Genomic Assessment of Clinical Variability in OGS

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): The diagnosis of osteogenic sarcoma (OGS) is determined by the histological appearance and anatomical location of the tumor. Despite improvements in the clinical, morphological and imaging tools used to classify OGS and the introduction of neoadjuvant chemotherapy, patients with the same histological diagnosis often experience markedly different clinical outcomes. Clearly, classification using current diagnostic techniques does not adequately capture the variability in chemotherapy response and in metastatic frequency, or the unpredictable mortality that characterizes disease progression in patients with this tumor. The current tumor biology literature indicates that clinical variability in histologically similar tumors is due to the frequent inability of current diagnostic and staging criteria to characterize the biological variability inherent in distinct molecular subgroups of tumors. Thus, there is a great need for tools that, by identifying molecularly homogeneous subgroups of OGS, will assist in developing more effective interventions. Recently published studies show that multivariate analysis of the multigene signatures generated by expression profiling (from DNA-arrays) will "cluster" tumors with clinically distinct phenotypes (i.e. distinct molecular subgroups) from groups of histologically identical tumors. These studies have established that expression profiling based on tumor-specific genes is significantly more effective in identifying prognostic determinants of tumor behavior than expression profiling based on randomly-chosen genes, or profiling based on genes where expression does not change between normal and tumor cells. We submit that DNA-arrays containing OGS-specific genes will generate multigene signatures that are informative for clustering osteogenic sarcomas into clinically distinct molecular subgroups, and we propose in this application to develop such DNA-array-based tools for treatment of OGS. Our strategy is, first, to identify genes (molecular markers) which are "tumor specific" for OGS and to use these genes as probes in a custom DNA-array (OGS-chip); second, to validate the ability of this OGS-chip to generate multigene signatures and, by multivariate analysis, bin OGS tumor samples into clinically distinct clusters; and, finally, to use these multigene signatures to prospectively predict clinical characteristics of OGS.