摘要目的:建立儿童肝母细胞瘤人源性异种移植（PDX）小鼠模型，并对建模成功的PDX肿瘤与患儿的原代肿瘤生物学一致性进行比较，同时比较分析影响PDX模型建模成功的关键因素。方法:通过对39例肝母细胞瘤患儿的新鲜肿瘤组织样本进行PDX小鼠模型的构建，详细记录建模的肿瘤生长时间及体积大小，同时收集39例患儿的临床资料，对实验数据及临床数据进行分析。不同参数的成瘤率差异分析采用 χ2检验(分类变量)，正态分布的连续变量比较应用 t检验。 结果:经过传代及病理诊断确认成功构建21例肝母细胞瘤PDX模型，成功率53.8%（21/39）。建模成功的PDX模型各代次肿瘤样本的病理结果与相应原代肿瘤一致。影响PDX成瘤率主要因素分析显示，原代肿瘤的转移瘤建模成功率高于肝脏原位肿瘤的成功率（7/8比14/31， P = 0.049）；而病理分型对成瘤率无显著性差异。按照原发肿瘤及转移肿瘤的PDX成瘤组比较，2组在成瘤时间及成瘤体积差异均无统计学意义。肝母细胞瘤的PDX小鼠模型移植瘤组织苏木精-伊红染色与原代肿瘤一致；肝细胞抗原（Hepatocyte）、磷脂酰肌醇聚糖3、β-连环蛋白和甲胎蛋白4种蛋白在原代肿瘤组织与PDX小鼠模型的免疫组化的阳性率分别为100%比100%、100%比95.24%、100%比100%、95.24%比85.71%。 结论:成功建立了儿童肝母细胞瘤PDX小鼠模型，成瘤率较高，转移肿瘤成瘤率高于原发肿瘤，而且移植瘤保持了原代肿瘤的生物学特点。

abstractsObjective:To establish a patient-derived xenograft (PDX) humanized mouse model for hepatoblastoma in children. In addition, compare the biological consistency between successfully modeled PDX tumors and primary tumors in children while comparing and analyzing the influence of PDX model modeling success as a key factor.Methods:A PDX tumor model was constructed from fresh tumor tissue samples from 39 children with hepatoblastoma. The tumor growth time and volume size were recorded in detail. Simultaneously, 39 children’s data were collected for experimental and clinical analysis. The difference in tumorigenesis rate between different parameters was analyzed by χ2 test (categorical variable). Continuous variables with a normal distribution were compared using the t-test. Results:After cell passage and pathological diagnosis, 21 cases of hepatoblastoma PDX models were successfully constructed, with a success rate of 53.8% (21/39). Tumor samples from each generation of successfully modeled PDX models had pathology results that were consistent with those of the corresponding primary tumors. The analysis of the key factors affecting the tumor formation rate of PDX revealed that the metastasis rate was more successful in primary tumors than in liver in situ tumors (7/8 vs. 14/31, P = 0.049). However, there was no significant difference between tumor formation rates and pathological subtypes. According to the PDX tumor formation group comparison between the primary tumor and the metastatic tumor, there was no statistically significant difference between the two groups in terms of tumor formation time and tumor volume. Hematoxylin-eosin staining in hepatoblastoma’s PDX mouse was consistent with the primary tumor. Immunohistochemistry positivity rates of four proteins, namely hepatocyte antigen (Hepatocyte), phosphatidylinositol glycan 3, β-catenin, and alpha-fetoprotein, in primary tumor tissues and PDX mouse models were 100% vs. 100%, 100% vs. 95.24%, 100% vs. 100%, and 95.24% vs. 85.71%, respectively. Conclusion:A PDX mouse model for hepatoblastoma has been successfully established in children. The tumor formation rate is high, with metastatic tumors having a higher tumor formation rate than primary tumors and transplanted tumors retaining the biological characteristics of primary tumors.