ASCO 2013：新型生物标记物可用于弥漫大B细胞淋巴瘤分子分型及预后

弥漫性大B细胞淋巴瘤（diffuse large B-cell lymphoma，DLBCL）是最常见的一类淋巴瘤，约占成人非霍奇金淋巴瘤的30% - 40%。DLBCL在形态学特征、细胞遗传学、临床表现及预后上都表现出高度的异质性。

复旦大学附属肿瘤医院杜祥教授通过基因表达谱研究发现，根据肿瘤细胞起源，可将DLBCL分为生发中心起源B细胞亚型（GCB型）和非生发中心起源B细胞亚型（non-GCB型，包括 ABC型和Type-3型）。临床研究证实上述分子亚型是一个独立于常规临床参数的预后因素，GCB型患者的预后显著优于non-GCB型患者。

作为DLBCL分子分型的金标准，基因表达谱分型具有高通量、敏感性好和分类准确的优点，但需同时检测数百个基因的表达、实验技术复杂、对样本要求高、费用昂贵等因素限制了其在临床上的使用。通过免疫组织化学方法检测CD10、BCL6、MUM1等免疫指标可对DLBCL进行分子分型，但少数几个抗原标记无法很好地呈现基因表达谱分析所揭示的DLBCL分子亚型的全貌。免疫组化分型与基因表达谱分型结果存在一定差异。迄今，DLBCL分子分型研究成果仍无法真正应用于临床诊断和评估预后。

研究者通过对1682例DLBCL患者的基因表达谱分析，筛选出对分子分型起关键作用的基因LIMD1和MYBL1，以此建立一个双基因的分子分型模型“L-M指数”。在包括东西方人群、新鲜组织和石蜡包埋组织、不同治疗方案以及多种基因芯片检测平台的研究数据中证实了L-M指数分型结果与金标准具有较好的一致性（分类准确性为87%）。

L-M指数分型结果对于评估患者预后具有指导意义；L-M指数的可靠性在实时荧光定量PCR研究中得到证实。通过对PCR检测系统的进一步优化，有望实现L-M指数从基因芯片平台向常规技术平台的转化，为DLBCL临床诊断和预后提供一种新型的分子标记物。

A two-gene signature for subtype classification and survival prediction in Diffuse Large B-Cell Lymphoma
Background
Diffuse large B-cell lymphoma (DLBCL) is a heterogeneous group of B-cell lymphomas with wide variations in patient survival. Through gene expression profiling, DLBCL can be stratified into two major cell-of-origin subtypes with distinct prognoses: the favorable germinal center B-cell-like (GCB) and the unfavorable activated B-cell-like (ABC) DLBCLs. However, the current cell-of-origin signatures are not suitable for use in routine clinical practice. Our study aimed to identify a novel biomarker to facilitate the translation of research into clinical practice.
Methods
We performed an integrative analysis of seven independent cohorts comprising 1682 patients. The DLBCL-1 cohort was used for signature identification. The identified signature was then tested in the DLBCL-2 to DLBCL-7 cohorts.
Results
Two genes (LIMD1 and MYBL1) were significantly differentially expressed between the ABC and GCB subtypes. We integrated these genes into a composite marker, the LIMD1-MYBL1 Index. In the seven independent cohorts, the average concordance rate between the LIMD1-MYBL1 Index and the cell-of-origin signature classification was 87% (range, 77% to 94%). Furthermore, the LIMD1-MYBL1 Index is an independent prognostic factor for patients from different populations and for patients treated with a variety of therapies (Table 1).
Conclusions
We identified the LIMD1-MYBL1 Index that has clinical value for DLBCL subtype classification and prognosis. Although little is known about the oncogenic roles of these genes, our findings prompt further research on the molecular mechanisms of DLBCL.