Despite significant improvements of survival in patients with localized osteosarcoma, about 30–40% of the patients still die on tumor progression or relapse. In order to improve therapeutic outcome we postulate the need for individualized intervention schemes based on biological characteristics of the tumor. Identification of molecular changes important for pathogenesis and tumor progression is complicated by the complex karyotype of the tumor with numerous structural and numerical alterations. Here we describe the use of Affymetrix single nucleotide polymorphism arrays in a genome wide high-resolution approach to assay both loss of heterozygosity and variations in DNA copy numbers in 46 osteosarcoma biopsy samples. We combined established histological response parameters with our genetic findings to predict prognosis.

We found that overall chromosomal changes in osteosarcoma are good predictors of response to chemotherapy and outcome. Analyzing the minimal recurrent regions harbouring chromosomal alterations we expanded our investigations towards identification of gains and losses of chromosomal material and found candidate genes as potential prognostic parameters and therapeutic targets. Identified genomic regions and genes were validated by mRNA-expression studies and correlated with proteom analysis by MALDI Imaging.

Thus, structural chromosomal alterations detected by SNP analysis may serve as a simple but robust parameter to predict response to chemotherapy. The results also indicate that we are able to identify several genomic loci with high potential to predict the outcome of the disease. Furthermore new potential target genes were identified by this genome wide screen.

The project is part of the Translational Sarcoma Research Network (TransSaRNet).