The identification and validation of mechanisms and biomarkers for relapse in diffuse large B-cell lymphoma

 描述（由申请人提供）：弥漫性大B细胞淋巴瘤（DLBCL）是非霍奇金淋巴瘤（NHL）的一种侵袭性形式，是成人中最常见的淋巴瘤亚型，每年有45，000例新发病例。标准治疗包括联合化疗（R-CHOP）。虽然高达60%的患者可以通过化疗治愈，但其余40%的患者通常会看到最初的肿瘤大小减小，但最终会复发化疗耐药疾病。目前尚不清楚一些患者复发而另一些患者没有复发的原因。这些肿瘤进化和适应治疗的机制尚不清楚。目前还没有生物标志物可以预测哪些患者会复发。如果能够找到这样的生物标志物，那么风险较高的患者可以采用更积极的策略进行治疗和/或采用量化微小残留疾病的技术进行更积极的监测。我们假设系统生物学方法可以识别复发的机制和生物标志物。为了检验这一假设，我们将在具有临床注释的DLBCL诊断-复发配对活检的患者的初始队列中进行免疫球蛋白重链（IGH）VDJ测序、外显子组测序、转录组测序和DNA甲基化分析。使用计算分析，我们将识别和验证DLBCL复发特征。我们将在更大的患者队列中验证复发特征的关键改变 使用靶向桑格测序和基于靶向MassArray的甲基化分析。使用我们最近发表的CRISPR-Cas9模型（Kasap et al，Nature Chemical Biology，2014），我们将验证复发标记中的关键基因参与复发相关表型，如化疗耐药性。然后，我们将确定新的生物标志物，在一组初次诊断后至少5年未复发的患者中，使用VDJ测序、外显子组测序、转录组测序和DNA甲基化分析来识别复发风险高的患者。我们将使用设计用于识别此类生物标志物的计算分析来生成候选生物标志物，并使用独立的验证队列对其进行验证。拟定研究是对RFA“用于早期检测造血系统恶性肿瘤的生物标志物”（PA-12-221）的响应。

 DESCRIPTION (provided by applicant): Diffuse large B-Cell lymphoma (DLBCL) is an aggressive form of non-Hodgkin Lymphoma (NHL) and is the most common lymphoma subtype in adults with 45,000 new cases per year. Standard therapy consists of combination chemotherapy (R-CHOP). While up to 60% of patients can be cured with chemotherapy, the remaining 40% usually see initial tumor size reduction but eventually relapse with chemoresistant disease. The reason why some patients relapse while others do not is currently not known. The mechanisms by which these tumors evolve and adapt to treatment are unknown. There are currently no biomarkers that can predict which patients will relapse. If such biomarkers could be found, patients at higher risk could be treated with more aggressive strategies and/or be monitored more aggressively with technologies to quantify minimal residual disease. We hypothesize that a systems biology approach can identify both mechanisms and biomarkers of relapse. To test this hypothesis, we will perform immunoglobulin heavy chain (IGH) VDJ-sequencing, exome-sequencing, transcriptome sequencing and DNA methylation profiling in an initial cohort of patients with clinically annotated diagnosis-relapse paired biopses of DLBCL. Using computational analysis, we will identify and validate the DLBCL relapse signature. We will validate key alterations in the relapse signature in a larger cohort of patients using targeted Sanger sequencing and targeted MassArray based methylation analysis. Using a CRISPR-Cas9 model we recently published (Kasap et al, Nature Chemical Biology, 2014), we will validate the involvement of key genes from the relapse signature in relapse-associated phenotypes such as chemoresistance. We will then identify novel biomarkers that identify patients at high risk of relapse using VDJ-sequencing, exome-sequencing, transcriptome sequencing and DNA methylation profiling in a cohort of patients that have not relapsed at least 5 years after initial diagnosis. We will generate candidate biomarkers using a computational analysis designed to identify such biomarkers and validate them using an independent validation cohort. The proposed study is in response to the RFA "Biomarkers for Early Detection of Hematopoietic Malignancies" (PA-12-221).