IVS lymphoid cells reached maximum density. IVS cells proliferated up to 14-fold during an 
average period of 12 days. 
A mean of 6.7 x 10 9 cells were infused in 10 patients (7 melanoma, 3 renal cell 
cancer) along with the concomitant administration of IL-2 (180,000 lU/kg q8h for 5 days). 
Phenotype analysis of the IVS-LN cells revealed 78% T cells which were predominantly 
CD4+. Seven of 9 patients that received IVS-LN cells developed delayed-type 
hypersensitivity (DTH) to autologous tumor suggestive that antitumor reactivity was passively 
transferred. Of the 10 patients treated with IVS-LN cells and IL-2, there was one partial and 
one minor response, and one patient remains with stable disease at 36+ months. 
This initial study indicated that vaccine-primed LN cells harbored tumor reactive 
lymphocytes and that additional studies to generate larger numbers of cells might be 
required to evaluate their efficacy. A limiting factor for generating larger numbers of IVS cells 
was the amount of available tumor for the culture procedure. Therefore, we developed an 
alternate in vitro method to generate immune cells from tumor-primed pre-effector cells 
without the requirement of tumor stimulator cells (see Section 2.1 ). This alternative method 
utilized the stimulation of pre-effector cells by an anti-CD3 mAb for 48 hours followed by their 
expansion in IL-2 (10 u/ml). From our animal studies, this activation procedure resulted in the 
expansion and maturation of highly specific effector T cells. 
Utilizing this approach we initiated a second clinical study (19). Patients were primed 
to autologous tumor utilizing the same vaccination procedure as was used for the IVS 
protocol. Tumor-primed LN cells were harvested 10 days after vaccination and activated by 
the anti-CD3/IL-2 method. Since it was anticipated that the primed LN would expand 
approximately 5 to 6-fold after the first activation procedure, further activation utilizing the 
identical procedure would be performed a second time. Animal studies in our laboratory 
have demonstrated that the antitumor efficacy of these activated cells were maintained during 
a second activation procedure. We anticipated that this would enable at least a 30-fold 
expansion of 10 9 primed LN cells which could be obtained from each patient. Not only would 
this represent a significantly larger number of immune cells generated for each patient, it 
would also increase the number of patients who could potentially be treated since large 
numbers of tumor cells were not required. 
For anti-CD3 activation, lymphocytes were suspended in CM at a concentration of 2 x 
10 6 cells/ml in flasks coated with immobilized anti-CD3 mAb (OKT3). After 2 days of 
activation the cells were harvested, washed and expanded in IL-2. Expansion in IL-2 was 
accomplished by resuspending 3 x 10 5 anti-CD3 activated cells/ml in fresh CM containing 10 
u/ml of IL-2. Cells were grown to maximum density and then re-cultured in a secondary anti- 
CD3/IL-2 activation procedure. This activation scheme is illustrated in Appendix A with a 
patient who had 5 x 10 8 LN cells activated in vitro over a 13-day period to a final cell number 
of 1 .4 x 1 0 1 1 effector cells. 
To date 14 patients (9 melanoma and 5 renal cell) have been treated on this protocol. 
A median of 2.8 x 10 10 activated cells were administered in conjunction with IL-2 (360,000 
lU/kg q8 x 5d). Phenotype analysis of the anti-CD3/IL-2 activated LN cells revealed that 
CD8 + T cells were preferentially activated. Among the 9 patients with melanoma, 1 had a 
partial and 2 minor responses to treatment. Of note, among the 5 patients with renal cell 
cancer, 1 had a complete and 3 partial responses to therapy. Among the 5 patients with 
complete or partial responses, 3 developed DTH reactivity to autologous tumor. By contrast, 
among the remaining 9 patients, only 1 developed DTH reactivity to autologous tumor. 
Recombinant DNA Research, Volume 18 
[789] 
