Objective To investigate the core symptom clusters and their network structure characteristics in patients with primary brain tumors (PBT), and to provide a reference for the development of precise symptom management strategies. Methods Conve-nience sampling was adopted to recruit 300 patients with PBT.A survey was conducted using the M.D. Anderson Symptom Inventory-Brain Tumor Module.Subsequently, exploratory factor analysis was employed to extract symptom clusters, and R software was utilized to construct a symptom network model.Key indicators including node strength, bridge strength, and predictability were calculated for further analysis. Results Exploratory factor analysis extracted three symptom clusters:the somatic-emotional symptom cluster, neurocognitive symptom cluster, and gastrointestinal symptom cluster.Network analysis revealed that distress (rs=1.63), nausea (rs=1.41), and drowsiness (rs=1.40) were core symptoms; difficulty in comprehension (rE=1.32), distress (rE=1.11), and drowsiness (rE=1.09) were the most influential symptoms; limb weakness (rbs=0.58), sadness (rbs=0.55), nausea (rbs=0.51), and amnesia (rbs=0.50) served as key bridge symptoms.Distress exhibited the highest predictability (78.1%), and the centrality indicators of the network model demonstrated good stability. Conclusion The symptom burden of patients with PBT presents a network structure centered on the somatic-emotional symptom cluster.Clinical management should shift toward precision stratified intervention based on the symptom network:group-based intervention targeting core symptoms should be implemented to alleviate the overall burden, and targeted blocking of bridge symptoms should be conducted to prevent cross-cluster symptom spread.