Friday, May 29, 2009

More Documents About the Selling of Seroquel Show How Research Was "Subordinated to Commercial Goals"

We posted earlier this year about how documents produced on discovery and recently unsealed during litigation suggested how AstraZeneca handled clinical research data in the marketing of its atypical anti-psychotic drug Seroquel (quetiapine). A new crop of documents has just been released, providing yet more insights, as reported by the St Petersburg (Florida) Times:

Behind the scenes at the global pharmaceutical company AstraZeneca, the team in charge of the blockbuster antipsychotic Seroquel had one mission: make the multibillion-dollar seller even bigger.

To that end, internal company documents released Wednesday show how the British drugmaker hid unfavorable study results, promoted unapproved uses and even considered pitching the drug as less likely to lead to suicidal thinking than competitors'.

In particular,

Documents suggest Seroquel studies were repeatedly subordinated to commercial goals.

When a study in 2002 failed to show that Seroquel's sustained release formula was any more effective than a placebo in treating schizophrenia, orders from the top were to keep the results 'in strictest confidence.'

When a scientist in England wanted to study the weight gain in rats on Seroquel, AstraZeneca declined to fund the research, saying 'we could wind up with results that are not clearly advantageous.'

In one discussion, it was suggested that authors of potentially helpful research reports who raised too many questions, slowing publication, should be asked to step down.

A researcher who pressed for results of an unfavorable trial was rebuffed for weeks before being given 'three or four sentences describing high-level results.'

AstraZeneca marketers were jealous of what they saw as competitor Lilly's ability to cast questionable study results in a positive light. 'They (Lilly) are able to spin the same data in many different ways through an effective publications team,' according to a 2003 memo. 'Negative data usually remains well hidden.'

As Seroquel's sales soared, documents reflect an ongoing struggle between the safety and marketing teams over the potentially damaging issue of weight gain. In 2000, the company's scientists said data did not support the marketing claim that Seroquel resulted in only 'limited' weight gain. Close to 23 percent of the people who took the drug gained more than 7 percent.

Despite the safety team's objections, the word 'limited' remained on Seroquel's label for two more years.


This is a reminder how beleaguered we advocates of evidence-based health care (EBHC) have become. The idea of EBHC was that health care decisions for individual patients, and policies for groups of patients ought to be guided by critical review of the best available evidence from clinical research, guided by knowledge of biology and the biopsychosocial context of health, and informed by patients' values and preferences. The idea still makes sense to me, but it only works if physicians, patients and policy-makers have access to an unbiased sample of clinical research studies, so that studies with are not selectively suppressed to support vested interests. Although critical review can account for inevitable trade-offs, compromises, and errors in how studies are designed, implemented, and analyzed, the clinical epidemiological methods it uses are really not designed to root out falsehoods and deliberate deception.

However, the ongoing story of Seroquel, and many other cases discussed on Health Care Renewal suggest that when clinical research is sponsored by those who can profit from the product or service it evaluates, that research is prone to suppression and manipulation. Although I believe there are many honest scientists who work for pharmaceutical, biotechnology, device, health information technology and other health care corporations, it seems they often have to answer to marketers whose only goal is to sell more product.

As long as clinical research is sponsored and run by the people who can profit directly from selling the products and services the research is meant to evaluate, the ideal of evidence-based health care becomes less attainable.

As we have said before, suppression and manipulation of research amounts to post-hoc abuse of research subjects who volunteered their participation believing that it would advance science and health care.

Furthermore, suppression and manipulation of research can deceive physicians into prescribing tests and treatments that will fail to help, or even harm patients, and deceive patients into thinking that they are getting the best possible tests and treatments, when, again what they are getting is ineffective or even harmful.

In my humble opinion, there is an increasingly strong argument that clinical research should not be controlled, and probably should not be done at all by organizations with vested interests in the research producing results favorable to their products.

2 comments:

Lee Green MD MPH said...

Sigh.

When I started teaching EBM, some 15 or more years ago now, the challenge was critically appraising studies. I could teach our residents good analytic skills and those skills would allow them to identify the information that should change their practices. I have no idea how to teach my residents now, in a world where the challenge is deliberate manipulation of the evidence base by aggressive and very sophisticated companies to serve their commercial interests. "Caveat emptor" just doesn't suffice.

Anonymous said...

From the Dayton Business Journal May 28 we learn that the State of Ohio has recovered $8.4M for overcharging Medicaid.

The companies involved were:
Warrick $4.3M
Roxane $2.4M
Abbot $1.3M
Pharmacia $400K

While not even a rounding error for these companies I feel it shows a mindset that is constantly not only bending, but breaking the rules, in pursuit of profits. While profits are a good thing, the drive to achieve higher profits while engaging in illegal activity, needs a more definitive punishment.

Pay the fine, walk away, is not working.

Steve Lucas