HEPATOCELLULAR TRANSPLANTATION AND TARGETING GENETIC MARKERS TO HEPATIC CELLS 
It should be recognized that there are critical differences in the anatomy of the 
human liver compared to various commonly used animal models. We have already observed 
significant differences in the pattern of engraftment of transplanted hepatocytes in the 
mouse model (Ponder et al, 1991) and in published rat and dog experiments (reviewed 
above). Retroviruses are known to have species-specific patterns of infectivity and 
pathogenicity (Teich, 1984). There are significant differences in the conditions for 
cultivation and cryopreservation of hepatocytes from even closely related species. 
Furthermore, it may not be possible to model the specific pathological entities that 
cause hepatic failure in children using available animal models. There is no reason to 
assume a priori that any particular species represents an appropriate model for human 
experiments and that the success or failure of these experiments in any species will 
presage human trials . 
Data from experimental animals is critical when specific questions can be addressed 
in appropriate animal models. For example, primate data has been crucial in assessing 
the safety of murine retroviral vectors. This proposal is predicated on the safety 
studies that have been performed in primates by the NIH group (Cornetta et al, 1990). 
Dogs have been used to develop technical expertise for portal vein, splenic vein, or 
intrasplenic injection since these animals are an appropriate size model for children 
(Brandt et al, unpublished data). Mice have been used to established the feasibility 
of attaining long term hepatocellular since this work benefited from the availability 
of transgenic markers. We believe that judicious clinical trials of gene transfer into 
hepatocytes at this time, in coordination with continuing investigations in animal 
models, may ultimately reduce the number, extent, and danger of human investigation by 
providing a baseline of information from human subjects. (Ledley, 1991). 
The Code of Federal Regulations defines limitations on research which may be 
performed in pediatric patients. Research with "greater than minimal risk" (defined as 
the "risks of harm anticipated in the proposed research are not greater, considering 
probability and magnitude, than those ordinarily encountered in daily life or during the 
performance of routine physical or psychological examinations or tests") may be 
performed in pediatric subjects even if there is no benefit to the study subject it the 
risks are "reasonably commensurate" with the course of conventional therapy of the 
patient's disease and a significant amount of "generalizable knowledge" may accrue 
(Code of Federal Regulations, 1983; Andrews, 1987; Ledley, 1991). We believe that there 
is little risk inherent in the use of these vectors in human subjects. Furthermore, we 
believe that the amount of "generalizable knowledge" which may accrue from these studies 
relative to the limited additional risk of gene transfer justifies the use of these 
vectors in the present experiment. The opportunity to further the critical assessment 
of hepatocellular transplantation techniques as a therapeutic modality, as well as the 
opportunity to test the applicability of gene transfer technologies in human subjects, 
warrants the small risk inherent in the use of recombinant retroviral vectors in human 
trials . 
Recombinant DNA Research, Volume 14 
[795] 
