Recombinant DMA Advisory Committee - 6/9-10/94 
By flushing the tumor sites with the adeno-p53 vector, no evidence of tumor at the p53 
treated site was observed after 30 days. Sequential injections were used in the murine 
model. The first injection of the adeno-p53 construct reduced the growth of the 30-day 
established tumor. The second injection with the combination of p53 and cisplatin 
completely inhibited tumor growth and induced apoptosis. These in vitro and in vivo 
experiments have been accepted for publication in the Journal of the National Cancer 
Institute. 
Patient eligibility will be limited to subjects with local or regionally advanced NSCLC. 
The intrapleural administration arm of the study involves subjects with malignant pleural 
effusion who are not eligible for surgery, radiation, or other local treatment. For 
intrapleural administration, the adenovirus vector will be delivered through a chest tube. 
Dr. Post asked if adenovirus vectors have been previously delivered to the pleural cavity 
in any human trial. Dr. Roth answered that human subjects have not received 
adenovirus vectors intrapleurally to date. Dr. Post explained that unlike intratumoral 
injection, normal cells in the pleural cavity will be transduced by the adenovirus 
construct. Dr. Post asked if biopsies of eligible subjects will be screened by PCR for the 
p53 gene mutation. Dr. Roth responded that demonstration of this mutation is included 
in the eligibility criteria. 
Dr. Roth explained that the rationale for injecting larger tumors and including pleural 
disease is that a more concentrated adenovirus vector preparation can be obtained than 
with retrovirus vectors. 
Dr. Parkman expressed concern that the transduction of normal cells within the pleural 
cavity might result in pleuritis or pleurisy. Dr. Post asked if the wildtype p53 gene might 
render normal cells in the pleural cavity more susceptible to cisplatin. Dr. Parkman 
asked if mouse or monkey experiments have been performed to determine toxicity of the 
adeno-p53 construct when injected into the pleural cavity following cisplatin treatment. 
Dr. Roth responded that these experiments have not been performed. Dr. Parkman said 
that although the majority of the cells lining the pleural cavity are nondividing, the lining 
is continuously renewed by the dividing stem cells underneath. Destruction of these stem 
cells may result in serious pleural disease. 
Dr. Roth presented PCR data in response to concerns about wildtype replication- 
competent adenovirus contamination of the vector preparations. One wildtype virus 
particle is detectable among 1 x 10^ vector particles. The patient dose will range 
between 1 x 10^° and 1 x 10" particles; therefore, a maximum of 100 wildtype virus might 
be expected. One order of magnitude higher (1,000 wildtype virus particles) is required 
to cause human disease. Dr. Miller commented that a plaque assay is a more 
biologically relevant assay than PCR. 
Dr. Wei-Wei Zhang of MD Anderson Cancer Center (co-investigator) explained that the 
wildtype helper virus has been assayed by two methods: (1) PCR of the Ela gene, and 
(2) DNA labeling of HeLa cells infected with the virus. Helper virus was not detected in 
a background of 1 x lO’ vector particles. 
Recombinant DNA Research, Volume 19 
[355] 
