Differential expression of rituximab responders vs. non responders on 3 different blood cell types. Homo sapiens

New and effective therapeutical options are available for the treatment of Rheumatoid Arthritis. One of such treatments is rituximab, and chimeric anti-CD20 antibody that selectively depletes the CD20+ B cell subpopulation. Similar to established anti-TNF alpha therapies, there is a subgroup of RA patients that do not experience significant clinical response. Therefore, one of the major necessities in actual RA therapeutical management is to identify reliable predictors of the response to this therapies. In the present study we have evaluated 3 blood cell types (i.e. whole blood, isolated B cells and isolated CD4 T cells) using microarray gene expression profiling to identify their potential use as biomarkers for rituximab response. In all three tissues evaluated, we have identified statistically significant differentially expressed genes. The most relevant candidates have been reevaluated using RealTime PCR. These genes were: TRAF1 and arginase 1 in whole blood, Toll-Like Receptor 4 (TLR4) in CD4+ T cells and AT-rich interactive domain 3A (ARID3A) in B cells. In the present study we have demonstrated the potential of different blood cell types for the prediction of the response to rituximab. In particular, we have found a set of relevant candidate genes that could be the basis for future treatment response prediction. Overall design: We have recruited 9 RA patients starting immunotherapy with rituximab. The same day of the first infusion we obtained the three different cell type samples: whole blood (preserved in PaxGene tubes), B cells (using negative selection) and CD4+ T cells (using negative selection). At week 24 of treatment we determined the clinical response using the RelDas score. All tissue samples were processed in parallel (i.e. RNA extraction and microarray profiling with Illumina WG6 Beadchip) in order to minimize the technical variability. The three datasets were normalized separately using quantile normalization. After performing differential expression analysis between Responders and Non-Responders we identified significantly expressed genes in all three cell types. The most significant gene candidates were validated using the Taqman Real-Time PCR assays.

类风湿关节炎（Rheumatoid Arthritis, RA）的治疗现已拥有新型且有效的可选方案。利妥昔单抗（rituximab）便是其中之一，它是一种可选择性耗竭CD20阳性B细胞亚群的嵌合抗CD20抗体。与已获批的抗TNF-α（anti-TNF alpha）疗法类似，部分RA患者无法从该类治疗中获得显著临床应答。因此，在实际的RA诊疗管理中，一项核心要务便是筛选出可可靠预测此类疗法应答的生物标志物。

本研究针对三种血细胞类型——全血、分离B细胞以及分离CD4阳性T细胞，开展微阵列基因表达谱分析，以评估其作为利妥昔单抗应答预测生物标志物的潜力。在三种受试组织中，我们均鉴定出了具有统计学显著性的差异表达基因。针对最具相关性的候选基因，我们采用实时荧光定量PCR（RealTime PCR）进行了复筛验证，所涉及的基因包括：全血中的TRAF1与精氨酸酶1（arginase 1）、CD4阳性T细胞中的Toll样受体4（Toll-Like Receptor 4, TLR4），以及B细胞中的AT富集相互作用结构域3A（AT-rich interactive domain 3A, ARID3A）。本研究证实了不同血细胞类型在预测利妥昔单抗应答方面的潜力，尤其筛选出了一组相关候选基因，可为未来的治疗应答预测研究提供基础。

【研究设计】本研究共招募9名初始接受利妥昔单抗免疫治疗的RA患者。在首次输注当天，我们采集了三类细胞样本：全血（保存于PaxGene采血管中）、B细胞（采用阴性分选法分离）以及CD4阳性T细胞（采用阴性分选法分离）。在治疗第24周时，我们采用RelDas评分评估临床应答情况。所有组织样本均采用平行流程处理（即同步开展RNA提取与Illumina WG6 Beadchip微阵列谱分析），以最大限度降低技术变异。三组数据集均采用分位数归一化（quantile normalization）进行标准化处理。在对应答者与无应答者开展差异表达分析后，我们在三种血细胞类型中均鉴定出了显著差异表达的基因。最终通过Taqman实时荧光定量PCR（Taqman Real-Time PCR）对最具显著性的候选基因进行了验证。