The dental community should be a little more wary of articles detailing clinical trials published in peer-reviewed journals, according to a new review of clinical trials in the Journal of the American Medical Association. The author of the paper discovered that the findings of clinical trials are sometimes published without any note of their scientific misconduct found by the Food and Drug Administration (FDA).
Author Charles Seife, a professor of journalism at New York University, and his investigative reporting class evaluated clinical trials with significantly objectionable conditions or practices, such as falsifying information or not reporting severe adverse effects, that were published in peer-reviewed journals. They found that about 96% of the tainted clinical trial articles, including ones for medications prescribed by the dental community, did not mention any sort of scientific misconduct.
"Such allegations [of serious research misconduct] are relevant to include in the peer-reviewed literature on which physicians and other medical researchers rely to help them choose treatments that they offer to patients and other participants," Seife noted in the paper (JAMA, February 9, 2015). "To better serve the public health, the FDA should make unredacted information about its findings of research misconduct more readily available."
Using documents accessible through the FDA's website and from Freedom of Information Act requests, Seife and his class looked for clinical trial sites that were issued an Official Action Indicated (OAI) rating from FDA investigations. The FDA gives clinical trial sites an OAI rating only when they find "objectionable conditions or practices significant enough to warrant regulatory action."
Through their research, Seife and his students identified 421 OAI-rated inspections between January 1998 and December 2013. However, only 101 trials could be identified from those because of the amount of text the FDA redacted to protect sensitive corporate information.
About one-third of clinical trials with OAI-rated inspections remained unpublished, while the rest were published in peer-reviewed journals. The researchers were able to identify 78 published clinical trial articles where the authors did not mention the FDA inspection violations, such as falsification of information and failure to protect patient safety.
While the FDA usually excludes data from studies with OAI-rated inspections when judging a drug's safety, it does not make that information readily available to the scientific community. Consequently, papers about clinical trials that have been labeled as unreliable by the FDA can get published as fact in peer-reviewed literature, as was the case with rivaroxaban (brand name Xarelto, Bayer HealthCare), an anticoagulant drug.
Unreliable rivaroxaban studies highlight serious problem
Rivaroxaban and other anticoagulants have been warranted as safe to use for nonextensive dental surgery by the University of California, Davis Health System.
When rivaroxaban was first set to enter the U.S.market, the FDA audited 16 sites that were involved with a Bayer-sponsored clinical trial of the drug called RECORD 4. Half of investigated sites resulted in OAI ratings because of grievances such as falsification of information, failure to protect patient safety, and improper record keeping.
The FDA ultimately ruled RECORD 4 completely unreliable. David Craig Loucks, MD, a key researcher at a Colorado location, was cited in 2009 for a number of egregious errors, including destroying documents, falsifying his signature, lack of proper oversight, not ensuring proper subject safety, and not mentioning 12 severe adverse effects in a timely manner for the 24-person study.
The FDA also audited a sister rivaroxaban clinical study, called RECORD 2, that proved almost as unreliable as RECORD 4. The FDA found evidence of misconduct in 4 of the 10 sites, according to Seife in a Slate article. Like RECORD 4, RECORD 2 was part of a Bayer-sponsored clinical trial of its drug.
At least four papers citing the pharmaceutical-sponsored clinical studies were published in peer-reviewed journals anyway. All four of the published papers touted the findings of the discredited studies, concluding that rivavoxaban was superior to the anticoagulant enoxaparin (New England Journal of Medicine, June 26, 2008, Vol. 358:26, pp. 2765-2775 and 2776-2786; Lancet, July 5, 2008, Vol. 372:9632, pp. 31-39, and May 16, 2009, Vol. 373:9676, pp. 1673-1680).
Perhaps unsurprisingly, the authors of all four papers were connected to Bayer, according to an analysis of the clinical studies presented at 2008 American Society of Hematology annual meeting and published in the journal Blood. Most notably, Alexander G.G. Turpie, MD, the lead author of the paper on RECORD 4, and Ajay Kakkar, MBBS, PhD, the lead author of a paper published on RECORD 2, sit on the Global Anticoagulant Registry in the Field (GARFIELD-AF) Steering Committee for the GARFIELD Registry, which is funded by an unrestricted research grant from Bayer. However, it is not known for certain if either of the papers' authors knew of the OAI-rated inspections before writing the articles.
Most concerning, wrongful death lawsuits are now being pursed against Bayer, the drug's manufacturer, and Janssen Pharmaceuticals, a Johnson & Johnson company that markets Xarelto. Rivaroxaban can cause uncontrolled bleeding from even small cuts, prompting questions of its safety for even minor dental work, according to some law websites.
Questionable IV ibuprofen study may impact future research
An FDA inspection of a site of a Cumberland Pharmaceuticals-sponsored clinical study also yielded an OAI rating. The trial looked at how well Cumberland's Caldolor intravenous ibuprofen drug reduced patient's morphine use and postoperative pain after abdominal and orthopedic surgery.
The FDA's warning letter cited the study location researchers for discrepancy in the reporting of how much morphine the patients were actually given, breaking the double-blind study through a record revealing that two patients were receiving ibuprofen and not the placebo, and not reporting that one patient was vomiting for 48 hours and required an IV.
Two papers citing the efficacy of Caldolor were published in peer-reviewed journals based on the questionable findings (Clinical Therapeutics, September 2009, Vol. 31:9, pp. 1922-1935; Pain Practice, January-February 2011, Vol. 11:1, pp. 23-32). Amy Rock, the current senior director for regulatory affairs for Cumberland Pharmaceuticals (then a senior manager), was the responsible party for the clinical trial and a co-author on both published papers. Additionally, either a lead or co-author from each publication presented their papers for Cumberland at the 2009 American Academy of Pain Management annual clinical meeting.
The promising results of this study prompted another researcher at Tufts University School of Dental Medicine to see if intravenous ibuprofen can pre-emptively prevent postoperative pain after third-molar extraction. The clinical trial, which cites one of the published papers in its description, is currently recruiting participants.
Archana Viswanath, BDS, the lead researcher of the new clinical study, could not be reached for comment by press time on whether the validity of the first clinical trial would affect her study.
Beginning to solve the problem
It is clear that when studies without good clinical practice get published as fact the entire medical community, from researchers to doctors to patients, gets hurt.
"The FDA has legal as well as ethical responsibilities regarding the scientific misconduct it finds during its inspections," Seife concluded. "Failing to notify the medical or scientific communities about allegations of serious research misconduct in clinical trials is incompatible with the FDA's mission to protect public health."
While the JAMA article was revealing, the author did admit the research was biased. Documents from certain regions of the country and time periods were easier to locate online, Seife noted. The documents the researchers found were also so redacted that they could only analyze a percentage of them. The shortcomings prevent researchers from making generalizations of their findings that apply to all clinical trials, he added.
The shortcomings, however, do not detract from the need for more transparency of clinical trials. Seife recommends that the FDA should notify journals when a site participating in a published clinical trial receives an OAI inspection. Furthermore, he recommends the agency publish more OAI-rated investigations online and to do so without the heavy redactions.
Seife's final recommendation is for peer-reviewed journals to require article authors to disclose any adverse findings during FDA inspections. While this may not solve the problem entirely, it may help the medical community to read the published material with a more critical eye.