A compendium of myeloma associated chromosomal copy number abnormalities and their prognostic value

To obtain a comprehensive genomic profile of presenting multiple myeloma cases we performed high resolution single nucleotide polymorphism (SNP) mapping array analysis in 114 samples alongside 258 samples analysed by U133 Plus 2.0 expression array (Affymetrix). We examined DNA copy number alterations and loss of heterozygosity (LOH) in order to define the spectrum of minimally deleted regions in which relevant genes of interest can be found. The most frequent deletions are located at 1p (30%), 6q (33%), 8q (25%), 12 (22%), 13q (59%), 14q (39%), 16q (35%), 17p (7%), 20 (12%) and 22 (18%). In addition, copy number-neutral LOH, or uniparental disomy, was also prevalent on 1q (8%), 16q (9%), and X (20%), and was associated with regions of gain and loss. Based on fluorescent in situ hybridisation (FISH) and expression quartile analysis, genes of prognostic importance were found to be located at 1p (FAF1, CDKN2C), 1q (ANP32E), and 17p (TP53). In addition, we identified common homozygously deleted genes which have functions relevant to myeloma biology. Taken together, the dysregulated genes from the myeloma genome indicate that the crucial pathways in myeloma pathogenesis include the NF-?B pathway, apoptosis, cell-cycle regulation and Wnt signalling. SNP data: 114 tumour samples (MM) analyzed by Affymetrix 500K array set (Nsp+Sty), of which 80 samples have matched peripheral blood (PB) (non-tumour) DNA performed on the same array types. Matched tumour and non-tumour samples have the same ID number, e.g. MM400 and PB400 Expression data: 258 expression samples from CD138+ cell selection