In 2016, Joe Biden, then vice president, launched the Cancer Moonshot Initiative, a project to identify new approaches to prevent and treat cancer. That year, he learned of the glaring failures of the National Institutes of Health and its grantees in managing the outcomes of cancer experiments, and he told an audience at a cancer research conference that he was scandalized.
Researchers using federal funds to conduct cancer trials – experiments involving drugs or medical devices that rely on volunteer subjects – sometimes took more than a year to report their results to the NIH, as needed. “If you don’t report, the law says you shouldn’t get funding,” he said, citing a survey I published in Stat with my colleague Talia Bronshtein. “Doc, I’ll find out if it’s true, and if it’s true, I’m going to cut the funding.” It is a promise.
It was true then. It’s true now. More than 150 trials completed since 2017 by the National Cancer Institute at NIH, which is leading the $ 1.8 billion Moonshot effort, are expected to have reported results so far. About two-thirds reported past their deadline or not at all, according to a University of Oxford website that tracks clinical trials regulated by the Food and Drug Administration and the National Institutes of Health. Some test results have been expected for nearly two years. Overall, government-sponsored scientists have complied less than half the time for expected test results since 2018. (An NIH spokeswoman said, “We are prepared to take all steps to ensure compliance with ClinicalTrials.gov results reporting. ”)
Congress began requiring researchers to report the results in 2007, after drug companies were found to have injured patients by withholding data that showed the lucrative drugs were dangerous or ineffective. GlaxoSmithKline is known to have suppressed data from trials suggesting that the antidepressant Paxil was ineffective and increased suicidal ideation in adolescents. (In 2012, the company agreed to plead guilty to criminal charges related to fraud and failure to disclose security data and pay $ 3 billion in penalties.) The law’s reporting requirements were clarified in 2017. Trial sponsors must now report results within one year of collecting the most critical data.
The National Institutes of Health operate ClinicalTrials.gov, but many projects the agency leads or funds do not report results for years past their deadline. These include drug trials for advanced cancers of the pancreas, lungs, brain and gastrointestinal tract. The breaches dishonor the trial volunteers, many of whom almost certainly died without knowing whether their sacrifices benefited anyone except the beneficiaries. With dozens of federally funded trials of Covid-19 treatments and vaccines underway, the stakes – and ethical imperatives – of transparency have increased.
Physicians, researchers and patients rely on accessible and clear data that is only provided by ClinicalTrials.gov. In 2016, Francis Collins, director of the National Institutes of Health, announced that the agency would begin penalizing researchers for not complying with its reporting obligations. “We’re serious about this,” he said at the time. Yet in the years that followed, neither the FDA nor the NIH enforced the law. A “wall of shame” that Dr. Collins pledged to create to name offenders never materialized.
The app’s failure comes at a cost: Oxford researchers calculated that the FDA could have penalized trial sponsors $ 18 billion for reporting failures since 2018. It has yet to collect the first dollar. The National Institutes of Health are allowed to withhold funds from recipients, as Mr Biden pledged to do five years ago – but it never has.
Bad press forced some improvement. Many pharma giants, including AstraZeneca, have recently followed the rules, although reporting the results is hardly enough for transparency – as evidenced by the company’s recent coronavirus vaccine data debacle, in which Independent data controllers have criticized the company for overstating the effectiveness of its vaccine.
With researchers participating in government-funded trials setting a bad example by violating requirements most of the time, it’s no wonder institutions treat the law as mere suggestion. Even some prestigious and well-funded research organizations admit that they do not always meet the reporting requirements. Among them, the MD Anderson Cancer Center at the University of Texas and the Dana-Farber Cancer Institute told me they were working to improve their adherence. The Massachusetts General and Brigham and Women’s Hospitals have said they are committed to reporting in a timely manner.
Some researchers complain about confusing and onerous reporting requirements, although the 2017 revision of the rule significantly reduced these issues. Some say that publication in scholarly journals should suffice. But these publications often require paid subscriptions, which limits the number of people who can read them, and these reports sometimes truncate key data, including side effects. And for thousands of trials, nothing is published at all.
Improved reporting by some large pharmaceutical companies and major universities and nonprofits – including Memorial Sloan Kettering Cancer Center and Duke and Johns Hopkins universities – also deny the specious accusations by offenders. Many institutions are simply not making reporting a priority. The recalcitation by others suggests a lingering impulse to distract from data that poorly reflects experimental treatments – the insidious problem ClinicalTrials.gov was supposed to solve. Scientific progress demands openness, even when it’s bad for business or for the next grant proposal.
At his first presidential press conference, Biden pledged to boost medical research. Adequate access to this important work now depends on the president’s ability to keep his promise about ClinicalTrials.gov.
Charles Piller (@pill) is an investigative journalist for Science. He has reported extensively on medical research, public health and infectious diseases.