Select Committee on Health Minutes of Evidence


Supplementary memorandum by AstraZeneca (PI 33A)

AN OVERVIEW OF THE CLINICAL TRIALS AND PUBLICATION PROCESSES

A.  THE CLINICAL TRIALS PROCESS

Introduction

  AstraZeneca is committed to conducting clinical trials to international standards of safety, scientific and medical integrity. There are currently within the UK in excess of 30,000 regulations and guidelines that apply to the drug research and development process. AstraZeneca adheres to these, as monitored through a comprehensive series of internal checks and external quality assurance checks.


  The MHRA conducts statutory Good Clinical Practice (GCP) inspections. AstraZeneca (UK and Global) underwent such an inspection in November 2004 and it received very positive feedback on the quality of its clinical research organisation.

The Research and Development Process

  Research and Development encompasses discovery, pre-clinical development, clinical evaluation (Phase I-IV), which leads to licence submission, and lifecycle development. It currently takes in the region of 10-13 years for a candidate drug to be licensed and to receive its marketing approval:


  Currently approximately 10,000 chemical entities are investigated to successfully develop one drug through to licence approval. Of these 10,000 chemicals, typically 1,000 demonstrate the necessary biological activity, and only 10 proceed through to clinical trial development, only one of which will successfully receive approval.

The Clinical Trial Process

  There are 4 key stages of the clinical trials process following successful pre-clinical development:

    —  Phase I: This stage is concerned most with safety of the candidate drug and examines the pharmacokinetics of the drug in healthy volunteers (usually 10-50 patient studies).

    —  Phase II: This stage is examining safety and efficacy of the drug in diseased patients (patient numbers in the 100's). Within this stage dose-ranging studies are conducted.

    —  Phase III: This stage examines the efficacy and safety of the drug in much larger populations of diseased patients, usually numbering thousands.

    —  Phase IV: These studies examine the licensed drug in large every-day settings and populations and may compare the drug with appropriate competitor drugs (possible in Phase III as well).

The Clinical Trial

  There are several key phases in a clinical trial from design (involving external investigators), approval from relevant authorities and ethics bodies, study set-up, patient consent and enrolment, treatment, data collection (throughout), data analysis, report writing, and publication (which is reviewed in depth below). Depending upon the treatment period a single study will take a minimum of several months to complete the cycle; most will take in excess of two years and many will take considerably longer:


  At each stage of the study design and clinical trial there are numerous patient safety, data quality and process review checks to ensure that high standards of scientific, medical and ethical integrity are being upheld. Herewith a schematic representation the quality control and quality assurance to AstraZeneca studies:


  It is normal practice for large-scale trials to have independent data safety monitoring boards (DSMB) that comprise independent statisticians and physicians of international renown. Their role is to monitor the safety and efficacy outcomes of the study on an ongoing basis, and to validate the statistical analysis and interpretation of results.

Publication

  AstraZeneca has had a long standing publication policy. It is AstraZeneca policy to encourage the appropriate communication of information on its products and research and development activities to the international medical and scientific community and AstraZeneca endeavours to publish the results of all its clinical trials.

  Below is an illustration of how a manuscript is developed for submission to the Lancet for publication. This demonstrates the lead and accountability for authorship taken by the study Steering Group and the input of AstraZeneca at both the review stage before submission and before Lancet editorial review.




  Further information is provided in the following section on the publication process.

B.  THE PUBLICATION PROCESS

The AstraZeneca Philosophy

  AstraZeneca is committed to providing healthcare professionals and patients with relevant information that enables them to make the best treatment decisions. To that end, scientific and medical publications are an important means of communicating the results of the company's research and development. Hence, AstraZeneca has put in place a mandatory Publication Policy that outlines the company's principles with regard to external communication of scientific and medical information. Specifically this policy describes the company's aim to publish the results of AstraZeneca-sponsored clinical trials; our commitment to maintain high standards of medical and scientific integrity by presenting results in an accurate, objective and balanced fashion; and states that selective publication that would misrepresent the medical profile of an AstraZeneca product is not acceptable (ie, we must not selectively publish only those trials that have a positive outcome for AstraZeneca products and suppress those that appear unfavourable). Compliance with this policy is formally monitored.

Relationship between AstraZeneca and External Investigators

  AstraZeneca works in collaboration with external investigators in the design and conduct of a clinical trial, and in preparing publications from that trial. Whilst AstraZeneca plays an important role in co-ordinating the publication process, the lead author (usually an external investigator) will play the major role in terms of publication content. This is only right and proper since it is that lead author who through authorship takes public responsibility for the overall design, data and conclusions in the publication.

External Publication Guidelines

  There are a number of well-established external guidelines developed by professional bodies to improve the quality and ethical transparency of publications. Probably the best known are the guidelines established by the International Committee of Medical Journal Editors (ICMJE) and published in its Uniform Requirements of Manuscripts Submitted to Biomedical Journals (www.icmje.org). Publications on AstraZeneca-sponsored trials in biomedical journals follow the ICMJE guidelines and in addition will comply with individual journal's policies and Instructions for Authors.

Data Access and Analysis

  The database for a clinical trial is usually created and maintained by the pharmaceutical company. Expert AstraZeneca statisticians and programmers are responsible for managing these trial databases. To give a sense of the size and complexity of these databases, for just a single AstraZeneca trial (Exanta, SPORTIF III) there were 12,500,000 pieces of data collected that resulted in 200,000 pages of study data.

  Once the trial is completed, AstraZeneca would supply authors with the statistical tables and figures that relate to any planned publication.

  It is normal practice for large-scale trials to have independent data safety monitoring boards (DSMB) that comprise independent statisticians and physicians of international renown. Their role is to monitor the safety and efficacy outcomes of the study on an ongoing basis, and to validate the statistical analysis and interpretation of results. In the Exanta study quoted earlier, the DSMB reviewed all data monthly on an ongoing basis throughout the trial.

  Despite the safeguards afforded by rigorous internal and external quality controls, some critics remain concerned that the pharmaceutical company "owns" the database. AstraZeneca believe they have nothing to hide. For registration trials, AstraZeneca routinely make the electronic database available to registration agencies such as the FDA, and have complied with requests from medical journals for a further additional independent statistical analysis as part of the journal's peer review process.

Authorship

  The authorship of publications carries significant responsibilities and must be approached in a rigorous manner. For any publications involving AstraZeneca-sponsored trials, authorship is determined by strict criteria contained in ICMJE guidelines. Each author should have participated sufficiently in the work to take public responsibility for appropriate portions of the content. In particular, authorship credit should be based only on (1) substantial contributions to conception and design, or acquisition of data, or analysis and interpretation of data; (2) drafting the article or revising it critically for important intellectual content; and (3) final approval of the version to be published. Conditions (1), (2), and (3) must be met. Hence, it should be clear that the practice often referred to as "gift" authorship is not acceptable practice within AstraZeneca.

  Where AstraZeneca scientists, physicians and statisticians have played an important part in a study (eg, the conception and design, or analysis and interpretation) they will also be listed as authors. We continue to do this because we believe in transparency despite disturbing evidence that declaring competing interests negatively affects readers' perceptions of studies in terms of their importance, relevance, validity and believability (BMJ 2004, 328, 742-743).

  AstraZeneca believes it is proper for company authors to engage with external authors in discussion on sound scientific grounds, but that there should be no attempt to influence inappropriately the scientific or medical opinions of these external investigators.

Role of Professional Medical Writers

  Ghostwriting is the practice whereby someone other than the named author (eg, a professional writer) has been responsible for preparing the publication, typically with little or no involvement from the named author, and the contribution of the actual writer—the "ghost"—is not disclosed. AstraZeneca does not support "ghostwriting". However, we do believe that professional writers have a legitimate role to play in assisting authors, providing any such collaboration follows ethically acceptable practice.

  In an ideal world, the scientists, statisticians and physicians who were involved in design, conduct and interpretation of a study should be the people preparing the publication. However, the best clinical investigators do not necessarily make the best writers. They may lack the time, expertise or language skills to produce a well-written publication promptly (for example, many of the investigators will not have English as their first language). In these circumstances, most pharmaceutical companies, including AstraZeneca, may use professional writers and editors (internal staff or agency/freelance) to assist in publication development. The use of professional writers may be particularly helpful to aid the process of publishing results from large multicentre studies involving many contributors.

  AstraZeneca believe that the named authors must retain responsibility for the article's content, this is achieved by the named authors being fully involved from the outset; the principal author should determine the extent of involvement of professional writer and, importantly, surrenders no responsibility for the content of a manuscript by accepting this help; and finally there must be no attempt by the writer to manipulate the opinions of the named authors.

  If pharmaceutical companies did not provide this service to external investigators, a number of negative consequences are likely. Namely, publications would take longer to get out into public domain; more publications would be rejected because of poor quality; some publications would never see the light of day—individual investigators would have little incentive to publish "worthy but dull" studies; critics believe there is publication bias in the medical literature now—this situation would worsen if pharmaceutical companies did not provide writing or editorial support to busy clinical investigators.

Acknowledgments and Disclosures

  It is AstraZeneca policy to ensure adherence to the principles of good publication practice that are described by International Committee of Medical Journal Editors (www.icmje.org) and hence all authors should provide details of any conflict of interest, or financial support/financial connections to the work. Again, we continue to do this because we support transparency despite evidence that this declaration unreasonably biases readers' perceptions.

Products Withdrawn from Development

  The pharmaceutical industry is often criticised for not publishing clinical studies that cease to be of direct commercial relevance to them. Since the merger of Astra and Zeneca in 1999, we have withdrawn a number of products from development. One such example is Viozan (sibenadet), a dual dopamine D2-receptor and beta2-adrenoceptor agonist that was being developed for treatment of chronic obstructive pulmonary disease. The lessons learned from this failed development were published in a supplement to Respiratory Medicine. The following quotation from Stephen Rennard MD in the introduction to that volume (Respiratory Medicine 2003, Volume 97, Supplement A, pages S1-S79) underlines our ethical policy to share information with the whole medical and scientific community.

    "Finally, many drugs that fall in development for one reason or another do not result in compiled publications such as this. AstraZeneca is to be commended for recognising the value of the lessons learned from the sibenadet development programme and for helping to make them readily accessible through the development of this supplement."





 
previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2005
Prepared 26 April 2005