cell proliferation or maturation, then a step towards 
malignant transformation could occur. However, the chance 
that such an integration event will occur is very low given 
the size of the genome and the use of replication- 
incompetent vectors. (For review see 55 ) Also, it is likely 
that the malignant transformation of most primary cells 
requires multiple steps. One transduced cell is very 
unlikely to have insertions in more than one growth- 
regulatory gene. 
Safety data from animal and human studies : The original N2 
vector, the safety-modified LNL6 vector and several related 
marking and gene therapy vectors have been used extensively 
in animal studies and in pioneering human clinical 
studies. 10 ' 55 None of the over 1500 mice transplanted with 
bone marrow transduced with MoMLV-based retroviral vectors 
have developed toxicity or malignancies attributable to the 
gene transfer protocol. No evidence for delayed engraftment 
due to the transduction procedure has been observed. 
Experience in primates and humans is less extensive. Four 
primates given IV orIP infusions of replication-competent 
helper virus and immunosuppressed with cyclosporin developed 
only transient viremia and were without adverse effects up, 
to 3 years later. 57 21 primates (rhesus and cynomologous 
monkeys) were transplanted with marrow transduced with N2 
and SAX (carrying the human adenosine deaminase gene) 
retroviral vectors. 46 None showed evidence of circulating 
helper virus or development of neoplastic disease. Many of 
these early animals developed complications secondary to the 
total body irradiation used for conditioning, but several of 
the animals have survived and been followed carefully five 
years post-transplant. 
More recently, three of seven rhesus monkeys transplanted by 
our laboratory with CD34-enriched bone marrow cells 
transduced with a very high titer (>10 lo /ml) but grossly 
helper virus-contaminated supernatant developed T cell 
leukemia/lymphoma 183-206 days after transplant. Details of 
the genesis of the helper-contaminated producer cell line, 
the gene transfer procedure, and the molecular evaluation of 
these animals are given in a draft of our paper reporting 
these events, and a letter to the FDA. (Appendix I) We 
believe the evidence suggests that these malignancies were a 
result of the large amount of competent amphotropic helper 
virus ( 10*-10*/ml ) the target cells were exposed to during 
transduction. The animals were chronically viremic and the 
tumor cells contained multiple copies of the helper genome. 
This syndrome mimics what is seen in rodents exposed to 
large amounts of wild-type retrovirus when they are 
immunocompromised (i.e. newborn). Insertional mutagenesis 
of growth regulatory genes such as mvc and rir is believed 
[ 16 ] 
Recombinant DNA Research, Volume 16 
