Molecular profiling in cancer research and personalized medicine

Abstract

Recent efforts by worldwide consortia such as The Cancer Genome Atlas and the International Cancer Genome Consortium have greatly accelerated our knowledge of human cancer biology. Nowadays, complete sets of human tumours that have been characterized at the genomic, epigenomic, transcriptomic, or proteomic level are available to the research community. The generation of these data was made possible thanks to the application of high-throughput molecular profiling techniques such as microarrays and next-generation sequencing. The primary conclusion from current profiling experiments is that human cancer is a complex disease characterized by extreme molecular heterogeneity, both between and within the classical, tissue-defined cancer types. This molecular variety necessitates a paradigm shift in patient management, away from generalized therapy schemes and towards more personalized treatments. This chapter provides an overview of how molecular cancer profiling can assist in facilitating this transition. First, the state-of-the-art of molecular breast cancer profiling is reviewed to provide a general background. Then, the most pertinent high-throughput molecular profiling techniques along with various data mining techniques (i.e. unsupervised clustering, statistical learning) are discussed. Finally, the challenges and perspectives with respect to molecular cancer profiling, also from the perspective of personalized medicine, are summarized.