Medical journals should adopt a unified approach requiring full data access and independent statistical analysis for industry-sponsored clinical studies, argues an editorial in the the Journal of the American Medical Association.

The editorial by Dr Catherine DeAngelis and Dr Phil Fontanarosa, respectively editor in chief and executive deputy editor of JAMA, is the latest salvo in the debate over the safety of GlaxoSmithKline’s diabetes drug Avandia (rosiglitazone) from the publisher of the meta-analysis that effectively sparked off the controversy in June 2007.

It refers to a commentary (‘Setting the RECORD straight’) in the same issue by Dr Steven Nissen, co-author of the meta-analysis and indefatigable critic of GSK’s conduct over Avandia, taking the company to task over its crucial RECORD (Rosiglitazone Evaluated for Cardiac Outcomes and Regulation of Glycaemia in Diabetes) safety trial.

According to JAMA, Nissen’s commentary “describes a disturbing example of inappropriate conduct surrounding an industry-sponsored clinical trial of rosiglitazone and reveals a situation in which concerns about preserving market share apparently trumped concerns about the potential for causing patient harm”.

The journal cites claims from Nissen, the recent staff report on Avandia by the US Senate Committee on Finance and published media accounts that GSK, among other things:

- Undertook non-pre-specified unblinding of the study data.

- Attempted to undermine the authority and responsibilities of the study steering committee.

- Expedited publication of an unscheduled interim analysis of the RECORD trial, “specifically to counter the publication of a meta-analysis by Nissen and Wolski that suggested increased cardiovascular risk associated with rosiglitazone”.

- Had the trial results reported by employees “extensively involved in statistical analysis and preparation of the manuscript”.

- Failed to fully acknowledge the “significant cardiovascular risk” associated with Avandia.

GSK says the commentary and the editorial “unjustly challenge the motives of physician scientists seeking to inform diabetes treatment and patient safety and mischaracterise the RECORD study on rosiglitazone”.

JAMA Policies

JAMA details a range of policies it has put in place to discourage incomplete or inappropriate data analysis, selective reporting or misrepresentation of published clinical trials.

For example, with all studies reporting original data, at least one author must indicate that she or he “had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis”. For industry-sponsored studies, this statement must be provided by an academic investigator or government researcher not employed by any commercial funding source.

Data collection and data management for industry-sponsored studies should be conducted “primarily or solely” by the academic investigators, independently of the study sponsor or other for-profit research organisation and with additional monitoring and oversight, such as under the auspices of an academic independent data and safety monitoring committee, JAMA says.

Industry studies in which the data analysis has been conducted only or primarily by statisticians employed by the company sponsoring the research are not accepted for publication in JAMA. Moreover, for these studies there must be an independent analysis of the data by an independent statistician at an academic institution, rather than by statisticians employed by the sponsor or a commercial research organisation.

The journal recognises these requirements entail additional time, effort and cost. “However, given concerns about the integrity and reporting of some industry-sponsored research, this independent assessment of the data and statistical analysis is essential and also provides an additional layer of institutional oversight for the integrity of the data analysis and reporting,” DeAngelis and Fontanarosa argue.

Given the circumstances surrounding Avandia and the RECORD trial, “and perhaps other industry-sponsored trials”, it is “now time for all editors to require that academic researchers have full access to all trial data and that all industry-sponsored trials include independent statistical analysis and assurance”, the authors write. “This approach would add powerful support to the fundamental principle that physicians must first do no harm.”

Independent verification

In response, GSK noted that RECORD was a regulatory commitment study and was run by an independent steering committee, an independent safety monitoring board and a clinical endpoint committee.

“The safety monitoring board reviewed all of the data during the course of the study for oversight of patient safety,” the company stated. “Both the interim and final analyses were independently verified at the London School of Hygiene and Tropical Medicine. The methodology and baseline characteristics and design were published well in advance of any analysis.”

The interim analysis of the RECORD trial was undertaken “as quickly as possible to gather additional information about the potential risk for patients, not only those in the study, but also those struggling with diabetes treatment in the real world”, GSK said. The company had asked for and obtained the endorsement of the RECORD steering committee before conducting the analysis, it added, commenting: “The allegation that GSK did not obtain this endorsement prior to unblinding the data is simply untrue”.

The authors of the JAMA editorial and commentary underline the importance of industry-sponsored studies meeting key criteria and being assessed by journals before publication to ensure high standards. “The desired criteria include: a detailed reporting of finance, a detailed accounting of roles for sponsors and academics, and clear rationale of design, conduct management and analysis of the study,” GSK said. “RECORD met all these criteria.”