A validated gene expression model of high-risk multiple myeloma is defined by deregulated expression of genes mapping to chromosome 1

作者：John D. Shaughnessy, Fenghuang Zhan, Bart E. Burington, Yongsheng Huang, Simona Colla, Ichiro Hanamura, Peter Stewart, Bob Kordsmeier, Christopher E. Randolph, David R. Williams, Yan Xiao, Hongwei Xu, Joshua Epstein, Elias J. Anaissie, Somashekar G. Krishna, Michele Cottler‐Fox, Klaus A. Hollmig, Abid Mohiuddin, Mauricio Pineda‐Roman, Guido J. K. Tricot, Frits van Rhee, Jeffrey R. Sawyer, Yazan M. Alsayed, Ronald C. Walker, Maurizio Zangari, John Crowley, Bart Barlogie · 发表于：Blood · 年份：2006 · DOI：10.1182/blood-2006-07-038430 · 被引用次数：931 · 研究领域：Multiple Myeloma Research and Treatments、Protein Degradation and Inhibitors、Cancer Mechanisms and Therapy

To molecularly define high-risk disease, we performed microarray analysis on tumor cells from 532 newly diagnosed patients with multiple myeloma (MM) treated on 2 separate protocols. Using log-rank tests of expression quartiles, 70 genes, 30% mapping to chromosome 1 (P < .001), were linked to early disease-related death. Importantly, most up-regulated genes mapped to chromosome 1q, and down-regulated genes mapped to chromosome 1p. The ratio of mean expression levels of up-regulated to down-regulated genes defined a high-risk score present in 13% of patients with shorter durations of complete remission, event-free survival, and overall survival (training set: hazard ratio [HR], 5.16; P < .001; test cohort: HR, 4.75; P < .001). The high-risk score also was an independent predictor of outcome endpoints in multivariate analysis (P < .001) that included the International Staging System and high-risk translocations. In a comparison of paired baseline and relapse samples, the high-risk score frequency rose to 76% at relapse and predicted short postrelapse survival (P < .05). Multivariate discriminant analysis revealed that a 17-gene subset could predict outcome as well as the 70-gene model. Our data suggest that altered transcriptional regulation of genes mapping to chromosome 1 may contribute to disease progression, and that expression profiling can be used to identify high-risk disease and guide therapeutic interventions.