iOAN-KETTERING INSTITUTE for GANGER RESEARCH 
May 25, 1979 
Dr. Jane Setlow, Chairman 
NIH Recombinant DNA Advisory Committee 
Brookhaven National Laboratory 
Upton, New York 11973 
Dear Jane: 
Since my comments on the significance of the new Martin-Rowe data 
(document 670, Item 2) showing tumor induction by recombinant DNA were 
partially obstructed by David Baltimore at the last RAC meeting, I 
would like to summarize them here. I would also like to call the com- 
mittee’s attention to a very incisive comment made by Baltimore in ans- 
wer to Sheldon Krimsky’s plaint, at the May meeting, about the need for 
help in digesting reams of scientific data. David said, in effect, that 
only the data, the "results" section, of a scientific paper are hard, 
while discussion of the meaning of the data is bound to be influenced by 
the outlook of the writer and should not be accepted uncritically. I 
offer here an interpretation which is not influenced by any interest in 
carrying out experiments with recombinant DNA. I trust that the com- 
mittee members will themselves evaluate the risk-assessment data from 
the viewpoint of the public interest and not from any sense of obliga- 
tion to a special interest group like the scientific community. 
In brief, the results show that 1) once in contact with mouse cells, 
recombinant DNA molecules containing polyoma DNA can induce tumors, and 
have roughly the same ability to do so as free polyoma DNA. The insertion 
of viral DNA into recombinant molecules is therefore not a safeguard. 
2) The results have no bearing on the question of whether recombinant 
molecules might be given new opportunities for contact with animal cells, 
different from those of polyoma virus, when they are propagated in bacteria 
of any kind. This still remains to be determined before the overall risk 
can be assessed. (Some studies along these lines are being carried out by 
others under contract to the NIH) . 
1. Table 2 (p . 10 of ms 2 included in document 670) gives the data 
concerning tumor induction by Injected DNA. The first column 6hows that 
19% of animals injected with intact, uncleaved polyoma DNA developed tumors, 
and 19% (3/16) of animals Injected with intact dimer-lnsSrt polyoma — X re- 
combinant DNA also developed tumors. The other entries are for intact monomer 
[ 276 ] 
RESEARCH UNIT OP MEMORIAL SLOAN-KETTE RING CANCER CENTER 
