while parotid recovery was slightly higher (67.2% versus 66.7%). The larynx showed improved recovery relative to IMRT (81.7% versus 75.9%) but remained below sTFRT (88.0%), whereas esophagus recovery was lower (84.8% versus 93.6% for sTFRT). Despite delivering higher doses to larynx and esophagus, oTFRT maintained comparable or improved recovery relative to IMRT. Conclusions: By considering nonuniform OARs' radiosensitivity, oTFRT can improve recovery and potentially reduce side effects, warranting further consideration.