Journal tightens its controls

The publication that ran controversial findings of a drug-trial study has acknowledged concerns about the results and is changing its review procedures, writes Phil Baty

A leading medical journal has issued a "statement of concern" about research findings it published on the multimillion-pound osteoporosis drug Actonel, following an investigation into how the research was conducted by The Times Higher .

The Journal of Bone and Mineral Research , which published the research in 2003, also confirmed this week that it would reform its publication procedures "to help ensure that future articles published in the JBMR are scientifically accurate and valid".

The Times Higher reported last November that findings on Actonel, manufactured by Procter & Gamble, had been released under the name of Sheffield University researchers despite the fact that the academics had not carried out their own independent analysis of the firm's drug-trial data.

Academics from Sheffield's Bone Metabolism Research Unit had instead left all data analysis to statisticians employed by the company that manufactured the drug. P&G also heavily funded the Sheffield research unit and was paying for the Sheffield study on Actonel.

When Aubrey Blumsohn, a senior lecturer and a member of the unit, finally gained limited access to some data after repeated requests, he raised concerns with The Times Higher about the validity of findings being made public.

Concern has focused on an article published in the JBMR in 2003 by Richard Eastell, head of the Sheffield unit, with a junior colleague, Rosemary Hannon. Both authors signed a declaration that "all authors had full access to the data and analyses".

Doubts about the validity of claims made in this paper have been raised by senior statisticians.

The JBMR "statement of concern" acknowledges that "questions have been raised" about the article, "Relationship of early changes in bone resorption to the reduction of fracture risk with risedronate" (the generic name for Actonel).

In a statement to The Times Higher , John Eisman, the journal's editor-in-chief, said that P&G had now released the relevant drug data to Sheffield University and that the JBMR had "requested that a formal reanalysis of the data be undertaken and that the reanalysis be overseen by Sheffield University".

The original authors are re-analysing the material, and their follow-up article will also address the issue surrounding the authors' declaration that they had had "full access to the data and analyses".

Dr Eisman said that the journal would also publish a statement from Dr Blumsohn that he wished to be dissociated from two research abstracts the JBMR published under his name, which report conclusions that he says he cannot verify.

"In addition, we have taken steps to help ensure that future articles published in the JBMR are scientifically accurate and valid," Dr Eisman said.

He said the journal would adopt the American Association of Medical Colleges' "principles for protecting integrity in the conduct and reporting of clinical trials" published last January.

"We have also reached a decision to require that a letter jointly signed by all authors and industry sponsors, stating that the authors had access to the data upon which the analyses were based, accompany any manuscript that reports the results of an industry-sponsored clinical trial."

G has stressed that the safety and fracture-reducing benefits of Actonel have been proven "in one of the largest clinical trials programmes for osteoporosis therapy" and said that the company stood by the original conclusions published in the JBMR , which are rightly open to scientific debate and further study.

P&G said it had been standard practice to limit external access to its drug-trial data, although it had since allowed access.

Professor Eastell has indicated to colleagues previously that he worked closely with the P&G statisticians to identify the best approach to analysing the data.

He told The Times Higher : "I can confirm that we are working on analysis of the data that is independent of the company. We plan to submit this... to the JBMR ."

phil.baty@thes.co.uk

WATCHDOG POWERLESS TO INTERVENE

The national drug watchdog has admitted that it has no legal powers or remit to investigate concerns about the conduct of Sheffield University's study of Actonel, six months after the agency was asked to look into the matter for the Government.

After The Times Higher first reported concerns that Sheffield academics had published research findings on the drug without having conducted their own analysis of the drug-trial data, the Government's Chief Medical Officer asked the Medicines and Healthcare products Regulatory Agency to investigate.

Liam Byrne, the Health Minister, asked the agency to examine "whether there has been any failure to report to the MHRA any change in the medicine's risk/benefit in relation to the allegations", and "whether there had been any breach of clinical trials regulations".

However, the research was an exploratory "post-hoc" study based on the clinical trial data carried out after the drug had been licensed.

Ian Oulsnam, operations manager of the MHRA, said: "The ministers' questions related to issues that might be covered by the powers of the MHRA. We have no authority to investigate this kind of exploratory research, which is not directly part of clinical trials and which is not submitted for the purposes of licensing medicines."