Human Gene Therapy Subcommittee - July 29-30, 1991 
unmodified tumor cells to deal with the question of the safety. This would not exactly 
parallel the experimental system. The safety concern is not just the possibility that the 
tumor cells will adversely affect patients, but the possibility that the secretion of high 
levels of potentially toxic cytokines from these tumor cells will have an untoward effect. 
The issue is that a small amount of tumor growth might be clinically important if there 
is a toxic reaction to the production of the cytokines it produces. On the other hand, 
the data in the literature show that the cytokines supply the best protection from 
proliferation of these cell lines in patients. The results of the preliminary study do not 
answer these questions. The IBCs principal concern had to do with the introduction of 
live tumor cell lines from the patient more than a recombinant DNA issue. 
Dr. Neiman said that when cells that cany inherent biological danger are introduced 
into patients, it would be desirable to have suicide genes in the vector construct. For 
instance, a TK gene included in this system would allow the elimination of any 
undesirable tumor population. The protocol requests authorization to use any end- 
stage cancer patient with any diagnosis. However, the demonstrated ability to establish 
cell lines from patients is restricted to melanomas, renal cell carcinomas, and colon 
cancer. Only the renal cell carcinomas and the melanoma lines have been transduced 
successfully and expression of the introduced gene demonstrated. Despite the request 
to study any tumor type, only these two tumor types have been characterized. It may 
be possible to inject tumor cell material and lymphokines together at a single site to 
generate the same response, as opposed to using a transduced cell line. The animal 
model is based on the rejection of established, transplantable tumor cells. However, 
the proposed treatment will establish a cell line from a primary tumor and attempt to 
reproduce the resistance to cell growth that has been shown with the established lines. 
It is important to note that the established cell lines from the patient's tumor are often 
phenotypically and sometimes genetically different from the original cell line. Also, 
there are situations in which cytokine secretion from a particular tumor type has been 
shown to enhance tumor growth rather than stimulate a resistance. One needs to be 
concerned about focusing the clinical experiments where there is a reasonable 
expectation of a protective effect rather than a pathologic effect. 
Dr. Neiman said the request for the authorization of 50 patients is too high. A pilot 
study with a limited number of patients, using defined cell types, would be more 
appropriate. The animal experiments are provocative, and it is possible that useful 
therapeutic modalities can be derived from this protocol. 
Ms. Areen stated that the National Cancer Institute IRB made changes to the consent 
forms. There was a recommendation for a separate informed consent document for 
starting cell lines that does not appear in the protocol. Also, there was a 
recommendation by the NIH IBC that it would be appropriate to have an initial 
experiment involving tumor cell lines that had not been transduced. 
[912] 
Recombinant DNA Research, Volume 14 
