SOUNDING BOARDS 
GENE THERAPY IN HUMAN BEINGS: 
WHEN IS IT ETHICAL TO BEGIN?* 
Ti ik recent rapid expansion of research in the area 
of molecular genetics, resulting in large part from the 
application of rccombinant-DNA technology, has pro- 
duced uncertainty in the minds of some physicians 
about how much has been accomplished in the area of 
genetic engineering. Successful introduction of genes 
into intact mice'- 2 and into mouse embryos’ indicates 
that the transfer of certain genes into human beings 
may now be technically feasible. Indeed, the Los 
Angeles Times reported on October 8 that an attempt at 
gene therapy in human beings was conducted this 
summer. 4 Is it ethical to carry out such a new thera- 
peutic procedure, aimed at alleviating a serious or 
lethal disease in a patient, before the probable bene- 
fits and risks of the procedure have been determined 
in animals? The ethical arguments that have guided 
therapy-oriented medical research in the recent past 
have indicated overwhelmingly that the answer is no 
(although exceptions may be permissible for criti- 
cally ill subjects). At what point, then, is it ethical to 
begin a revolutionary new treatment procedure? In 
this paper, we briefly summarize some of the relevant 
ethical arguments and the current state of the art of 
gene therapy; we then attempt to explain what still 
needs to be learned from animal studies before exper- 
iments with human beings should begin. 
Ethical Background 
One of the cornerstones of the ethics of clinical in- 
vestigation is the rule governing when to start experi- 
mental therapy in human beings. A persistent prob- 
lem is embodied in the question, “When is it ethical to 
begin the experiment?” Beginning with the Nurem- 
berg Code,’ a body of ethical guidelines for investi- 
gators and peer-review groups has accumulated. As 
stated by Rule 3 of the Nuremberg Code: 
The experiment should be so designed and based on the results of 
animal experimentation and a knowledge of the natural history of 
the disease or other problem under study that the anticipated re- 
sults will justify the performance of the experiment.' 
This rule is repeated or paraphrased in every major 
code of ethics governing clinical investigation and 
developed by international groups, 6 medical associa- 
tions, 7 and single institutions.' Henry Beecher sum- 
marized the ethical considerations involved in pre- 
paring to begin experiments with human beings by 
stating: “A study is ethical or not at its inception; it 
does not become ethical because it succeeds in pro- 
ducing valuable data. . . .” 9 The fundamental prin- 
ciple is that it should be determined in advance that 
the probable benefits outweigh the probable risks. 
•The views expressed in this article are those of the authors and jre not in- 
tended to represent official policies of the National Institutes of Health or 
the Department of Health and Human Services. 
These statements were made in an era when the 
first (and only) line of defense in clinical research was 
the conscience of the individual investigator. Two 
additional procedures intended for the protection of 
subjects have been broadly accepted: prior review of 
research protocols by a local Institutional Review 
Board (IRB), and procurement of the written con- 
sent of the subject or proxy. The Belmont Report, 10 
the most definitive recent statement of the ethical 
guidelines that should govern IRB review, stressed 
that the adequacy of preparation for research in 
human beings is vitally related to answering the 
paramount question of the review process: Are the 
benefits and risks in a favorable ratio? A systematic- 
analysis of risks and benefits requires an examination 
of information about all aspects of the proposed re- 
search, including consideration of whether additional 
animal research might be more appropriate than sub- 
jecting human beings to the possible risks, given the 
existing state of knowledge. The investigators are 
obliged to explain how they know that the pre- 
suppositions of the research have probable validity 
and whether the information can be obtained in any 
safer way than by trials in human beings. Such con- 
siderations are inherent in the requirements of the 
Food and Drug Administration for treatment of a pa- 
tient with a new drug. In short, the moral justifica- 
tion for initialing therapeutic research in human be- 
ings rests on an assessment of the validity of the hope 
of success. T his assessment is usually based on re- 
sults of experiments with animals or other studies in 
human beings that suggest, on balance, that the pro- 
posed research question may be answered with ac- 
ceptable risk to the subject. 
T he definition of acceptable risk is complex and de- 
pends in part on the severity of the patient’s illness. 
Current standards for medical research in the area of 
cancer therapy, for example, permit experimental use 
of highly dangerous therapeutic agents in patients 
with fatal illnesses for which no effective conventional 
therapy is known. These agents include toxic muta- 
gens that are capable of killing or altering the genetic 
composition of somatic and germ cells. The use of 
such treatments obviously requires very thoughtful 
prior assessment. Similarly, considerable caution 
should be exercised before giving a patient genetic 
material in an attempt at therapy. The severity of the 
illness, despite existing therapy, would have to be so 
great that a potentially considerable risk from the new 
therapy might be justified. There are a number of 
genetic diseases that fall into this “severe-illness” cat- 
egory. Two of them, d-thalassemia and sickle-cell 
anemia, have been extensively studied at the molecu- 
lar level and are known to result from an abnormality 
in the amount (in d-thalassemia) or type (in sickle-cell 
anemia) of beta-globin chains. In theory, these two 
seriously debilitating genetic anemias could be treat- 
ed by insertion of a beta-globin gene into the pa- 
tient's bone marrow; this approach would direct the 
[ 314 ] 
Recombinant DMA Research, Volume 12 
Reprinted from the Xew HnelanJ Journal u/ Medicine 
303:129.1-1297 (November 27). 19X0 
