P. I . : B . Gansbacher 
SCIENTIFIC ABSTRACT OF PROPOSED STUDIES 
We propose to study whether immunization with cytokine 
secreting tumor cells will induce host anti tumor responses in 
cancer patients. Our first protocol will be a phase I study in 
melanoma patients and will address whether allogeneic HLA-A2 
matched melanoma cells expressing a recombinant human IL-2 gene 
can be used as vaccine without causing major toxicity. 
Furthermore, to determine whether such a vaccination protocol is 
applicable for non-melanoma tumors a similar approach will be 
used in renal cell carcinoma patients. Toxicity studies will be 
done in stage IV renal carcinoma patients vaccinated with HLA-A2 
matched IL-2 secreting renal carcinoma cells. Serologic and 
cellular host antitumor responses will be assessed in both 
groups . 
Autologous tumor cell lines are difficult to establish and 
not available for many patients. There is substantial experience 
at MSKCC and at other institutions that has established the 
general safety of immunization using irradiated allogeneic tumor 
cells. The choice of allogeneic HLA-A2 -positive tumor cells in 
our initial studies is based on the following considerations, 
first, HLA-A2 molecules expressed on allogeneic cells are capable 
of presenting identical peptides derived from the same tumor 
associated antigen to HLA-A2 positive effector populations. 
Second, the use of a single established well characterized 
immunizing cell line bypasses the necessity to introduce the 
cytokine gene into autologous tumor cells for each patient. 
Growing autologous tumor cells in tissue culture can be a 
difficult and time-consuming procedure and the success rate for 
establishing autologous tumor cell lines can be quite low. Third, 
since approximately 40% of North Americans are HLA-A2 positive, a 
tumor vaccine produced from HLA-A2 matched allogeneic tumor cell 
lines could be used in a substantial subpopulation of melanoma 
patients and renal carcinoma patients. 
Evidence exists that tumor specific cytotoxic cell 
precursors are present in melanoma, renal cell carcinoma, lung 
and breast carcinoma patient. These effector cells have been 
found in PBMC and can be expanded in vitro and in vivo. Several 
laboratories, including our own have shown in animal studies that 
introduction of cytokine genes into tumor cells is feasible and 
that such modified tumor cells induce potent host anti-tumor 
responses. Presumably, by allowing the tumor cells to present 
both the antigenic peptide by the MHC- class I molecule and the 
second signal in form of the secreted IL-2, allows a substantial 
clonal expansion of tumor specific effector cells. By having 
access to systemic circulation in vivo these effector populations 
have access to residual tumor at distant sites. 
Recombinant DNA Research, Volume 15 
[943] 
