vol. 353, i. 9167, page 1873

Abstract: Genetic manipulation of germ cells should be prohibited. Alterations of the genetic code in sperm cells, ova, and fertilized ova could be used to prevent disease or to induce desired traits. In comparison to gene transfer in other body cells, germ cell modification may produce effects that were not predicted, and may not reveal themselves until the child is grown, or in future generations. Germ cell modification not only influence the single offspring, but create unique DNA that will persist into subsequent generations. Prenatal diagnosis permits specific pregnancies to be evaluated for genetic fitness. Manipulation of germ-cell DNA is unnecessary and potentially dangerous.

Human germline gene modification has been foreseen but not yet accomplished.1-6 It can be defined as the genetic manipulation of human germ cells, or of a conceptus, resulting in inherited changes in DNA. With the development of advanced in-vitro fertilisation (IVF) methods, preimplantation DNA analysis, improved techniques for gene transfer, insertion, or conversion, and of embryo implantation procedures, the technical barriers to such an intervention seem easily surmountable. Unintended changes in DNA may occur when gametes are manipulated or stored.7,8 Inadvertent germline mutations. therefore, may have already occurred as a result of reproductive technologies in current use, such as artificial insemination and IVF. There are unpublished reports that researchers in the USA have already carried out a manipulation involving the exchange of a mitrochondrial genome in an IVF protocol. If true, this human experimentation involving intentional hereditary changes was probably conducted without federal oversight of safety, since there are no discussions of this protocol in the available public record.

Tsukui and colleagues9 used viral vectors in somatic gene therapy protocols to infect mouse eggs in vitro, leading to germline transmission of a transgene in the progeny. Although removal of the zona pellucida is a prerequisite for infection of the eggs in vitro, the early oocytes of postnatal ovaries also lack zonas. These experiments thus raise the possibility that modification of gametes may occur in vivo, and constitute a germline hazard in the 200 or more somatic gene therapy protocols now in use. Any such alterations would be difficult to detect. Intentional or inadvertent germline modifications may pose significant burdens. Although there are restrictions on experimentation that might result in human modifications,10 and opposition to its implementation has been voiced,11-15 some leading scientists and other commentators have begun to advocate the development and application both of techniques that may increase the risk of inadvertent alteration of the germline, and of methods that would alter it deliberately.6,16-18

W French Anderson and his colleagues have developed an experimental protocol for the treatment of adenosine deaminase deficiency during fetal development; although their therapeutic intent is directed towards somatic cells, they acknowledge that the technique may modify germ cells as well. They have submitted this proposal to the National Institutes of Health (NIH) for review (panel). By introducing a genetic construct in utero, which knowingly allows for the alteration of germinal tissue, their attempt at a potentially transmissible correction could be used to erode opposition to germline genetic manipulation since germline modification would be achieved, though unintentionally.

Opposition to germline modification is based on several lines of reasoning.19-22 First, as we have already suggested, germline DNA modifications may affect gene function in ways that are not immediately apparent, so their occurrence may not be recognised for a generation or more-for example, germline introduction in mice of an improperly regulated normal gene resulted in progeny with unaffected development but high tumour incidence during adult life.23 Furthermore, interactions among genes and their products are highly integrated, have been refined over evolutionary time scales, and often serve to stabilise developmental pathways and physiological homoeostasis.24-26 Through experimental error, unanticipated allelic interactions, or poorly understood regulatory mechanisms such as imprinting, there is a risk that germline genetic manipulation will alter sensitive biological equilibria. Disruption of these interactive systems is likely to have complex and uncertain biological effects, including some that appear only during the development or functioning of specific cells or tissues.27 Many of these effects could be undesirable.

Second, this sort of intervention is not needed. With available methods of prenatal diagnosis, virtually all interested couples can choose not to transmit specific identifiable genes. Other reproductive options (artificial insemination, egg donation) and adoption are available to those not able or willing to use prenatal or preimplantation selection methods. An exception might be when, rarely, two individuals have the same recessively inherited disorder. If such couples chose to reproduce, it could be argued that they would "need" germline or very early genetic interventions since all their progeny might inherit a disease-associated genotype. Yet, even these children may differ genotypically and phenotypically from their parents and the development of a new mode of treatment for this unusual occurrence does not seem justifiable. Although available alternative procedures are invasive, germline modifications would also require similar interventions since they would probably involve IVF. Moreover, the associated risks with existing procedures are not as serious as those created by introducing a hereditary genetic "error" into a family. People who oppose prenatal diagnosis on philosophical or religious grounds would be unlikely to want to take part in germline modification if they were aware of its intrinsically experimental nature and of the numbers of human embryos that would have to be expended during the development of the technology. No unmet need balances the risks of germline interventions to mothers, fetuses, and future generations. Moreover, the costs associated with the general development and implementation of germline manipulation would be formidable.

If there is no clinical need for germline modifications, the primary reason for using this intervention would be human enhancement.28 Apart from the uncertainties about its ultimate outcome, enhancement is a form of eugenics. Though not a recrudescence of overtly coercive, public-health-based eugenics popular earlier this century, germline manipulations represent an individual or familial form. Seemingly private personal decisions and "choices" about medical or non-medical programmes for enhancement would, nevertheless, reflect prejudices, socioeconomic and political inequalities, and even current fashion. Though enhancement procedures now in use (eg, cosmetic surgery or orthodontics) also change according to fashion, germline intervention would intentionally subject later generations to modifications undertaken on the basis of existing values and conditions. The chance that "desirable" manipulations might later be viewed as disastrous makes germline enhancement "therapies" unacceptable.

Human germline interventions would necessarily alter the lives of individuals who are yet to be born. Informed consent by the affected individuals is not possible. Extension of the parental right to consent for minors would be required.29 Such legal permission to specifically alter the lives of generations of unborn individuals would be unprecedented and unjustified.

If germline manipulation is attempted, there will be mistakes or errors in its application. Neither social acceptance nor the necessary range of protections and care for accidentally damaged individuals can be guaranteed.30 Unexpected alterations in family relationships will occur, and "wrongful life" disputes could arise.31 Irrespective o whether such interventions were to take place in research or clinical settings, these issues mean that germline modifications cannot be approved by existing standards for the protection of human beings.32 No benefits to and future individual would justify abrogating or curtailing these restrictions.

For these biomedical reasons, as well as others based in legal,33 philosophical,19,34,35 cultural, and spiritual/religious traditions,36,37 human germline modifications should be opposed and prohibited. Experimentation that may gradually make human germline modification more feasible is under way; it may require further review. Further study is needed of the safety of somatic gene therapy protocols to ensure that they detect, with adequate sensitivity, germline alterations. Many individuals and groups that monitor developments in human genetics can be expected to mount vigorous opposition to the development of human germline protocols, involving direct action, legal manoeuvres, and organising among interested public groups. Unlike man other countries, including those of the EU, which have prohibited germline manipulation in principle,38,39 restrictions on the procedure in the USA are mainly based on practical considerations (see, for example, the summary of the January 1999, RAC-sponsored conference at http://www.nih.gov/od/orda/gtpcconc.htm. Site accessed March 20, 1999) and are subject to revision as the state of the science changes. Although debate about human germline modifications should continue and, indeed, be broadened to include representation of a diverse cross- section of viewpoints and backgrounds, such discussion should not be construed as suggesting that such a method would ever be appropriate or acceptable.

We thank Jacque Bradley for technical assistance, Jon Beckwith, Felipe Cabello, Suzanne Bodor, Vernon Chong and Parris Burd for helpful suggestions and critical comments on drafts of this paper.

References

1. Splicing life. In: President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioural Research. Government Printing Office, Washington DC, 1982: 1-115.
2. Friedman T, ed. Therapy for genetic disease. Oxford: Oxford University Press, 1991.
3. Fletcher JC, Anderson WF. Germ-like gene therapy: a new stage of debate. Law Med
Healthcare 1992; 20: 2-39.
4. Wivel NA, Walters L. Germ-like gene modification and disease prevention: some medical and ethical perspectives. Science 1993; 262: 533-38.
5. Cook-Deegan RM. Germ-like therapy: keep thewindow open a crack. Pol Life Sci 1994; 3: 217-20.
6. Walters L, Palmer J. Ethics of human gene therapy New York: Oxford University Press, 1997.
7. Bouquet M, Selva J, Auroux M. Cryopreservation of mouse oocytes: mutagenic effects in the embryo? Biol Reprod 1993; 49: 764-69.
8. Vishwanath R, Shannon P. Do sperm cells age? A review of the physiological changes in sperm during storage at ambient temperature. Reprod Fertil Dev 1997; 9:321-31.
9. Tsukui T, Kanegae Y, Saito I, Toyoda Y. Transgenesis by adenovirus- mediated gene transfer into mouse zona-free eggs. Nat Biotechnol 1996; 14: 982-85.
10. McDonough PG. The ethics of somatic and germline gene therapy. Ann NY Acad Sci 1997; 816: 378-82.
11. Krimsky S. Human gene therapy: must we know where to stop before we start? Hum Gene Ther 1990; 1: 171-73.
12. Lippman A, Billings P, Bereano P, et al. Position paper on human germ line manipulation. Hum Gene Ther 1993; 4: 35-37.
13. Bonnicksen AL. The politics of germline therapy. Nat Genet 1998; 19: 10-11.
14. Ewbank J. Problems of germline therapy. Nature 1998; 392: 645.
15. McLaren A. Problems in germline therapy. Nature 1988; 392: 645.
16. Anderson WF. Uses and abuses of human gene transfer. Hum Genet Ther 1992; 3: 1-2.
17. Wadman M. Germline gene therapy "must be spared excessive regulation". Nature 1998; 392: 317.
18. Kolata G. Scientists brace for changes in path of human evolution. New York Times. March 21, 1998: A1.
19. Rifkin J. Algeny. New York: Viking Press, 1983.
20. Fowler G, Juengst ET, Zimmerman BK. Germ-line gene therapy and the clinical ethos of medical genetics. Theor Med 1989; 10: 151-65.
21. Lappe M. Ethical issues in manipulating the human germ-line. J Med Phil 1991; 16: 621-41.
22. Juengst ET. Can enhancement be distinguished from prevention in genetic medicine. J Med Phil 1997; 22: 125-42.
23. Leder A, Pattengale PK, Kuo A, Stewart TA, Leder P. Consequences of widespread deregulation of the c-myc gene in transgenic mice: multiple neoplasms and normal development. Cell 1986; 45: 485-95.
24. Strohman R. Ancient genomes, wise bodies, unhealthy people: limits of a genetic paradigm in biology and medicine. Persp Biol Med 1993; 37: 112-45.
25. Newman SA. Interplay of genetics and physical processes of tissue morphogenesis in development and evolution: the biological fifth dimension. In: Beysens D, Forgacs G, Gaill F, eds. Interplay of genetic and physical processes in the devlopment of biological form. Singapore: World Scientific, 3-12.
26. Wilkins AS. Canalization: a molecular genetic perspective. BioEssays 1997; 19: 257-62.
27. Jaenisch R, Breindl M, Harbers K, Jahner D, Lohler J. Retroviruses and insertional mutagenesis. Cold Spring Harb Symp Quant Biol 1985; 50: 439-45.
28. Miller HI. Gene therapy for enhancement. Lancet 1994; 344: 316-17.
29. Robertson JA. Children of choice: freedom and the new reproductive technologies. Princeton: Princeton University Press, 1994.
30. Billings PR, Kohn MA, de Cuevas M, et al. Discrimination as a consequence of genetic testing. Am J Hum Genet 1992; 50: 476-82.
31. Cohen CB. "Give me children or I shall die!" New reproductive technologies and harm to children. Hastings Cent Rep 1996; 26: 19-27.
32. Protection of Human Subjects. Federal Register 1981; 46: 6386 and 1983; 48: 9269.
33. Annas GJ, Elias S. Somatic and germline therapy. In:Annas GJ, Elias S, eds. Gene mapping: using law and ethics as guides. New York: Oxford University Press, 1992: 142-54.
34. Davis BD. Germ-line therapy; evolutionary and moral considerations. Hum Gene Ther 1992; 3: 361-63.
35. Neel JV. Germ-line gene therapy: another view. Hum Gene Ther 1993; 4: 127-28.
36. World declaration on our responsibilities towards future generations. In: Agius E, Busuttil S, eds. What future for future generations? Valletta, Malta: Foundation for International Studies, 1994: 305-11.
37. World Council of Churches. Manipulating life: ethical issues in genetic engineering. Geneva: World Council of Churches, 1983.
38. Gene therapy in man. Recommendations of the European Medical Research Councils. Lancet 1988; i:1271-72.
39. Wadman M. European states outlaw permanent changes. Nature 1998; 392: 317.

Lancet 1999; 353: 1873-75 Veterans Health Administration, Heart of Texas Health Care System, 1901 North Highway 360 Grand Prairie, TX 75050

(P R Billings MD); Biological Laboratories, Harvard University, Cambridge, MA (R
Hubbard PhD); and Department of Cell Biology and Anatomy, New York Medical
College, Valhalla, NY, USA (S A Newman PhD)