Recombinant DNA Advisory Committee - 12/3-4/92 
as more experience is gained. Ultimately, statistical analysis will be the key factor. The 
RAC must define the probability that there will be no greater that n RCR particles in a 
total of 1 x 10* particles. Establishing this n may imply that further analysis must be 
performed on significant dose levels. In addition, patients should be informed that there 
is always going to be a finite risk of RCR, even though that risk is extremely small. 
Presentation-Dr. Anderson 
Dr. Anderson stated that he appreciated the RAC's efforts in the review of this report, 
and that their comments have been extraordinarily helpful. Dr. Anderson explained that 
the majority of the reviewers questions will be addressed in the presentation. 
Dr. Anderson acknowledged that the report did not define the number of RCR particles 
required to produce pathology. He presented in vivo data derived from experiments 
involving 19 monkeys. Four of the monkeys were part of a bone marrow transplantation 
protocol that was performed over 5 years ago; the animals were severely immune 
compromised. Five of the monkeys were part of a safety study in which they received 
cyclosporin and prednisone; these animals were moderately immune compromised. Ten 
of the monkeys were part of a study conducted by Dr. Neinhuis, 3 of which developed 
lymphomas. 
How many viral particles did each animal actually get? The 5 monkeys that were 
moderately immunocompromised received 3 x 10 s PFU. None of these 5 animals had 
any evidence of pathology. The 4 severely immune compromised monkeys received 1.2 x 
10 9 PFU of RCR, not retrovirus vector particles. None of these 4 animals had 
pathology. The 10 animals in Dr. Nienhuis' study were immune compromised due to 
lethal irradiation, i.e., the monkeys were T cell depleted. In addition, these animals 
received either 5-fluorouracil (5-FU) or stem cell growth factor. Subsequently, their 
CD34( + ) progenitor cells were exposed to 80 to 86 hours of continuous virus exposure 
by 4 separate additions of virus. These animals were exposed to 2 x 10 7 PFU. Three out 
of 10 of these monkeys developed lymphoma. This result demonstrated that there is a 
30% chance of getting lymphoma in a severely immune compromised animal in which 
their cells were exposed for 80 to 86 hours at an exposure rate of 2 x 10 7 PFU. The 
actual number of vector particles was 2 x 10 12 . Severely immune compromised patients 
have a 10% chance of getting lymphoma from the suppression itself. 
How do these results translate into safety factors for clinical protocols? Basically, there 
are 4 barriers. First, there has to be breakout. There was no instance of breakout in the 
first 50 production runs performed by Genetic Therapy, Inc. (GTI). Each run produces 3 
lots of vector. Since that time, a breakout has occurred. The second barrier is that an 
RCR would have to be missed. Under the current testing conditions, the level of 
detection is 1 RCR per ml. How are these assays performed? Approximately 100 ml of 
Recombinant DNA Research, Volume 16 
[587] 
