What next for human gene therapy?

BMJ 2003; 326 doi: https://doi.org/10.1136/bmj.326.7404.1410 (Published 26 June 2003) Cite this as: BMJ 2003;326:1410
  1. Eric T Juengst, associate professor of bioethics
  1. Department of Bioethics, School of Medicine, Case Western Reserve University, Cleveland, Ohio 44106, USA

    Gene transfer often has multiple and unpredictable effects on cells

    The high hope of genetic medicine for 30 years has been to develop a way of using recombinant DNA techniques to treat patients through the genes involved in their diseases. As Richard Roblin, scientific director of the Council on Bioethics of the President of the United States, put it in 1979: “There is something aesthetically compelling about cutting to the heart of the problem, by treating the disease at the molecular level, where it originates.”1 Since 1990, this vision has generated a modest industry of bench research and animal studies, culminating in almost 1000 clinical trials in humans around the world, for a wide variety of diseases.2 In the past few years, however, the field has learned that in genetic medicine, as in war, the “surgical strike” is rarely as clean and effective as theory implies it should be.

    After almost a decade without much clinical success,3 the field has experienced in quick succession its first iatrogenic death,4 its first apparent “cures,”5 and then among those cured patients the first instances of serious downstream disease traceable to the main theoretical genetic risk of gene transfer—“insertional mutagenesis”.6 In the first such incident the development of T-cell acute lymphoblastic leukaemia in two of nine research subjects three years after (otherwise successful) gene therapy for their X-linked severe combined immunodeficiency disease was deemed by investigations in France and in the United States to be caused by “insertional mutagenesis at or near the LMO-2 gene with aberrant production of LMO-2 protein.”7 As a result in January 2003 the US Food and Drug Administration placed a “clinical hold” moratorium on all similar studies in the United States. To date this hold has not been lifted although individual protocols continued to be considered on a case by case basis at both the National Institute of Health and the FDA.8

    Gene transfer research is at a crossroads, and to avoid a quagmire of setbacks that can only undermine international support for its efforts a broader coalition of its neighbours is required in genomics, proteomics, (study of proteomes or the protein complements of genomes), and cell biology, to know which way to turn. Meanwhile it is not too early to plan for the important social policy challenges that will emerge once the field's objective is achieved.

    So far the compelling aesthetics of gene therapy have been successful in raising high expectations outside the field, and that is part of its dilemma. Advocates for patients with rare diseases continue to press for the right of families to take the risks of gene transfer research on behalf of their loved ones.9 The World Anti-Doping Association, which monitors performance enhancing drug use in international sports, has begun seriously to entertain the prospect that undetectable gene transfer interventions could help elite athletes cheat.10 And science policy organisations, such as the American Association for the Advancement of Science, have begun to highlight the risk that human germ line genetic modifications could occur first in the relatively unregulated clinical sphere of international infertility medicine.11 In these contexts, desperation, national pride, and professional foolhardiness may thrust gene therapy unwillingly into political and ethical minefields that its current regulatory structures cannot help navigate.

    One of the early lessons of the international effort to map and sequence the human genome has been the discovery that the human species makes do with far fewer genes than had been predicted: roughly 35 000 rather than 150 000.12 This discovery might seem to simplify the gene therapists' mission, but it actually has the opposite implications. To produce the vast host of proteins used by human cells, our genes must be capable of “multitasking,” by having their component coding regions rearranged and recombined with subunits of other genes for transcription. This versatility means that mutations in a gene—therapeutically induced or otherwise—will ramify through the cellular proteome in multiple directions, depending on how the modified coding region is used by the cell. Moreover, the orchestration of this complex dance is primarily in the hands of the cellular environment, not the nuclear genome. This clouds the surgical metaphor for genetic modification considerably. Gene transfer is more like introducing rabbits to Australia than it is like a heart transplant: it makes change in a cellular ecosystem that will almost always be pleiotropic in its effects, and often in unpredictable ways.

    To date, gene transfer researchers have invested heavily in building “rabbit proof fences,” designed to contain the molecular effects of misdirected genetic insertions. It may be prudent now to focus on learning more about the cellular dynamics of correctly targeted gene transfers, taking advantage of the tools and knowledge that the next few years of research in genomics and proteomics will bring. In the meantime, now is the time for the gene transfer research community to become more involved in shaping the social policy context for their work.13 Desperate parents, unscrupulous coaches, and adventuresome infertility specialists will all need the aid of skilled scientists to take the steps that could compromise further medical progress in genetic medicine. If gene transfer researchers are alert to these issues, accurate in their advertising, and articulate about the moral convictions that guide them, the field can go far to secure its pride of place in the new world of genetic medicine now beginning to emerge.


    • Competing interests None declared.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    View Abstract

    Sign in

    Log in through your institution

    Free trial

    Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
    Sign up for a free trial