Predict cytogenetic abnormalities with gene expression profiles

Cytogenetic abnormalities (CA) are important clinical parameters in various types of cancer, including multiple myeloma (MM). We developed a model to predict CA in patients with MM using gene expression profiling (GEP) and validated it by different cytogenetic techniques. The model was shown to have an accuracy up to 0.89. These results provide proof of concept for the hypothesis that GEP could serve as a one-stop data source for clinical molecular diagnosis and/or prognosis. 92 paired RNA-DNA samples were hybridized to Affy U133Plus2 and Agilent 244K aCGH arrays and used as training set. Another 23 paired samples as test set.

细胞遗传学异常（Cytogenetic abnormalities, CA）是包括多发性骨髓瘤（multiple myeloma, MM）在内的多种恶性肿瘤的关键临床参数。本研究基于基因表达谱（gene expression profiling, GEP）构建了用于预测多发性骨髓瘤患者细胞遗传学异常的模型，并通过多种细胞遗传学技术完成了模型验证。结果显示，该模型的最高准确率可达0.89。本研究结果为“基因表达谱可作为临床分子诊断及/或预后评估的一站式数据来源”这一假说提供了概念验证。
本研究共纳入92对RNA-DNA配对样本，将其与Affy U133Plus2芯片及安捷伦244K阵列比较基因组杂交（array comparative genomic hybridization, aCGH）芯片进行杂交后作为训练集；另有23对配对样本作为测试集。