Washington

Gene therapy researchers at the Baylor College of Medicine in Houston, Texas, who recently reported success with the ‘gutless’ adenoviral vector in animals may not get to test the vector in humans.

The pharmaceutical company, Merck, which owns the licence on a technique used to produce the vector, is refusing to extend its material transfer agreement (MTA) with Baylor, saying that it does not want to be liable for any problems the vector might cause in clinical trials.

Conventional adenoviral vectors usually have only one or two small regions deleted, but ‘gutless’ (also known as ‘fully deleted’ or ‘helper dependent’) adenoviral vectors have a large chunk of their viral genome excised. This means that they could be used to avoid two problems that have limited the usefulness of other adenoviral vectors: inflammation and lack of long-term expression.

Viral genomes in adenoviral vectors trigger immune responses, including inflammation. This is thought to have played a role in the death of 18-year-old Jesse Gelsinger, the first death directly attributed to gene therapy (see Nature 401, 517; 1999). In a later hearing, Merck's former senior vice-president Thomas Caskey testified that the ‘gutless’ vectors might provide a safer alternative to earlier generations of adenoviral vectors.

Animal experiments reported by Arthur Beaudet, chairman of molecular and human genetics at Baylor, at a conference in Colorado last month, reinforce that view, says Beaudet. In these experiments, Lawrence Chan, professor of molecular and cellular biology and medicine at Baylor, tested the vector in mouse models for inherited high cholesterol.

The experiments resulted in gene expression and lowered plasma cholesterol levels for at least six months. Whereas controls treated with a conventional adenoviral vector had marked inflammation in the liver, says Chan, mice that received the ‘gutless’ adenoviral vector experienced no significant inflammatory response.

After the experiments but before Gelsinger's death, Beaudet asked Merck to extend the MTA for a tool used to manufacture the ‘gutless’ vectors. The tool removes a large chunk of the vector's genome, yet still allows it to replicate. Merck denied the request, citing its concerns about liability.

The tool in question emerged from efforts in the early 1990s by a number of scientists — including Caskey, who was then at Baylor — to develop ‘gutless’ viruses that needed a second, ‘helper’ virus to replicate. A research group headed by Frank Graham, professor of biology and pathology at McMaster University in Canada, subsequently solved the problem of eliminating the helper virus in the final vector preparations by excising the viral packaging signal.

Merck, which bought a licence on Graham's patent, retains intellectual property rights on it. Even if Baylor researchers sign a liability waiver, the US Food and Drug Administration could still hold the company to “ultimate regulatory responsibility”, says Larry Hirsch, vice-president of public affairs at Merck Research Laboratories in Whitehouse Station, New Jersey.

Merck's refusal to extend the Baylor MTA for the vector-producing tool beyond preclinical use is “not unique to gene therapy”, says Hirsch. “It's the way we handle all products in early stages of development.” He adds that the vector is “nowhere near” ready for testing in a clinical trial.

Chan says the refusal is “frustrating”. The Baylor team's only alternative, beyond abandoning the ‘gutless’ adenoviral vector, is finding a new way to create one. While that remains a possibility, it would probably require years of work. Producing large enough quantities of the vector for humans remains a problem, adds Graham. Meanwhile, Merck plans to continue developing and testing the vector on its own.