(HSVf/c) gene. These studies and others have demonstrated that introduction of the HSVf/c gene into 
mammalian cells renders them highly sensitive to subsequent treatment with the normally non- 
toxic nucleoside analog ganciclovir (GCV) (Moolten et al, 1990; Moolten et al, 1992; Freeman et 
al., 1993; Hasegawa et al, 1993; Caruso et al, 1993; Ram et al, 1993). One key feature of this 
system is a bystander effect where tumor growth may be slowed, or tumor tissue eliminated, after 
GCV administration, even when only 10-15% of cells are transfected with HSVf/c. The mechanisms 
of this in vivo bystander effect are currently unknown, but are under active investigation. 
To date, all of the studies using transfer of HSVf/c (except one-see below) have used recombinant 
retrovirus (Culver et al, 1992; Freeman et al, 1993, Hasegawa et al, 1993; Caruso et al, 1993; 
Ram et al, 1993). Although potentially effective, certain features of retrovirus-mediated gene 
transfer are not optimal for in vivo gene therapy. These include difficulties in producing and 
purifying high titer viral stock, the fact that only actively dividing cells allow retroviral 
integration and transgene expression, and the possibility of insertional mutagenesis (Mulligan, 
1993). In addition, because recombinant retroviral supernatant may not reliably infect tissues in 
situ, syngeneic "producer cells" are required (Ram et al., 1993), confounding interpretation of 
subsequent effects and raising potential safety issues. 
For these reasons, our group has begun to explore the possibility of using adenoviruses to deliver 
the HSVf/c gene to treat malignant mesothelioma (Smythe et al, 1994a; 1994b; 1994c- See 
Appendix). In contrast to retroviruses, adenoviruses have the ability to directly transduce tissues 
in vivo, can easily be produced in high titer, do not require cell division for transduction of cells, 
and have an established record of safety as vaccination material (Kozarsky and Wilson, 1993). 
Although the duration of gene expression may be short, this limitation may be less important where 
the goal of therapy is rapid eradication of tumor cells. 
V.A.4. Gene Therapy for Mesothelioma: A Model to Study Gene Therapy for 
Localized Malignancy? 
Although our protocol will focus on a group of patients with mesothelioma, we feel that this tumor 
will serve as a valuable paradigm for other tumors that exert morbidity and mortality by growth 
within a defined anatomic space and for the application of gene therapy to tumors, in general. Such 
cancers would include those of the brain, ovary, bladder, as well as metastatic disease to the 
leptomeninges. Other investigators have demonstrated the feasibility of this approach in brain 
tumors and leptomeningeal metastasis using HSVf/c-producing retroviruses (Culver et al, 1993; 
Ram et al, 1993; Ram et al., 1994). Recently, Woo's group has shown that the size of brain tumors 
in mice could be reduced after stereotaxic implantation of an Ad.RSVf/c vector (Chen et al, 1994). 
Efforts are currently beginning at the University of Pennsylvania to apply this technology towards 
the treatment of brain tumors. 
Because there are currently no approved protocols using adenovirus for cancer therapy, we feel it 
will be important to obtain clear biologic endpoints of response. For this reason, we have 
incorporated a biopsy procedure designed to measure the efficacy of viral transduction and the local 
effects of such transduction. The ease with which pleural biopsies can be obtained compared to brain 
biopsies is an advantage of this protocol. 
V.A.5. Gene Therapy of Mesothelioma: Short and Long Term Objectives 
The need for alternative approaches to the treatment of mesothelioma are clear. Our preliminary 
data suggest that Ad.RSVf/c can penetrate deeply within established tumors and effect tumor 
eradication. Our short term goals will be to 1) determine the safety and efficacy of instilled 
Ad.RSVf/c vector followed by treatment with GCV and 2) to analyze some of the biologic and 
immunologic features of this drug delivery system. Based on these results, we will design studies to 
more closely approach our long term goal, the cure of patients with malignant mesothelioma. This 
may involve multiple treatments. It is also possible, if not likely, that because of the large amounts 
Recombinant DNA Research, Volume 20 
[395] 
