Recombinant DNA Advisory Committee - 12/2-3/93 
cells do not express this secondary B7 signal. The investigators propose to introduce the 
B7 gene into melanoma cell lines to enhance immunogenicity. Preclinical in vivo data 
indicates introduction of the B7 gene into non-immunogenic tumor cells enhances tumor 
cell immunogenicity and induces CTL activity, which results in rejection of both B7- 
modified and non-modified tumor cells. As compared to cytokine gene transfer 
protocols, this protocol has no possibility of adverse effects associated with systemic 
cytokine toxicity. As a point of clarification. Dr. Parkman explained that the B7 antigen 
proposed for this study is different from the HLA-B7 antigen previously approved by the 
RAC for Dr. Gary Nabel's protocol (University of Michigan #9306-045). B7 is not an 
HLA antigen, but belongs to a group of accessory molecules that bind to the CD28 
receptor. 
Dr. Parkman explained that the patient population will be limited to either HLA-A1 or 
HLA-A2 individuals to ensure the appropriate HLA match with the corresponding 
melanoma cell lines. Approximately 50 patients will be divided into 3 dose escalation 
groups to assess toxicity. Each group will receive 6 injections of 10 7 , 10 8 , or 10 9 cells that 
have been lethally irradiated at 20,000 rads. The cDNA of the B7 gene will be expressed 
in the bovine papilloma virus (BPV) vector, BCMGNeo-B7. The B7 gene is regulated by 
a CMV promoter and the neo* gene by the SV40 promoter. No other gene products 
will be expressed from this construct. This vector will be transfected into 3 melanoma 
cell lines using DNA-liposomes (lipofection). Only plasmid DNA will be used for the 
lipofection procedure; therefore, there is little risk of transmission of the transduced 
gene to host cells. High dose irradiation will render the tumor cells nonviable. Patients 
will be evaluated for immunological responses and evaluated for any toxicity associated 
with the transduced tumor cells. 
Dr. Parkman inquired whether the HLA-A2 patients will receive injections of all 3 
transduced melanoma cell lines on a rotating basis, and the HLA-A1 patients will receive 
only the HLA-A1/HLA-A2 cell line. Since 40% of Caucasians are HLA-A2, perhaps the 
protocol should be limited to this population in order to simplify interpretation of the 
data. He inquired about the necessity to accrue 50 patients on the proposed study. If a 
total of 6 patients are proposed for each dose group, 18 patients should be accrued. 
What is the rationale for injection of untransduced cells? Is there additional data 
demonstrating in vitro CTL responses? Since the scientific end point of this study is 
production of a CTL response, in vitro preclinical CTL data must be provided. The PI 
has submitted additional data in response to the written primary review; however, there 
are several remaining questions regarding specificity of cell killing and the HLA type or 
subtype of the normal donor cells. 
Review-Ms. Grossman 
Ms. Grossman noted that many of her concerns were addressed by Dr. Parkman. The 
preclinical data demonstrating in vitro CTL activity are inadequate, the vector sequence 
is inadequate, and there are no preclinical animal studies. She expressed concern that 
ectopic B7 expression might lead to the development of autoimmune disease due to 
immunogenicity to nontumor antigens. Based on the lack of scientific and immunologic 
Recombinant DNA Research, Volume 18 
[545] 
