Recombinant DNA Advisory Committee - 12/1-2/94 
will be forthcoming from the ongoing hepatic infusion primate studies. The gonads were not examined for 
vector distribution in the murine experiment but will be examined in the primate study. Dr. Dan Maneval 
(Canji, Inc.) stated that the ovaries were negative for (S-gal marker expression. 
Dr. Warren stated that the starting dose for the human study is 2,000-fold lower than the dose that caused 
murine liver toxicity. The mouse is a relevant model to examine the toxicity of the replication-deficient 
adenovirus vector because it is nonpermissive for virus replication but permissive for virus infection. Dr. 
Samulski commented that adenovirus toxicity can be observed even in nonpermissive hosts. This second 
generation adenovirus vector has reduced potential for toxicity because it is a temperature-sensitive mutant. 
Dr. Maneval said that serum liver enzyme levels were not examined in the hepatic artery toxicity study. 
Dr. Venook said that 27 patients are proposed for study was based on a dose-escalation scheme with 3 patients 
in each cohort. The number of dose-escalation cohorts will be designed according to the results of the primate 
toxicology study. If toxicity is encountered in the human studies, higher dose cohorts will not be initiated; 27 
subjects is the projected number at the present time. 
Dr. Glorioso asked if the toxicology studies involved preimmunized animals. The severity of an immune 
reaction may vary depending on the immune status of the patient. Dr. Miller noted there is no mention of 
immune status in the eligibility criteria and expressed concern about the potential for a stronger immune 
response in preimmunized patients. Since most patients are expected to have prior adenovirus exposure, 
experiments involving preimmunized animals is pertinent. Dr. Samulski remarked that in the CF protocols, 
the exclusion criterion of antibody positive patients was deleted because most patients have prior virus 
exposure. Dr. Glorioso said that the possible consequence of immune reaction in the CF studies is the loss of 
some transduced cell in the lung; however, the entire liver could be rejected in this proposed study. Dr. 
Par km an stated the issue of the preexisting immune response is twofold: (1) the possibility of decreased 
transduction efficiency, and (2) the more serious concern of increased inflammatory response in the liver. 
Preimmunized cotton rat studies are critical for determining hepatocellular damage. Dr. Miller expressed 
concern that a severe reaction might occur during the dose-escalation schedule. Dr. Par km an said important 
information can be obtained from the experiment comparing the toxicity in the imm uniz ed versus non- 
immunized cotton rats. 
Dr. Warren presented syngeneic Buffalo rat tumor model data demonstrating preferential uptake and 
expression of the transgene following hepatic artery vector delivery by tumor cells as compared to the 
surrounding liver parenchyma. This differential uptake and expression can be attributed to the preferential 
blood supply from the hepatic artery to the tumors. Dr. Samulski commented that an alternative explanation 
for this phenomenon could be that these tumor cells have acquired the property of preferential virus infectivity 
during in vitro cell culturing, and similar preferential virus infectivity may not exist in the natural liver cancer 
cells. 
Dr. Venook stated not all patients are expected to receive both hepatic pump placement and subsequent 
chemotherapy. These patients will provide information about long-term side effect of ACN53 treatment alone. 
Patients undergoing pump placement will receive chemotherapy 28 days following ACN53 treatment. This 
timeframe will allow for evaluation of the safety and biological efficacy of ACN53 treatment. The normal 
toxidty pattern associated with chemotherapy is predictable and well characterized; therefore, any unusual 
occurrence due to gene transfer should be readily recognizable. Dr. Venook emphasized that the liver 
function can be sensitively monitored by assaying serum liver enzymes rather than depending on the less 
specific symptoms such as nausea and vomiting. This Phase I study will provide further toxicity data from 
humans. 
Dr. Samulski asked if anim al toxicity studies will be performed on both immunized and non-immunized 
animals. Dr. Par km an said if the cotton rat experiment demonstrates no difference in toxicity, then there will 
be no need to conduct preimmunized primate studies; however, if differential toxicity is observed, additional 
primate studies will be required. Dr. Finkle (Canji, Inc.) said that a formal toxicology study plan has been 
Recombinant DNA Research, Volume 20 
[467] 
