Science in Society Archive

Gene Therapy's First Cancer Victim

I-SIS was almost a lone voice warning of cancer from foreign genes inserting into the genome from ‘gene therapy’ and other exposures to transgenic DNA. Regrettably, this has now become reality. Dr. Mae-Wan Ho raises unanswered questions and calls for a comprehensive safety review of gene therapy and other transgenic technologies.

The French team that made news in 2000 treating children with severe combined immune deficiency (SCID) had to call a halt to the gene therapy trial. One of the 11 children treated has developed leukaemia.

Of the 11 children treated by Alain Fischer and Marina Cavazzana-Calvo in Neckar Hospital in Paris, 9 have been cured. The children suffered from a form of severe combined immune deficiency (SCID) associated with a defective gene on the X chromosome. The gene encodes a cell-surface receptor, g-cchain, that binds to growth signals called cytokines. Without it, children fail to produce mature T cells, B cells and natural killer cells needed to protect them from infections. SCID has been cured with transplants of perfectly matched bone-marrow cells. Of those who receive imperfectly matched transplants, up to 30% die.

Doctors at Great Ormond Street Children’s Hospital in London have since treated four patients. In US and Italy, gene therapists targeted another form of SCID associated with a defect in the gene for the enzyme adenosine deaminase (ADA). Yet another US group has treated a patient whose SCID is associated with a defect in the gene JAK3, which produces a protein involved in transmitting the signals received by cytokine receptors.

Fischer and Cavazzana-Calvo’s success over previous gene trials depended on the ex vivo procedure they pioneered, in which bone marrow cells taken from the patient are transformed outside the body, using a vector carrying a normal copy of the mutated gene to insert the gene into the genome. The transgenic cells are then re-introduced into the patient.

This was hailed as a breakthrough for gene therapy, as it avoided most of the risks of in vivo treatments in which the transgenes are directly introduced into the patient. That has put patients at immediate risks from toxicities of the vector, which killed teenager Gelsinger in 1999. In addition, infectious viruses could be generated from the vector, and cancer could result from the insertion of the vector into the wrong places in the genome.

As in the genetic modification of plants and animals, the genetic modification of human beings is severely hampered because precisely targeted gene insertion is still not technically feasible. It is hoped that by modifying the cells outside the patient’s body, the patient will not be directly exposed to high doses of the vector, and any cells that develop cancer, or any infectious viruses that are generated could be selected out.

Unfortunately, a routine check of their 4th patient last Spring revealed that the child had a high number of T cells in his blood, and by August, the T cell count reached 200 000 cells per litre. The child was admitted to hospital.

Molecular studies carried out since has left little doubt that the gene therapy was to blame (see box).

What went wrong?

Molecular biologist Christof von Kalle at the University of Cincinnati Children’s Hospital, using a special polymerase chain reaction (PCR) technique to amplify the inserts, found that the retrovirus vector and the new gene were in the T cell clone. The foreign DNA had inserted itself, in reverse, in the initial coding region of a gene (LMO2) on chromosome 11, which is essential for the early development of blood cells. Abnormal expression of this gene was linked to leukaemia. An additional anomaly in these T cells was that part of chromosome 6 was duplicated and attached to chromosome 13.

Fischer wants to use the same technique to make a catalogue of the insertion sites. So far, he and von Kalle have examined two other patients form the French trial, and already identified more than 100 insertion sites. None of them are in genes involved in cancer. But carrying out successive analysis on individual patients should tip them off about potential leukaemia.

In successive samples taken from the patient that has developed leukaemia, von Kalle found that the patient’s T cells carried about 50 different insertion sites. But eventually all of the T cells were clones of one cell with the LMO2 insertion.

Fischer accepts that the vector caused an "insertional mutagenesis", splicing itself into a "dangerous gene", causing it to become over-expressed.

A planned clinical trial by researchers in the National Institutes of Health (NIH) in the United States using the same procedure was cancelled. Four other groups, including the Children Hospital in London, has been using or planning similar trials. The United States, France, and Germany (see later) have suspended trials that use the same gene-transfer technology, but not the UK.

"Everyone was aware" of the theoretical risk, he said, but believed it was "very small", claiming that the phenomenon did not turn up in animal experiments.

He was wrong on both counts and more. "Insertional carcinogenesis" is an identified, if not established clinical entity in the cancer literature (reviewed in Slipping Through the Regulatory Net, ‘Naked’ and ‘Free’ Nucleic Acids, TWN Biotechnology and Biosafety Series, 2001). At least one experiment with a retroviral vector had caused leukaemia in all the experimental animals, and the risks of cancer not just restricted to retroviral vectors either. On the basis of those experiments, Germany had already suspended a range of gene-therapy trials involving retroviral vectors.

Another experiment with the most commonly used adeno-associated vector (AAV) also caused high incidence of cancers in animals (see "Failures of gene therapy", Science in Society 16) Furthermore, the mouse Moloney leukaemia virus vector used was among the very first gene therapy vectors, and has been phased out by many gene therapists due to safety concerns.

The molecular details are only now being unravelled, that should have been worked out long before the trials have been so widely taken up. It confirms what critics including us have been stressing, there is simply no control over where and in what form the inserts end up in the genome. More than 100 sites have been identified in just two patients.

A number of questions are raised concerning gene therapy trials in general: the safety of retroviral vectors in use, other than the mouse Moloney leukaemia virus involved in the SCID trial. Another concerns the safety of ‘gene-marking trials’, in which retroviral vectors are used to mark and follow the fate of cells injected into the boy. Forty-one such trials are currently under way in the US. But some experts are wondering whether such trials should be discontinued.

Another question that has not even been raised is, did the insert move subsequent to the transgenic cells being put back into the patient, as some of us have been predicting it could?

An expert panel of the US Food and Drug Administration (FDA) met in an emergency session and urged the FDA to lift the suspension of the trial. The panel concluded that the cancer was almost certainly caused by the gene therapy.

Stuart Orkin of the Dana-Farber Cancer Institute at Harvard University in Boston read a warning from a report he had co-authored in 1995, noting an inherent risk of leukaemia in retrovirus-based gene therapy. He said that there are "potentially numerous sites within the genome that could contribute to leukaemia", adding that the more he learns about the genome, the more possibilities he finds.

In summing up, the Chair, molecular biologist Daniel Salomon of the Scripps Institute in La Jolla, California, said there is no avoiding it – the most successful gene therapy trial also appears to have been the first to induce cancer.

Salomon and other panel members said FDA should ask clinics to step up their monitoring of patients who have been treated with retroviruses. FDA estimates that about 300 clinical trials have provided therapy using retroviruses and 150 are still active. The Panel also recommend that SCID patients be excluded from this therapy if they can get a bone marrow transplant from a matching (HLA-identical) donor, and that clinicians warn volunteers that retrovirus therapy can cause cancer.

One panel member, Abbey Meyers, president of the National Organization for Rare Disorders – made a pitch for placing all retrovirus-based trials on hold because no one can judge the risks. But a woman who identified herself as the grandmother said her grandson has failed bone marrow transplant four times and has been waiting 3 years to be enrolled in a trial, now on hold at the NIH.

NIH’s Recombinant DNA Advisory Committee (RAC) met in December and recommended US trials should proceed "with appropriate monitoring and informed consent", but stopped short of firm ruling on monitoring patients after trials.

It did not consider retroviral vectors apart from those used to treat SCID.

Current guidelines do not require researchers to monitor their patients’ health for any set period. RAC member, David Sidransky, cancer geneticist at Johns Hopkins says it would be too expensive and impractical. Theodore Friedmann, gene-therapy researcher at University of California, San Diego, and chair of RAC said the Committee will take up the issue at its next meeting in March. He hopes that "the language we come up with will be relevant not just to SCID, but to gene-transfer trials more generally."

This episode raises safety concerns over exposures to artificial vectors and constructs of transgenic technology in general that must now be addressed. There must be a comprehensive review, not just of gene therapy but also of other transgenic technologies, ie, the genetic modification of animals and plants for biomedical and agricultural uses, as the methods and constructs used are similar, and so are the risks involved (See Science in Society 2002, 16).

Article first published 10/04/03


  1. Science, News of the Week, 4 October 2002;
  2. Ho MW and Cummins J. Failures of gene therapy. Science in Society 16, 13-5;
  3. "What to do when clear success comes with an unclear risk?" Science 2002, 298, 510-1;
  4. "A tragic setback" by Erika Check, Nature 2002, 420, 116-8;
  5. "Safety panel backs principle of gene-therapy trials" Erika Check, Nature 2002, 420, 595.

Got something to say about this page? Comment

Comment on this article

Comments may be published. All comments are moderated. Name and email details are required.

Email address:
Your comments:
Anti spam question:
How many legs does a spider have?

Recommended Reading

search | sitemap | contact
© 1999 - 2019