Quantifying the evolutionary dynamics of extra-chromosomal DNA in human cancers.

癌症是一种基因组疾病。细胞在其遗传密码中积累错误，这可能导致细胞不受控制的增殖、癌症生长和死亡。我们现在明白，癌症的形成和发展是衰老身体中持续进化过程的自然结果。量化这些进化过程，提前获得洞察力，以期改进早期发现和治疗，是癌症研究界的主要目标。在过去的十年里，我们在揭开癌症基因组的进化方面取得了实质性的进展。然而，每当我们接近完全理解时，新的发现就会出现。最新的惊喜是所谓的染色体外DNA(EcDNA)在癌症中的作用。人类健康的基因组被分割成23对染色体。在癌症中，染色体片段可以断裂并形成ecDNA，环状DNA结构，长度在几百到数百万个碱基对之间。在健康细胞中，在细胞分裂时，两个子细胞都继承了一套完整的染色体。这是由每条染色体上称为着丝粒的区域实现的。EcDNA缺乏着丝粒。因此，ecDNA在子代细胞中不是平均分离的，一些癌细胞根本不遗传任何ecDNA。因此，我们认为ecDNA只是扭曲的癌症基因组的随机副产品，没有实际的因果功能。令人惊讶的是，最近的研究表明，情况恰恰相反。环状染色体外DNA(EcDNA)是一些最难治疗的癌症的特征，例如脑癌和肺癌。可检测到ecDNA的患者预后最差，对治疗的反应也较小。然而，人们对肿瘤内ecDNA的进化过程知之甚少。在这里，我建议开发一个框架来量化环状ecDNA在人类癌症中的进化过程。因为ecDNA在细胞分裂过程中不能正确地断言，我们不能仅仅应用现有的工具来测量癌细胞中染色体DNA的进化。我们需要一个新的理论框架和新的计算技术来解释和量化实验和临床观察。该奖学金的主干和我的主要专长是利用数学和计算机模拟对ecDNA进化动力学进行全面的理论理解。在前期工作中，我证明了这在严格的数学分析中是可行的(Pichugin，Huang&Werner，BioRxiv 2019)，这个计划的继续将为我们提供工具，以识别其肿瘤是由ecDNA引起的个别患者，预测ecDNA如何适应治疗，以及如何调整治疗策略以优化反应。此外，将我们的理论与实验数据相结合也很重要。我将继续与英国、美国和德国的专家合作(见所附的支持信)，以便将理论与尖端实验技术相结合(请参阅图表以获得概念证明和初步数据)。其次，在团契期间，我将在我的团队内建立关于ecDNA进化的独立实验的专业知识，这将使我能够发展我自己的干湿能力。我将得到我的导师的密切支持，他们在我们中心内负责癌症基因组学计划，并通过我的学生和技术人员对我的国际合作者的实验室进行交流和访问。总而言之，染色体外的DNA导致了一些最难治疗、预后最差的癌症。然而，人们对ecDNA的进化过程知之甚少。我建议开发数学/计算工具箱来描述ecDNA的进化，并在现有的和新产生的癌症基因组和想象数据中测试这些理论的预测。量化ecDNA进化对于癌症基因组的解释、知情预测和治疗策略的优化设计至关重要。

Cancer is a disease of the genome. Cells accumulate errors in their genetic code, which can cause uncontrolled cell proliferation, cancer growth and death. We now understand that the formation and progression of cancer is a natural consequence of an ongoing evolutionary process in an aging body. It is a major goal of the cancer research community to quantify these evolutionary processes, to gain insight ahead of time in the hope of improved early detection and treatment. We made substantial progress to unravel the evolution of the cancer genome in the last decade. Yet, whenever we were getting close to a complete understanding, new discoveries emerge. The most recent surprise is the role of so called extra-chromosomal DNA (ecDNA) in cancer. The healthy genome of a human is partitioned into 23 pairs of chromosomes. In cancer, fragments of chromosomes can break and form ecDNA, ring like circular DNA structures with a length of a few hundred to millions of base pairs. In healthy cells, upon cell division, both daughter cells inherit a complete set of chromosomes. This is made possible by regions on each chromosome called centromeres. ecDNA lack centromeres. Consequently, ecDNA is not segregated equally amongst daughter cells and some cancer cells do not inherit any ecDNA at all. We therefore thought that ecDNA is just a random by-product of distorted cancer genomes with no actual causal function. Strikingly, recent studies show that the opposite is true. Circular extra-chromosomal DNA (ecDNA) are a feature of some of the most difficult to treat cancers, e.g. in brain and lung. Patients with detectable ecDNA have worst prognosis and are less likely to respond to treatment. Yet very little is known about the process of ecDNA evolution within tumours. Here I propose to develop the framework to quantify the evolutionary process of circular ecDNA in human cancers. Because ecDNA do not properly assert during cell divisions, we cannot just apply established tools that measure the evolution of chromosomal DNA in cancer cells. We need a new theoretical framework and new computational techniques to interpret and quantify experimental and clinical observations. The backbone of the fellowship and my major expertise is to develop a comprehensive theoretical understanding of ecDNA evolutionary dynamics, using mathematics and computer simulations. In preliminary work, I showed that this is feasible within a rigorous mathematical analysis (Pichugin, Huang & Werner, BioRxiv 2019) and the continuation of this programme will provide us tools to identify individual patients whose tumours are caused by ecDNA, predict how ecDNA adapt to treatment and how treatment strategies can be adjusted to optimise response. Furthermore, it is important to integrate our theory with experimental data. I will continue my collaborations with experts in the UK, USA and Germany (see attached support letters) in order to integrate theory with cutting-edge experimental techniques (see Figures for proof of concept and preliminary data). Secondly, during the duration of the fellowship I will establish the expertise for independent experiments on ecDNA evolution within my group that will allow me to develop my own dry-wet capacity. I will be closely supported by my mentors, who run cancer genomics programmes within our Centre and via exchanges and visits of my students and technicians to the labs of my international collaborators. In summary, extra-chromosomal DNA drives some of the most difficult to treat cancers with worst prognosis. Yet, little is known about the evolutionary process of ecDNA. I propose to develop the mathematical/computational tool-box to describe the evolution of ecDNA and test predictions of these theories in existing and newly generated cancer genomic and imagine data. Quantifying ecDNA evolution is essential for the interpretation of cancer genomes, informed forecasting and optimal design of therapeutic strategies.