Nick DeVito and Ben Goldacre
The US FDA Amendments Act (FDAAA 2007) requires certain clinical trials to report their results onto ClinicalTrials.gov within one year of completion. Our FDAAA TrialsTracker shows all individual trials that breach this legal requirement. Once a week, we write about one unreported clinical trial: you can read more background here and read past entries here.
This week’s unreported trial is titled “A phase 2 safety study in which ischemic stroke patients will be randomized within 24 hours of symptom onset to placebo or oral lovastatin 640 mg per day for 3 days” (NCT01976936). This trial enrolled 164 adult ischemic stroke patients. The study was randomized and was masked for participants, care providers, and investigators. The experimental intervention was high-dose lovastatin (640 mg) for 3 days following stroke; the active comparator was low-dose lovastatin (80 mg) for 3 days following stoke; the control group was given placebo. The primary outcome was Liver Function Tests with safety outcomes examining liver toxicity or evidence of rhabdomyolysis; the secondary outcomes were score on NIH Stroke Scale, Barthel Index Score, and Modified Rankin Scores.
Recent estimates of the global burden of disease show that stroke is the second most common cause of death and third most common cause of morbidity (measured by DALYs) worldwide. Statins are one of the most widely prescribed classes of medicine worldwide.
We intend that this series should occasionally shed light on interesting issues around transparency rules, and how registry data are used. You can read some general background about the FDA Amendments Act 2007—and why a trial is considered “due”—here and here.
Searching on our FDAAA TrialsTracker shows that this trial was “reported late” in March. Checking the ClincialTrials.gov record shows that results have been submitted. So why is this trial being featured this week? As the ClinicalTrials.gov record shows: although results have been submitted, they still have not yet been posted, as they are “stuck” in quality control.
When results are submitted to ClinicalTrials.gov, they undergo quality control review by staff at the National Library of Medicine. The staff check for issues or inconsistencies with the submitted results and return them to the sponsor if there are any concerns. This usually happens within roughly a month of submission. The “Results Submitted” tab on ClinicalTrials.gov shows that results were submitted to clinicaltrials.gov for this trial on 29 March 2018, and then returned to the sponsor with queries on 30 April 2018.
There is no indication of any response from the sponsor to these queries. This is an apparent violation of section §11.64 of the FDAAA 2007 Final Rule. This section is titled “When must clinical trial information submitted to ClinicalTrials.gov be updated or corrected?” It states:
“…the Director may provide electronic notification to the responsible party of apparent errors, deficiencies, and/or inconsistencies in the submitted information identified during procedures for quality control review established by the Director, as specified at https://prsinfo.clinicaltrials.gov. The responsible party must correct or address all apparent errors, deficiencies, and/or inconsistencies identified in the notification not later than…25 calendar days for clinical trial results information, after the date of the electronic notification sent to the responsible party.”
This trial is well past the 25 calendar day limit for resubmission of corrected results: as of writing, it is 78 days since the results were returned to the sponsor with queries. Notably the sponsor was already 57 days late in submitting results.
This delay in response is not an isolated incident. Of the 84 trials that have submitted results after being “overdue” on our FDAAA TrialsTracker as of 16 July 2018, 18 (21.4%) have surpassed the 25 calendar day deadline to re-submit corrected results.
Particularly troubling is that the authors of the Final Rule anticipated this issue and built in mechanisms to counter the abandoning of submitted results during the quality control process. §11.52 clearly states that results information submitted should be posted “not later than 30 calendar days after the date of submission.” It is clear, from the rationale given in the text of the Final Rule itself, that publication after this time is supposed to occur regardless of QC status:
“Commenters expressed concern about the potential to misinform those using the public record and suggested only posting sections that have fulfilled quality control criteria. Some commenters suggested that the harm of posting information before the quality control review process has concluded is greater than the benefit of posting the information in a timely manner. While we understand these concerns, we interpret the statutory posting deadline to be a clearly delineated timeline between submission and posting… Taking into consideration the commenters’ concerns and the statutory requirements for posting clinical trial results information, we maintain the NPRM proposal in the final rule”
ClinicalTrials.gov have previously informed us, in December 2017, that this 30-day posting requirement had “not yet been implemented”. This appears to still be true. We have heard anecdotes from sponsors that the ClinicalTrials.gov quality control procedure can be cumbersome and frustrating. Overall, it seems that trial reporting laws continue to not be implemented, even ten years after the FDA Amendments Act 2007 was passed, and over a year after the publication and implementation of the Final Rule on that Act.
This unreported trial was sponsored by the study PI, Dr. Mitchell S Elkind of Columbia University in collaboration with the National Institute of Neurological Disorders and Stroke. We hope the investigators will submit corrected results for this unreported trial soon; and that the systems set out in the FDA Amendments Act 2007 will be implemented and enforced.
Competing interests: BG has received research funding from the Laura and John Arnold Foundation, the Wellcome Trust, the Oxford Biomedical Research Centre, the NHS National Institute for Health Research School of Primary Care Research, the Health Foundation, and the World Health Organization; he also receives personal income from speaking and writing for lay audiences on the misuse of science.
Nicholas J DeVito is a researcher at the EBM Datalab at the University of Oxford.
Competing interests: ND is employed on BG’s LJAF grant and is a Naji Foundation Scholar at the University of Oxford.