To address these concerns we have been developing ways to deliver to the 
synovial lining of joints genes which encode potentially therapeutic proteins (fig. 6). 
When successful, cells within the synovium now synthesize the proteins of interest which, 
in the case of secreted proteins, accumulate intraarticularly, that is at the site of disease. 
In this way, the joint now becomes the site of synthesis of its own therapeutic agent. 
Thus in the sense on which we are applying it, gene therapy is being used as a drug 
delivery system (8,24,27). Depending upon the turnover of cells within the synovium, the 
persistence of the transferred gene and promoter activity, there is the potential for 
prolonged expression of the therapeutic DNA. 
Genes may be delivered to synovium by in vivo or ex vivo approaches (fig. 7). 
Our group is actively investigating both of these. However, at the present state of 
development of the field, with safety as a paramount consideration, we strongly favor the 
ex vivo method for human trials. There are several reasons for this. With the ex vivo 
approach no viral particles are introduced into the joint. This means that there can be 
no sensitization of the patient to viral protein. In addition, it gives the investigator full 
control over the cell type which becomes transduced. Injection of vector directly into the 
joint risks the transduction of lymphocytes as they traffic through the joint and out again 
to other parts of the body. The long-term effects of transgene expression at untargeted 
organs are unknown. Furthermore, with the ex vivo method, all genetic manipulations 
occur outside the body and thus permit all and any required tests of the transduced cells 
to be carried out prior to their reintroduction into the body. 
As described in the following section, we have used the rabbit knee as a model 
system for developing an ex vivo method for delivering genes to the synovial lining of 
joints (28,29). It involves the surgical removal of synovium from the donor, its growth in 
nutrient medium in vitro and its transduction with a replication-defective retrovirus 
known as MFG. An almost identical virus has already been approved by the RAC 
committee for use in ex vivo gene therapy for cancer (Somatix group) and for Gaucher 
disease (P.I.-J.A. Barranger). The principal investigator on the latter protocol is a 
consultant to the present one. One of the co-investigators of the present protocol (Paul 
D.Robbins, Ph.D.) developed MFG in the laboratory of Dr. R. Mulligan of the 
Whitehead Institute. 
In the present protocol MFG will be used to carry into human synovial fibroblasts 
(type B synoviocytes) two genes. One codes for IRAP and the other for a so-called 
"suicide" gene, herpes simplex thymidine kinase C which renders cells expressing it 
sensitive to the antibiotic ganciclovir. This is used as an additional safety measure 
(q.v.). By convention, an MFG virus carrying two transgenes is called DFG. Thus the 
vector we propose to use in our clinical trial is designated DFG-IRAP-tk. 
E. Proposed Clinical Study 
Flow diagrams outlining the proposed clinical trial are shown in figures 8 and 9. 
[4301 
Recombinant DNA Research, Volume 19 
