Adapting Functional Precision Oncology for pediatric brain cancer

摘要 癌症通常由异质性肿瘤细胞群组成，其特征在于突变， 将每个细胞亚群彼此区分开。父母资助利用DNA测序和一套 重要的新分析算法和可视化工具，以识别癌症患者的突变并跟踪 病人肿瘤在治疗过程中的演变我们已经开发的工具，或者说 开发过程，形成了我们正在实施的功能性精确肿瘤学方法的基础， 犹他州的大学，从母项目中拨出的资金和机构资金相结合。 当前的补充申请将调整母项目资助的工具，以有效用于儿科 癌的本补充提案的主要目标是调整我们的功能性精确肿瘤学 一种为脑肿瘤儿童的治疗选择提供信息的方法， 在没有精确指导的情况下，很难选择合理的治疗方法。有迫不得已 然而，有理由相信这种适应将需要算法修改。这是因为， 我们的方法依赖于肿瘤中的基因组突变，但以前的研究表明， 比成人癌症的突变量要低得多。因此，有必要扩大我们的功能， 在儿科病例中采用这种方法，并使其适应儿科肿瘤中较低的突变负荷。完成这样 适应，我们将，首先，进行功能性药物筛选和基因组/转录组表征在两个 儿科脑癌指数患者，所以我们有适当的测试案例来驱动我们的工具开发和测试。 然后，我们将分析儿科脑肿瘤指数患者数据集，并调整我们的功能精度 用于具有预期较低基因组突变负荷的儿科脑肿瘤的信息学方法。我们预见 发展的两个特定领域：由于儿科病例中的突变数量基本上较低， 我们将整合同时分析体细胞拷贝数变异（CNV）突变的能力， 用体细胞点突变来（1）重建构成肿瘤的基因组亚克隆;和（2）携带 从用于研究的单细胞RNA测序数据中将细胞外分配到基因组定义的亚克隆 肿瘤亚克隆特异性基因表达行为。我们已经证明，CNVs可以用于 肿瘤亚克隆重建和细胞分配，并预期通过整合CNV和点突变- 基于单一工具的分析将允许儿科癌症患者在，或接近， 成人患者可达到的水平。 1

ABSTRACT Cancers are typically composed of heterogeneous populations of tumors cells characterized by mutations that distinguish each cell subpopulation from one another. The parent grant leverages DNA sequencing and a suite of important new analytical algorithms and visualization tools to identify mutations in cancer patients and track the evolution of a patient’s tumor over the course of treatment. The tools we have developed, or are in the process of developing, form the foundation of a functional precision oncology approach we are implementing at the University of Utah with a combination of set-aside funding from the parent project, and institutional funding. The current Supplemental Request will adapt the tools funded by the parent project for effective use in pediatric cancers. The main objective of this Supplemental Proposal is to adapt our functional precision oncology approach to inform treatment selection in children with brain tumors, a patient cohort that experiences extremely dire prognoses in which rational treatment choice is difficult without precision guidance. There are compelling reasons to believe, however, that such adaptation will require algorithmic modifications. This is because key aspects of our approach rely on genomic mutations in the tumor, but previous studies show that pediatric tumors have a substantially lower mutation load than adult cancers. Therefore it is necessary to expand our functional approach in pediatric cases, and adapt them to lower mutation loads in pediatric tumors. To accomplish such adaptation, we will, first, perform functional drug screening and genomic/transcriptomic characterization in two pediatric brain cancer index patients, so we have appropriate test cases drivin our tool development and testing. We will then, second, analyze the pediatric brain tumor index patient datasets, and adapt our functional precision informatics methods for use in pediatric brain tumors with expected lower genomic mutation loads. We foresee two specific areas of development: because of the substantially lower number of mutations in the pediatric cases, we will integrate the ability to simultaneously analyze somatic copy number variation (CNV) mutations, together with somatic point mutations to (1) reconstruct the genomic subclones that make up the tumor; and (2) to carry out cell assignment to the genomically defined sublones from single-cell RNA sequencing data used to study tumor subclone-specific gene expression behavior. We have already shown that CNVs can be used both for tumor subclone reconstruction and cell assignment, and anticipate that by integrating CNV and point mutation- based analyses in a single tool will allow functional precision analysis of pediatric cancer patients at, or close to, the level achievable for adult patients. 1