The concept of antisense oligonucleotides in cancerology is a direct consequence of the results of molecular biology. These small synthetic DNAs (often with chemical modifications) are used in tumor cells to target genes which are active to support the cell proliferation. These genes can be proto-oncogenes, growth factors, transcription factors, factors involved in the signal transmission from the cell membrane to the nucleus The aim is to prevent specifically the protein synthesis of one gene through a Watson Crick interaction between the oligonucleotide and the RNA transcript. Therefore with molecular weights comparable to the ones of some classical anticancer agents, oligonucleotides are expected to be much more specific and less toxic. Many oligonucleotides have been described in the last 10 years as efficient against transformed cells in culture. In the last 4 years they have also been shown to inhibit, with efficiency, the growth of human tumors grafted to mice after local or systemic administration. These results already demonstrate that in various instances it is possible to greatly reduce the tumor growth by targeting one single genetic event. However we have still to learn a lot at the basic level about how oligonucleotides work and how to improve their efficiency. Among other questions a major one is how are they delivered to their site of action in the cell? Vectorization and/or serum deprivation are required in cell culture to obtain gene inhibition. However oligonucleotides already display activity in animal as free injected molecules. Actually 4 clinical trials are taking place in phase 1 with oligonucleotides targeting specific genes involved in cancers (PKC, p53, c-myb, c-raf).