Systematic analyses of the cancer genome: lessons learned from sequencing most of the annotated human protein-coding genes.

PURPOSE OF REVIEW The availability of a reference human genome sequence has enabled unbiased mutational analyses of tumor genomes to identify the mutated genes that cause cancer. This review discusses recent insights from such analyses of protein-coding genes in breast and colorectal cancers. RECENT FINDINGS Mutational analyses of approximately 18,000 human protein-coding genes in breast and colorectal cancers have identified 280 candidate cancer genes. These include known cancer genes, but most had not previously been linked to cancer. There are few frequently mutated cancer genes among hundreds of less frequently mutated candidate cancer genes, and the compendium of mutated genes differs among tumors of the same tissue origin. SUMMARY Recent work has shown the feasibility of coding cancer genome sequencing, and new technologies promise to facilitate these mutational analyses. Whereas cancer genetics can identify candidate genes in a rapid and scalable fashion, careful functional studies of mutated genes are required for ultimate proof of cancer gene status and translation into clinical utility. The rapid progress of cancer genetics has yielded novel diagnostic and therapeutic modalities, and cancer genome sequencing will accelerate this development to the benefit of cancer patients.