Recombinant DMA Advisory Committee - 2/4/91 
Dr. Rosenberg said that the protocol had undergone a modification as a result of a 
request from the Food and Drug Administration (FDA) to lower the initial starting dose 
to 3 X 10® cells and to eliminate co-administration of 11^2. However, the FDA said that 
if this dose escalation was used, a weekly escalation was permissible rather than once 
every three weeks. When a maximum tolerated dose is reached, the dose is reduced to 
one-third of this and then a new triweekly escalation using this one-third dose, in 
combination with IL-2, is started. This modification was brought about by the desire on 
the part of the investigators to use a vector from GTI, but because the safety studies 
were carried out with a Cetus-produced vector, the FDA had thought it necessary to 
drop the starting levels. 
Therefore, the protocol had begun with a weekly dose of 3 X 10® TNF^jl cells, and that 
the first 2 patients had received these doses. On February 8, 1991, the first escalation to 
3 X 10^ cells will take place. This will cause a delay of about 5 weeks in terms of getting 
the doses of TNFjil in combination with IL-2, but that it is being done in the interest of 
patient safety. 
Because of the extensive disease in the patients currently under treatment, Dr. 
Rosenberg thought the experiment would put the investigators in a position to make a 
conunent on the effectiveness of the cells. However, because 40% of patients respond to 
TILs without the TNF gene it will require a larger series of patients to see whether or 
not this modification represents an improvement. He hoped to treat 50 patients with the 
TNFtil cells over the course of the next year. 
Dr. Mclvor asked if there was any change in the amount of expression the investigators 
were getting from the transduced TILs before they are infused into the patients. Dr. 
Rosenberg said that they were not capable of getting the gene into all of the TILs being 
produced by the patients and that it was difficult to do, but by following cell cycle 
kinetics of the TILs and adding the vector exactly when the cells were in S phase, the 
transduction efficiency was being improved. 
Dr. Mclvor asked if negative control samples were being taken from non-tumorous 
tissues to indicate homing to tumor relative to normal tissue. Dr. Rosenberg said that 
they had tried hard to get this data but that so far the only option to get normal tissue 
has been at autopsy. Biopsies have been taken from two patients who had died. Using 
PCR there was no finding of TILs in normal tissue in one case, but in the other case 
there was a positive finding in one side of the renal cortex. This was thought to be due 
to contamination and that PCR is being repeated. There was no evidence of continued 
survival in normal tissues at a sensitivity of 1 in 100,000 cells. 
Dr. Atlas noted that the original protocol was designed to determine a maximum 
tolerated dose, and he wondered if the change in the protocol, starting at a lower dose 
Recombinant DNA Research, Volume 14 
[457] 
