Evidence suggests that some clinical trials may have more to do with marketing than answering important research questions. The ethics of conducting these so-called ‘seeding trials’ is questionable.
What is this paper about?
It has been suggested that so-called seeding trials are designed to appear as if they address important research questions, but in fact are motivated by marketing objectives. By ensuring that a large number of prescribers become familiar with a new drug, the sponsoring manufacturer hopes that their prescribing habits will change. A secondary aim is to raise the prescribers’ estimations of the company, and even to make them feel beholden to it. An editorial accompanying this study describes the process, at least in the United States:
“The company flatters a physician by selecting him because he is “an opinion leader” and incorporates him in the research team with the title of “investigator.” Then, it pays him good money: a consulting fee to advise the company on the drug’s use and another fee for each patient he enrolls. The physician becomes invested in the drug’s future and praises its good features to patients and colleagues. Unwittingly, the physician joins the sponsor’s marketing team. Why do companies pursue this expensive tactic? Because it works.”
Possible indicators of this type of paper are given below (see #). The term “seeding trial” was used in an editorial 14 years ago, and has appeared several times since then, but is there any evidence that such practices actually go on?
What does this paper claim?
The authors had access to Merck & Co‘s internal and external communications, including emails, presentations, memoranda, etc, as (paid) consultants for plaintiffs in actions against Merck & Co* over the cardiovascular effects of Vioxx® (rofecoxib).
After reviewing these communications, they conclude that the ADVANTAGE trial was a seeding trial, designed by Merck & Co’s marketing division to fulfil a marketing objective. They also say that Merck & Co hid the marketing nature of the trial from participants, physician investigators, and institutional review board members (the equivalent of research ethics committees in the UK).
ADVANTAGE recruited 5,557 patients with osteorarthritis from 600 GP practices, an average of only just over nine patients per practice. It compared rofecoxib 25 mg daily with naproxen 500 mg daily for three months, with the primary endpoint being discontinuation due to gastrointestinal (GI) adverse events. The VIGOR study was launched around the same time as ADVANTAGE and was the study required by the FDA to support Merck & Co’s claims of enhanced GI safety with rofecoxib.
The head of the research division at Merck had argued against ADVANTAGE, describing it as a large marketing clinical study which was “intellectually redundant”. But in an internal email, an employee in the marketing division had said “[ADVANTAGE] may be a seeding study, but let’s not call it that in our internal documents”.
Note: in a rapid response to this paper the Executive Director, Global Center for Scientific Affairs, Merck Research Laboratories denied what he called the ”numerous inaccuracies” in it and stated that the evaluation by the marketing division of the “commercial possibilities” of ADVANTAGE “does not change the fact that the primary intent of the study was to answer scientific questions of importance to primary care physicians.”
What is wrong with a seeding trial? The Declaration of Helsinki (the touchstone of research ethics) states that “medical research involving human subjects should only be conducted if the importance of the objective outweighs the inherent risks and burdens to the subject” and “medical research is only justified if there is a reasonable likelihood that the populations in which the research is carried out stand to benefit from the results of the research.” It also states that “each potential subject must be adequately informed of the aims, methods, sources of funding, any possible conflicts of interest, institutional affiliations of the researcher, the anticipated benefits and potential risks of the study and the discomfort it may entail.”
So a study must address a genuine, unanswered, clinically important question and have a good chance of answering it. It is unethical to expose people to risk solely or primarily to help the marketing of a drug, and doubly so if potential subjects are not informed of the true reasons behind the study. In addition, participation in any trial and the experience gained with the drug, the kudos that comes from being called “an opinion leader” and being given the status of a “clinical investigator” may distort the prescriber’s perceptions of the drug in particular and the company in general. Other authors have noted that “when a gift or gesture of any size is bestowed, it imposes on the recipient a sense of indebtedness. The obligation to directly reciprocate, whether or not the recipient is conscious of it, tends to influence behaviour. . . . Food, flattery and friendship are all powerful tools of persuasion, particular when combined”.
But many studies probably arise from a mix of motives. As the Merck official writes in his rapid response, “a scientifically sound, rigorously conducted study that shows the benefits of a drug for patients and doctors would also benefit the company that produced the drug”. Clinical trials are expensive, and pharmaceutical companies are businesses not charities: a company sponsoring any trial would be foolish if it did not consider the potential effect on its sales. It was only the court-required access to internal documents that enabled the alleged true motivation for ADVANTAGE to be recognised. It is not possible to say how widely seeding trials are used.
The accompanying editorial suggests that the important consideration for potential investigators and subjects, and research ethics committees, is whether or not the proposed study addresses an already settled question. We suggest there are others as well, including the importance of the question being addressed and the likely ability of the study to answer it.
# The article, accompanying editorial and references cited by them suggest that possible indicators of a seeding trial may include:
- a lack of a control group
- an open label/unblinded design
- a very large projected enrolment relative to the importance of the question, with a large number of centres each recruiting only a small number of subjects
- a study with disease-oriented outcomes (i.e. surrogate markers) instead of patient-oriented outcomes (i.e. looking at whether the drug helps people live longer and/or better), especially in already-marketed drugs
- a short-term study in a chronic disease
- a trial not aimed at gaining a new licensed indication
- a study which does not address important clinical questions or addresses questions which are being substantially addressed in studies already started
- a study where there is a long delay between its completion and its publication
None of these is specific: there may be good reasons why a study has one or more of these characteristics and none would necessarily invalidate a study scientifically. Equally, a seeding study might have only a few of them. But their presence should perhaps raise the question of the true motives behind the study.