Search for a command to run...
The childhood eye cancer retinoblastoma is initiated by the loss of both alleles of the prototypic tumor suppressor gene, RB1. However, a large number of cytogenetic and comparative genomic hybridization (CGH) studies have shown that these M1 and M2 mutational events--although necessary for initiation--are not the only genomic changes in retinoblastoma. Some of these subsequent changes, which we have termed M3 to Mn, are likely crucial for tumor progression not only in retinoblastoma but also in other cancers. Moreover, genes showing genomic change in cancer are more stable markers and, therefore, possible therapeutic targets than genes simply differentially expressed. In this review, we provide the first comprehensive summary of the genomic evidence implicating gain of 1q, 2p, 6p, and 13q, and loss of 16q in retinoblastoma oncogenesis, including karyotype, CGH, and microarray CGH data. We discuss the search for candidate oncogenes and tumor suppressor genes within these regions, including the candidates (KIF14, MDM4, MYCN, E2F3, DEK, CDH11, and others), plus associations between genomic changes and clinical parameters. We also review studies of other regions of the retinoblastoma genome, the epigenetic changes of aberrant methylation of MGMT, RASSF1A, CASP8, and MLH1, and the roles microRNAs might play in this cancer. Although many candidate genes have yet to be functionally validated in retinoblastoma, work in this field lays out a molecular cytogenetic pathway of retinoblastoma development. Candidate cancer genes carry diagnostic, prognostic, and therapeutic implications beyond retinoblastoma.