Identification of a diagnostic gene signature for the differentiation of the organ of origin for adeoncarcinomas at the esophagogastric junction (AEG).

在目前的第7版TNM分类中，食管胃交界处（AEG）的UICC腺癌被视为一个单一的实体，导致统一的治疗算法。起源器官是远端食道或近端胃。流行病学资料指出，由于影响这种恶性疾病预后的不同生物学危险因素，每个器官的致癌机制各不相同。到目前为止，器官起源的分类仅通过某些形态学特征来完成，如男性肿瘤肿块的定位或替代物，如巴雷特化生的存在。一个有效的标记面板，允许区分食管或胃源性肿瘤，不仅有助于流行病学研究评估AEG患者的个体风险特征，而且可以进一步为个性化治疗算法的未来方法提供背景。基于组织的阵列技术的引入使得同时分析大量不同基因的差异表达成为可能。为了评估和进一步评价由这些技术产生的计算数据，需要使用生物统计方法。对共表达网络中基因-基因相互作用的系统评估使分层成为功能模块，这些功能模块根据功能和形态标准分类为明确定义的术语（基因本体术语）。因此，不仅有可能通过器官起源来鉴别癌症的特定基因特征，而且还可以通过功能标准来选择这些基因，以确定候选基因，这些基因是治疗化合物的假定靶点。对于初级生物统计分析，并不强制生成从头阵列数据。还可以访问公开可用的、已经发布的数组数据集。建议的项目包括三个主要部分。最初，通过生物统计学方法评估公开可用的数据集，以证明食管癌和胃腺癌的差异基因表达。然后，这些数据被用来生成一个基因标记，通过基于组织的阵列评估来区分AEG的起源位置。第二步包括在一小部分患者中对该基因特征进行原理验证，然后在AEG患者的大型独立队列中对该特征进行主要验证。在第三步中，所包括的靶基因进一步表征其分子功能，以阐明其作为治疗靶点的假定作用。在体外环境中，每个基因对相关信号通路的调节作用将被评估。这主要包括调控途径，负责细胞增殖和分化。

In the current 7th edition of the TNM classification of the UICC adenocarcinomas at the esophagogastric junction (AEG) are regarded as one single entity resulting in uniform treatment algorithms. Organ of origin is either the distal esophagus or the proximal stomach. Epidemiological data point towards distinct mechanisms of carcinogenesis for each organ due to different biological risk factors that influence the prognosis of this malignant disease. Up to today, classification of the organ of origin is only done by certain morphological features like localisation of the man tumor mass or by surrogates like presence of Barretts metaplasia. A valid marker panel that allows the differentiation of tumors of esophageal or gastric origin, would not only facilitate epidemiological studies assessing the individual risk profile for AEG patients but could furthermore deliver the background for future approaches of individualized therapeutic algorithms.The introduction of tissue-based array-techniques enables the simultaneous analysis of the differential expression of a high number of different genes. For assessment and further evaluation of the by these techniques generated data computational, biostatistical methods a required. The systematic assessment of gene-gene-interactions in co-expression networks enabled the stratification into functional modules which are classified by functional and morphological criteria into clearly defined terms (gene ontology terms). Thus, it is not only possible to identify a specific gene signature that differentiates carcinomas by there organ of origin, but also to select these genes by functional criteria in order to identify candidate genes that are putative targets for therapeutic compounds. For the primary biostatistical analysis, it is not mandatory to generate de novo array data. One can also take access publicly available, already published array data sets.The proposed project comprises three main parts. Initially, publicly available data-sets are evaluated by biostatistical methods to demonstrate differential gene expression in adenocarcinomas of esophageal and gastric origin. Then, these data are used to generate a gene-signature that allows the differentiation of AEG by the site of origin via tissue-based array assessment. The second step includes the proof-of-principle confirmation of this gene-signature in a small subset of patients, before the main validation of the signature is performed in a large, independent cohort of patients with AEG.In a third step, the included target genes are further characterized concerning there molecular function in order to elucidate there putative role as therapeutic target. In an in vitro setting the modulatory effect of each gene on relevant signalling pathways will be assessed. This includes mainly regulatory pathways, responsible for cell proliferation and differentiation.