Recombinant DNA Advisory Committee - 12/1-2/94 
without some formal analysis of these results, any conclusion of efficacy (even for a Phase I study) is 
unfounded. Dr. Marcus stated that in the new protocol the study design has been improved to assess the 
survival data, taking into account the factors mentioned by Dr. Chase. Dr. Straus asked if any patient is 
disease-free. Dr. Marcus said that one patient in Dr. Oldfield’s protocol remains free of the disease. 
Dr. Mitchell Berger (Protocol #9303-037) stated that the survival rate of these 4 patients is encouraging. Dr. 
Smith cautioned that referral-based studies automatically eliminate the sickest patients; similar bias has been 
observed in leukemia studies. Dr. Smith agreed with Dr. Chase’s comments that these results do not provide 
statistically significant data. Dr. Saha asked about the stage of tumors these patients have. Dr. Berger 
responded that most patients have Grade 4 tumors. Ms. Meyers inquired if there is any magnetic resonance 
imaging (MRI) indication that the tumors have disappeared in these 4 surviving patients. Dr. Berger answered 
that a residual abnormality is normal for the long-term survivors and stressed that the stable condition of the 
subject should be observed as a positive outcome. 
Dr. Berger addressed the adverse effects observed in Protocol #9303-037. There are 2 possible causes for 
these adverse effects: (1) the rate of VPC infusion, and (2) contiguity of the resection cavity with the 
ventricular system. Rapid injection of VPC may cause leakage into the third ventricle that could irritate the 
hypothalamus causing a hypertensive crisis and fever. The protocols has been modified to include an isotope 
tracer to monitor for leakage and to administer the VPC at a slower rate. 
Commenting on the seizure that occurred in one patient. Dr. Berger said if the routine anticonvulsant had 
been given to that patient after brain tumor surgery, this episode of seizure might have been avoided. He said 
that other complications are unlikely to be related to gene therapy, i.e., pulmonary embolism, infarct in the 
contralateral brain, and infection of the Ommaya reservoir area. 
Ms. Meyers asked if autopsy results indicated the presence of vector sequences in the surrounding tissue. Dr. 
Marcus responded that 2 autopsies have been conducted; however, the preserved tissue has not yet been 
analyzed, and the brains are being fixed in formalin but not yet examined. 
Dr. Smith asked the investigators to comment on the report of finding vector sequences in the lung during a 
hemorrhagic episode. Dr. Ed Otto (GTI) presented PCR data of blood samples from all 15 patients on 
protocol #9206-019. PCR can differentially detect the vector and potential recombinant RCR generated from 
the vector. The sensitivity of this assay is at a level of detecting 10 vector copies per 500,000 blood cells with 
an assay competence of 99.99%. There is no evidence that any blood samples contained vector or RCR 
sequences. A false positive result was obtained using a single neomycin resistance (neo R ) gene marker; 
however, when a second HSV-TK gene marker was included, none of the samples tested positive for both 
markers. 
Dr. Smith asked whether there was any evidence in the preclinical animal studies that GCV alone evoked a 
reduction in tumor size, an effect described in Dr. Grossman and Woo’s protocol. Drs. Berger and Miller 
(GTI) said they do not have definitive data in regard to this question. 
Dr. Brinckerhoff asked if an immune response has been observed to the murine VPC. Dr. Yawen Chiang 
(GTI) explained that immune responses to VPC were documented in several patients who have received 
multiple VPC treatments. Some patients had VPC antibodies after a single treatment; however, these patients 
apparently did not have any adverse hypersensitivity reactions. No immune responses, either humoral or 
cellular, were observed in the rat model. Rats were treated with multiple intradermal VPC injections, and no 
obvious hypersensitivity reactions were observed. Dr. Parkman commented that the rat is not a good model 
for human hypersensitivity reactions; the guinea pig is the appropriate animal model and should be challenged 
with subcutaneous injections, not intradermal injections. 
Dr. Straus asked whether any retroviral sequences had been identified in the brain tissue obtained at autopsy. 
Dr. Otto said that one autopsy analysis was incomplete due to failure to extract DNA from formalin fixed 
tissue. A second cryopreserved specimen revealed no retrovirus sequences or RCR. Investigators have been 
[444] 
Recombinant DNA Research, Volume 20 
