1 . 0 INTRODUCTION AND BACKGROUND 
Local therapies, such as surgery and radiation, rarely cure patients with 
disseminated cancer... Lamentably, our ability to alter the natural history of 
metastatic cancer using available chemo therapeutics is also limited and it is 
apparent that additional therapeutic strategies are needed. We, and others, have 
previously demonstrated that interleukin- 2 (IL-2) -based immunotherapy induces 
clinically significant tumor regression in 20 to 30 percent of treated patients 
who have metastatic melanoma or renal cell carcinoma, respectively (1). 
Responses have been noted, albeit much less frequently, among treated patients 
who have other malignancies including colon carcinoma, ovarian carcinoma, breast 
carcinoma and non-Hodgkin's lymphoma (2). The cost to the individual patient is 
significant toxicity. Although a small proportion of patients manifest durable 
clinical responses, prolonged survival, cure, or even palliation have yet to be 
clearly demonstrated. 
An alternative strategy for immunotherapy is suggested by in vitro and 
preclinical murine data showing that the cytokine microenvironment within tumors 
may determine the outcome of the immune response. In particular, the 
immunomodulatory cytokine inter ieukin-4 (IL-4) has a central role in the immune 
response to tumors in certain murine models. Implantation of the murine 
mastocytoma P815 into the anterior chamber of the eye results in progressive 
growth of this tumor while implantation into the subconjunctival sac of the eye 
results in tumor rejection. Cytolytic precursor cells residing in both sites 
secrete IL-2. However, only those isolated from the rejecting site 
(subconjunctival sac) elaborate IL-4 upon exposure to tumor (3). Bosco, et al 
(4), have injected IL-4 into the region of tumor-draining lymph nodes of mice 
bearing either a fibrosarcoma or a mammary adenocarcinoma. As little as 0.1 
picogram per day induced tumor regression and protective immunity to subsequent 
tumor challenge. Tepper et al, (5) demonstrated that an invariably lethal murine 
B-cell lymphoma or a mammary adenocarcinoma, both transfected with the IL-4 gene, 
regressed after a brief period of growth. Golurabek et al (6) have observed a 
similar effect with a spontaneous renal cell carcinoma (Renca) transfected with 
the gene for IL-4. They extended these observations by demonstrating both 
protective immunity against tumor rechallenge in mice vaccinated with these gene- 
modified tumors and tumor regression in a 6 to 9 day tumor model. In contrast 
to the demonstrated lack of therapeutic efficacy of IL-4 when administered 
systemically to patients with cancer (unpublished observations) , these studies 
suggest that the constant local production of IL-4 is associated with development 
of antitumor immunity that may translate to regression of established cancer. 
Some of these antitumor immunologic effects have been observed with as little as 
0.44 units (U) IL-4/10 6 cells/48 hours (5). However, regression of established 
tumor has only been associated with a vaccine that elaborated 1500 U/10 5 cells/24 
hours (6). In this study, 10 6 cells used for vaccination would have been 
expected to release 15,000 U of IL-4 daily at the vaccine site. Such high, 
constant intratumor concentrations of IL-4 can not be attained with systemically 
administered IL-4 due to the associated toxicity. 
The mechanism underlying the antitumor effects of IL-4- transduced tumor vaccines 
has not been clearly defined. Tumor regression has been associated with a 
mononuclear infiltrate consisting predominantly of macrophages (5, 6). 
Recombinant DNA Research, Volume 16 
[499] 
