Recombinant DNA Advisory Committee - 12/1-2/94 
Investigator Response— Dr. dayman 
Dr. dayman stated that 20 tumor cell lines have been assayed in his laboratory, all of these cell lines can be 
equally transduced by the adenovirus vector. The "bystander" effect, i.e., cell killing without direct 
transduction, has been characterized by electron microscopy, DNA fragmentation, and nuclear fluorescence 
studies. All of these observations are consistent with the apoptosis mechanism. With regard to transduction 
efficiency, an experiment was conducted in which a 1.5 cm solid tumor nodule was injected with 100 microliters 
of 1 x 10'° pfu of vector. 25% of the tumor cells in that nodule expressed the transgene. 
In response to Dr. Zallen’s concerns about the Informed Consent document. Dr. dayman said that there will 
be no costs incurred to patients for surgery, post-operative follow-up, injections, or vectors. However, 
treatment and procedures not directly related to the protocol will not be covered by MD Anderson Cancer 
Center, e.g., standard therapy and long-term follow-up. Medical Care will be provided for residents of Texas 
regardless of their insurance coverage. 
Dr. dayman said that in vitro experiments were conducted using head and neck tumors and non-small cell 
lung cancer cells. The "bystander" killing effect of nontransduced cells was demonstrated for all experiments. 
Subjects will be characterized for p53 mutations; however, specific mutations will not be included as an 
eligibility criterion. In vitro experiments with cultured tumor cell lines suggest that this "bystander" effect is not 
p53 mutation-specific. The adenovirus vector is an episomal vector that produces transient expression of p53; 
therefore, overexpression should not be a concern for subjects with limited life expectancy. 
Dr. dayman stated that tumor accessibility is a major advantage with regard to autopsy information. Autopsy 
analysis will include examination of the tumor and surrounding tissue. He agreed to perform PCR screening 
assays to identify p53 mutations in the adenovirus vector stocks. 
Dr. Par km an inquired whether 2 copies of the wild-type p53 gene can affect normal cell growth. Dr. dayman 
responded that no growth effect was observed on either normal human fibroblasts or oral keratinocytes 
transduced with the adenovirus vector; however, transient p53 expression was observed for 5 to 7 days. Dr. 
Roth added that no morphological or proliferative effects were observed in non-immortalized human epithelial 
cells. Dr. Glorioso asked about the multiplicities of infection (MOI) used for the preclinical experiments. Dr. 
Roth answered that these studies were conducted using an MOI of 100 pfu per cell. A titer of 1 x 10 9 pfu was 
analyzed. 
Dr. dayman said that adenovirus is tropic to epithelial tumor cells but the cytomeglovirus (CMV) promoter is 
not specific for the cell type that is targeted. Dr. Samulski mentioned that there are promoters that are 
specific for prostate or liver cells. 
Dr. dayman presented data involving 17 cell lines which have either homozygous mutations of both p53 alleles 
or have two wild-type alleles transduced at an MOI of 100 to 1. Although the wild-type parental cell lines 
show a slight delay in cell death, killin g occurred equally in all of the cell lines tested. There was no effect on 
the normal fibroblast cell lines. Dr. dayman showed data demonstrating apoptosis and infection efficiency. In 
a dose response experiment, expression of the p53 protein was shown by an immunohistochemistry technique. 
There was some inflammatory response to virus infection at high multiplicity. Antitumor effect was 
demonstrated in murine experiments. 
Dr. Wei Zhang (MD Anderson) presented data derived from consecutive HeLa cell infection experiments. No 
replication competent adenovirus was detected in the vector stocks. Similar results were obtained using a 
sensitive PCR assay specific for the Ela region of the wild-type adenovirus. Dr. Miller commented that PCR 
data is acceptable since the assay has been properly validated by spiking experiments. The sequential 
amplification assay has not been validated. Dr. Samulski commented that radioactive label uptake by newly 
replicating adenovirus is a delayed phenomenon, requiring up to 48 hours to observe uptake. The experiment 
demonstrating lack of replication at high MOIs is invalid because the data were collected at 24 hours. 
Recombinant DNA Research, Volume 20 
[461] 
