neurologic function. One animal was euthanized after one week (monkey #1). Lymphocytes and foamy 
macrophages were found within the injected areas, and the numbers decreased with increasing distance 
from the injection site. This is consistent with a localized area of inflammation. Monkey #2 was 
treated in the same manner as monkey #1 but was clinically followed. He remained well, and MRI and 
PET scans after one month revealed no abnormalities. In vitro tests of peripheral blood lymphocytes 
showed no evidence of a increased T cell response to adenovirus infected fibroblasts. Of the two other 
monkeys studied, both remain clinically well. One monkey #4, developed MRI changes at the injection 
site consistent with localized edema. This finding persisted for greater than one without an increase or 
decrease in size. Monkey #3 that was similarly treated at a lower dose developed no MRI changes over 
the same time period. 
These studies suggest that recombinant adenovirus administered into the CNS of primates produces a 
localized dose dependent inflammatory response which is clinically inapparent. These studies suggest 
that the toxicity of intracranial adenovirus administration will not be of a magnitude that precludes its 
use in humans. 
II. B. 7 Adenovirus-Mediated TK Gene Transfer In Rats with CNS Tumors 
Intracerebral gliomas were induced by injection of 10 5 9L gliosarcoma cells (5 pi) in the forebrain of 
Fischer rats. The location and size of the tumor was determined by MRI imaging on approximately day 
16 post implantation. Using the MRI scan information, the glioma mass was injected with 
H5.010RSVTK'. Treatment with ganciclovir (15mg/kg) was initiated 24 hours post virus injection and 
was repeated twice daily thereafter. The growth of the untreated glioma has been previously shown to be 
exponential over the entire growth period (B. Kim and B. L. Davidson, unpublished data). The tumor 
doubling time was determined before H5.010RSV7X administration and GCV treatment and is shown in 
Table I. Marked growth retardation during the treatment period was observed. All of the rats treated 
with Ad.RSVIacZ and GCV succumbed to their tumors by day 28. In contrast, the rats treated with 
HS.OIORSVTK and GCV, showed a significant delayed in mortality, with 2/8 of the rats surviving to day 
40. 
rat # 
Tdfhrsl pre- 
Td fhrsl Dost- 
Volume (mm 3 ) 
pre-treatment 
1 
44 
964 
1 25 
2 
59 
276 
1 00 
3 
5 1 
206 
1 50 
4 
59 
• 
1 28 
5 
43 
213 
200 
6 
37 
140 
200 
Table I. Tumor doubling time (Td) before and after H5.010RSV7X/GCV treatment 
II.B.8 Adenovirus-Mediated TK Gene Transfer In a Human Brain Tumor 
Xenograft Animal Model. 
In Vitro Studies: We have conducted a survey of primary human brain tumor cell lines, each of which is 
tumorigenic in athymic mice or rats. Human glioma cell lines included U251, U87, U373, WF, JHG31, 
and JHG10, and A172. Medulloblastoma cell lines included D324, D425, D283, and CHOP 707. All cell 
lines are continuous (i.e. passaged >70 x) and are well-characterized. Adenovirus transduction 
efficiency was evaluated by in vitro transfection of glioma or medulloblastoma cells with an adenovirus 
vector containing the p-galactosidase reporter gene (H5.010CMV/acZ). At infection ratios of 10 viral 
particle per cell, 48-70% of cells demonstrated nuclear staining. At higher infection ratios, 100 
[358] 
Recombinant DNA Research, Volume 20 
