thus protecting the epithelium from the damaging effects of neutrophil 
elastase) , aerosolized reduced glutathione (to augment the antioxidant 
protective screen of the airway to protect the epithelium from the inflam- 
matory cell derived oxidant burden) and aerosolized amiloride (to rees- 
tablish the ionic milieu of the epithelial surface) (Hubbard et al.,1992; 
McElvaney et al.,1991; McElvaney et al. , 1992; Knowles et al . , 1990; Roum, 
J. and Crystal, R.G. unpublished observations). Although there are encour- 
aging results from all of these agents, all have been evaluated only in 
phase I or II studies, and none attack the primary abnormality in the 
disease. Lung transplantation, including heart-lung, double lung and single 
lung procedures has been carried out in 312 individuals with CF (worldwide 
through 1991). The three year survival rate in this cohort is 52%. The 
average cost of these procedures is $150,000-200,000, with an estimated 
cost of $28, 000/year for follow up treatment. Finding suitable donors is a 
major problem, with typical waiting periods of 9 to 12 months. 
1.5 Rationale for Gene Therapy to Treat the Respiratory Manifestations of 
Cystic Fibrosis 
The identification of the CFTR gene in 1989 (Kerem et al . , 1989; Riordan et 
al., 1989; Rommens et al., 1989) opened the door to strategizing therapies 
for CF in which the normal gene would be transferred to somatic cells of 
individuals with CF, thus reversing the biologic abnormalities consequent 
to mutations of the two parental CFTR genes. Because the respiratory mani- 
festations of CF dominate the clinical picture, it is the most rational 
target for gene therapy for this disorder. The concept that the respiratory 
manifestations of CF are a good candidate for gene therapy is based on 
several facts. First, the respiratory disease is confined to the epithelium 
of the airways. In vitro studies have demonstrated that the CFTR protein is 
a Cl“ channel that modulates the secretion of Cl - in response to elevations 
of intracellular cAMP (Anderson et al., 1991a; Anderson et al., 1991b; Bear 
et al. , 1992; Drumm et al . , 1990; Rich et al . , 1990). Mutations of the CFTR 
gene render epithelial cells unable to modulate Cl - permeability through 
the cAMP pathway (Frizzell et al., 1986; Hwang et al . , 1989; Li et al., 
1988; Li et al., 1989). It is this biologic abnormality that is believed to 
cause the respiratory manifestations of the disease (Collins, 1992; Welsh 
et al. , 1992; Welsh and Fick, 1987). Second, in vitro studies have shown 
that transfer of the normal CFTR cDNA to epithelial cell lines derived from 
individuals with CF can override this abnormality and permit the cells to 
secrete Cl - in response to increased intracellular cAMP (Drumm et al . , 
1990; Rich et al., 1990). Finally, the lethal consequences of mutations of 
the gene occur almost exclusively in the lung (Welsh and Fick, 1987) . 
Together, these concepts suggest the feasibility of somatic gene therapy 
for CF, i.e., it should be possible to correct the pulmonary manifestations 
caused by mutations of the CFTR gene by directly transferring a normal CFTR 
cDNA to airway epithelial cells. 
The architecture of the airways demands that if gene therapy for the respi- 
ratory manifestations of CF is to be successful, the transfer of the normal 
CFTR cDNA to the airway epithelium will have to be carried out in vivo via 
the air side of the epithelium. On clinical and technical grounds it is not 
[626] 
Recombinant DNA Research, Volume 16 
