Disease Progression Modeling of Bladder Cancer

项目摘要/摘要 癌变可以被视为一个多步骤的进化过程，其特征是积累 遗传和表观遗传改变，由微环境施加的选择压力驱动。这个 肿瘤进化的描绘将提供对肿瘤生物学的宝贵见解，并为 改进的诊断学、预后学和靶向治疗的发展。 时间序列数据是推导动态级数模型的理想选择，但这是不可能收集的 人类癌症，因为需要及时的手术干预和系统治疗，这改变了 疾病的自然病史，并施加影响肿瘤进化的选择压力。要克服这个问题 人类序列抽样问题，我们设计了一种计算策略来理解癌症的进化 从“静态”样本(切除的组织样本)中得出伪时间序列数据。该设计基于 每个样本可以提供疾病过程的快照的原理，以及如果样本的数量 足够大，我们可以恢复疾病进展的可视化。我们演示了 开发了一种称为CancerMapp的管道，将其应用于分析来自 超过9000个乳房组织样本。乳腺癌进展模型确定了两个主要轨迹 恶性肿瘤--早期分裂为基底肿瘤，并通过管腔肿瘤形成连续体。计算性的 方法和乳腺癌模型概念已经在独立研究中得到验证，我们的 这些发现为我们研究所的一些调查以及 菲尔德。 建立在我们以前工作的逻辑基础上，我们现在提出一个大规模的跨学科研究计划，以得出一个 膀胱癌进展模型(BLCA)。胆囊癌是五种最常见的恶性肿瘤之一。 全世界。仅在美国，2018年的新增病例估计为72,500例，估计死亡人数超过 15,000人，预计在不久的将来还会增加。BLCA分类为多个 分子亚型是最近提出的，有可能影响临床治疗。 然而，显著的生物亚群异质性仍然存在，在统一之前还需要更多的工作 分类系统可以得到广泛的接受。更重要的是，到目前为止，人们还不了解 子类型之间的相互关系。对亚型如何相关以及癌症如何进化的见解 对观察到的分子病理表型变化的影响是下一步需要进行的分析 也是这项提案的重点。 这项拟议的工作将为一系列以前无法实现的研究方向提供信息。这个 BLCA路线图的推导和推动逐步癌症的关键分子事件的识别 进展将为肿瘤生物学提供新的见解，并指导改良癌症的发展 诊断学、预见学和靶向治疗。带注释的进度图也可以指导设计 临床试验和动物研究的重点是癌症发展的关键点，这可能会产生 以有限的资源获得最佳回报。

PROJECT SUMMARY/ABSTRACT Carcinogenesis may be viewed as a multistep evolutionary process characterized by accumulation of genetic and epigenetic alterations, driven by selective pressures imposed by the microenvironment. The delineation of tumor evolution would provide invaluable insights into tumor biology and lay a foundation for the development of improved diagnostics, prognostics and targeted therapeutics. Time-series data are ideal for deriving models of dynamic progression, but this is impossible to collect in human cancer because of the need for timely surgical intervention and systemic therapy, which alter the natural history of the disease and exert selection pressures that affect tumor evolution. To overcome the human serial sampling issue, we have devised a computational strategy to understand cancer evolution by deriving pseudo time-series data from ‘static’ samples (excised tissue specimens). The design is based on the rationale that each sample can provide a snapshot of the disease process, and if the number of samples is sufficiently large we can recover a visualization of disease progression. We demonstrated the utility of the developed pipeline - referred to as CancerMapp - by applying it to the analysis of gene expression data from over 9,000 breast tissue samples. Breast cancer progression modeling identified 2 major trajectories to malignancy – an early split to basal tumors, and a continuum through luminal tumors. The computational approach and the breast cancer model concept have since been validated in independent studies, and our findings have provided the impetus for a number of investigations at our institute and by colleagues in the field. Built logically on our previous work, we now propose a large-scale interdisciplinary research plan to derive a progression model for bladder cancer (BLCA). BLCA is among the five most common malignancies worldwide. In the US alone, new cases for 2018 are estimated at 72,500 with estimated deaths at over 15,000, figures that are anticipated to increase in the near future. Classification of BLCA into multiple molecular subtypes has recently been proposed and has the potential to impact clinical management. Nonetheless, significant biologic subgroup heterogeneity remains, and more work is needed before a unified classification system can gain wide acceptance. More importantly, there is, as yet, no understanding of the inter-relationships between subtypes. Insights into how subtypes are related and how cancer evolution influences the observed changes in molecular pathologic phenotype is the next level of analysis required and is the focus of this proposal. The proposed work will inform a range of research directions that were previously unattainable. The derivation of a BLCA roadmap and the identification of pivotal molecular events that drive stepwise cancer progression will provide new insights into tumor biology and guide the development of improved cancer diagnostics, prognostics and targeted therapeutics. Annotated progression maps can also guide the design of clinical trials and animal studies to focus on pivotal points of cancer development, which may yield the best return with limited resources.