growing tumor contain T cells that were immunologically sensitized but functionally deficient 
in expressing antitumor reactivity. This population of cells designated as "pre-effector" cells 
could nevertheless be further stimulated in vitro to differentiate into potent immune effector 
cells (3,4). Initially an in vitro sensitization (IVS) procedure was used for the generation of 
effector cells by the culture of tumor-draining LN cells with excessive numbers of irradiated 
tumor cells in the presence of IL-2. In IVS, the function of tumor cells was to provide specific 
antigenic stimulation while IL-2 promoted proliferation of T lymphocytes. In extensive animal 
studies with the immunogenic MCA 105 and MCA 106 sarcoma tumors, we have 
documented that IVS cells generated from tumor-bearing mice were capable of curing a 
significant proportion of animals with grossly visible metastatic disease in the lung and liver 
(5). 
The success of adoptive immunotherapy was critically dependent upon the 
immunogenicity of the tumor. In animal studies, therapeutic IVS cells could only be 
generated when the tumors were immunogenic and capable of eliciting transplantation 
resistance in their syngeneic hosts. It has not been possible to apply this approach to tumors 
which do not induce systemic immunity. A major concern for cancer immunotherapy in 
humans is whether or not immunogenic animal tumors represent sufficient models because 
human tumors originate spontaneously and may be poorly immunogenic. To address this 
issue, we examined the feasibility of generating immune effector cells with reactivity against 
poorly immunogenic murine tumors (6,7). The MCA 102 was selected for study because it 
failed to elicit systemic immunity in syngeneic B6 mice by a variety of immunization 
procedures including tumor growth and excision, immunization with irradiated tumor cells 
and with various combination of tumor cells admixed with C. parvum. Initial experiments with 
the MCA 102 tumor indicated that unlike the MCA 105 and MCA 106 tumors, therapeutic 
effector cells could not be induced from lymph nodes draining a progressive tumor. 
However, utilizing alternative methods, we found it was possible to generate sensitized 
lymphocytes with therapeutic efficacy against the MCA 102 tumor. The procedure to achieve 
this required both in vivo tumor priming and IVS. Unlike pre-effector cells sensitized to the 
weakly immunogenic MCA 105 tumor which were elicited naturally during progressive tumor 
growth, the process of inducing pre-effector cells to the MCA 102 tumor was an artificial one 
and required the use of a bacterial adjuvant, C. parvum . in addition to tumor cells 
administered as a vaccine. If properly primed pre-effector cells are induced, the IVS method 
can provide an environment for them to mature into functional immune cells. Indeed, using 
this approach, transfer of as few as 1.5 x 10 7 vaccine-primed IVS cells not only reduced 
established pulmonary MCA 102 metastases but also prolonged survival and cured tumors in 
a majority of treated animals (6). Analysis of the distribution of pre-effector cells in 
various lymphoid organs revealed that their occurrence was regionally 
restricted and only detectable in lymph nodes draining the tumor vaccine. 
During the peak response, no evidence of the existence of pre-effector cells in the spleen, 
mesenteric lymph nodes, PBL or bone marrow was seen. Based on these studies, we have 
completed a clinical study to evaluate the immunobiology and efficacy of similarly generated 
primed LN cells in patients with advanced cancer (see Section 3.2). 
The requirement for a relatively large number of autologous tumor cells for generating 
IVS cells in culture was found to be a limiting factor in its clinical application. Hence, we 
realized that alternate techniques to stimulate and propagate primed pre-effector cells 
needed to be identified. Since antigen recognition by T cells involves the TCR/CD3 complex 
for signal transduction and antibodies binding to the complex activate T cells, we examined 
whether in vitro antigenic stimulation could be achieved by using mAb to CD3 (8-10). This 
possibility was examined with several murine tumor models. A hamster mAb, 145-2C11, 
directed against the CD3e chain of the murine TCR/CD3 complex has been characterized 
[786] 
Recombinant DNA Research, Volume 18 
