Genzyme Corporation Ad2/CFTR-2 Gene Transfer Protocol: Single 
One Kendall Square, Cambridge, MA 02139 Aerosol Administration to the Lung of CF Patients 
one of most concern relates to the possibility that administration of adenovirus may induce an inflammatory or immune 
response. An additional concern is the possibility of virus replication. 
Evaluation of Clinical Efficacy of Gene Transfer in CF : A number of potential problems can be anticipated in the 
evaluation of clinical efficacy of gene transfer in the lung. As a result, it may be difficult to determine whether clinical 
benefits outweigh any adverse effects associated with vector administration. Because CF is a chronic disease, tests of 
severity made over a short period of time may not be sensitive or may be subject to error. If one chooses to study 
patients with severe disease [FEM^ of 20-30% predicted), then there will be a large amount of irreversible bronchiectasis 
and other pathology that would not be expected to be affected by gene transfer. Moreover, the 2 year mortality rate for 
those patients is 50% [14] . In contrast, those patients with mild lung disease (FEV] > 61%) have mortality rates of less 
than 5% in 2 years. Thus, assessment of mortality is impractical as an endpoint in evaluation of therapy. 
Pulmonary function tests are frequently helpful as an endpoint but they are not without problems. They are subject to 
fluctuations by factors other than those being investigated. Perhaps evaluation of pulmonary anatomy by high- 
resolution chest CT, and function evaluated by assessment of mucociliary clearance, and cardiopulmonary exercise 
testing may be more sensitive and specific. In any case, it is doubtful that clinical efficacy will occur with 1 
administration of Ad2/CFTR*2. As the great majority of cells within the respiratory epithelia are terminally differentiated 
and turn over eventually, continued CFTR expression within airways will require periodic reapplication of Ad2/CFTR-2. 
Measurement of changes in clinical endpoints will possibly require restoration of CFTR function over a period of several 
months at least. However, we must first establish the safety of the vector in the lung prior to attempting to establish its 
efficacy by repeat administration. 
Design of Ad2/CFTR-2 : Ad2/CFTR-2 is diagrammed in Figure 1. Aside from the difference in the virus serotype, this 
vector, like our previous vector, Ad2/CFTR-1 , differs from Ad5-based CF vectors by retaining the E3 region. E3 
sequences were retained in these vectors because E3-deleted, but otherwise wild-type adenoviruses have been shown 
to be more pathogenic for rodents. It should be noted however that expression of E3 in El -deleted vectors has not 
been reported. The Ad2/CFTR-2 DNA construct was derived from a full length copy of the Ad2 genome from which the 
El genes had been deleted and replaced by an expression cassette encoding CFTR. The expression cassette included 
the promoter for phosphoglycerate kinase (PGK) and a poly A addition site from the bovine growth hormone gene. PGK 
is not a strong promoter, but the endogenous levels of CFTR in human airway cells are not high perhaps meaning that 
strong promoters are not required. In addition, the E4 region of Ad2 was deleted and replaced by open reading frame 6 
(ORF6) of the Ad2 E4 region. The minimal E4 function required for virus replication in vitro has been retained. The 
bovine growth hormone poly-adenylation signal has been incorporated into Ad2/CFTR-2 downstream from the CFTR 
cDNA. 
Figure 1 : Ad2-Based CFTR Gene Therapy Vectors 
^^^^^JMaioM^teJYanscn^tion^ 
^ El a El b Ji3 
PGK 
Promoter 
Ad2/CFTR-2 
BGH poly A' 
ryyyyi 
aryyy^ 
I Ela 
* Promoter 
CFTR 
Ad2/CFTR-1 
pIX 
IXXXxKX>l IKXX>» 
CFTR 
pIX 1 P< 
E4 
AE4 
+ORF6 
Poly A 
Recombinant DNA Research, Volume 20 
[549] 
