Recombinant DNA Advisory Committee - 12/1-2/94 
many questions remain regarding safety and potential efficacy. The "bystander" effect must be more clearly 
understood since the mechanism appears to be different from the effect observed for brain tumors. There is 
no rationale for exposing subjects to possible risks of the retroviral vector since PA317 (no vector) are equally 
effective as VPC in combination with GCV therapy. The investigators must demonstrate: (1) VPC are 
clinically superior to PA317, and (2) persistence of VPC for an appropriate period of time in a large anim al 
model or equivalent in vitro experiments involving appropriate lysing antibodies to murine cells. 
Dr. Gerard McGarrity (GTI) made several comments: (1) The blood brain barrier in glioblastoma patients is 
not absolutely intact; therefore, similar immune reactions would be expected for both studies. (2) Rat serum 
does contain anti-murine antibodies that inactivate VPC; therefore, the rat data are valid. (3) The proposed 
head and neck protocol is an ideal system to obtain data regarding VPC persistence since tumor and tissue 
samples will be readily accessible. (4) Regarding the "bystander" effect, the 9L rat glioblastoma data was 
shown because this system is where most data is available; whereas, the hepatocellular carcinoma is the worst 
case scenario. 
Dr. Parkin an noted that the biology of one tumor type does not predict response in another tumor type; 
therefore, preclinical data derived from an appropriate tumor model is critical The rat model is not an 
adequate model to determine VPC persistence; larger animals, Le., primates are more appropriate. The 
ongoing porcine experiments will be acceptable if the serum can be shown to kill the murine fibroblasts and 
the VPC are persistent for greater than 30 minutes. Dr. Haselkorn agreed with Dr. Parkman’s comments 
about the necessity to conduct the proper experiments. Dr. Miller stated that data should be provided 
demonstrating that transduction of the TK gene is responsible for the therapeutic effect of GCV. In addition, 
Dr. Parkman stated the data should be provided to explain the mechanism of the "bystander" effect on tumors 
at other sites. Dr. Erickson said that the data must demonstrate that VPC are clinically superior to PA317. 
A motion was made by Dr. Parkman and seconded by Dr. Erickson to defer the protocol submitted by Dr. 
Gluckman by a vote of 16 in favor, 0 opposed, and no abstentions. The protocol was deferred until the 
investigators return to the full RAC with additional preclinical data in a squamous cell carcinoma/large animal 
model demonstrating the persistence of PA317/GlTklSvNa.7 VPC and that administration of 
PA317/GlTklSvNa.7 VPC is clinically superior to PA317 cells alone. The RAC strongly recommended that 
the mechanism of the distant "bystander" effect should be established. 
ADDITION TO APPENDIX D OF THE NIH GUIDEUNES REGARDING A HUMAN GENE TRANSFER 
PROTOCOL ENTITLED: PHASE I TRIAL OF INTERLEUKIN-2 PLASMID DNA/DMRIE/DOPE UPID 
COMPLEX AS AN IMMUNOTHERAPEUTIC AGENT IN SOUD MALIGNANT TUMORS OR 
LYMPHOMAS BY DIRECT GENE TRANSFER JDRS. HERSH, AKPORIAYE, HARRIS, STOPECK, 
UNGER, AND WARN EKE 
Review-Dr. Miller 
Dr. Walters called on Dr. Miller to present his primary review of the protocol submitted by Drs. Evan Hersh, 
Emmanuel Akporiaye, David Harris, Alison Stopeck, Evan Unger, and James Warn eke, of the Arizona Cancer 
Center, Tucson, Arizona. Dr. Miller explained that the proposed study involves direct intratumoral injection of 
a liposome/plasmid DNA vector encoding human IL-2. The investigators hypothesize that local IL-2 
production will stimulate an antitumor response. A similar strategy has been employed for other RAC- 
approved protocols. Although there is little evidence that this strategy works in humans, there is sufficient 
preclinical data to support the protocol. The vector is liposome based; therefore, concerns about virus spread 
and transmission are negligible. 
One potential concern is the possibility that the plasmid could enter and integrate into normal immune cells. 
If such integration were to occur, constitutive IL-2 production (a growth stimulatory cytokine) could promote 
uncontrolled cell proliferation. The argument that the proposed DNA administered does not integrate is 
probably incorrect; stable transfectants can be isolated in cell culture by using this DNA transfer technique. 
Recombinant DNA Research, Volume 20 
[455] 
