News | Published:

Virus treatment questioned after gene therapy death

Nature volume 401, pages 517518 (07 October 1999) | Download Citation


San Francisco

Verma: new caution is “a good thing”. Image: SALK

Researchers at the University of Pennsylvania are investigating the first death in a gene therapy experiment, which was revealed last week. Their enquiries centre on the adenovirus vector used to deliver potentially therapeutic DNA to the liver.

Jesse Gelsinger, an 18-year-old, high-school graduate from Arizona, developed a fever and blood clots throughout his body within hours of treatment to correct partial ornithine transcarbamylase (OTC) deficiency, a rare metabolic disease that can cause a dangerous build-up of ammonia in the body. He died four days later.

US officials immediately began notifying the 100 or so gene therapy experimenters using adenovirus vectors, which are made using a disarmed version of the virus that causes the common cold. Both the Food and Drug Administration (FDA) and National Institutes of Health (NIH), however, stopped short of recommending any policy changes or clinical holds.

According to the protocol, researchers at the University of Pennsylvania's Institute for Human Gene Therapy used an ‘E1-deleted, E2A-temperature-sensitive’ adenovirus vector to infect liver cells with the normal OTC gene, which codes for a urea-cycle enzyme that removes excess nitrogen from the body.

Gelsinger, the eighteenth and final patient in the Phase I experiment, was the second person to receive a dose of 3.8×1013 virus particles, believed to be the highest so far with an adenovirus. The virus was delivered by a catheter inserted into the groin artery and advanced into the vessel that feeds the liver.

Researchers at the university are reviewing their notebooks and toxicology data. They are studying the vector, treating it with restriction enzymes and testing it in primates.

They have also conducted an autopsy and are examining tissues and liver cells, looking in particular for vector-related problems because of the speed of Gelsinger's reaction. “So far, we haven't seen anything that we'd do drastically different,” said Nelson Wivel, deputy directory of the University of Pennsylvania's Institute of Gene Therapy.

Inder Verma, professor of genetics at the Salk Institute in La Jolla, California, praises the Pennsylvania team for their openness, saying it will help to preserve public confidence and allow scientists to learn from the incident, “I'm sure it will introduce a note of caution in every experimentalist who does gene therapy, and that's a good thing,” he says.

Verma describes vectors as the “Achilles heel” of gene therapy, and says that dose-escalation studies using adenoviruses should be re-examined. Most gene therapy involves retroviral vectors, but adenoviruses are popular for cancer and cystic fibrosis.

Verma says he thinks vectors like the one used in the OTC trial will soon be abandoned in favour of ‘gutless’ adenoviruses, retroviruses, AAV or lentivectors.

A major problem of adenoviruses is that even inactivated versions can stimulate an immune response. Sustained expression of the gene is therefore impossible, and the immune system may destroy infected cells—the very cells targeted for help. The severe inflammation associated with these vectors, especially in the liver, is particularly dangerous for OTC patients.

Members of the NIH's Recombinant DNA Advisory Committee (RAC), which at the time held regulatory authority over gene therapy experiments using federal funds, approved the protocol 11 to one, with four abstentions. They raised concerns about the risks of the treatment and its use—for the first time in gene therapy—in asymptomatic patients.

Individuals without the targeted gene cannot break down nitrogen, which can lead to a fatal build-up of ammonia soon after birth. The gene is X-linked—female carriers usually lead normal lives, but up to ten per cent of them could experience dangerous symptoms. Patients with partial enzyme production have done well under dietary and drug treatment.

The investigators had studied their vector in mice and primates. In the December 1995 review of the proposal for a trial in humans, RAC members discussed the potential for lethal liver inflammation based on toxicity results in Rhesus monkeys and one animal's death after an extremely high dose of a first-generation vector.

The likelihood of efficacy and the importance for this and other liver conditions convinced them to approve the study, with the recommendation that the researchers use a less invasive route of administration through a peripheral vein.

Because of concerns about infection of reproductive cells, FDA regulators made the researchers go back to treating the liver directly. Under current rules, such an experiment might be reviewed by the RAC, but only requires approval from the FDA. An FDA spokesman said that the agency cannot discuss experiments or their results without the consent of the investigators concerned.

Verma and others point to the technology's success in thousands of patients up to now, and say that the death should not be seen as a setback for gene therapy as a whole. “We would obviously prefer that this tragedy had not happened,” says Wivel. “But these things do occur in cutting-edge research.”

About this article

Publication history




  1. Search for Sally Lehrman in:

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing