Editor's Column:
“Leopards and Ghosts in the Temple”

Martin McKneally

“Leopards break into the temple and drink to the dregs what is in the sacrificial pitchers; this is repeated over and over again; finally it can be calendared in advance, and it becomes part of the ceremony.” (Franz Kafka)

With this vivid image, bioethicist and UofT Law Professor Trudo Lemmens characterized the current state of engagement of industry in academic science. Trudo and Frank Cunningham from the UofT Centre for Ethics recently organized an intriguing workshop at the Law School on the Ethics of Ghostwriting. This widely used technique for marketing surgical devices and pharmaceuticals was described in detail. Talented writers prepare manuscripts designed to help market commercial medical products. These papers are offered to prominent clinicians, who serve as guest writers, sometimes called host writers, lending their name and reputation to the marketing plan after review and token editing. Honoraria, advisory board salaries and research support are often used to enhance interest in these transactions. While still widely prevalent, guest writing is regarded as fraud in some jurisdictions. To deal with this problem, some journals’ policy on accepted manuscripts is to require the names of all contributors, some of whom might otherwise remain in the shadows as unacknowledged but well-paid ghosts. In response to such policies, I anticipate that there will be more transparent use of commercial medical writers as acknowledged authors and contributors.

I recently helped adjudicate a controversial decision on the publication of an article about the kallikrein inhibitor Ecallantide, a recombinant peptide in the same class as Aprotinin. Aprotinin had been widely used to reduce surgical bleeding until it was withdrawn because of treatment related complications and deaths. The Ecallantide trial was stopped because the drug was also associated with unacceptable mortality. Importantly, the study also demonstrated a beneficial effect on postoperative blood loss in the comparison group of patients treated with Tranexamic Acid, a fibrinolysis inhibitor made by Pharmacia-Upjohn.

There was understandable concern about publication, because of the extent of involvement of industry in the trial itself, and in the production of the manuscript. Cubist Pharmaceuticals, the manufacturer of Ecallantide, funded the study and editorial assistance from PharmaWrite, a commercial medical communications company. In addition, the first author is a company employee and stockholder in Cubist, and eight of the nine authors disclosed financial ties to the company. The data were monitored by Global Research Services, a commercial contract research organization (CRO), and analyzed by InVent Clinical, another independent CRO. Nevertheless, three careful reviewers, the associate editor, and the editor-in-chief were impressed by the value of the clinical contribution.

Should journals publish papers so extensively linked to commercial companies? While the intuitive initial response might be negative, analysis at the rational and reflective levels leads to a different conclusion. Rationally, we should not exclude any class of authors categorically on the basis of their affiliation – anymore than we would exclude a manuscript from a particular hospital, university or organization. No policy, precedent or principle precluded acceptance.

A reflective decision about publication should be based on values. The salient values that guide editorial decisions are: scientific merit, including freedom from bias; interest to readers; and usefulness to the profession. The article in question was judged to be scientifically sound, interesting, and useful. So, what were the ethical issues that disturbed the editorial board?

1. Commercialization of science. This is a concern, but prevention is inappropriate and unrealistic. Collaboration with industry has a long history, and is now widespread and highly productive. Industry supports virtually all pharmaceutical research, and most of the research on medical and surgical devices. When this collaboration is functioning optimally, the core values of science (such as accuracy, authenticity, logic, and truthfulness) are not compromised by commercial interest in profit.

2. Commercial bias in scientific trials and reports. When problems arise at the margins of data management, such as decisions about exclusions, time boundaries on sampling, and interpretation of adverse events, “the profit motive [risks] compromising standards of scientific truthfulness to a greater degree than in traditional academic science.” Authors and their institutions bear the principal responsibility for the integrity of their work, including its conduct and preparation for publication. Journals lack the resources to police the publication component of this vast enterprise, though reviewers and editors are advised to be alert against commercial messaging. Safeguards imposed externally include government mandated registration of all clinical trials, publication of protocols, and for some journals, obligatory access for reviewers to the full data. These measures create options to monitor, verify, and strengthen the oversight of research. The integrity of scientific publications is still largely dependent on trust in the reliability of all who contribute to the development of scientific progress. When egregious misrepresentation or fraud is discovered, the consequences are public disgrace and loss of reputation, as in the notorious cases of John Darsee and Hwang Woo-suk. Fines and other legal sanctions ensue if there is intent to defraud and evidence of harm. Commercial firms may suffer financial consequences from reputational loss, but the stakes are so high with blockbuster drugs and devices that even million dollar settlements and fines are not always a sufficient deterrent.

3. Commercial editorial assistance in the preparation of the manuscript. This merits discussion, as there are over 182 medical publication companies writing scientific articles for various clients. At the UofT workshop on the ethics of ghostwriting, experienced ghost writers told about the development, by the International Society of Medical Publications Professionals, of guidelines and a certifying examination aimed at establishing an aura of professionalism in this expanding field. Some critics view this as strategic marketing, but harmonizing these activities with the values of the science community could make a welcome contribution to clarity that these writers bring to the scientific literature. Alastair Matheson, a scholarly free-lance consultant and writer, told us that a good medical communications team requires substantial technical and academic expertise, including scientists, statisticians, commercial analysts, publication planners, trial designers, professional writers, and journal specialists who know the stylistic and format preferences of particular journals, editors and reviewers. It will not surprise readers to learn that professionally prepared articles have a rate of acceptance that is four times higher than the usual submitted manuscript.

The Ecallantide paper turned out to be an exemplary model of commercially sponsored clinical science. I spoke to one of the authors who helped design the trial. I learned that all authors had contributed significantly to the study and to the report. Contributions from the commercial medical communications company were minor and acknowledged. The manuscript was not written by ghost authors. The findings are distinctly unfavorable to the manufacturer of Ecallantide, and the report of a beneficial effect of Tranexamic Acid favors a competitor. Often, a negative study like this would be buried by a commercial sponsor. The commendable decision to submit the study for peer review certainly earned my respect and endorsement of publication.

The question remains – is this a Trojan horse, designed to establish a level of trust in commercial publication that will facilitate the acceptance of subsequent studies of less merit or integrity? Time and experience will answer. Like the ancient city of Troy, our scientific journals cannot remain permanently behind a defensive wall, but should be cautious, alert, and circumspect as we open the gates.

Acknowledgements: I am grateful to Trudo Lemmens, Alastair Matheson, Eleanor and Miles Shore, and a cottage full of McKneallys for helpful comments that improved earlier versions of this column. I have no relevant financial relationships to disclose.

M.M.