Extensive gene therapy studies in preclinical models and in clinical trials underscore the relative safety of onco-retroviral vectors. Up until recently, no adverse effects have been reported in nearly 2000 patients that were enrolled in gene therapy clinical trials involving onco-retroviral vectors. However, the main safety concern of using onco-retroviral vectors is related to the risk of malignant transformation following oncogene activation due to random onco-retroviral genomic integration. Based on primate studies, there is an apparent low risk of malignancy that is predominately associated with the occurrence of chronic retroviremia resulting from replication-competent retroviruses (RCR), particularly in immunosuppressed recipient hosts. However, in the latest packaging cell lines and vectors, the risk of RCR-generation has been drastically reduced, primarily by minimizing the homologous overlap between vector and helper sequences. Nevertheless, results from a recent preclinical study in mice and a clinical trial in patients suffering from SCID-X1 strongly suggest that onco-retroviral vectors devoid of RCR can contribute to lymphomagenesis by insertional activation of cellular oncogenes. The risk of inadvertent germline transmission of onco-retroviral vectors appears to be low, especially relative to the endogenous rate of germline insertion, which is known to occur naturally in the human population via transmission of endogenous retro-transposons. The strict dependency of onco-retroviral gene transfer on cell division is an important safety advantage that significantly limits the risks of horizontal transmission. Since improved onco-retroviral vectors or transduction protocols may result in an increased number of retroviral integrations per cell, this may concomitantly increase the risk of malignant transformation. The use of suicide genes, self-inactivating vectors and/or chromosomal insulators is, therefore, warranted to further enhance the safety features of onco-retroviral vectors. Detailed analyses of insertion sites combined with long-term clinical follow-up may contribute to a more accurate risk assessment.