The loss of control over the cell cycle and the disruption of cascade mechanisms for programmed cell death are major factors in tumorigenesis. Defects in thep53 gene and in the regulation of genes of the retinoblastoma pathway such as p16 or cyclin D1 occur in a large percentage of tumors and have been well studied. Reintroduction or overexpression of genes suppressing proliferation or promoting apoptosis offers a potential for selective suicide of tumor cells. Transfer of tumor suppressor genes into growing tumors with mutations in the respective gene can indeed reduce tumor growth (1). The reason for this effect is probably not just a reestablishement of the normal phenotype, but rather a severe disturbance of the cancer cell's regulatory balance of life and death, which can result in apoptosis. Retransfer of two or more cancer genes can have synergistic effects on apoptosis induction. The choice, which gene combination will be particularly efficient, depends on the pattern of mutated genes. We have recently reported that the cotransfer of p53 and p16 leads to a better induction of apoptosis and reduction of tumor growth than the transfer of either gene alone (2). In addition, it seems that normal cells with an intact genotype are more resistant to the action of tumor supressor genes than tumor cells with mutations in the respective genes. Several approaches are now underway to exploit those gene combinations which are the most efficient for tumor cell-specific apoptosis on a given genetic background.