Comprehensive Prostate Cancer Characterization by Genomic and Transcriptomic Prof

描述（由申请人提供）： 大约55%的前列腺癌（PCA）具有癌症特异性基因融合。这是一个重要的机会，以开发一种非侵入性的高度具体的测试，早期检测PCA。用于检测TMPRSS 2-ERG基因融合的尿液检测试剂盒的实施已在进行中，证明了其高度特异性。生物标志物开发的合乎逻辑的下一步是探索新的癌症特异性基因融合和体细胞改变。该提案的总体目标是发现和验证可以在尿液中非侵入性检测的PCA特异性生物标志物。我们采用了系统生物学方法，将最先进的RNA测序（RNA-seq）和DNA 6.0单核苷酸多态性（SNP）阵列与新的计算分析方法相结合。在目标1中，我们将用RNA-seq鉴定新的基因融合体，在目标2中，我们将确定与目标1的RNA-seq所确定的基因转录本表达改变相关的体细胞拷贝数改变（获得和损失）。我们将在第1阶段的发现步骤中使用100个冷冻PCA样本和25个配对的良性前列腺组织，以初步发现PCA特异性生物标志物。PCA样本将根据Gleason评分（肿瘤分级）平均分为6（3+3）、7（3+4）、7（4+3）和8-10。我们还将尝试平衡活检时的年龄、PSA水平和肿瘤分期（器官局限性（pT 2）与前列腺外扩展（pT 3））。将评价肿瘤的已知ETS基因融合（例如，TMPRSS 2- ERG、SLC 45 A3-ETV 1等）。我们还将尝试根据目前对基因融合的了解，平衡基于已知融合的样品与不融合的样品。第1阶段的验证步骤将使用FISH进行新型基因融合，使用免疫组化（IHC）和/或FISH进行体细胞改变，以确定组织微阵列上的PCA特异性，其中包含200多个PCA样本和400个良性组织。我们还将探索PCA特异性在100个尿液样本收集的男性和不PCA使用严格的EDRN协议对当地样本。将使用PSA的mRNA作为内部对照，检测融合基因和体细胞改变基因的过度表达或表达不足的RT-PCR试验。生物统计小组将监督分析的所有步骤，并与计算生物学小组一起帮助优先考虑主要候选人。我们预计每年从每个生物标志物类别中提名20个候选物，并将移交5-10个用于哈佛/密歇根州/韦尔康奈尔的EDRN临床验证中心的正式2期验证（超出本申请范围）。本提案中的发现将成为早期检测研究网络PCA计划临床验证的有力候选者。这项研究还将对PCA的分子生物学产生重要的见解，这对理解疾病进展和确定潜在的药物靶点具有广泛的意义。最后，计算工具和数据集将成为整个研究界的重要资源。

DESCRIPTION (provided by applicant): Approximately 55% of prostate cancers (PCA) harbor cancer specific gene fusions. This represents an important opportunity to develop a non-invasive highly specific test for the early detection of PCA. Implementation of a urine assay to detect the TMPRSS2-ERG gene fusion is already underway, demonstrating a high specificity. A logical next step in biomarker development is the exploration of novel cancer specific gene fusions and somatic alterations. The overall goal of this proposal is to discover and validate PCA specific biomarkers that can be non- invasively detected in the urine. We have taken a Systems Biology approach integrating state-of-the-art RNA-sequencing (RNA-seq) and DNA 6.0 single nucleotide polymorphism (SNP) arrays with novel computational approaches for analysis. In Aim 1, we will identify novel gene fusions with RNA-seq and in Aim 2 we will determine somatic copy number alterations (gains and losses) associated with altered gene transcript expression as determined by RNA-seq from Aim 1. We will use 100 frozen PCA samples and 25 paired benign prostate tissues in the discovery step of Phase 1 for initial discovery of PCA specific biomarkers. The PCA samples will be equally divided based Gleason Score (tumor grade), 6 (3+3), 7 (3+4), 7(4+3), and 8-10. We will also attempt to balance for age at time of biopsy, PSA levels, and tumor stage (organ confined (pT2) versus extraprostatic extension (pT3). Tumors will be evaluated for known ETS gene fusions (e.g., TMPRSS2- ERG, SLC45A3-ETV1, etc.). We will also attempt to balance samples based on known fusions versus no fusions based on the current knowledge of gene fusions. The verification step of Phase 1 will occur using FISH for the novel gene fusions and immunohistochemisty (IHC) and/or FISH for somatic alterations to determine PCA specificity on tissue microarrays with over 200 PCA samples and 400 benign tissues. We will also explore PCA specificity in 100 urine samples collected from men with and without PCA using a rigorous EDRN Protocol on local samples. RT-PCR assays for fusion genes and over or under expression of somatically altered genes will be tested using mRNA for PSA as an internal control. The biostatistical group will supervise all steps of the analysis and help prioritize the lead candidates in conjunction with the computational biology group. We anticipate nominating 20 candidates from each biomarker category per year and will hand off 5-10 for formal Phase 2 validation by the EDRN Clinical Validation Center of Harvard/Michigan/Weil Cornel (outside the scope of this application). Discoveries made in this proposal will be strong candidates for clinical validation as part of the Early Detection Research Network PCA program. This study will also generate important insight into the molecular biology of PCA, which has broad implications in understanding disease progression and identifying potential drug targets. Finally, the computational tools and dataset will be important resources for the research community at large.