A patient with arthritis who died in July during a gene-therapy trial may have succumbed to an infection she had before the viral vector was administered, experts said on Monday at a meeting of an advisory panel in Bethesda, Maryland, investigating the incident.

Little of the evidence presented to the panel seemed to indicate that the injected viral vector had a key role in 36-year-old Jolee Mohr's demise. Although DNA sequences from the vector were found in her liver and spleen, “the detection of the sample is very low and below the limit of quantification in these assays”, said Jeffrey Bartlett, a member of the National Institutes of Health (NIH) Recombinant DNA Advisory Committee (RAC). “It really indicates the absence of ongoing replication of the vector in these tissues.”

The panel is still awaiting the results of tests on Mohr's tissue samples, expected by December, and the results of blood studies looking for the vector and protein produced by the transgene. It will then reach a firmer conclusion about whether the gene therapy she received three weeks before her death was in any way responsible for it. But even with all the information in hand, “there will I think still be some uncertainty”, said RAC chair Howard Federoff.

Mohr died on 24 July, 22 days after her right knee was injected with an adeno-associated virus (AAV) vector made by Targeted Genetics, a Seattle company. This vector was engineered to contain a gene for an anti-inflammatory protein, TNFR:Fc, which inhibits tumour-necrosis factor-α (TNF-α). TNF-α causes inflammation in rheumatoid arthritis.

Autopsy results and the clinical history of the patient presented at the 17 September meeting indicated that the immediate cause of her death was massive bleeding of unknown origin in the tissue space behind the kidneys. This displaced her abdominal organs and ultimately compressed her lungs, making them unable to function. She was also overwhelmingly infected with histoplasmosis, an environmental fungus that can cause serious infections in immunocompromised individuals. She began to feel ill with fatigue and a low-grade fever three days before the gene therapy was administered, the panel heard.

TNFR:Fc suppresses the immune system, and so it is possible that the gene therapy caused the infection. However, the design of the trial, in which some patients were also taking similarly acting immunosuppressant drugs, means that it will be difficult to ascertain what caused the infection to take hold. Mohr was also taking a TNF-α-inhibitor drug for the arthritis, which suppresses the immune system. Carol Kauffman, an infectious-disease specialist at the University of Michigan, Ann Arbor, told the panel that overwhelming histoplasmosis infection had killed several patients taking such TNF-inhibitory drugs.

Of the 871 gene-therapy trials registered with the NIH, 4% have used AAV. Officials at the NIH and the US Food and Drug Administration reported at the meeting that they had reviewed all the AAV trials to date and found no pattern of adverse events associated with AAV, or any increased incidence of adverse events compared with trials using other gene-therapy vectors. The trial Mohr took part in was unusual in that it involved repeated injections. It was halted after her death.