NONTECHNICAL ABSTRACT 
No consistently effective therapy exists for metastatic renal cell 
carcinoma (RCC) . Currently, only 2% of patients with advanced RCC live five 
years following diagnosis. Interest in immunotherapy for RCC has been stimulated 
by the observation of spontaneous, short-term regressions of metastases without 
treatment in up to l%-3% of patients. We have conducted extensive laboratory 
studies using a new strategy for inducing anti-tumor immune responses to mouse 
tumors including RCC. By inserting immunostimulatory genes into mouse RCC tumor 
cells, and injecting them under the skin, systemic antitumor immune responses 
have been reproducibly induced, resulting in eradication of small amounts of 
implanted tumor at distant sites. The Granulocyte-Macrophage Colony Stimulating 
Factor (GM-CSF) gene in these model studies conferred the most potent antitumor 
effects. Efficient introduction of this gene in the model cancer vaccine cells 
was accomplished with the retroviral vector MFG . This efficiency made feasible 
the generation of genetically engineered human cancer vaccines. Using the MFG 
vector, we were able to prepare kidney cancer cell vaccines from 20 of 22 cases 
in a preliminary study. Such transduced cells secrete 10 to 50 fold more GM-CSF 
than cultured, non-transduced RCC cells. 
Lethal irradiation of the genetically engineered RCC cells did not diminish 
therapeutic effects of vaccine cells genetically engineered to secrete the GM-CSF 
gene in our mouse studies. Irradiation of tumor vaccines affords a measure of 
safety for human studies without compromising potential therapeutic efficacy. 
Otherwise, patients might have to be injected with live genetically engineered 
cancer cells for treatment. To develop this new strategy for the treatment of 
human cancer, safety of tumor vaccines produced by our procedure must be 
established in a phase I (toxicity) study. 
The overall objective of this phase I study is to evaluate the safety and 
tolerability of RCC vaccine cell skin injections using a vaccine derived from a 
patient's RCC cells prepared either with or without human GM-CSF gene transfer. 
To help ensure safety, all tumor cell vaccines will be irradiated prior to 
injection. For evaluation of toxicities both from cancer vaccine cells alone, 
as well as from GM-CSF gene transfer, the trial will have two dose escalating 
studies within it. In one arm of the study, advanced kidney cancer patients will 
be treated with irradiated cells prepared without GM-CSF gene transfer. In the 
other arm of the study, patients will be treated with escalating doses of 
irradiated RCC cells transduced with the human GM-CSF gene. Patients will be 
randomly assigned to each arm. In all cases, surgical removal of the kidney will 
provide the source of the vaccine cells. The patients to be enrolled in this 
study will have greater than 1 billion cancer cells in their bodies as evidenced 
by the presence of visible metastases on X-rays. Unfortunately, the mouse RCC 
genetically engineered tumor vaccines were effective at eliminating up to about 
10,000 cancer cells at distant sites. While it may be unlikely that the 
injections will benefit each patient enrolled in this trial with advanced 
disease, if safety can be established for these procedures, future studies 
involving larger number of RCC patients with occult metastases could be 
conducted. 
Recombinant DNA Research, Volume 17 
[285] 
