A Sjogrens Syndrome Diagnostic

描述（由申请人提供）：Sj" green 's Syndrome （SS）是一种自身炎症性疾病，以大量症状为特征，包括眼睛和嘴巴干燥。尽管这种情况相对常见，在美国有多达310万人受到影响，但准确的诊断一直很复杂，而且在很大程度上是无效的。如果不进行诊断和治疗，SS可能会发展为显著的发病率，并使某些癌症的风险增加100倍。因此，SS患者的识别是一个迫切的未满足的需求。目前的SS诊断工具有限，部分原因是症状范围广泛，并且经常与其他自身炎症性疾病重叠。在这项研究中，我们将使用外周血白细胞全基因组表达谱结合多变量分析来鉴定SS诊断性疾病的生物标志物。我们的初步数据支持这样的假设，即这些生物标志物可以被识别出来，正确地对患者进行分类，并为当前的SS诊断方法提供有力的辅助。我们的初步数据表明，基于多变量基因表达的算法在区分SS、类风湿关节炎（RA）和未受影响的对照组方面比目前的诊断实验室测试具有更好的性能。在本提案中，将开发一种算法，利用标准SS差异中来自其他常见风湿病的更广泛的样本集。多变量算法的开发将使用最先进的多变量统计，利用充分的动力队列，以及通过独立队列中的串行关联分析进行生物标志物识别和验证的迭代过程。我们的团队利用这一通用方法创建了与临床结果高度相关的多变量生物标志物算法，并将其转化为临床实验室测试。

DESCRIPTION (provided by applicant): Sj"gren's Syndrome (SS) is autoinflammatory condition, characterized by a plethora of symptoms, including dry eyes and mouth. Despite the fact that the condition is relatively common, with up to 3.1 million persons affected in the US, accurate diagnosis has been complex and largely ineffective. If left undiagnosed, and thus untreated, SS can progress to significant morbidity and increase the risk of certain forms of cancer up to 100 fold. The identification of SS patients is, therefore, a pressing unmet need. Current SS diagnostic tools have been limited, in part because of the wide range of symptoms and frequent overlap with other autoinflammatory disorders. In this study, we will use genome-wide expression profiling of peripheral blood leukocytes coupled with multivariate analysis to identify SS diagnostic biomarkers of disease. Our preliminary data support the hypothesis that such biomarkers can be identified that correctly classify patients and provide a powerful adjunct to current SS diagnostic methods. Our preliminary data demonstrate that a multi-variate gene expression-based algorithm has superior performance to distinguish among SS, rheumatoid arthritis (RA), and unaffected controls than current diagnostic laboratory tests. In this proposal, an algorithm will be developed utilizing a broader set of samples from additional common rheumatologic diseases within a standard SS differential. Multi-variate algorithm development will be done using state-of-the-art multi-variate statistics, utilizing adequately powered cohorts, and an iterative process of biomarker identification and verification though serial associative analyses in independent cohorts. Our group has utilized this general methodology to create multivariate biomarker algorithms highly correlated to clinical outcome, which have been translated into clinical laboratory tests.