Recombinant DNA Advisory Committee- 9/12-13/94 
that a theoretical risk still exists that VEGF DNA could circulate and produce VEGF to 
induce neovascularization particularly in Type I diabetic patients. He asked if this 
problem has been investigated in the animal models. 
Dr. Miller said that the investigators stated that a modification has been made to the 
plasmid DNA vector after discussion with FDA officials. The modification involved 
change of the selectable marker and deletion of the SV40 replication origin. Does the 
new vector function as well as the old one? He asked the investigators to explain why 
there are many blank spaces in the submitted vector sequence. 
Investigator Response-Drs. Isner and Walsh 
Responding to Dr. Miller’s question, Dr. Walsh said the SV40 origin of replication has 
been deleted according to FDA’s suggestion to eliminate any chance of autonomous 
replication of vector DNA in animals. As to why there are N’s in the DNA sequence, Dr. 
Walsh said that is due to sequences unreadable by the automatic sequencer. But the 
sequence is 98% in agreement with the predicted sequence, which accounts for the entire 
plasmid DNA. Dr. Miller commented that if there is no compelling reason to change the 
vector, it should not be changed since all other animal data were obtained from the 
original vector. Dr. Walsh said that the FDA routinely requests removal of the SV40 
origin of replication and the /3-lactamase gene which confers ampicillin resistance in this 
type of vector. The concern relates to potential ampicillin contamination of the plasmid 
DNA preparations. Dr. Miller commented that it is a remote possibility. 
Dr. Noguchi of FDA remarked that the FDA’s position has been that the aforementioned 
sequences should be removed if they are not needed. He agreed in this case, that if 
removal affects the activity, the vector should not be changed. 
Dr. Samulski said that it was entirely possible that deleting the plasmid vector sequences 
could affect gene expression. He suggested that the human study should be performed 
with the original plasmid with which all the preclinical data were obtained. 
Dr. Isner stated that a reasonable compromise would be to move into this initial clinical 
study with the original plasmid. In the future, if the study progresses into a product 
development phase, the more stringent safety issues will be readdressed. Dr. Miller 
agreed it is a reasonable compromise. 
Dr. Samulski suggested that FDA issue some guidance to the investigators early on in the 
study proposal so that the investigators would not have to repeat all the experiments. Dr. 
Miller asked if FDA would provide consultation when the project is started. Dr. Noguchi 
said it is a good idea to start FDA negotiations before the project is started. He said in 
this case, the preclinical studies would be better performed with a plasmid without the 
SV40 replication origin. 
Dr. Saha said this is an exciting protocol. He asked the investigators to compare their 
plasmid study with the recent paper published in Science by Gary Nabel and his co-worker 
Recombinant DNA Research, Volume 20 
[69] 
