Abstract: Background　Radiation-induced brain injury is a common and serious complication after whole-brain radiotherapy. Although FLASH radiotherapy has shown a sparing effect on normal brain tissue, the underlying mechanisms remain incompletely understood. Neutrophil extracellular traps (NETs) can amplify inflammatory responses and have been implicated in various neurological disorders; however, their role in radiation-induced brain injury remains unclear.Objective　To investigate the role and potential mechanisms of neutrophil extracellular traps (NETs) in mitigating radiation-induced brain injury (RIBI) following wholebrain irradiation using ultra-high dose rate (FLASH) radiotherapy. Methods　Wild-type C57BL/6J mice and PAD4⁻/⁻ mice each received a single 10 Gy dose of whole-brain X-ray irradiation and were assigned, according to dose rate, to either the conventional dose rate radiotherapy group (conventional dose rate radiotherapy, CONV-RT, 4 Gy/min) or the ultrahigh dose rate radiotherapy group (FLASH-RT, 200 Gy/s). H&E staining, NeuN immunofluorescence, and the Morris water maze test were used to evaluate acute histopathological injury after irradiation, long-term neuronal preservation, and spatial learning and memory function, respectively. MPO/CitH3 immunofluorescence and Western blotting were used to detect intracerebral NETs levels, and IBA-1/CD68 and IBA-1/iNOS immunofluorescence were used to analyze early microglial activation and pro-inflammatory polarization. Primary bone marrow neutrophils were isolated from mice, and NET formation was induced by in vitro irradiation. NET-enriched preparations were then collected, co-cultured with BV2 microglia, and the pro-inflammatory phenotypic conversion of BV2 cells was assessed by flow cytometry. In addition, PAD4⁻/⁻ mice were used to verify the role of NETs in the FLASH effect.Results　At24 hpost-irradiation, both FLASH-RT and CONV-RT induced mild acute damage in the hippocampus, with no significant difference between the groups (P＞0.05). One month post-irradiation, the FLASH-RT group showed greater neuronal preservation in the DG and CA3 regions (P＜0.05 and P＜0.01, respectively) and significantly better spatial learning and memory than the CONV-RT group (P＜0.05). FLASH-RT significantly reduced the number of MPO/CitH3 double-positive structures and CitH3 protein expression in the hippocampus compared with CONV-RT (P＜0.001). Both irradiation regimens induced early microglial activation; however, FLASH-RT attenuated the increase in pro-inflammatory IBA-1 ⁺/iNOS ⁺ microglia at 72h post-irradiation (P＜0.05). In vitro experiments demonstrated that NETs promoted upregulation of CD86 expression in BV2 cells (P＜0.001). PAD4 deficiency inhibited NETs formation, reduced microglial pro-inflammatory polarization (P＜0.05), and narrowed the behavioral differences between FLASH-RT and CONV-RT (P＜0.05). Conclusion　FLASH-RT mitigates long-term neuronal damage and improves cognitive function following whole-brain irradiation in mice. Its neuroprotective effects may involve reduced NETs formation, suppression of secondary microglial pro-inflammatory polarization, and alleviation of neuroinflammation. NETs may contribute to the neuroprotective effects of FLASH-RT and represent a candidate mechanism worthy of further investigation.