Having determined that antidepressants can cause suicidality in children and should bear black box warning labeling, the U.S. Food and Drug Administration has now responded to a personal injury attorney's petition by asking 14 manufacturers to reanalyze anticonvulsant drug trial data for behavioral signals of suicidality.
Russell Katz, M.D., director of the FDA Division of Neuropharmacological Drug Products, wrote a letter in April to New York attorney Andrew Finkelstein according to the Boston Globe, which broke the story April 20. The letter assured Finkelstein that the agency is "taking this matter very seriously," and said the drug manufacturers will be required to complete the reanalysis within six months.
One and a half years before, Finkelstein had purchased a 15-second national television ad seeking individuals who had become suicidal while on Neurontin (gabapentin). (One of his clients with bipolar depression had attempted suicide while receiving Neurontin.) According to the Globe, Finkelstein subsequently compiled and submitted to the FDA more than 200 adverse-event reports of patients who became suicidal while receiving this medication and filed more than 70 personal injury lawsuits against the drug's manufacturer, which remain unresolved. In May 2004, with lawsuits pending, Finkelstein filed a citizen's petition for the FDA to require a black box warning on the gabapentin drug label.
Although an association of the drug with suicide, if it exists, may reflect a lack of efficacy for bipolar disorder rather than a causative drug effect, previous findings that Neurontin was promoted for unapproved and unproven uses could leave the manufacturer vulnerable to additional civil actions. It also appears to some that the FDA investigation may be prompted less by indications of drug effect than by congressional pressure on the agency for lapses in drug safety and the long-term vacancy of an FDA commissioner (Psychiatric Times July, p1).
Approximately 8% to 10% of patients with untreated bipolar affective disorder die by suicide, according to Frederick K. Goodwin, M.D., professor of psychiatry and director of the Center on Neuroscience at George Washington University. Goodwin's research has revealed different suicide rates with different mood stabilizer treatments, but he implicates the illness rather than the treatments. "Anything that [effectively] treats the illness should be associated with reduction in the suicide rate," Goodwin told Psychiatric Times.
Goodwin suggested that the FDA investigation of the anticonvulsants could reflect a heightened susceptibility of the agency to outside pressure as an acting commissioner awaits congressional confirmation and the U.S. Congress investigates the FDA record on drug safety. "This whole thing ... got started with a trial lawyer, and the trial lawyers have gotten very smart at using the FDA," Goodwin asserted.
Asked whether his research into different rates of suicide in treated patients could suggest provocation by a mood stabilizer rather than lack of protective effect, Goodwin replied, "I've never seen anything in the research literature or in my clinical experience that would have led me to that conclusion.
"Now, in the case of [selective serotonin reuptake inhibitors], there is reason to ask the question," Goodwin added. "People who are treated with SSRIs-it's a well known clinical reality that the risk period for suicide is highest as they start to recover."
Goodwin cited a meta-analysis that ascertained the risk of completed suicide in patients with bipolar disorder to be 13 times less with lithium (Eskalith, Lithobid) treatment than without it (Baldessarini et al., 2001). "Lithium, perhaps more than the others, reduces [suicide rates], but it does not mean the others increase it; it just means that lithium may decrease it more," Goodwin said.
In the FDA investigation of suicidality in children receiving antidepressants, behavioral data from the clinical trials were not consistently classified by the study sponsors to distinguish suicidal behavior from other self-injury or psychodynamic. The FDA had referred the data to a group at Columbia University with expertise in adolescent suicides for assistance in classifying the emergent behaviors.
Asked whether the agency would provide the resultant classification guidelines to the manufacturers of the anticonvulsant drugs to facilitate more useful initial data assessment, FDA spokesperson Karen Mahoney told PT, "We are asking sponsors to use the same approach to searching for and classifying adverse events that might be considered possibly suicide-related that was used in evaluating the suicidality data in children. Once we have the data from manufacturers, we will do the data analysis, as we did with the pediatric data."
Mahoney added, "Columbia's only role was in classifying cases, and we are expecting manufacturers to use a similar approach to that used by Columbia."
To the question of whether the FDA anticipates the drug companies' reports will be sufficient to reveal or rule out suicidality drug effect, without additional hearings or outside analysis, Mahoney replied, "The database should be large enough to detect a signal if there is one. Of course, it's not possible to ever absolutely rule out an effect. The absence of a signal would provide some reassurance."
The manufacturer of Lamictal (lamotrigine) was among those receiving the request for reanalysis of emergent behaviors in clinical trials. Ann Rhyne, the manufacturer's spokesperson, told PT in April, "We're still at early stages and working with the FDA, deciding which studies should be included and what data reported."
Lamotrigine was approved in 2003 for long-term treatment of bipolar disorder and has demonstrated benefit for bipolar depression. Of the clinical trials conducted for that indication, Rhyne commented, "We found no evidence that Lamictal is associated with increased risk of suicide, suicide attempt or suicidal ideation."
The risk for suicide in bipolar disorder was characterized in a JAMA editorial as "extraordinarily high" (Baldessarini and Tondo, 2003). The authors estimated such risk at 0.4% per year, in contrast to the international general population average of 0.017% per year. The highest suicide risk in bipolar disorder, they noted, occurs with dysphoric-agitated (mixed) states, and with the more severe, recurrent depression of bipolar II disorder.
The authors added (Baldessarini and Tondo, 2003):
Moreover, suicide attempts by patients with bipolar disorder have an increased lethal potential. Among patients with bipolar disorder, rates of suicide attempts are only 5 times the rates of completed suicides, whereas in the general population, rates of suicide attempts are 10 to 20 times the rates of completed suicides.
In their meta-analysis of 33 studies with sufficient data to estimate annual rates of suicide, lithium treatment appeared to reduce the risk, with a 0.197% annual rate of attempted or completed suicide, compared to a 13-fold greater rate of 2.57% without lithium treatment (Baldessarini et al., 2001). Even with lithium treatment, Baldessarini and Tondo (2003) emphasized, the rate of completed suicides in bipolar disorder is approximately 10 times that in the general population.
With little data to differentiate mood stabilizers for affecting suicide risk, Goodwin and colleagues (2003) undertook a retrospective study of more than 20,000 patients with bipolar disorder in two comprehensive health plans. The patients had been treated with either lithium, divalproex (Depakote), carbamazepine (Equetro, Tegretol) or a combination in the period of 1994 through 2001. The investigators assessed the incidence in this population of suicide mortality, suicide attempts resulting in hospitalization, and attempts or suicidal behavior resulting in emergency department visits without inpatient admission.
Goodwin and colleagues found that at both health plan sites, rates of suicide attempts and suicide death were substantially greater during treatment with divalproex than with lithium. Comparisons of lithium treatment with carbamazepine or combinations were less consistent, with a statistically significantly higher hazard ratio (2.9) with carbamazepine only in attempts resulting in hospitalization (Table).
From these and other data, lithium might be considered a candidate to join clozapine (Clozaril) as an FDA-approved treatment to prevent suicidal behavior. Baldessarini and Tondo (2003) observed in their JAMA editorial, "Hopefully, such renewed interest in the potentially treatment-modifiable lethality of major mental disorders will be sustained and increasingly successful."
A putative protective effect of lithium, however, does not imply that other mood stabilizers heighten risk. Nor do the adverse-event reports of suicidal behavior in patients receiving gabapentin necessarily evidence causality, Robert Temple, M.D., associate director for medical policy at the FDA Center for Drug Evaluation and Research explained to the Boston Globe. These reports, he said, "Can't really tell you whether the suicidal event is because of the drug or despite the drug."
Baldessarini RJ, Tondo L (2003), Suicide risk and treatments for patients with bipolar disorder. [Published erratum JAMA 291(2):186.] JAMA 290(11):1517-1519 [editorial].
Baldessarini RJ, Tondo L, Hennen J (2001), Treating the suicidal patient with bipolar disorder. Reducing suicide risk with lithium. Ann N Y Acad Sci 932:24-38; discussion 39-43.
Goodwin FK, Fireman B, Simon GE et al. (2003), Suicide risk in bipolar disorder during treatment with lithium and divalproex. JAMA 290(11):1467-1473 [see comments].