By Pallab Ghosh
BBC News science correspondent
The adverse side effects that occurred in a drug trial that hospitalised six healthy volunteers could have been predicted, experts have told the BBC.
Trial took place in north London
They said there was enough worrying evidence in the scientific literature on the effects of the kind of drug that was tried out to ring alarm bells.
More tests should have been carried out before trying it out on people for the first time, they argue.
TeGenero, the makers of the drug, have dismissed the claims.
The drug trial took place at London's Northwick Park Hospital in March.
The six volunteers had a severe reaction to the monoclonal antibody TGN1412. Several ended up in a critical condition.
An interim report by the Medicines and Healthcare products Regulatory Agency (MHRA) - which gave the trial the go ahead - said the reaction was unpredicted.
Dr David Glover is a member of an industry task force set up to investigate how trials of this kind should be carried out in the future.
He is one of a number of experts that have decided to speak out for the first time, to detail their concerns about the events leading up to the accident.
Potential for danger
He told the BBC Radio 4 PM programme: "It may be that it was unpredicted by the tests that were done. I believe from the basic science it was predictable."
Professor Greg Winter, of the Laboratory of Molecular Biology in Cambridge, was a key figure in the first development of monoclonal antibodies 20 years ago.
He said those testing TGN1412 might have been lulled into a false sense of security by the fact that it did not seem to harm monkeys - but it was wrong to make too many assumptions based on animal experiments.
Tests of the drug had the potential to go wrong because it was known to stimulate the immune system, and there had to be a possibility of triggering an uncontrolled chain reaction.
Professor Winter said: "I'm not saying that these experiments shouldn't be done but I'm saying that one should have started from the principle that this was probably rather more dangerous - in fact a lot more dangerous - than the typical experiments which are done with antibodies in clinical trials.
"If anything, I would have expected to see the kind of response those patients had."
Drug manufacturers TeGenero said in its application to do the trial that this effect had not been seen in monkeys.
It planned to give the volunteers a dose 500 times smaller than that given to the animals.
Regulators accepted that this would provide an "adequate safety margin".
But industry consultant Dr Andrew Millar said the theoretical risk meant more care should have been taken.
"My view is that if there is a 0.1% chance of it happening that is unacceptable for a healthy volunteer."
The drugs firm AstraZeneca is also developing drugs of this kind.
Harsuk Parmer, its director of discovery medicines, said the company carried out a wider range of tests than normal before proceeding to human trials when working with monoclonal antibodies.
Even then the first step they take is to put ultra low concentrations just under their skin rather than straight into the blood stream.
A report by the Academy of Medical Sciences also suggests there was evidence that the trial was potentially risky.
In a statement, TeGenero said it was an "oversimplification" to suggest that the side effects could have been predicted in advance.
It had carried out experiments that showed that the drug behaved the same in monkeys and humans.
This was closely scrutinised by the German Regulators prior to approval. That and additional information was passed on to the MHRA.
However, Dr Glover said details of the experiments were being kept out of the public domain by the MHRA for reasons of commercial confidentiality.
He questioned that decision, and said sharing information would aid the formulation of guidelines about appropriate starting doses in human trials.
Dr Parmar questioned whether the MHRA had the expertise to properly judge applications to test drugs such TGN1412.
The MHRA said it had re-reviewed the data and even with the benefit of hindsight could see nothing that could have predicted these effects in humans.