A signature of chromosomal instability inferred from gene expression profiles predicts clinical outcome in multiple human cancers

• Published in: Nature genetics Vol. 38; no. 9; pp. 1043 - 1048
• Main Authors: Carter, Scott L, Eklund, Aron C, Kohane, Isaac S, Harris, Lyndsay N, Szallasi, Zoltan
• Format: Journal Article
• Language: English
• Published: New York Nature Publishing Group US 01.09.2006; Nature Publishing Group
• Subjects: Agriculture; Aneuploidy; Animal Genetics and Genomics; Biological and medical sciences; Biomedical and Life Sciences; Biomedicine; Breast Neoplasms - diagnosis; Breast Neoplasms - pathology; Cancer; Cancer Research; Chromosomal Instability; Chromosome aberrations; Chromosomes; Clinical outcomes; Cohort Studies; Female; Fundamental and applied biological sciences. Psychology; Gene expression; Gene Expression Profiling; Gene Function; Genetic aspects; Genetic transcription; Genetics of eukaryotes. Biological and molecular evolution; Human Genetics; Human subjects; Humans; letter; Neoplasm Metastasis; Neoplasms - diagnosis; Neoplasms - genetics; Neoplasms - pathology; Oncology; Physiological aspects; Prognosis; Tumors; United States; Human; Instability; Malignant tumor; Gene expression; Cancer
• ISSN: 1061-4036; 1546-1718
• DOI: 10.1038/ng1861

We developed a computational method to characterize aneuploidy in tumor samples based on coordinated aberrations in expression of genes localized to each chromosomal region. We summarized the total level of chromosomal aberration in a given tumor in a univariate measure termed total functional aneuploidy. We identified a signature of chromosomal instability from specific genes whose expression was consistently correlated with total functional aneuploidy in several cancer types. Net overexpression of this signature was predictive of poor clinical outcome in 12 cancer data sets 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 representing six cancer types. Also, the signature of chromosomal instability was higher in metastasis samples than in primary tumors and was able to stratify grade 1 and grade 2 breast tumors according to clinical outcome. These results provide a means to assess the potential role of chromosomal instability in determining malignant potential over a broad range of tumors.