patients with a variety of metastatic cancers with their own resected tumors 
in graded doses (27,28). This revealed that the majority of patients could 
grow tumors at these inoculation sites if an adequate inoculum was 
administered. These were all patients with widely metastatic disease or 
unresectable advanced cancers and no impact from the inoculations on their 
overall clinical course was identified. For patients with widely metastatic 
cancer, a very small local cutaneous tumor inoculation (representing a 
fraction of their progressive metastatic disease) and subsequent resection of 
any growing tumor, is unlikely to significantly accelerate the course of their 
systemic disease. In support of this, one can cite the extreme case of tumor 
auto-inoculation which occurs when large numbers of malignant cells are 
intravenously infused as a result of peritoneal-venous shunting to palliate 
malignant ascites. Multiple clinical and post-mortem studies fail to show 
significant decreases in survival or alterations in the pattern of metastatic 
disease in shunted versus non-shunted patients (29-31). Such shunted patients 
can develop microscopic metastatic implantation, but these studies show that 
these implants fail to reach a significant size prior to the patient's death 
from their pre-existing known metastatic disease. Certainly the cells which 
might escape from a small, cutaneously-implanted tumor site into the systemic 
circulation (if this occurs at all), would be far less than that from the 
intravenous auto- inoculation which occurs on a daily basis in these shunted 
patients or in any patient with widely metastatic cancer. 
Thus, data suggests that auto-inoculation of a small amount of viable 
tumor at an isolated cutaneous site will often generate tumors for TIL growth, 
but that in the setting of widely metastatic cancer, such an approach is 
unlikely to significantly affect survival or the disease course of such 
patients. In the experiments in this protocol, however, the introduction of 
the IL-2 gene into tumor is designed to increase the immunogen icity of the 
tumor (as reviewed earlier) and will prevent tumor growth in most cases. The 
use of draining lymph node lymphocytes can thus provide a source of cells for 
use in adoptive immunotherapy of these patients. 
Recombinant DNA Research, Volume 15 
[1111 
