associated with malignancy in mice will not occur in humans. 
4) Inability to transduce murine tumor Infiltratine T lvmphocvtes 
At the time of the Subcommittee meeting we felt that our inability 
up to that point to transduce murine CD8 TIL was a fluke considering our 
experience in successfully transducing murine CD4 T cells, and the ease with 
which we were able to transduce human CD4 and CD8 TIL. In fact, we thought we 
had murine TIL transduced at the time of the meeting. Consequently, we readily 
accepted the recommendation that we repeat our murine T cell data with murine TIL 
even though we all recognized that it was exceedingly unlikely that any new 
information would arise from the repeated experiment. To our consternation, 
murine TIL are not transduced with either ecotropic- or ampho tropic -packaged 
retroviral vectors. Table 1 lists the variables we have tested. Our conclusion 
is that murine C57 BL/6 (and BIO) CD8 T lymphocytes either do not possess the 
surface receptors that recognize murine retrovirus or the cell machinery itself 
prevents internalization, replication, or integration of the retroviral vector. 
We are now making a major effort to develop a murine TIL model that can be 
used to study re troviral-mediated gene transfer since the next step in our 
program is to insert therapeutic (i.e., anti-tumor) genes into retroviral vectors 
in order to improve this approach to the adoptive immunotherapy of cancer. 
However, we do not believe that the present protocol specifically requires murine 
TIL experiments to satisfactorily answer the questions raised by the 
Subcommittee . 
B) Human TIL : 
1) Recommendations from the Subcommittee: 
a) Data demonstrating that the human TIL that are marked by the vector 
are representative of the relevant cell populations 
b) Data from a "dry run" with human TIL which demonstrate a lack of 
infectious helper virus by the most sensitive assays available 
2) Introduction 
We have no problem with the two human TIL recommendations. The 
critical one is the second (l.b.) since this is the central risk issue in the 
risk/benefit analysis. 
3) Data with human TIL 
a) Our data indicate that vector-marked TIL are representative of the 
TIL population. We have obtained several types of data (see below) which 
indicate that vector -marked TIL are representative of the whole TIL population 
but to a different extent in different patients. There does appear to be a 
greater tendency to recover CD8 TIL rather than CD4 TIL following transduction 
and growth under selective conditions although clearly both CD4 and CD8 TIL 
populations can be infected with our vector and express G418 resistance (Table 
[ 210 ] 
Recombinant DNA Research, Volume 13 
