By Lisa Seachrist

Washington Editor

BETHESDA, Md. - Gene therapy experts convened by the Recombinant Advisory Committee (RAC) agree that Jesse Gelsinger died as a result of the experimental therapy he received in September at the University of Pennsylvania. But no one can explain why the treatment felled the 18-year-old.

A thorough review of the gene therapy protocol Jim Wilson and his colleagues undertook establishes only that Gelsinger died after receiving an altered adenovirus, or cold virus. However, the incident also highlights a number of protocol violations by the Penn researchers, which may contribute to an alteration of how adverse events are reported in gene therapy trials.

"The matter of Jesse Gelsinger's death is a separate issue from reporting events to us," said Philip Noguchi, FDA's director of the division of cellular and gene therapies for the Center for Biologics Evaluation. "We are first looking at the delays in reporting to determine whether human error played a role in his death. But, these are protocol violations and that is very serious."

Noguchi said the agency is still in the early process of its investigation on what happened at Penn. The RAC, however, will entertain a proposal today to institute a more public adverse event reporting system. That proposal is opposed by several industry groups.

Thursday's agenda, however, focused on a detailed account of Gelsinger's death and a discussion of the recommendations the RAC will publish for continued use of adenovirus vectors in gene therapy.

Gelsinger suffered from partial ornithine transcarbamylase (OTC) deficiency. The inherited disorder results from an inability to process ammonia into urea. Most affected males usually die shortly after birth or within the first few years of life. Patients with less serious deficiency can survive provided they stick to a highly restricted diet free of protein. Even with medication and dietary interventions, patients are susceptible to severe hyperammonemia (high blood levels of ammonia) as a result of viral infections or slips from their diets. Without medical intervention, hyperammonemia leads quickly to coma and death.

The Penn researchers were conducting a dose-escalating Phase I safety study of a replication-incompetent adenovirus engineered to carry the ornithine transcarbamylase gene. The study was the first in an attempt to establish if the vector could restore, temporarily, the ability to process ammonia.

"If children are in coma for more than 72 hours they will suffer permanent brain damage," said Mark Batshaw, chief academic officer at Children's National Medical Center. "[This vector] was designed as a rescue for children in coma."

The researchers employed a protocol testing six escalating doses of vector testing three adult patients at each dose. Despite agreeing at a RAC meeting in 1996 to use an intravenous route of administration of the viral vector in order to reduce hepatic toxicity, the researchers in conjunction with FDA reviewers administered the vector directly through the hepatic artery in order to lessen the likelihood the virus would travel to the gonads and result in germline alterations.

"We recognize we should have come back to the RAC to discuss this change," Batshaw said. "But, this all happened at a time when the RAC's responsibilities were changing. We apologize for that."

The Penn researchers, however, didn't simply fail to discuss issues with the clinical trial with the RAC. They also failed to alert FDA of some serious adverse events in other patients. These events, while not life-threatening, would have resulted in the agency placing a clinical hold on the project in order to ascertain whether these toxicities would result in more serious events at higher doses. In addition, the Penn team promised FDA to dose two women with each dose before trying the therapy on a male. Because OTC is an X-linked disorder, women are less seriously affected than males. Gelsinger was the second patient in the highest dose group treated with the therapy, rather than the third.

Even though the entry criteria for the study required patients to have been stable for one month prior to enrolling in the study, Gelsinger was dosed with the gene therapy vector one day after the Penn doctors switched him to an intravenous therapy because his ammonia level was slightly elevated. After receiving the gene therapy, Gelsinger developed an expected fever. However, the next morning he had "altered mental status" and appeared jaundiced. He progressively became worse and by the second night was in a coma. He eventually died from adult respiratory distress syndrome.

Animal studies had highlighted liver toxicities and the potential for disseminated intravascular coagulation. Wilson said the team was surprised by the lung problems. Upon autopsy, the researchers were in for more surprises when they discovered Gelsinger's bone marrow was devoid of any precursors to red blood cells and the viral vector could be found in most tissues of his body.

"The thing that is most disturbing about the results of the autopsy is the biodistribution," Wilson said. "I'm not sure depositing a lot of vector into the spleen is a good thing."

The FDA and Penn researchers are still exploring why Gelsinger died. They are looking at a flood of various cytokines in response to the virus as a mediator of his death. In addition, they are going to look for human parvovirus infection to explain the bone marrow abnormalities.

In response to the information presented on Wednesday and Thursday, the RAC has begun to work on an Adenovirus Safety and Toxicity report. Suggestions include standardization of vector assays, a database of adverse events and a more specific definition of study endpoints. RAC will produce the report sometime next year.