4.4 Preparation and Injection of Fibroblasts. 
When the total population of fibroblasts reaches about 1 X 10^ cells, the 
medium will be changed and the amount of IL-12 appearing in the culture 
supernatant of representative flasks over 24 hours will be determined by ELISA 
for the heterodimeric human IL-12 and in a bioassay. This will be used to 
estimate the amount of IL-12 that may be elaborated at each injection site per 
10' cells per 24 hours. Fibroblasts will then be harvested by brief 
trypsinizatlon and prepared for injection with diluted in appropriate amount 
of saline. 
Accessible cutaneous tumor sites will be measured and recorded. If there 
is only one or two cutaneous lesions accessible in patient's skin, the 
injection sites will be restricted to the circumference of the tumor so that 
control fibroblasts are Injected at 12 0' clock and 1/3 of the preparation 
will be injected at 1, 5, and 9 0' clock. Similarly, the second Injection will 
be at 2, 6, and 10 O'clock as described in Fig. 5. Subsequently, Injection 
sites will be marked with India ink. Injection sites can be identified and 
evaluated after the treatment. If there are many (more than four) accessible 
cutaneous lesions which are smaller than 1 cm in diameters, one or two lesions 
could treated in each injection day. 
Patients will first undergo primary peritumoral injection at tumor sites 
using transduced fibroblasts or non- transduced fibroblasts at a single control 
site. A preparation of fibroblasts will be suspended in 0.1 to 1 ml of sterile 
saline depending on the tumor size and amount of fibroblasts. The cells will 
be irradiated with 5,000 rads prior to ini action. At the time of 
administration, a small aliquot of fibroblasts will be set aside to test for 
lack of growth and for production of IL-12 following irradiation. 
The number of transduced fibroblasts to be injected will be determined 
by the estimated amount of IL-12 secreted from them based on the results of 
vitro assay. The estimated amount of secreted IL-12 will be 10 ng/24 hours in 
total for each weekly injection for the first three patients. The value of 10 
ng/24 hours is the achievable amount of IL-12 secreted from 1 x 10° 
fibroblasts transfected with our current vector system and about 1/5 of the 
dose used in our day 7 established murine tumor models. This amount is about 
1/70 of the amount of the proposed initiating dose in phase I clinical trial 
of Intravenous injection of recombinant human IL-12 (10 ng/kg) , and it is 
1/350,000 of the maximal tolerated dose in primate models when given 
systemlcally . Thus, we consider this dose reasonable and safe as an 
initiating value. After evaluation of adverse effects, the amount will be 
increased to 30 ^ after at least two weeks of observation of the initial 
three patients. This sequence will be followed by 100, 300, 1,000 and 3,000 
ng/24 hours after the completion of the evaluation of three patients at each 
dose level. 
Each week after injection, the size and characteristics of the Injected 
tumors will be measured in two diagonal diameters and recorded. This injection 
will be repeated provided that the patient has not suffered a severe local 
reaction such as abscess or ulceration. The same dose of IL-12 producing 
fibroblasts per tumor will be Injected, each week for a total of four weeks. 
The local level of IL-12 in regressing lesions has not been directly measured. 
Although direct determination of this value is difficult, previous study of 
IL-4 transfected fibroblasts in clinical trials of IL-4 gene therapy showed 
that mRNA expression of transfected IL-4 can be detectable up to seven days 
with a gradual decrease over this time period. It is a general understanding 
that in vivo expression of the transfected gene is less than that observed Jji 
vitro . Thus the amount of IL-12 secretion in the regressing tumor lesions 
should be smaller than the calculated amount based on the results in vitro 
which would be estimated as 130 ng/48 hours. 
( 654 ) 
Recombinant DNA Research, Volume 19 
