Recombinant DNA Advisory Committee - 6/9-10/94 
toxicity study. Dr. Post stated that it may be preferable to examine the data as a whole 
and determine if there is clear rationale for the proposal rather than defining specific 
experiments. Dr. Parkman expressed concern that IL-7 has never been administered to 
humans; therefore, toxicity is unknown. Dr. Chase stated that there is no statistical 
rationale for the proposed study. What conclusions can be drawn from a study involving 
5 patients in each experimental group? 
Dr. Walters asked Dr. Noguchi to comment on this issue. Dr. Noguchi responded that 
the 219 patients entered in RAC approved trials to date do not provide a sufficient base 
by which to address common toxicity, i.e., 1 in 100 subjects. The primary goal of a Phase 
I study is to estimate toxicity. Toxicity in drug trials is usually not observed until the 
Phase n study. He said that 5 subjects per group is preferable to 3, particularly for rare 
genetic diseases. 
Dr. Economou explained that he has consulted with Dr. William Cumberland, Professor 
of Biostatistics at UCLA School of Public Health, regarding statistical analysis and the 
size of the treatment groups. Because of the number of variables in the proposed study, 
the number of subjects to be entered in each group is statistically difficult to determine. 
It is the opinion of the investigators that 5 subjects would yield useful data regarding 
toxicity and immunological responses. 
Responding to Dr. Smith's question on patient selection. Dr. Economou explained that 
metastatic melanoma patients who have failed conventional treatment are referred to 
UCLA and are eligible for several protocols that are being conducted simultaneously. 
Usually subjects are triaged with at least 2 cycles of chemotherapy. Subjects with 
subcutaneous or regional disease are then offered the option of entering one of the 
adoptive immunotherapy trials. Subjects that are not entered on an adoptive 
immunotherapy study are offered other systemic trials. The vaccine trials are offered as 
a possible third level of treatment. Entry on one research protocol does not preclude 
entry into another study. Some subjects can be entered onto more than one study in a 
sequential fashion. Dr. Haselkom recommended that a patient advocate should be 
assigned to assist subjects in understanding the studies and making informed decisions. 
Dr. Chase stated that the choice of the number of patients based on budget and duration 
of a protocol is a reasonable one if there is no other statistical criteria. Entering subjects 
sequentially on studies could lead to unrecognized consequences that could cloud results. 
Dr. Economou responded that melanoma patients usually demonstrate very clear 
responses to beneficial treatment. Given sufficient time between entry on individual 
protocols it is possible to determine the effectiveness of each treatment. The major 
objectives of the present study are toxicity and immunological endpoints which are easily 
discernible. 
Dr. Parkman asked about the specific criteria for entering patients on either the IU7 or 
the IU2 protocol. Dr. Economou responded that subjects will be assigned in a random 
fashion- Dr. Chase commented that randomization of patient entry is an acceptable 
procedure. 
[348] 
Recombinant DNA Research, Volume 19 
