The transfer of genes into tumors of the central nervous system has been touted as a novel treatment. However, several scientific and technological hurdles will have to be resolved before such strategies become useful clinical tools. This review summarizes the current knowledge in the field. Some of the gene delivery vectors employed both preclinically and clinically are those based on retroviruses, herpes simplex viruses, adenoviruses, adeno-associated viruses, and reoviruses. Cells such as fibroblasts and neural progenitor cells may also provide therapeutic value. These vectors are used to deliver into the tumor cell a variety of anticancer genes, such as those that activate chemotherapy agents, increase tumor immunogenicity, modulate tumor apoptosis and/or angiogenesis. One of the issues confronting such therapeutic strategies revolves around the blood-brain-barrier that may limit the penetration of vectors and genes form the circulation into the tumor. Results from a variety of clinical trials are becoming available. While the safety of this treatment strategy appears to have been established, therapeutic efficacy has been lacking. Additional refinements in the basic technology of vector construction and further understanding of the basic biology of gene transfer and expression will help in establishing gene therapy as clinically useful against brain tumors.