M.J. Welsh and A.E. Smith, RAC Application 
in hamsters at doses that in some cases greatly exceed the proposed human dose, both in 
terms of total virus added and multiplicity of infection per cell. 
6.4 Will the Virus Recombine? 
Homologous recombination occurs at reasonable frequency within adenoviruses of the 
same subgroup (105), but the products of such recombination would be very similar to the 
starting materials. Recombination between subgroups does not appear to occur. The only 
viable products of non-homologous or illegitimate recombination would necessarily 
contain El -like sequences and delete other sequences in viral or CFTR coding regions. 
Such products would likely be overgrown by wild-type virus and subsequently become self 
limiting. To ensure this, we will only treat patients seropositive for Ad2. 
6.5 Will Virus Growth Be Complemented? 
The growth of the El-deleted vector could be complemented by provision of El gene 
function. Growth of the present vector is so disabled, however, that even under the most 
favorable conditions, wild-type virus rapidly overgrows the defective virus. 
Complementation by provision of a cellular protein with El -like activity is a theoretical 
possibility (101,102). We have not detected virus replication in tissue culture cells or in 
animals, but this possibility cannot be discounted. For this reason we plan as an exclusion 
criterion the presence of El sequences detectable by PCR in the nasal brush samples taken 
prior to treatment. 
6.6 Will the Virus Be Sufficiently Pure? 
To alleviate the possibility of introducing another human virus along with the adenovirus, 
the parent cell lines and the viral seed stock will be extensively tested for biological agents. 
We will also test for cellular transforming activity in the preparations. A PCR procedure 
capable of detecting 1 part wild-type virus in 10 8 El deleted virus will also be used. The 
presence of wild- type virus below this level perhaps originating from recombination 
between Ad2/CFTR-1 and the endogenous Ad5 El sequences in 293 cells should become 
apparent in the extended in vitro viral tests on the virus seed stock. 
6.7 Will Administration Be Immunogenic? 
Our data in monkeys and hamsters show that they mount an immune response to introduced 
adenovirus vector. This could lead to a reduction of efficacy in protocols requiring 
multiple administrations and we plan to test this in animals. However, for the present 
human protocol we plan only a single application in each patient. 
6.8 Does the Protocol Minimize Risk? 
The present protocol proposes to treat 3 patients and to use minimal amounts of virus 
consistent with the gathering of safety and limited efficacy data. We believe that the 
provisions for the health of the patients, their care givers, visitors, and the public do 
minimize risk. 
Recombinant DNA Research, Volume 16 
[869] 
