American Enterprise Institute for Public Policy Research 
1150 Seventeenth Street, N.W., Washington, D.C. 20036 
(202) 362-5800 
Telex: 671-123° 
28 October 1988 
Dr. James B. Wyngaarden 
Director 
National Institutes of Health 
9000 Rockville Pike 
Bethesda, MD 20205 
Dear Dr. Wyngaarden: 
I am writing to express my disappointment at news reports questioning RAC's 
approval of the Rosenberg-Blaese-Anderson experiment. They suggest that the 
RAC did not have access to all the data it needed to make an informed decision. 
This is inaccurate. Had important data been withheld from the RAC simplv to 
accommodate publication, I am confident the RAC would not have approved the 
experiment. I can only speak for myself, but I felt we had, before our meeting, all 
the data we needed to give you a recommendation to proceed. 
The information we were given was persuasive on several points: 
First, although the use of recombinant DNA in this experiment is unlikely to be 
of much immediate benefit to the patients, the experiment is not at all frivolous. 
With further optimization and refinement, the TIL immunotherapy protocol ap- 
pears to hold great promise for millions of future cancer patients. This line of re- 
search deserves a high priority. 
Second, the risk to the individual patients is very small. That is, it is well within 
the range of risks that are tolerated routinely in clinical research, even among 
subjects far less ill than these individuals. Indeed, the risk associated with the ge- 
netic labeling is inestimably small compared to other therapy-related risks that the 
patients are accepting at the same time. Note also that, without having to go 
through any elaborate approvals, these same researchers have performed the 
same experiment using radioisotopic cell labels, rather than genetic labels. They 
demonstrated convincingly that the genetic label will be a far more effective re- 
search tool. I believe it is also true that the hypothetical risks associated with the 
genetic label-such as insertional mutagenesis— are far smaller and more controll- 
able than the risks associated with autoirradiation from radioactive cell labels. 
Third, these experiments present no plausible risk to the general population— 
which is properly RAC's chief concern. The theoretical possibility that was 
considered— reassembly of a replication competent retrovirus from homologous 
host sequences--is plausible, if at all, only in mice. It cannot reasonably be used 
to argue against human trials. 
Recombinant DNA Research, Volume 13 
