基于转录组测序大数据的肺癌PDX模型评估方法研究

PDX (Patient-derired Xenograft) is now an important method to study cancer. Compared with cell line model, PDX holds better clinical and molecular pattern and can more accurately reflect the drug response of the tumor. Lung cancer is one of the most popular and deadly cancer in China. Establish lung cancer PDX model is important for the precision therapy of lung cancer. But the time of establish a PDX model is long, the success rate is low and it is difficult to study metastasis. Therefore, we want to find the biomarker than can evaluate the effectiveness of PDX model and predict the metastasis of lung cancer by integrating the transcriptomic data of PDX tissues, the lung cancer tissues, adjacent tissues and various human tissues. First, we will compare the transcriptomic data of lung cancer and adjacent tissues to build a lung cancer predictor. Such predictor can evaluate how well the PDX models represent the lung cancer tissues. Then, we will compare the human tissues and identify the tissue specifically expressed genes. By calculating the similarity between PDX model and human tissues, we can predict which tissues the tumor may metastasis to. The method of PDX model evaluation will enable the personalized drug response study, prevent metastasis and prolong survival.

病人源性异种移植培养（Patient-derired Xenograft，PDX）技术是目前研究肿瘤的重要手段。与细胞系模型相比，PDX更好的保留了肿瘤组织的临床形态及分子生物学特征，更能准确的反映肿瘤组织对药物的反应。肺癌是我国高发病率和死亡率的癌症，建立肺癌的PDX模型对于肺癌的精准治疗意义重大。但是PDX模型存在建模时间长，成功率低，难以研究转移三大问题。因此，我们希望通过整合PDX组织，肺癌和癌旁组织，各种人体组织的转录组数据，找到能够评估PDX模型有效性和转移倾向性的生物标志物。首先，我们基于肺癌和癌旁组织转录组数据，构建预测肺癌的方法，评估PDX模型能够在多大程度上代表肺癌组织。然后，我们从人体组织数据中识别组织特异表达的基因，通过PDX模型与人体组织的相似性，预判肿瘤可能向哪些组织转移。本课题希望建立的PDX模型评估方法将有助于实现病人个体水平的药效研究，预防转移，延长生存。

病人源性异种移植培养（Patient-derived Xenograft，PDX）技术是目前研究肿瘤的重要手段。与细胞系模型相比，PDX更好的保留了肿瘤组织的临床形态及分子生物学特征，更能准确的反映肿瘤组织对药物的反应。但是PDX模型存在建模时间长，成功率低，难以研究转移三大问题。因此，我们希望通过整合转录组数据，对PDX模型进行系统深入的研究，评估PDX模型的有效性。首先，我们比较了人体25个不同组织的基因表达差异，构建了组织预测模型，该模型的预测准确率高达99.4%；然后，我们比较了人体内肿瘤组织和移植到小鼠体内后肿瘤组织的表达模式差异，证明了PDX保留有原肿瘤组织的绝大部分癌变信息，PDX模型可以有效的代表原肿瘤组织；然后，我们比较了不同部位肿瘤的PDX模型，证明了PDX保留有原肿瘤部位的组织特异性信号，PDX模型可以区分不同部位的肿瘤；最后，通过比较原发灶和转移灶肿瘤移植后的PDX模型，证明了PDX模型可以用于肿瘤转移研究。本课题证明了PDX模型的有效性，有助于实现病人个体水平的药效研究，预防转移，延长生存。