adjacent to the injection site. Rare foci of perivascular infiltrate was seen in the left 
hemisphere, brainstem, and cerebellum. Multifocal subarachnoid lymphocytic infiltration 
was present predominantly over the right hemisphere, but was present to a lesser extent 
over the left hemisphere and around the brainstem and cerebellum. No choroid plexus or 
ependymal destruction was present although focal choroid plexus inflammation was seen on 
the right in the animal surviving for 10 days. No viral intranuclear inclusions were seen. 
Luxol fast blue stained sections disclosed no loss of myelin except in the necrotic cavities. 
Systemic examination disclosed congested lungs in the animal dying on day 5; and 
microscopic examination demonstrated eosinophilic material filling the alveolar spaces 
suggestive of neurogenic pulmonary edema. No pulmonary inflammation was present. The 
animal euthanized at 10 days showed no systemic pathology. 
Group 3- High dose ADV/RSV-tk. no GCV treatment. 3- and 6-week survival. To 
differentiate between the effect of virus injection and virus injection plus GCV administration 
2 male baboons (10.5 and 1 1.5 years-old) were injected with 3 x 10 10 particles of ADV/RSV- 
tk in the same manner as above. These two animals were not treated with GCV. Samples 
were taken and analyzed as described above. One baboon was necropsied at 3 weeks 
after the first post-injection MR. The other baboon was necropsied at 6 weeks after the 
second MR. Tissue and fluid samples were analyzed as above. 
The animal examined at three weeks had a 1.5 cm cystic cavity in the right centrum 
semiovale. No mass effect, herniation, or hemorrhage was grossly present. 
Histopathologically, the cavity was filled with macrophages containing lipid debris. Adjacent 
to the cavity was mild gliosis, persistent lymphocytic inflammation, perivascular lymphocytic 
cuffing, and minimal white matter edema. No viral inclusions were seen, and the choroid 
plexus and ependyma were intact. Mild focal subarachnoid space collections of 
lymphocytes were present over the right hemisphere, but not the left or around the 
brainstem. No systemic pathological alterations were present. 
The animal examined at six weeks after treatment had a 1.6 cm slightly irregular cystic 
cavity in the right posterior centrum semiovale. The lining of the cavity wall was light brown, 
and no mass effect, herniation, or hematoma was evident. Microscopic examination 
demonstrated lipid-laden macrophages within the cavity wall, persistent lymphocytic infiltrate 
in the wall and around nearby blood vessels, and minimal edema and gliosis immediately 
adjacent to the cavity. No choroid plexus, ependymal, and or residual leptomeningeal 
infiltrate was seen. No systemic pathological alterations were present. 
In summary, there appears to be a dose dependent neurotoxic effect of vector with 
high dose treatment groups having coagulative necrosis at the injection site which over time 
is cleared by macrophages to produce cystic cavities. The process persists for six weeks 
although the inflammatory component is resolving by this point. GCV appears to potentiate 
the clinical toxicity of the high dose vector since both animals receiving high dose vector 
plus GCV died or became ill enough to require euthanasia. The moderate dose vector plus 
GCV animals exhibited only microscopic evidence of neurotoxicity at the injection site. The 
adenoviral vector exerts dose dependent cytopathic effects via the penton structural 
proteins paralleling the cytopathic effect of wild type adenovirus. In the this experiment, the 
fact that the high dose groups both with and without GVC had necrosis while the moderate 
dose animals receiving GVC did not, strongly suggests direct cytopathic effect of vector 
rather than toxicity arising from tk conversion of GVC to toxic compounds. This experiment 
establishes the dosage threshold for cytopathology of 1 .5 x 10 9 particles. Vector dosage 
Recombinant DNA Research, Volume 20 
[729] 
