Should drug companies publicly release all data from the tests they conduct on human volunteers? A growing number of parties say yes, and last week, a group of academics unveiled a scorecard to spotlight companies that have recently kept their clinical data hidden from public view—and those that have made the data available.
As with all shorthand systems that boil down complicated scenarios, the clinical trial scorecard, developed at New York University, Harvard University, and Yale University, requires explanation.
I’ll get to that. But what struck me like a bolt was the organizing principle of the report’s authors. It was simple, perhaps even unassailable: In exchange for the privilege of testing for-profit drugs on human subjects, companies have an obligation to make all their clinical data available for the common good. Science and medicine should work best when research can be examined, criticized, and serve as a foundation for more studies.
But when drug companies hold back results, positive or negative, there is no common foundation to build upon. “How can we practice evidence-based medicine if we don’t have all the evidence?” said Jennifer Miller, the report’s lead author and an assistant professor in the division of medical ethics at New York University School of Medicine.
Miller, who said she has been working on the project for six years, expects her scorecard to become an annual tally, but last week’s release in the journal BMJ Open started with a small sample size. She and her team assessed 318 trials used to test 15 drugs, all approved by the FDA in 2012. This is a subset of all new drugs approved that year; the authors decided to investigate only drugs brought to market by large companies. With their size and financial muscle to put behind the transparency requirements, “if anyone would score well, it would be [them],” said Miller.
Not all scored well. Before I get to the names, it’s important to describe the scorecard. It has two parts. The first part measures how companies complied with FDA regulations, implemented in 2007, which require some disclosure of data that leads to a drug approval. But those regulations are a low bar, said Miller—“a minimal set of standards” that only require a fraction of the body of evidence to be disclosed.
Instead, what Miller and colleagues see as the gold standard of clinical disclosure is the second part of the scorecard, which tallies the percentage of trial results that are either reported in summary or published in full—a much broader set of data than required by FDA. In other words, by emphasizing this part of the scorecard, Miller and colleagues want drug companies to follow the spirit of the idea of data for the common good, not just the narrow letter of the 2007 law. That said, this is not just an academic idea: Miller noted that the FDA’s parent department, Health and Human Services, has ethical language in its “common rule,” which in turn was based on a 1979 report on the ethics of research with human subjects.
Some companies fared quite well. Under the scorecard’s ethical gold standard, three companies published all trials—perfect scores—that led to their 2012 drug approvals: Pfizer, for the drug axitinib (Inlyta, approved for renal cell carcinoma); the Janssen division of Johnson & Johnson for bedaquiline (Sirturo, for tuberculosis), and GlaxoSmithKline, for an infants’ vaccine branded MenHibrix.
At the bottom of the list, the three worst scores belong to Gilead Sciences, for its work on the HIV four-drug combination sold as Stribild (only 21 percent of trial results are publicly available); Sanofi, for the multiple sclerosis drug alemtuzumab (Aubagio, only 22 percent of results available); Sanofi again, for the colorectal cancer drug ziv-aflibercept (Zaltrap, 40 percent); and Pfizer and its development partner Protalix, for the Gaucher disease treatment, taliglucerase alfa (Elelyso, also 40 percent).
A Gilead spokesman said the company had no comment. A Sanofi spokeswoman responded that the company looks forward to reviewing the study, and that it complies with the drug industry’s own guidelines for sharing clinical trial data. And a Pfizer spokesman said Elelyso data disclosure was the responsibility of Protalix.
The Pfizer spokesman also noted the high marks Pfizer received not just for axitinib but for trials related to two other drugs approved in 2012. He said Pfizer has taken multiple steps in what he called “a longstanding commitment to clinical transparency.” For example, Pfizer is working on several data-sharing projects and collaborative clinical trials.
Bayer, which made public only 42 percent of trial information related to its colorectal cancer drug regorafenib (Stivarga), sent a statement via a spokeswoman that said the company makes data from its trials available on its website and on the trial registry clinicaltrials.gov “regardless of the outcome of the study.”
Nearly every company pays lip service to a commitment to transparency. (Search on a company’s name and “clinical data transparency” to find their webpages.)
But which ones have gone beyond lip service? Miller went out of her way to praise Pfizer and GSK. “They’ve cleaned up their act,” she said. “They’ve gone above and beyond the legal requirements.”
For GSK, that “act” once included the cover-up of negative data around the antidepressant paroxetine (Paxil). It was one of GSK’s predecessor companies, SmithKline Beecham, which had paroxetine approved for adults in 1992 and continued to study it as a potential treatment for teenagers.
That study, dubbed Study 329, was originally positive and “influential in the literature supporting the use of antidepressants in adolescents,” wrote researchers who revisited the notorious study—long the target of retraction efforts and lawsuits. Earlier this year, those crunching the 329 data again published a very different story.
Not only was there no benefit compared to placebo, but also, they wrote, “there were clinically significant increases in harms, including suicidal ideation and behaviour and other serious adverse events in the paroxetine group and cardiovascular problems in the imipramine group.” (Imipramine is an even older type of antidepressant that was used as a comparator in the trial.)
Miller called the company’s behavior a “cover-up” that increased the risk of suicide in teenagers. GSK has already paid for its sins, including $3 billion in fines in 2012, and is now a vocal proponent of data disclosure.
Might the new scorecard prompt other companies to shed more sunshine on their trial data? Miller hopes so, but she stressed that giving negative ratings “aren’t about shaming bad practices,” said Miller. The scorecard is “about providing opportunities and letting every company understand how they’re performing related to ethical and legal standards.”
She also acknowledged that the 2007 FDAAA rules for disclosure—the legal letter of the law—contain unclear language, which could be contributing to some of the low scoring.
In the end, will commitments to transparency help anything beyond a company’s PR image? If a paroxetine-like situation can be avoided, the answer is clearly yes. And less dramatically, the slow but steady accumulation of information about a drug could lead to new ideas and new uses for it.
When I asked Miller where else transparency has helped, she mentioned restaurant inspection ratings leading to less foodborne illness outbreaks. It’s unclear whether a comprehensive study has been conducted, but there are reports here and there of local successes. An oft-cited 2005 study showed that Los Angeles County, which includes the city of Los Angeles and many other towns, had a decrease of foodborne illness in the first few years after it began its program.
In Toronto, officials credit their city’s inspection system for a 30 percent drop in cases of foodborne illness, according to the Canadian Broadcasting Corp.
Miller’s transparency report is part of a larger effort to hold the drug industry accountable for practices that make it a nearly perennial loser in the court of public opinion.
For the 15 approved drugs studied, there were nearly 100,000 participants involved. Nearly half the drugs had at least one undisclosed Phase 2 or Phase 3 study. In following years, the researchers hope to expand the scorecard beyond the biggest drug companies, a wise move considering drug programs are increasingly targeting smaller groups of patients, often with rare diseases, which could let companies with less financial clout afford to take drugs all the way through to market.
So entrepreneurs, scientists, and investors in the biotech space, take heed. Clinical transparency isn’t just a Big Pharma issue. As the movement gains momentum, there will be fewer excuses to hide behind trade secrets or competitive advantage. The world will want to know your results, and how you achieved them.