QuarterWatchT: 2010 (Quarter 3)
New signals for liraglutide, QUEtiapine, and varenicline
From the May 19, 2011 issue
QuarterWatch™ is an ISMP surveillance program that monitors all serious, disabling, and fatal adverse drug events (ADEs) reported to the US Food and Drug Administration (FDA) by drug manufacturers and the public through its MedWatch reporting program. The primary goal of the program is to improve patient safety through the identification of signals that may represent serious safety issues with medications. The term signal means evidence that, in our judgment, is substantial enough to warrant further investigation to determine frequency of occurrence and establish a causal relationship to the suspect drug.
Reporting totals and trends
In the third quarter of 2010, FDA received 36,679 domestic reports of serious, disabling, or fatal injuries associated with drug therapy. This represents a 22% increase from the same quarter in 2009, and an 11% increase when compared to the second quarter in 2010. Reports that healthcare professionals and consumers sent directly to FDA declined by 17% compared to the same quarter in 2009, while reports initiated by drug manufacturers increased by 33%. The largest single factor in this overall increase was a surge in ADE reports for rosiglitazone (AVANDIA), to which FDA restricted access in September 2010 because of its cardiovascular risks. (The drug was withdrawn from the market in Europe.)
Selected ADE signals
Liraglutide (VICTOZA) and pancreatitis. We observed signals for pancreatitis and other adverse effects within 9 months of the 2010 launch of liraglutide. FDA approval of this drug was controversial because of uncertainty about its cardiovascular risks, and because animal studies showed an increased risk of thyroid cancer. Early ADE reporting did not speak to these still unresolved issues but did reveal a marked signal for pancreatitis.
There are now a total of four drugs on the market for Type 2 diabetes that lower blood glucose levels through their effects on glucagon-like peptide-1 (GLP-1). Evidence accumulates that all may share an elevated risk of pancreatitis, although possibly to differing degrees. Novo Nordisk, the manufacturer of liraglutide, noted that epidemiological studies show a higher risk of pancreatitis in diabetic patients, independent of treatment. Because the risks and benefits of these drugs only become apparent over many years, pre-approval testing is too short to determine if the drugs’ benefit-risk profile is favorable. Some experts fear if long-term pre-approval testing was required, drug development costs might be prohibitive. As a result, uncertainty about the long-term risks and benefits continues with a new generation of agents for Type 2 diabetes.
QUEtiapine (SEROQUEL) and irreversible injury. QUEtiapine shares with other antipsychotic drugs the risk that some of its most common side effects can be irreversible, notably some cases of diabetes and certain movement disorders. In the third quarter of 2010 we noted hundreds of new reported cases of diabetes associated with QUEtiapine, together with a smaller number of reports of dyskinesia (involuntary muscle movements), dystonia (spasms or prolonged contractions), and parkinsonism (tremors or muscle rigidity).
Because of their side-effect profile, powerful antipsychotic drugs, including QUEtiapine, were once largely restricted to the most severe mental illnesses, schizophrenia and psychosis. Despite evidence that these risks have never been reduced, ADE evidence shows QUEtiapine is now being used as a general purpose psychiatric drug for a wide variety of disorders. FDA approved some expanded use of QUEtiapine on the basis of 3-6 week tests that cannot assess serious long-term risks. However, the ADE data show that both newer FDA-approved indications and off label uses of the drug for conditions such as sleep disorders, anxiety, autism, and agitation have resulted in serious injuries reported in wider patient populations where much less is known about the risks and benefits.
Varenicline (CHANTIX) risks underestimated. Evidence continues to accumulate showing that varenicline has an unacceptable safety profile. We discovered FDA had not been aware of hundreds of older serious psychiatric ADE reports initiated by Pfizer, because they had not been entered into the FDA Adverse Event Reporting System (AERS). Most notable were 150 cases of completed suicides, some dating back to 2007.
When we first examined varenicline data in 2008, ADE reports for this drug outnumbered reports for all other US prescription drugs, although dispensed prescriptions and reported events later declined. In 2009, FDA required strong warnings about psychiatric side effects for both physicians and patients. Now, with a spike of 1,055 serious ADEs reported in the third quarter of 2010, varenicline again surpassed all other drugs we regularly monitor and also ranked first in reported deaths—more than twice as many as any other drug.
The quarter totals had been boosted by 589 reports about serious and fatal ADEs, including 150 completed suicides, which had occurred in prior years but had been entered into the AERS for the first time in July 2010. In the QuarterWatch™ tally, we could only identify 37 completed suicide reports initiated by the company and previously submitted to AERS through June 2010. Healthcare providers and consumers directly reported an additional 85 suicide cases, which were entered into AERS without manufacturer involvement. That meant that, as of June 2010, fewer than half the known suicides identifying varenicline as the suspect drug had been reported to the AERS (see Table 1 in PDF version).
Although Pfizer declined to comment, FDA provided a response explaining in part why this happened. In brief, when a drug company learns of a patient’s death, an “expedited” report is normally submitted to FDA within 15 days and flows automatically into AERS. For unknown reasons, Pfizer classified these suicides among less serious injuries and submitted them amid hundreds of other less serious reports as an appendix to a quarterly text report intended for page-by-page reading. Thus, the reports were not entered into the AERS as FDA expected.
This problem apparently continued for years and was not limited to varenicline or to Pfizer. FDA identified the problem in 2010 and told Pfizer to submit all its periodic reports for varenicline in the correct electronic format so they would automatically be entered into its AERS. Pfizer did so in July 2010, and the additional 150 suicides were included in its submission.
We believe varenicline is unsafe for widespread clinical use and encourage FDA to reassess this drug’s safety profile based on this new information and take the necessary steps to mitigate harm. In addition, FDA should investigate further why so many serious ADEs were not promptly reported. Also, they need to address whether manufacturers are identifying expedited events correctly, reporting them within 15 days, and ensuring that the reports are entered into the AERS.
Additional signals from other drugs are described in the full QuarterWatch™ Report at: www.ismp.org/quarterwatch.