Recombinant DNA Advisory Committee - 10/7-8/91 
Dr. Kelley moved that the protocol be deferred. The committee has given the 
investigators very good advice, and they are to be complimented on developing an 
experimental model. On the other hand, there are a number of things that need to be 
done prior to proceeding to human experiments. Dr. Gellert seconded the motion. 
Dr. McGarrity called for further discussion. Mr. Capron asked for a summary of the 
points on which deferral rests. Dr. Kelley said while the animal model is a good one, 
there are many things that still need to be done. Particularly, an examination at autopsy 
to determine exactly what tumor cells have been killed needs to be done. Also, safety 
tests represent another major area where there are still some questions. Measures of 
efficacy represent another area that bears discussion; such as, how to assess the patient's 
tumor burden before, during, and after therapy. Laparoscopy is a very subjective 
measure. Dr. Freeman said CAT scanning may also be performed. 
Dr. Leventhal disagreed with the assessment by laparoscopy. She said that the most 
sensitive assay for residual tumor is the surgeon actually looking; it is better than a CAT 
scan. A patient who looked negative on CAT scan could be positive by laparoscopy, and 
certainly by laparotomy. However, one cannot require a patient to have two 
laparotomies in order to participate in an experimental protocol. 
Dr. Leventhal said the committee would like to have some further evidence that tumor 
growing intraperitoneally can be cured with the proposed treatment. She suggested that 
Dr. Freeman treat the mice and sacrifice some untreated animals at the same time. This 
will show that the animal model is relevant. The investigator could start treatment at a 
measurable fraction of the residual survival of the individual mouse. If it has 15 days left 
to live, the mouse could be treated at five, six, or seven days. The investigator needs to 
observe untreated mice at five days to illustrate the outcome of the untreated mice. Dr. 
Leventhal asked what in vivo studies would be performed in that animal model, on the 
residual tumor cells, that might supply a way to evaluate the treatment. She expressed 
her concern that since the mechanism of the therapy is not understood, there may be a 
quantitative barrier in the human model because there is no end product to determine if 
the gene was transferred. 
Dr. Haselkorn said that the system is a black box in the mouse. This black box will be 
transferred to humans, treating the humans as mice. Further investigation should be 
done on the immunological component of the mechanism. The hypothesis is that in the 
TK+ cell that is treated with ganciclovir, there is an accumulation of ganciclovir 
diphosphate, or ganciclovir triphosphate, in some kind of membrane enclosure which can 
be transferred to a susceptible cell. There have been no experiments that test that 
hypothesis directly. This experiment should be done before treating humans because it 
may not be necessary to administer whole cells; one may be able to merely inject the 
vesicles. 
Recombinant DNA Research, Volume 15 
[19] 
