Unequivocal delineation of clinicogenetic subgroups and development of a new model for improved outcome prediction in neuroblastoma
Unequivocal delineation of clinicogenetic subgroups and development of a new model for improved outcome prediction in neuroblastoma
复制标题
DOI：
10.1200/jco.2005.06.104
复制
发表时间：
2005-04-01
影响因子：
45.3
通讯作者：
Speleman, F
中科院分区：
文献类型：
--
作者：
Vandesompele, J;Baudis, M;Speleman, F
目的神经母细胞瘤是一种遗传异质性的儿科肿瘤，临床表现多样，从广泛播散到自发消退。在这项研究中，我们的目的是全面的遗传亚组发现和评估的独立预后标记的基础上，全基因组的畸变检测比较基因组杂交（CGH）。材料和方法公布的CGH数据从231原发未经治疗的神经母细胞瘤转换成数字化格式，适用于全球数据挖掘，亚组发现，和多变量生存分析。其中只包括一些遗传参数，我们在这里首次提出了一种策略，允许无偏评估所有的遗传不平衡检测CGH。所提出的方法牢固地建立了三种不同的临床遗传学亚组的存在，并表明17号染色体状态和肿瘤分期是患者预后的唯一独立的重要预测因素。重要的新发现是：（1）正常的17号染色体状态作为具有高度增加的风险的假定可切除期肿瘤的亚组的描绘器;（2）识别赋予呈现全17号染色体获得的1期、2期和4S期肿瘤100% 5年存活率的存活者签名;和（3）3 p缺失作为诊断时年龄较大的标志的鉴定。结论我们提出了一种新的用于改善患者预后预测的回归模型，将肿瘤分期，17号染色体，和扩增/缺失状态。这些发现可能在更可靠的风险评估、临床结果评价和当前治疗方案的优化方面证明是非常有价值的。
Purpose Neuroblastoma is a genetically heterogeneous pediatric tumor with a remarkably variable clinical behavior ranging from widely disseminated disease to spontaneous regression. In this study, we aimed for comprehensive genetic subgroup discovery and assessment of independent prognostic markers based on genome-wide aberrations detected by comparative genomic hybridization (CGH).Materials and Methods Published CGH data from 231 primary untreated neuroblastomas were converted to a digitized format suitable for global data mining, subgroup discovery, and multivariate survival analyses.Results In contrast to previous reports, which included only a few genetic parameters, we present here for the first time a strategy that allows unbiased evaluation of all genetic imbalances detected by CGH. The presented approach firmly established the existence of three different clinicogenetic subgroups and indicated that chromosome 17 status and tumor stage were the only independent significant predictors for patient outcome. Important new findings were: (1) a normal chromosome 17 status as a delineator of a subgroup of presumed favorable-stage tumors with highly increased risk; (2) the recognition of a survivor signature conferring 100% 5-year survival for stage 1, 2, and 4S tumors presenting with whole chromosome 17 gain; and (3) the identification of 3p deletion as a hallmark of older age at diagnosis.Conclusion We propose a new regression model for improved patient outcome prediction, incorporating tumor stage, chromosome 17, and amplification/deletion status. These findings may prove highly valuable with respect to more reliable risk assessment, evaluation of clinical results, and optimization of current treatment protocols.