精准医疗是根据患者肿瘤的分子学和遗传学特点进行针对性治疗的新兴医疗模式。目前用于癌症研究的模型主要是肿瘤细胞系（Patient-derived cancer cell lines，PDC）和人体肿瘤组织小鼠移植模型（Patient-derived xenografts，PDX）。近年来，体外构建的多种癌症患者来源的肿瘤类器官（Patient-derived organoids，PDO）成为了研究热点。类器官是体外三维（3 dimensional，3D）培养构建的多细胞团，具有自我更新、自我组织能力和维持其来源组织的生理结构和功能特点。PDO预测患者对抗癌药物的反应潜力已经得到很好的验证，但其构建效率低和体外扩增耗时限制了在临床上的应用。针对临床癌组织样本，我们开发了一套双重压滤法，这种方法能够在3天内从肺癌患者手术切除或活检肿瘤组织中构建得到肺癌类器官（Lung cancer organoids，LCOs），LCOs能够重述亲本肿瘤的组织学和基因组学特征，并具有体外无限扩增的潜力。本文中，我们首先对本课题组开发的一款超疏水微坑阵列芯片（Superhydrophobic microwell array chip，SMAR芯片）进行了多种功能性验证，证明该芯片可以作为高通量基因传递平台应用于化学转染、病毒感染和电穿孔转染。通过将病毒包装和靶细胞感染过程高效地集成化到SMAR芯片，证明该平台可在多种类型细胞中进行基因传递和大规模的基因功能研究的潜力。我们将SMAR芯片和与其匹配的电极阵列电穿孔芯片配合使用，对原位生长的细胞进行电穿孔转染以递送不同遗传物质，并利用CRISPR/Cas9基因编辑技术对其进行了功能验证，证明该平台能够在高通量的遗传物质筛选中发挥重要作用。鉴于SMAR芯片的优良特性和构建获得的P0代LCOs，我们搭建了一个利用少量LCOs在一周内可以获得临床上有意义的药物反应的药敏测试平台。数据证明，一周药敏测试结果与PDX体内药敏结果、肿瘤的遗传突变特征以及临床治疗结果吻合度较好。基于双重压滤法和药敏测试平台，我们又初步探索了卵巢癌（Ovarian cancer，OC）PDO的构建和P0代OC-PDO的培养与药敏测试。利用SMAR芯片平台可以进行单一样本和多样本的高通量快速药敏测试，与该芯片耦合使用的PDO模型为预测临床患者的特定药物反应提供了一种技术上可行的手段，有助于促进癌症患者的个性化治疗。
Cancer treatments are increasingly being targeted to specific patient populations based on the molecular and genetic features of their tumor, so called precision or personalized cancer medicine. Preclinical cancer models are essential tools for cancer research, commonly used human cancer models include patient-derived cancer cell lines and primary patient-derived tumor xenografts. Recently developed 3D culture technologies have opened new avenues for the development of novel cancer models in vitro for human diseases, such as patient-derived organoids. The organoids are a multicellular mass constructed by three-dimensional (3D) culture in vitro, which has ability of self-renewal and self-organization and maintains the physiological structure and function of its derived tissue. The potential of patient-derived organoids (PDOs) to predict patients’ responses to anti-cancer treatments has been well recognized, however the lengthy time and the low efficiency in establishing PDOs hamper the implementation in clinics. we developed a sample processing method of mechanical force based double pressure filtration to dissociate the tumor clusters from tumor tissue efficiently, which can be adapted to generate lung cancer organoids (LCOs) from surgically resected and biopsy tumor tissues in three days. The LCOs can recapitulate the histological and genetic features of the parental tumors and have the potential to expand indefinitely.In this thesis, we first performed a variety of functional validation of a superhydrophobic microwell array chip (SMARchip) developed in our group. We verified that the chip can be used as a high-throughput gene delivery platform for chemical transfection, viral infection, and electroporation infection. By integrating the process of virus packaging and target cell infection onto SMARchips efficiently, gene delivery and large-scale gene function studies can be carried out in multiple types of cells. By combining the SMARchip with its matched electrode array electroporation chip, the in situ grown cells can be electroporated and transfected to deliver diverse genetic material. Additionally, CRISPR/Cas9 gene editing techniques was adapted to verify its function and proved that the platform can play an important role in the screening of genetic material.By employing a nanoliter-scale superhydrophobic microwell array chip, we demonstrated hundreds of LCOs at passage 0, a number that can be generated from most of the samples, were sufficient to produce clinically meaningful drug responses within a week on a rapid drug sensitivity test platform. Experimental data proved our one-week drug tests are in good agreement with patient-derived xenografts, genetic mutations of tumors, and clinical outcomes. Based on the processing method and drug sensitivity test platform, we have preliminarily explored the construction, 3D culture and drug sensitivity tests of ovarian cancer organoids. The SMARchip platform can be exploded to performed high-throughput rapid drug sensitivity tests with a single sample and multiple samples, and the PDO model coupled with the microwell device provides a technically feasible means for predicting patient-specific drug responses in clinical settings and facilitates the individualized treatment for patients.
