BMJ  2007;334:551 (17 March), doi:10.1136/bmj.39153.352234.DB

News

UK report recommends better planning for phase I drug trials

London

The first trials of new drugs in humans should be planned much more carefully, with the same clarity of purpose, design, and analysis as for studies supporting drug licensing, a report published this week recommends.

The report, published by a working party of the Royal Statistical Society, was prompted by the TGN1412 drug trial last year in which six healthy volunteers experienced severe immune reactions (BMJ 2005;332:683, doi: 10.1136/bmj.332.7543.683). It found that "the trial design was not well suited to its objectives of testing the safety and tolerability of the drug."

Stephen Senn, professor of statistics at the University of Glasgow and chairman of the working party, said, "The fact that so many volunteers simultaneously suffered severe reactions clearly signalled that the design of the TGN1412 trial might have been deficient."

He considered that many early phase studies lack clear aims or plans for analysis. "Researchers are uncertain about what they are going to find, so they don't document in detail the study design or how they will analyse what they find."

The working party recommended that phase I, "first in man," studies should be designed and described much more carefully. This should include quantitative justification of the starting dose based on appropriate preclinical studies and relevant calculations; assessment of the risk level for the recommended study dose; and appraisal of the uncertainty about these recommendations.

Documentation of this information should be given to the ethics committee, study participants, and insurers. Professor Senn said that volunteers taking part in trials should be fully informed about possible risks and uncertainties as part of gaining their informed consent.

Researchers should take a precautionary approach to study design for any experimental medicine that is "first in class" (and therefore high in terms of risk) or which has a high dose specific risk. The trial design should allow sufficient time between giving the drug to subsequent subjects so that it can be halted without risk to subsequent subjects if any evidence of toxicity emerges.

Studies with even a remote possibility of complications, such as the cytokine storm immune reaction that TGN1412 caused, should be done only in hospitals with full facilities for giving tertiary care.

The report also called for much better communication of information about early clinical trials. The working party found deficiencies in the linkage of databases held by the Medicines and Healthcare Products Regulatory Agency on phase I trials and associated serious adverse events.

Professor Senn commented, "Collection of data on risk, including their relation to study design, type of experimental medicine, and whether volunteers are patients or healthy subjects, should be a routine matter. Drug regulatory bodies must take a leading role in analysing, sharing and communicating data on risk to all parties."

The working party recommended that drug regulatory authorities should provide a mechanism for the drug industry to collect and share data on serious adverse reactions in phase I studies to improve risk assessment.

---

The Report of the Working Party on Statistical Issues in First-in-Man Studies is available at www.rss.org.uk/first-in-man-report.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?

Relevant Article

Severe adverse reactions prompt call for trial design changes

Susan Mayor
BMJ 2006 332: 683. [Extract] [Full Text] [PDF]