Researchers have outlined lessons that can be learned from a drug testing trial in which a person died.
The experts are urging a review of how tests for new medicines - so-called first-in-human clinical trials - are assessed, following the tragedy in France in January that also left four people seriously ill.
Writing in the British Journal of Clinical Pharmacology, the researchers - from Edinburgh and the Netherlands - make several recommendations to improve the safety of clinical trials.
The report authors argue that a comprehensive risk assessment should be carried out for every new drug.
Under current guidelines, only compounds that are considered to pose a major risk to patient safety are assessed.
They recommend that all clinical trials should adopt a sequential dosing method, in which new medicine is given to just one person first, followed by a delay to assess any side-effects.
The researchers have called for urgent release of study data from the fatal clinical trial so that lessons can be applied.
Unless additional information becomes apparent to show that there were unique features about this study, we are going to need to be much more careful in designing studies to sequentially dose patients. Currently this is typically just done for one or a few groups. It is possible that all groups will need to be sequentially dosed.
Five volunteers who were taking part in the clinical trial became seriously ill and subsequently one died.
The trial of a new pain relief drug was the first time that it had been given to people, although it had been extensively tested in animals in line with regulatory procedures.
This type of study - known as a phase one clinical trial - is designed to test the safety of a new drug in healthy volunteers and to identify any potential side-effects.
Thousands of these clinical trials in humans have been conducted in Europe over the past decade with very few serious incidents.
Although the pain drug trial had been subject to safety reviews, the medicine was not considered to be high risk because it had been shown to be safe in monkeys.
These tests did not find any ill effects even at higher doses than were used in the human trial.
In addition, the drug is chemically similar to other compounds that have already passed clinical trials.
Because the drug was not deemed to be high risk, additional safety precautions - such as sequential dosing - were not implemented.
The Edinburgh researchers - backed by an expert from the Centre for Human Drug Research in Leiden - argue that assumptions about what constitutes a high-risk drug need to change.
We can never entirely remove the risks involved with initial testing in humans, but improvements have been made in the past and should continue to be made in the future to safeguard volunteers. A pharmacological approach to dosing and a comprehensive assessment of risk for each drug – not just those thought to be high risk – is needed.