Gene Therapy for CF using Canonic Liposome Mediated Gene Transfer: Phase I Trial 
doses, broad ranges can usually be defined for most cell types in which efficient gene 
transfer occurs with minimal cytotoxicity. Our in vivo studies, as well as studies from 
other laboratories, have failed to show significant local respiratory mucosal toxicity over 
the range of lipid/DNA conjugate concentrations which will be used in this protocol. The 
risks also include long term scaring of the mucosa beneath the inferior turbinate, 
although the risk of chronic scarring occurring after a single administration seems very 
small. Humoral immune response to lipid/DNA conjugates could occur, and it is possible 
that immune response against liposomes containing DNA might limit repeated 
administration of the CFTR. Because DNA in large amounts is constituitively present 
within the lung (and appears to contribute to CF sputum viscosity) it seems less likely 
that a single administration of plasmid DNA within the nose will lead to a clinically 
significant or limiting immune response. In addition, lipid carrier molecules (for use in 
delivery of drugs, recombinant proteins, etc.,) in general have not been highly antigenic 
in human or other in vivo studies. Nevertheless, possible systemic response to localized 
nasal administration of lipid/DNA conjugates (including antibody generation to these 
conjugates), will be monitored during this study. The characteristics of any such 
responses will be carefully documented. Because some evidence suggests that the Delta 
F508 CFTR may not appropriately target to the apical cell membrane, it is possible that 
over-expression of a wild-type CFTR on the cell surface could evoke antibodies directed 
against the CF protein itself. However, 1) cellular synthesis without membrane targeting 
does not preclude immune tolerance and 2) experiments in transgenic mice in which 
human CFTR was over-expressed in respiratory mucosa failed to show any adverse 
effects, suggesting that expression in an animal naive to wild-type human CFTR did not 
evoke a significant anti-CFTR immune response. Finally there is 1) a theoretical risk of 
integration of plasmid within the host genome, resulting in malignant transformation of 
the host cell, and 2) development of replication competence by transfected plasmid DNA. 
In general, these risks are much smaller than those observed with the viral based DNA 
delivery systems (e.g., retroviral, adenoviral). Such events have never been observed 
using lipid/DNA conjugates as reviewed by Nabel (55). 
b. Risks associated with procedures to test safety and efficacy. 
1. Blood drawing. Venipuncture and arterial sampling (for blood gases) can produce 
modest discomfort and bruising as well as vasovagal symptoms. Local anesthesia 
(with Xylocaine) will be used during arterial blood gas sampling if requested. 
Risk and damage to the vascular system of the hand after radial artery puncture 
is very small and can be minimized further by assessing collateral circulation to 
the palm using an Allen test. These procedures are without substantial long term 
risk. 
2. Measurement of nasal potential difference. This is associated with minimal 
discomfort at the site of the reference needle placed subcutaneously in the 
forearm. Potential risks include bruising, bleeding, and infection, but these have 
never been observed at the reference needle site in over 50 individual measure- 
ments performed by Dr. Sorscher. No cutaneous infection has been observed in 
over 10 years of experience as reported by the group which developed the nasal 
potential difference protocol at the University of North Carolina, Chapel Hill. 
Perfusion of Ringer’s lactate onto the nasal mucosa is usually very well tolerated 
Recombinant DNA Research, Volume 18 
[835] 
