Recombinant DNA Advisory Committee - 12/1-2/94 
has an antitumor effect in these brain tumor models, and the addition of "gene therapy" acts in an adjunctive 
fashion. In addition, adenovirus vector per se is directly cytopathic. The design of the present human study 
has shortcomings and is unable to sort out whether any efficacy seen would be either due to GCV alone, to 
the vector without HSV-TK gene, or to the combination of the vector with GCV. 
Since adenoviruses have been reported to recombine with or reactivate members of polyoma/papovavirus 
family such as SV40 and JC viruses. Dr. Smith asked the investigators to elaborate if there is an additional risk 
of reactivating the latent JC, SV40, or polyoma virus in the brain of patients who have latent infection. Since 
the present construct has potential toxicity, the health care workers, the investigators, as well as the patients 
should be monitored for virus shedding. Dr. Smith asked the investigators to explain the reason why a 
previously approved protocol (#9105-008) in which Dr. Woo is a co-investigator, has never been initiated. 
Dr. Smith reiterated that the major concern of this protocol is the toxicity observed in the baboon experiment. 
Dr. Samulski credited the investigators for performing this baboon experiment at a high vector dosage in order 
to demonstrate the potential toxicity. Similar caution should be taken in regard to the protocol by Drs. Eck 
and Alavi (#9409-089) previously approved by the RAC that used a similar adenovirus vector to treat brain 
tumors. Dr. Glorioso stated the previous approval of Drs. Eck and Alavi’s protocol was made without the 
benefit of the baboon toxicity data. In addition, the immune status of the patients might further aggravate the 
toxicity. Dr. Par km an said the question of immune status was discussed during the review of Drs. Eck and 
Alavi’s protocol, and one of the stipulations of the approval was to submit data from a pre-immunized cotton 
rat experiment. 
Review-Dr. Zallen 
Dr. Zallen raised the same concern about the vector toxicity observed in the baboon experiment. She fully 
endorsed the comments made by the previous reviewers. Dr. Zallen stated that the Informed Consent 
document is quite clear and addresses all the concerns that have been identified by the RAC in the Points to 
Consider. The few changes recommended by Dr. Zallen have been accepted by the investigators. Once the 
toxicity issue is clarified, such information should be included in the Informed Consent document. 
Other Comments 
Ms. Meyers said the Informed Consent document is well written but the potential risks need to be stated more 
explicitly to include the adverse effects seen in the retroviral HSV-TK/GCV protocols and the toxicity of 
adenovirus vectors in baboons. Dr. DeLeon said Dr. Harold Ginsberg (NIH) made a statement during a 
previous RAC meeting that the adenovirus vectors with E3 deletion potentially has more toxicity, she asked 
the investigators to explain the reason for such deletion in the present vector construct. 
Investigator Response-Drs. Grossman and Woo 
Dr. Grossman identified himself as the Chief of Neurosurgery at Baylor College of Medicine. The toxicity 
issue is of the greatest concern to the investigators, and it is the reason to perform the baboon experiment. 
There is no treatment that will not have toxicity at a certain dosage level. The purpose of the baboon 
experiment is to determine the dosage level that will be toxic. The adenovirus vector at a dosage of 3 x 10 10 
pfu is toxic in the baboon, and the clinical human trial will be conducted below this dosage level. Liquefactive 
necrosis at the injection sites is not unexpected since all treatments of brain tumors are to destroy cells and to 
cause necrosis. An important factor to consider when comparing toxicity of baboon and human brains is that 
the baboon brain is very tiny and the necrotic reaction produces increased intracranial pressure, while the 
human brain is much larger and more tolerant to such necrotic lesions. Radiotherapy produces a necrotic 
lesion at the center of the tumor mass but does not kill the peripheral brain tumor cells. Hopefully, the gene 
therapy owing to its "bystander" effect will be able to kill the peripheral cells as well The investigators 
theorize that the necrotic reaction will be controllable, and it will be tolerated by the patients. The starting 
dose in humans is 300 times smaller than the toxic dose in baboons. Furthermore, the investigators will be 
[472] 
Recombinant DNA Research, Volume 20 
