Human Gene Therapy Subcommittee - July 29-30, 1991 
transduced with one of two retroviral vectors containing either the human interleukin-2 
(11^2) gene or the human tumor necrosis factor (TNF) alpha gene. The transduced 
cells will be injected subcutaneously into the patient's thi^ without the cells being 
irradiated. They are viable tumor cells that presumably have a relatively normal 
capacity to grow; however, the investigators do not expect that significant local tumor 
growth will occur. Since these patients already have significant tumor burden, 
additional tumor cells that grow locally would not represent a significant increase. The 
inguinal nodes at the site of tumor injection will be excised on day 21. Tumor 
infiltrating lymphocytes (TIL) cells will be expanded in the presence of 11^2. The 
expanded TIL cells will be administered to the patient with standard 11^2 therapy. 
Because the initial stimulus of these TIL cells is the cytokine gene-modified tumor 
cells, the expanded TIL cells may have a greater capacity to kill tumor cells. This will 
result in a more effective form of therapy. 
Dr. Parkman said the investigators have submitted two separate protocols with four 
primary groups of data as their preclinical justification. An animal model was used to 
study metastases resulting from pre-existing tumors by injecting tumor cells three days 
before TIL cells. The TIL cells were able to reduce the number of metastatic cells that 
grow. This is a demonstration that the source of the TIL cells appears to affect the 
efficacy of the TIL cells. TIL cells that were developed from a subcutaneous tumor 
reduce metastatic tumor growth better than TIL cells from a parenchymal organ. This 
is one of the reasons why the tumor cells will be administered cutaneously and TIL 
cells will be derived from a draining lymph node rather than a lymph node that is more 
visceral. Experiments performed by Dr. Fearon, in which the IL-2 gene was transduced 
into a relatively non-immunogenic tumor cell, showed that the introduction of these 
modified tumor cells into an animal allowed the animal to develop cytotoxic T cells 
that could kill the tumor. The work further showed that if unmodified tumor cells and 
modified tumor cells were introduced into different sites of the same animal, tumor 
cells at both locations would not grow. Presumably, the cytotoxic T cells that were 
generated from the modified tumor cells were not only able to kill the modified tumor 
cells, but were also able to inhibit or kill the unmodified tumor cells that were at the 
other anatomical site. TNF-modified tumor cells administered to animals showed that 
the tumor stopped growing and actually regressed. Data show that TNF and T cells 
contributed to this effect. Unlike the IL-2 experiments where there was some systemic 
response, the TNF experiments did not show a systemic response. 
Dr. Parkman stressed that there are no preclinical data to support the latter part of the 
proposed clinical trials, i.e., the demonstration that expanded TIL cells will have a 
greater capacity to inhibit or kill tumor cells. The investigators have shown that they 
can study the trafficking of TIL cells and the homing to metastases. 
Dr. Neiman noted that the protocol had received conditional IBC approval and that 
the IBC had required a pilot study be done with the injection of live patient-derived 
Recombinant DNA Research, Volume 14 
[911] 
