Development and implementation of multiplex methods to understand the biology and heterogeneity of patient-derived cancer models

摘要 癌症的实验模型提供了破译癌症的分子基础和发展的手段。 新的治疗剂。到目前为止，大多数癌症研究都采用了已建立的癌细胞系和 转基因小鼠模型。尽管这些模型提供了对许多 癌症发生和发展的各个方面，这些模型中的每一个都有重要的局限性，包括适应性 对培养(细胞系)，缺乏基因组不稳定(小鼠模型)，以及不能充分代表光谱 人类癌症的突变和亚型。下一代癌症模型(NGCM)，如有机化合物 最近已经开发出了一些模型。NGCM解决了以前模型的许多缺陷，并承诺加快 癌症研究和实验性治疗努力。 最近的方法学进步使创造患者来源的癌细胞株和 效率更高的有机化合物。当与基因组分析相结合时，这些新模型可能会促进新的 对人类癌症的洞察。然而，有机化合物需要复杂的培养条件，并表现出明显的 这些特性对标准分子和细胞生物学技术的实施构成了挑战。至 促进有机化合物模型在研究界的广泛使用，我们必须开发创新的 克服这些挑战的技术，使针对一系列癌症表型的有机化合物研究成为可能。 在这个项目中，我们将基于我们在基因组规模和信息学方法发展方面的专业知识，如 以及我们派生许多HCMI模型的工作，目的是开发高吞吐量方法来 在由人类癌症模型创建的患者衍生的有机化合物中进行遗传和小分子筛选 主动性(HCMI)。此外，我们将使用创新的方法来询问细胞状态的可塑性和异质性 在这些模型中。这些研究将使癌症研究界能够同时进行高吞吐量和低吞吐量 在患者衍生模型中的分析，并提供对稳定性和表型的深入洞察 这些模型。虽然我们将集中精力使用胰腺癌有机化合物进行技术开发，但我们 预计本提案中开发的方法将广泛适用于以下许多不同的模型 一系列癌症类型。 在目标1中，我们将开发和实现一种高度多元化的方法来筛选患者衍生的有机化合物。 既有小分子又有遗传试剂的模型。这些研究将提供一种强有力的方法来 在高吞吐量下询问HCMI模型。在目标2中，我们将以我们的初步研究为基础，表明 患者来源的有机化合物表现出异质性和表达表型的快速变化。我们会审问 这些状态的动态变化并评估这些模型中的异质性程度 制定了物理和测序方法。在目标3中，我们将建立在阿喀琉斯项目和深度地图的基础上 (www.DepMap.org)创建和实现优化的基因组规模CRISPR-Cas9文库，该文库允许 对患者衍生的有机化合物中的遗传依赖性进行系统的遗传询问。 我们预计，这些研究将创造新的方法，允许对HCMI模型进行严格评估，如 以及胰腺癌新生物标志物和治疗靶点的发现。更广泛地说，这些 研究将提供关键的原则证据，证明这些方法可以被其他人用来研究特定的 下一代癌症模型中的表型，如有机化合物。

Abstract Experimental models of cancer provide the means both to decipher the molecular basis of cancer and to develop new therapeutic agents. To date, most cancer research has employed established cancer cell lines and genetically engineered mouse models. Although these models have provided tremendous insight into many aspects of cancer initiation and progression, each of these models has important limitations, including adaptation to culture (cell lines), lack of genomic instability (mouse models), and inadequate representation of the spectrum of mutations and subtypes of human cancers. Next generation cancer models (NGCMs) such as organoid models have recently been developed. NGCMs address many deficits of prior models and promise to accelerate cancer research and experimental therapeutic efforts. Recent methodological advances now make it possible to create patient-derived cancer cell lines and organoids with increased efficiency. When coupled with genomic analysis, these new models may facilitate new insights into human cancers. However, organoids require complex culture conditions and display distinct properties that pose challenges for implementation of standard molecular and cell biology techniques. To facilitate widespread use of organoid models within the research community, we must develop innovative technologies to overcome these challenges and enable study of organoids for a range of cancer phenotypes. In this Project, we will build on our expertise in the development of genome scale and informatic methods as well as our work to derive many of the HCMI models with the goal of developing high throughput approaches to perform genetic and small molecule screens in patient-derived organoids created by the Human Cancer Models Initiative (HCMI). In addition, we will use innovative methods to interrogate cell state plasticity and heterogeneity in these models. These studies will allow the cancer research community to perform both high and low throughput analyses in patient-derived models and to provide deep insight into the stability and phenotypes represented by these models. While we will focus our technology development efforts using pancreatic cancer organoids, we anticipate that the approaches developed in this proposal will be widely applicable to many different models from a range of cancer types. In Aim 1, we will develop and implement a highly multiplexed method to screen patient-derived organoid models with both small molecules and genetic reagents. These studies will provide a powerful approach to interrogating HCMI models at high throughput. In Aim 2, we will build on our preliminary studies that indicate that patient-derived organoids exhibit heterogeneity and rapid shifts in expressed phenotypes. We will interrogate the dynamics of these state changes and assess the degrees of heterogeneity in these models using newly developed physical and sequencing methodology. In Aim 3, we will build on Project Achilles and the DepMap (www.DepMap.org) to create and implement an optimized genome scale CRISPR-Cas9 library that permits the systematic genetic interrogation of genetic dependencies in patient-derived organoids. We anticipate that these studies will create new methods that permit rigorous evaluation of HCMI models as well as the discovery of novel biomarkers and therapeutic targets in pancreatic cancer. More broadly, these studies will provide critical proof of principle that these methods can be used by others to study specific phenotypes in next generation cancer models such as organoids.