A preliminary investigation of a UK trial in which six healthy volunteers became critically ill said this was probably due to effects of the drug in humans not predicted by animal studies. But the report stopped short of questioning how the study was carried out.

The investigation into the phase I trial of the monoclonal antibody TGN1412, carried out by the Medicines and Healthcare Products Regulatory Agency (MHRA), the body that approves clinical trials in England, said the trial was run according to the agreed protocol, using the correct dose. It also found no evidence of a manufacturing problem or contamination of the product given to the trial volunteers.

The MHRA's report did not comment on suggestions that it might have been more prudent to have tested the drug in one volunteer at a time, leaving enough time to observe any adverse effects before giving it to another person, rather than giving it to the whole group at once (BMJ 2006;332: 683, 25 Mar).

The protocol for the trial …