July 30, 1990, Human Gene Therapy Subcommittee 
tumor site without systemic toxicity, the same antitumor benefits might be achieved. 
This is the rationale for this proposed protocol. 
Dr. Rosenberg presented a series of slides. He began with slides of the ongoing N2/TIL 
protocol showing that 5 patients had TIL cells bearing the marker gene after 21 days, 
which is the time at which IL-2 supplementation is discontinued. Further, the cells were 
found in tumor biopsies of the patients. He noted that all safety studies required by the 
subcommittee had been performed and were negative without exception. Further, safety 
studies of the viral supernatant that transits to TILs (tests of patient exposure to virus) 
have also been negative. He said that these same tests would be performed in the 
TILpNF protocol. 
Dr. Rosenberg said the current attempt to introduce the TNF gene is based on the 
observation that TIL recirculate and localize to cancer deposits, both from scans with 
indium^^^ and from actual measured accumulation of radioactivity in tissue. He also 
presented data showing that human tumors in nude mice are made to regress due to 
introduction of transduced cells using the same vectors that are expected to be used in 
the protocol. Further, he presented data from three studies that showed that local 
injection of TNF into human tumors results in tumor regression. 
Dr. Rosenberg noted that an addendum to the protocol contains results of experiments 
by Dr. Attan Kasid which show that the viral vector has stable expression of the TNF 
gene and that TILs can be produced which will make 100 times the normal amount of 
TNF. 
He said the protocol will be a classic Phase I study which has been used for evaluation 
of a very large number of chemotherapeutic agents in patients with advanced cancer. 
The study will begin with a dose of non-selected TIL that has been determined to be 
non-toxic and then will be escalated at 3-week intervals until toxicity is reached. Only 
after a maximum tolerated dose has been reached will selected cells be given, dropping 
back to a level of 1/10 the tolerable dose of non-selected TIL, and dropping back to 1/3 
the concentration of IL-2 to be administered. 
Dr. Rosenberg said the initial dose of TILtnf to be infused will be .07 micrograms of 
TNF per kilogram. This is approximately 1/100 the dose of TNF which is maximally 
tolerable in humans by i.v. administration, which is 8 mg/kg per 24 hours. This amount, 
coupled with a half-life of a few minutes, should be less than one percent of the safe 
tolerated dose. Further, non-transduced TIL make 20 picograms of TNF per 10^ cells, 
and with high doses of IL-2, as many as 6 X 10^^ TIL have been infused into patients, 
thereby producing .2 mg/kg per day of TNF per patient, which is 3 times the starting 
dose to be used in the TIL^nf protocol. 
Insofar as organ-specific toxicity is concerned, patients have already been treated with 
Recombinant DNA Research, Volume 14 
[189] 
