The Complexities of Physician Interactions with Pharmaceutical Companies

In addition to these concerns related to clinical practice, and of particular relevance to research societies and academic centers, are issues arising during the conduct of pharmaceutical-company-sponsored clinical trials. Because pharmaceutical companies are the major source of novel and interesting therapeutic compounds, it is of critical importance to efficiently and collegially conduct pre-clinical and clinical research in collaboration with these companies. Drug-company-sponsored trials have become a key source of clinical research submissions to the ASH and ASCO annual meetings. For example, I analyzed that approximately a third of the oral presentations in the hematologic malignancies sessions at ASH's annual meeting in 2006 were sponsored by a pharmaceutical company or had co-authors from companies. Similarly, 44 percent of 332 clinical trials published in the Journal of Clinical Oncology in 2005 reported funding by pharmaceutical companies.¹ 

A number of issues can complicate these collaborations:

• It can be difficult to distinguish the presentation of new information at the annual meeting from marketing. It is common to see multiple presentations from the same company rehashing or "updating" older data, frequently focusing on subgroup or interim analyses, often with essentially the same information presented at ASCO the following spring. Judging by the similarities in format, these are often (usually?) prepared by the company, and, not infrequently, the abstracts/slides/posters are sent to the authors with short notice without the opportunity to review the primary data or have substantive input into the content. At the other end of the spectrum are submissions containing data that are too preliminary for meaningful interpretation, and company executives have acknowledged to me that such presentations are sometimes aimed more at investors than the scientific audience.

• Abstracts/manuscripts are often written by investigators from companies or by writers hired by the companies, a point which is sometimes, but not usually, acknowledged in the publication. Although this is not necessarily "bad," the authors of the abstract/paper have the critical responsibility for careful review of the data and the language used to describe the conclusions, particularly given the recent report that evaluated articles published in the Journal of Clinical Oncology and concluded that "conflicts of interest are associated with highly positive conclusions that use superlatives to promote the experimental arm."²  It has been suggested that investigators and companies adhere to recently proposed AAMC Guidelines for Protecting Integrity in the Conduct and Reporting of Clinical Trials to help provide consistency and transparency in the conduct and reporting of such trials.

• There has been considerable interest in the lay press and academic journals about the effects of financial conflicts of interest (COI) posed by consultancies and stock ownership, with some articles documenting excessive honoraria to physicians for unbalanced presentations of clinical data. Of perhaps greater importance than what hopefully represents outliers, are other more subtle influences, including the potential effect on investigator objectivity of a career-advancing association with "positive" studies. This can result in the traditional rewards of publication, academic promotion, subsequent grant funding, and the ability to direct/attract the next major trial to one's institution, but it can also provide opportunities for remunerative consulting and extensive travel for lectures at restaurants and other non-academic venues around the country, often using slides prepared by the burgeoning industry of medical education companies. Few can claim indifference to these side benefits, although the temptations might be less if younger academics were paid more competitive wages. Professional societies, regulatory agencies, and universities have guidelines governing these interactions in place, although it remains unclear whether these are really realistic and functional and capture the impact (if indeed there is an impact) of these less obvious COI.

• There are also issues about interactions with clinical research organizations, including the emphasis on the excessive amount of data collected on industry-sponsored trials and the large number of frequently meaningless data queries, which decrease the efficiency of the conduct of trials, consume the time of data managers, and, thereby, reduce the number of trials that institutions can open. This is an area requiring more attention and new methodological research focused on whether data minimization might be sufficient for most clinical trials.

It is always easier to harp on the more negative and polarizing aspects of contentious issues such as this. However, pharmaceutical companies have and will continue to contribute enormously to the improved care of our patients, and these interactions are therefore likely to increase. Whatever new regulations may be forthcoming, in the end, the ultimate responsibility will reside with the individual physician and investigator who will, at times, have to react to some of the situations described above. Most will continue to respond with intelligence and integrity, although certainly they should maintain a questioning and critical attitude about these relationships.