Abstract
BACKGROUND: Brain metastases (BMs), among the most prevalent intracranial malignant tumors, frequently induce acute symptoms such as headaches and motor deficits due to elevated intracranial pressure. Although hippocampal-avoidance whole brain radiotherapy (HA-WBRT) has demonstrated superior preservation of cognitive function compared to standard WBRT-despite comparable intracranial progression-free survival and overall survival outcomes-its clinical application remains restricted by the complexity and time-intensive nature of conventional planning and delivery protocols. These limitations hinder its utility as an emergent intervention for patients requiring urgent symptom relief. PURPOSE: To address this challenge, we developed a novel HA-WBRT-specific all-in-one (AIO) radiotherapy integrated with online adaptive techniques, and systematically evaluated its operational feasibility, dosimetric precision, and therapeutic effectiveness for emergency HA-WBRT in patients with BMs through preliminary clinical outcomes and related findings. METHODS: The HA-WBRT-specific AIO workflow featured by rapid auto-segmentation of the target and hippocampal, HA-constrained planning using volumetric modulated arc therapy (VMAT) with three coplanar lock-field full arcs ache ensuring a high first-pass success rate for treatment plans, and real-time in vivo quality assurance (QA) enabling safe beam delivery without delay was applied to the eleven emergency patients with BMs receiving HA-WBRT during their initial treatment fraction. Rotational setup errors in subsequent fractions were documented. For these fractions, online adaptive radiotherapy (ART) was implemented using the AIO-generated plan as the baseline. Time requirements for each workflow step and in vivo 3D gamma passing rates were analyzed to evaluate efficiency and dosimetric accuracy. Dosimetric differences between the ART-optimized plan (ART-Plan) and the image-guided radiotherapy plan (IGRT-Plan) were compared to quantify the impact of rotational deviations on dose distribution. Cognitive function (via the Mini-Mental State Examination [MMSE]) and functional independence (via the Basic Activities of Daily Living, BADL) were assessed before and after radiotherapy. RESULTS: The average duration of the AIO radiotherapy was 24.3 ± 0.6 min. The majority of AIO-Plan parameters satisfied RTOG 0933 criteria. For the AIO-Plans, in vivo 3D γ pass rates exceeded 94% (3 mm/3% gamma criteria with a 10% dose threshold). Recorded rotational errors in emergency patients exhibited irregular and substantial variations, highlighting the need for online adaptation. Notably, 94% of fractions required adaptation to meet the target coverage (TC) criteria specified by NRG CC001. All ART-Plans demonstrated 3D γ pass rates exceeding 93% (3 mm/3% gamma criteria with a 10% dose threshold). The average duration of ART was 22.9 ± 1.6 min. Post-radiotherapy, both MMSE and BADL scores showed measurable improvements, while acute symptoms were effectively controlled and alleviated. CONCLUSIONS: The AIO workflow for managing emergency patients with BMs is feasible and safe, facilitating prompt initiation of highly efficient radiotherapy during the first treatment session, eliminating delays. Subsequent online ART effectively corrects rotational deviations, ensuring adequate prescription dose coverage of target volumes while optimizing dose distribution to normal tissues. HA-WBRT incorporating the AIO workflow with online adaptation demonstrates potential for timely control and alleviation of acute symptoms.