Does the pharmaceutical industry have too much control over clinical trials?

November 15, 2002

Fri, 15 Nov 2002 22:00:00

New York, NY - Selective presentation of data, pressure on investigators to tow the company line, suppression of negative datathese are some of the claims made by clinical trial investigators against the pharmaceutical industry. With the companies trying to control clinical trials to an ever-greater degree, are cardiologists losing sight of the science? And are investigators themselves doing enough to ensure that the true findings really are reported?

Dr Nancy Olivieri (courtesy of Iron Disorders Institute)

Many cardiologists will be familiar with the story of Canadian company Apotex and clinical investigator Dr Nancy Olivieri, which has recently been the subject of an editorial in the New England Journal of Medicine. In short, Olivieri became concerned about possible serious adverse events with Apotex's drug during a clinical trial and wanted to go public with her findings, but in an effort to stop her, the company terminated her involvement in all its clinical trials and filed 2 lawsuits against her for breach of confidentiality. Olivieri stood up to the company and released her data, but in the process she was put through years of angst. Even worse, rather than support her, the hospital that she worked forwhich was expecting a large donation from Apotexaccused her of research misconduct, for which she was subsequently completely vindicated.

How many investigators would stand up to sponsors of trials in the way she did? If companies come across too much opposition from the investigators in their trials, they may well decide to go with a different group. And as attracting funding for clinical research is more and more difficult, investigators are finding themselves under pressure to keep the sponsors happy.

The Apotex-Olivieri case is a dramatic scenario, but subtler influences of industry over clinical research are much more common. Among these are the ways new data are presented.

Here's the outcome, see you later

Everyone knows the routine. The latest megatrial is to be reported at the American College of Cardiology or American Heart Association meetings. The hall is packed. Everyone is waiting with bated breath to hear whether drug x worked in condition y. The chief investigator spends 15 minutes reporting the resultsto relate what could be the most important trial in the field for years. Although procedures have improved at the recent cardiology meetings, with the introduction of a discussion from another leader in the field following immediately after each hotline presentation, the whole presentation is still limited to just a few minutes, which is hardly enough to do justice to important new clinical results.


The chief investigator spends 15 minutes reporting the resultsto relate what could be the most important trial in the field for years.


When the session is over, the presenters generally rush offto other commitments or even to the airport, as these trials are often presented at the end of major meetings. The press releases have already been written, the strategy prepared. It is more like a political exercise than a scientific discussion.

It is impossible to truly understand the full results and implications of a major new trial under such circumstances. But this is the normal way in which the data are released, and quite often no more information will become available before the paper is published, which can take months or, in some cases, even yearsor, for negative trials, sometimes not at all.

Even when the final paper is published, how is it ensured that the correct data are shown and the right conclusions reached? Who decided exactly how the information should be presented and how the conclusions should be worded? The academic investigators, it is assumed, but this is not always the case. More and more, it appears, the pharmaceutical company sponsoring the trial has an influence, and this may distort the true picture. How are practitioners to know if this is the case or not and how much influence the company has had?

In short, they don't. It is hoped that the chief investigators are telling the truth, and trust is put in them and their academic groups. But what remains unknown is how much their arms have been twisted or how much pressure has been put on the groups to give out a particular message.


It is hoped that the chief investigators are telling the truth and trust is put in them and their academic groups.


Several years ago at the European Society of Cardiology, the final results of a major cardiology study were presented by someone other than the chief investigator, even though he was attending the meeting. At first glance, the results seemed positive, but on further investigation it became apparent that the drug involved was not going to have a healthy future as it had a serious side effect that would prevent its widespread use. This side effect and its implications were played down at the presentation. Afterward, the chief investigator unofficially revealed that he had not presented the results because the sponsor had insisted that mention of the side effect be minimized, and he had refused to do this. This person had reached a position in his career where he could do this without too many adverse consequences, but less-well-known investigators who need to raise their clinical research profile for career progression are in a much more difficult situation.

There are many other instances of pharmaceutical companies pressuring clinical investigators, but the doctors involved are very reluctant to go on record about such matters for fear of losing research funds, or even worse, becoming the victim of a campaign to discredit them, as happened to Olivieri. Thus, these practices are allowed to continue.

In some cases even the chief investigator and steering committee of the trial may be working in the dark, as they have not always had access to the raw data from the trial to conduct their own analyses. Some companies like to analyze the data from their trials themselves and are reluctant to allow anyone else, including the investigators, access to these analyses or to the raw data. This creates an atmosphere of mistrust.

Journals set new rules on disclosure

The situation has been addressed recently by the major journals, which last year together published new rules on the disclosure of industry's roles in clinical trials. These rules state that from now on, authors must disclose full details of their own and the sponsor's role in studies. Moreover, some journals are insisting that the responsible author sign a statement indicating that he or she accepts full responsibility for the conduct of the trial, has had full access to the data, and controlled the decision to publish.

This should improve matters regarding publication of trials, but there are still few controls over presentation of data at meetings, and even at the publication level, companies can still pressure individual groups of investigators to write the paper in a certain way.

Dr Steven LaRosa

One clinical investigator who takes a particularly dismal view of industry's role in clinical research is Dr Steven LaRosa, an infectious disease doctor at the Cleveland Clinic. He believes the industry is harming clinical research by taking too much control. "Some companies are worse than others. The smaller ones tend to be more open to outside analysis, but the majority of the big pharmaceutical companies want to keep complete control, themselves," LaRosa says. "If we as physicians want to use the data to ask another question it depends on whether the company deems this okay or not. If they say no, we lose the opportunity to answer important scientific questions."

He also believes that many clinical investigators contribute to the problem by not standing up to the companies enough. "Sponsors will not use investigators whose opinions differ from theirs. Pharmaceutical companies surround themselves with consultants who tell them what they want to hear. Academia and industry are much the same. It's an old boys' club and exists in every specialty. It's a big problem. It's getting to the point where you can't really believe anything anybody who is involved with a trial says because it is likely they are reflecting the sponsor's views. It's not lies, it's just not hearing the whole story. But if we haven't got the data we can't do anything about it."


Sponsors will not use investigators whose opinions differ from theirs.


LaRosa also takes issue with the amount of money that investigators are paid. "The amounts involved are quite shocking and this contributes to the problem. Everyone accepts that the investigators need to be reimbursed for their time, but people don't realize how much is involved. Money for clinical trial work goes to the hospital, but consulting fees go straight into the physician's pocket. The bigger the name, the higher the fees. That is bound to influence them. They may not lie, but they won't be as forthright as they would be if they were not receiving these fees. At a consulting meeting, if there is 1 dissenting view, that person probably won't be asked back again. It would help if physicians were not allowed to take consulting money; this happens in the NIH."

Are things really this bad?

LaRosa certainly paints a dreary picture. Are things really that bad? A few leading cardiologists, asked what they thought, offered divided opinions. Some said there was no problem and everything was completely above board, but they were in the minority. Most had some concerns about too much industry control, with many complaining that their academic freedom is being taken away or at least is under threat. And while the new rules from the top journals are a good start, most believe there is still a long way to go to.

Two cardiologists who have been involved in many industry-sponsored trialsDrs Eric Topol (Cleveland Clinic) and Paul Armstrong (University of Alberta, Edmonton)both say the situation has worsened in recent years, with the consolidation in the industry creating larger companies that are more difficult to deal with.

Dr Eric Topol

Freedom of thought under threat

Topol says he finds LaRosa's comments frightening. "I don't know that it is that bad in cardiology, but there is certainly cause for concern. My main concern is that the ability of academics to think on their own is being suppressed. The ability for them to have autonomy is far from optimal. The pendulum is swinging in the wrong direction. During the late 1980s and early 1990s it was much easier to have full autonomy. Now, the ability to design, conduct, and present trials exactly how we like is getting more difficult."

Dr Paul Armstrong

Armstrong concurs: "It is an issue on everyone's minds. The parameters have definitely changed. The freedom we had 10 years ago is much less prominent now. Competition between companies is fierce and has been made worse by mergersthere is much more emphasis on the bottom line than ever before."

He points out that there is often also a control struggle within each company between the marketing and research departments, but it is essential to maintain freedom of thought and publication.


We academics want total freedom, but in many respects that is unrealistic for companies that are laying out millions of dollars to fund a trial.


"There are outstanding individuals in the R&D departments of companies that have often switched over from academia, but often the marketing budget controls clinical research. We academics want total freedom, but in many respects that is unrealistic for companies that are laying out millions of dollars to fund a trial. The situation has become worse as cost of trials has risen. But whatever happens we still expect to have freedom of thought."

One clinical investigator who cannot be chastised for bowing to industry pressure is Dr Curt Furberg (Wake Forest University), who several years ago famously took a stand against the calcium blockers, which included 1 of the best-selling drugs in the world at the timeamlodipine (Norvasc® - Pfizer).

Dr Curt Furberg

He agrees that industry has too much control over clinical research and believes this situation has come about for several reasons. "There are not many breakthrough drugs in the pipeline at present, and for most medical conditions, there are multiple choices, which means companies have to try harder to make their drug look attractive. These conditions, coupled with public pressures on cost, all have an impact on their bottom line. This is an unhealthy situation in general and lies behind the actions they are taking. They have to survive so perhaps they are doing things they shouldn't."

Furberg notes that companies used to do 10 trials and then just picked the 2 they liked the best to submit to the regulatory authorities. Then this was stopped, and the agencies demanded to see all the data. "The companies then needed to have more control, and the whole issue of academic freedom hit them. So to avoid this they have taken 2 routesthe use of clinical research organizations (CROs) and developing countries."


Almost every trial conducted in a developing country has a positive finding.


Furberg says neither of these options is helping matters: "CROs bypass the issue of academic freedom altogether, as the CRO wants to please the company it is working for and so is no safeguard whatsoever. And developing countries have no money, so an industry dollar goes a long way there. Investigators are very anxious to please. Almost every trial conducted in a developing country has a positive finding. So the companies will keep doing that as they get what they want."

Spinning results

Furberg, Armstrong, and Topol all say that "spinning" results of trials unquestionably occurs. Armstrong notes that sponsored symposia are the worst culprit in this regard but that it also happens in late-breaking sessions where important new results are released in just a few minutes. He believes these late-breaking sessions should be changed significantly, with longer talks followed by a question-and-answer session. "Fifteen minutes is no good. This just encourages spinning, and there is no time to ask questions. We need invited discussion groups led by people with impeccable credentials who have no relationship to the sponsor and who represent both sides of the argument if there is one."


Fifteen minutes is no good. This just encourages spinning, and there is no time to ask questions.


Furberg adds that sometimes the "spinning" begins even before the trial does, with the company deciding on a "less-than-optimum choice" for the comparator or control drug, maybe at a nonoptimal dose. "There should be an organization that approves trials before they are conductedrecommends design changes to ensure the right trial is done to answer a necessary question," he suggests. Suppression of results that don't come out right is another bad practice, Furberg says, giving the example of the PRAISE-2 trial of amlodipine in nonischemic heart failure, which has never been published. "This showed absolutely no effect. It was reported at the ACC in March 2000, but it is still not published."

Will the new journal rules solve the problem?

While agreeing that the new rules from the journals should improve things, Armstrong suggests that more papers should be accompanied by editorials, which he describes as "beacons of light in the search for the truth." However, he points out that even the journals are not completely independent, as they are usually owned by a commercial enterprise that makes its money from the pharmaceutical industry. Referring to the case a few years ago when the editor of the New England Journal of Medicine was removed, he says that editors of journals have very difficult jobs, striving for independence in a commercial world.

Topol says the new recommendations from the journals "is the most important thing to have happened in recent years," but he warns that the companies are looking for ways to circumvent the journals' requests.

Control of the database

Access to the data seems to be 1 of the most problematic areas. Some companies like to analyze the data themselves and have been reluctant to let even the study investigators have access to the raw numbers. But this is no longer allowed if the trial is to be published in a leading journal. Under the new rules, journal editors are now checking whether investigators had full access to the data. Topol explains: "The journals have stated that access to data must be full and assured. The companies may now be forced to comply with these guidelines. The situation could therefore be 'rehabilitated.' But the companies are trying not to comply with this. They want the data held by a clinical research company. This is obviously not what was intended. They are trying to think of ways to stop free access to the data."

"They are frightened that someone will use their data in a away unfavorable to them. Indeed, there have been cases of this in the past," Topol continues. But most sponsors still demand that the manuscript be sent to them to be cleared before being submitted to a journal. "This is very difficult for us to live with. Very often our opinion differs from theirs. Bowing to the wishes of the sponsor is not necessarily in the best interest of the medical community. In many of our trials we have gotten harsh words from companies for not doing this."


They [the pharmaceutical companies] are frightened that someone will use their data in a away unfavorable to them.


Armstrong says the analysis of data is very variable. A good scenario, and 1 practiced by the VIGOUR group to which he belongs, is where the academic group collects and holds the data but the company has access to it. "In this case the investigators and the company generally analyze the data together." However, he also believes that academic groups must be allowed to discuss the results without the sponsor being present, and this is where disagreements may occur. "If someone is putting up $50 million they will want some access to the data."

He says that many companies want control of the database. "They are afraid to lose control. They are frightened that by letting other people loose with the data, it may be used in a way that will not be good for the bottom line of the company. But if a company controls the database, it will generate skepticism. When presented at a meeting or published in a journal there should be a statement about transparency around analysis of the data. Also the exact protocol should always be published so that everyone knows that the study did what it said it would do."

Rent-a docs

Topol and Armstrong also agree that opinion leaders are not doing enough. Many are happy to be "rent-a-docs," Topol says. "If industry has a hard time getting them to 'say the right thing' they can always find someone else who will. There are not very many who will stand up to the companies." He adds, however, that money is not the driving factor in this situation but rather academic status. "It all depends how much you are willing to compromise."


There are some opinion leaders who behave more like sales reps, being paid by a company to spread a certain message around various meetings and institutions.


Armstrong says that clinical investigators must be careful not to have too many ties with just 1 company. "There are some opinion leaders who behave more like sales reps, being paid by a company to spread a certain message around various meetings and institutions. These doctors always give the company linethey are very influential and very persuasive, very impressivebut in actual fact they are little more than traveling salespeople. Most of the other opinion leaders know who they are, but often the practicing cardiologist may not." He believes the academic center hosting a meeting should check up on this more than they do. Armstrong adds that it helps if the investigators conduct trials for many companies. "Then we have multiple conflicts of interest. If these are all properly declared people will then see that we are being honest."

Furberg admits that few doctors would stand up to the companies the way he did. "Most academics need company money to perform clinical research. I am lucky in that I have been successful in securing independent funding and so have been able to take a stance against companies on issues I feel strongly about. But many investigators don't have enough background work behind them. If you have already made itthen maybe you can afford to say what you think. But on the way up very few people would risk their careers."

Government funding best but difficult to attain

All agree that the best model is when funds for clinical trials are provided from the government, but such funding is scarce. The NIH is "very difficult to get money out of," Topol notes. "I have recently had an application turned down from the NIH for an inflammation trialhow many years will we have to wait for this to be done now?" He suggests that a good way forward may be an independent clinical trial entity run by the government, toward which companies contribute funds. "The more distance they keep, the more authentic the findings will seem. But most companies don't seem to realize this."


The more distance [the companies] keep, the more authentic the findings will seem.


Government funding of trials in Canada is similarly thin on the ground. Armstrong notes that: "In Canada we should be spending 1% of the healthcare dollars on research, but we have a healthcare budget of well over $100 billion and only $750 million gets spent on research. This is a long way from the ideal."

Furberg has been more successful than most in finding government funds for his trials, which has allowed him to be so outspoken. At present, none of his funding comes from industryhe receives support from the NIH and his academic institution. "The last study I did with industry funds was HERS with HRT, and we all know what happened to that. The company was not happy with the findings and the way they were reported. The relationship between the company and the investigators changed after we got the results."

He says he is still contacted by companies to do research, but as chair of the large government-funded ALLHAT hypertension trial he is not allowed to accept much industry money. "As chairman of ALLHAT, I have to declare all extra income from commercial sources, and if this exceeds a certain amount I have to take my name off the paper. This is a solutionto put a limit on how much 1 researcher can take from a drug company."

Private practice trials

Everybody interviewed for this article had serious concerns about research conducted in private practice. "Research from private practice is very dodgy," Armstrong says. "The money may go straight into the doctors' pockets, and money earmarked for research funds often becomes doctors' bonuses. This should not be allowed to happen without appropriate oversight. The situation in academic institutions is better as research funds can be spent only within the department or faculty with more transparency."


Money earmarked for research funds often becomes doctors' bonuses.


Topol was more outspoken on this issue: "Private-practice trials are a despicable situation. Fee-per-patient funding is intended to pay for clinical work, but in certain private practices it can be pocketed by the investigator, leading to another hit to clinical research. There is already a public mistrust, but this sort of thing could take down clinical trials as we know it. It is unacceptable and a completely unethical exploitation of patients."

FDA saves the day

But there does seem to be a rock in the stormy seas of clinical researchthe US FDA. At the end of the day, the clinical trials for a new drug must be approved by this agency or the drug cannot be marketed, and the FDA is not going to be swayed by company influences. Armstrong says the FDA is the "real saving grace" in the clinical research process. "You can't get a drug through the FDA on false credentials. The agency has extraordinary access to data, unlike the journals or many investigators, and it is a truly independent organization."


You can't get a drug through the FDA on false credentials.


Furberg agrees: "One thing that we have gotten right is the regulatory control. The FDA will not allow through a drug based on dodgy data." But he adds that not all studies are conducted for FDA approval, and companies now regularly perform phase 4 studies for marketing purposes once a drug is already available. "These do not have to get through the FDA, so the 1 real safeguard against data manipulation has been removed."

Some hope on horizon?

Despite all their negative comments, these 3 researchers are still continuing to conduct trials and are finding ways to try to conduct the trials they want and report the findings honestly.

Armstrong says he is not ready to give up yet. "There are still some far-sighted companies and I'm not ready to throw in the towel yet, but it is tough and we don't know how it will end up."

Topol is leading a new trial of clopidogrel in 15000 patients. "We are calling the shotswe have told the companies exactly what we want to do and how we want to do it. It can be done. But I've had so many tough battles and it whittles you down. We have to go through this repetitively. It is very tiring and overall I'm quite discouraged."

Furberg says he has no regrets that he spoke out against the calcium blockers. "It is something that I felt strongly about, and it hasn't harmed my career." On the contrary, he is now chair of ALLHAT and sits on an FDA committee on drug safety. "I am in good position. I only wish I had challenged the status quo earlier in my career," he comments.

Related link

1. [Education > Heartbeat; Oct 23, 2002]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: