Another Death in a Gene Therapy Trial

The case of the death of a research subject in an early trial of gene therapy for arthritis raises some eerie parallels to previous cases. So far the case has sparked minimal media attention, so although the news so far raises far more questions than answers, it appears worth summarizing on Health Care Renewal.

As reported by the Seattle Times,

Shortly after suspending a major clinical trial because of serious illness in a patient, Targeted Genetics and federal regulators said Thursday that the person had died.

Whether it was the experimental arthritis drug being tested that caused the illness is not yet known. But the condition 'was related in time to the receipt of a second injection of the product,' the U.S. Food and Drug Administration said Thursday in a statement. The agency is investigating whether the illness was related to the treatment.

'We do not yet understand the cause of death,' said Barrie Carter, Targeted Genetics' chief scientific officer.

Some 127 patients have received an initial dose of the active drug or placebo, including 74 that received a second dose of the drug, Targeted Genetics said. The drug, known as tgAAC94, is designed to be used along with other therapies.

The Scientist provided a little more detail about the therapeutic intervention.

Researchers in the field say the treatment's delivery vector, an adeno-associated virus (AAV), was unlikely to be the culprit.

'Vectors in this class have been used on hundreds of patients over the last 12 years, and are not associated with acute toxicity,' said Terence Flotte of the University of Massachusetts Medical School in Worcester. Flotte has been a principal investigator in several clinical trials of AAV-based gene therapy, but is not associated with the trial in question. 'I've never seen anything like this,' he told The Scientist. 'Whatever this is, it's unusual.'

In the trial, the AAV vector delivered a transgene encoding the receptor for tumor necrosis factor (TNF)-alpha, a cytokine that causes joint swelling in arthritis patients. The receptor, secreted by the target cells, binds to TNF-alpha, to reduce inflammation and protect the joint.

Usually prescribed for autoimmune conditions such as rheumatoid arthritis, the TNF-alpha receptor works by suppressing the immune system. But the protein could also have opened the door to an infection, suggested Flotte.

The Washington Post reported that a US National Institutes of Health (NIH) panel that reviewed the initial project protocol had some serious doubts.

Unlike the vast majority of such proposals, all of which aim to treat diseases by giving patients new genes, the plan to inject trillions of genetically engineered viruses into the joints of patients with arthritis was flagged for a special public review by the federal Recombinant DNA Advisory Committee, part of the National Institutes of Health.

At that Sept. 17, 2003, meeting, representatives of the sponsoring company, Targeted Genetics Corp. of Seattle, listened as a panel of experts wondered aloud why such a novel and possibly risky approach was to be offered to patients who were not especially ill, including some who had not even tried standard treatments.

Reviewers questioned the justification for the study, given that animal studies had found only a 'limited correlation' between the treatment and any improvements in subjects' condition. And they asked for more assurance that the engineered viruses were not going to spread around the body or cause untoward immune system reactions in patients.

Some also expressed concern that the informed consent document the researchers planned to use to describe the risks and benefits to participating patients was not upfront enough about the fact that the study was unlikely to help them and was designed merely to test the new approach's safety.

Even more eerie was the study's ties to a previous failed study of gene therapy. Per the Seattle Times,

The death is the second in the history of gene-therapy trials, experts said. In 1999, 18-year old Jesse Gelsinger, part of a University of Pennsylvania study on an inherited liver disease, died of a massive immune reaction to a gene-therapy treatment.

There's a business link to Targeted Genetics from that case: The UPenn experiment was led by Dr. James Wilson, whose company, Genovo, licensed the technology used in the trial. Targeted Genetics acquired Genovo in 2000 for $66 million in stock.

The company, however, says that there's no Genovo technology in its inflammatory-arthritis drug.

Note further that Wilson, who owned 30% of Genovo stock, reportedly made $13.5 million from the sale of the company to Targeted Genetics, per an article in the Seattle Times "Uninformed Consent" series.

It's too early to tell what this case means, but I do have some comments. Just because a treatment has been used on "hundreds" of patients does not mean that it has no serious acute side-effects. It only means that such side-effects may be relatively infrequent.

I may be overly conservative, but it seems we know a lot less about the sort of treatment used in this trial than we do about many other treatments. Occasionally, other novel treatments that seemed as if they ought to have been safe have not. Recall the TGN 1412 case (see post here) as well as the case of Jesse Geisinger noted above. Thus, in retrospect, it is easy to say that extreme caution in the first trials of the treatment was indicated. Instead, it often seems that people who work at the cutting edge of biotechnology are in a big hurry to develop their products. It is not clear from current reporting the extent that the design of the trials of tgAAC94 reflected the concerns of the 2003 NIH panel.

So we will keep an eye on this one. Some questions to keep in mind, based partly on the TGN 1412 story, are: how was the trial actually designed, what illnesses did subjects actually have, and what sort of informed consent did subjects give? Who actually carried out the trial (I assume it was not Targeted Genetics personnel)? Did previous animal or in vitro studies of tgAAC94 suggest the possibility of any bad adverse effects?