Recombinant DNA Advisory Committee - 09/14-15/92 
demonstrates that these retroviral constructs are capable of effecting long-term 
expression out to six months, whereas oligonucleotides are subject to degradative 
processes with extremely high concentrations and frequent infusions required. 
Dr. Roth presented in vivo murine data demonstrating that introduction of the normal 
p53 gene into cells that have a mutant p53 gene reverses the critical features of the 
malignant phenotype. Data was presented which was derived from experiments with 
immunosuppressed nude mice. When human lung cancer cells were injected into these 
mice, extensive local growth of the tumor occurred. However, when mice with three day 
established tumors were treated with three intratracheal injections of the retroviral 
supernatant, tumor burdens were significantly reduced. This very closely replicates the 
proposed clinical protocol. Data was presented demonstrating that injection of in vitro 
transduced cells greatly reduced the size of the tumor burden in mice; however, this data 
is not directly applicable to the proposed in vivo human experiments. 
Dr. D. Miller noted that an antitumor effect was observed when mice were treated with 
10 4 in vitro LNSX transduced cells. What effect would be observed if these animals were 
injected with five times this number of cells? Dr. Roth explained that data suggests that 
10 4 transduced cells is the threshold for the murine experiments. He stated that 10 4 
transduced cells prevents tumors in approximately 50% of the mice, and increasing this 
number to 10 5 cells will prevent 80 to 100% of the tumors. 
In vivo data was presented demonstrating that mice receiving the retroviral supernatant 
alone exhibited a significant reduction in tumor development; and in the few animals 
that did develop tumor, the volume of tumor was small in comparison to control mice. 
Microscopic examination of mice that did not develop tumors indicated no evidence of 
cancer cells. 
Dr. Roth addressed the issue of the effect of the transduced cells on the nontransduced 
cells known as the bystander effect. Based on transduction efficiency data one would 
expect 30 to 50% reduction in the growth of unselected tumor cells transduced with 
LNSX/p53. However, an 80 to 90% reduction in growth was observed suggesting that 
the transduced tumor cells may have an effect on untransduced tumor cells. This 
observation led to more controlled mixing experiments in which 50% transduced cells, 
either p53 or K-ras, were mixed with nontransduced cells. Data demonstrates that this 
mixed population of cells exhibit a growth pattern identical to a population of cells that 
have been transduced and selected in G418. 
Dr. Roth hypothesized that the transduced cells could possibly be releasing a factor that 
affects the nontransduced population. Evidence for this hypothesis is derived from the 
supernatant experiments in which untransduced cells were mixed with culture 
supernatant of p53 transduced cells. A 50% reduction in growth was observed with the 
Recombinant DNA Research, Volume 16 
[269] 
