Suicide in Depression: Balancing Risk Factors, Identifying Vulnerable Patients

This May, the FDA called for a black box warning on antidepressants to indicate that patients aged 18 to 24 years are at heightened risk for treatment-emergent suicidality. But a member of the FDA advisory committee that recommended that warning has issued his own warning, saying that the "real killer in this story is untreated depression and the possible risk from antidepressant treatment is dwarfed by that from the disease."

Andrew Leon, PhD, professor of biostatistics in psychiatry at Weill Cornell Medical College, New York, and member of the FDA Psychopharmacology Drugs Advisory Committee, made this point with coauthor Richard Friedman, MD, in a Perspective article in the June 7 issue of New England Journal of Medicine.¹ They affirmed that the data indicate an absolute risk of antidepressant treatment-emergent suicidality of 0.01%; and the committee previously estimated a 1% to 3% occurrence in children receiving antidepressants. Leon and Friedman emphasize, however, that these rates are substantially lower than the estimated 2.2% to 15% lifetime risk of suicidality for patients with major depression.

In an audio interview accompanying the article on the New England Journal's Web site, Leon commented, "Given the data that are available, I think it's important that the FDA has so strongly advised physicians to monitor suicidality in patients who are treated with antidepressants. Yet, at the same time, the FDA warns that untreated depression is a risk for suicide in and of itself."

Leon was among the 6-2 majority of the committee that voted to recommend expanding the warning about children and adolescents to indicate that young adults are also at risk for antidepressant-related suicidality. The committee determined that the meta-analyses of trials involving almost 100,000 patients supported the association between antidepressants and suicidality as well as the trend of heightened occurrence in younger age groups. Although the somewhat elevated rate in patients aged 18 to 24 was not statistically significantly higher than in older adults, it was a sufficiently strong signal for the committee to include this age group in the warning.

"The threshold for threat to safety is generally lower than that for efficacy, and the data did not provide strong evidence of an absence of risk."

Leon is also concerned, however, about the recently documented trend of reduced antidepressant prescribing that followed the 2003 black box warning of suicidality in children.² (For a discussion of this issue regarding children, see "Update on Antidepressants and Suicide" by Karen Dineen Wagner, MD [Psychiatric Times, July 2007, page 29]).

"The public health experiment has just begun," Leon declared. "What we want to see is if the black box reduced the distribution of the antidepressants in the United States so much [that] the rates of suicide increase . . . over the next several years. That will give us an indication that the black box warning was a mistake.

"If, on the other hand, suicide rates remain constant or go down," Leon said, "the FDA cannot be criticized for the decision to introduce the black box warning."

With suicide a hazard of untreated depression, and with antidepressant medication a generally effective treatment, Friedman and Leon ask the rhetorical question of why the FDA did not emphasize in the antidepressant label warning that the medication can lessen the greater risk of untreated illness. They point out, however, that such a statement would resemble a treatment recommendation, which is outside the agency's purview.

Friedman and Leon do recommend that clinicians provide this context to patients, explaining that while there is a risk with antidepressant medications, there is a larger risk to patients with untreated depression. They encourage clinicians to carefully monitor patients early in treatment, particularly during the first 4 to 6 weeks after starting an antidepressant.

Such recommendations are increasingly prevalent and are evolving into standards of practice. The recently updated Texas Children's Medication Algorithm Project (CMAP) on medication treatment of childhood major depression, for example, recommends assessing suicidal ideation and behavior at every visit, including frequency, places, intentions, and potential dangerousness. The CMAP consensus panel states that "careful monitoring and systematic assessment of suicidal thoughts and behavior during medication initiation and dose changes are essential."³

Seeking susceptibility

In addition to the determination that those in younger age groups are at higher risk and the recommendation that drug initiation and dosage change warrant close monitoring, efforts are being made to identify individuals who may be particularly susceptible to treatment-emergent suicidality. Several new NIMH-sponsored research projects were reviewed in these pages earlier this year ("FDA, NIMH Scrutinize Antidepressant-Linked Suicidality" by Kenneth J. Bender, March 2007, page 1).

The possibility of a genetic vulnerability to this adverse reaction to antidepressants was recently investigated in a cohort of the large, multicenter STAR*D study. In the June issue of Archives of General Psychiatry, Roy Perlis, MD, of the Depression Clinical and Research Program, Massachusetts General Hospital, Boston, and colleagues reported finding that particular genetic polymorphisms in men, but not women, were associated with emergence of suicidal ideation with antidepressant treatment.⁴

The researchers had previously reported finding that single nucleotide polymorphisms (SNPs) near the CREB₁ (cyclic adenosine monophosphate response element binding) gene are associated with anger expression, with its possible direction toward self as well as others.⁵ Perlis and colleagues note that the transcription factor CREB protein, coded by the CREB₁ gene, has been implicated by other researchers in mediating antidepressant response, suicide, and mood disorders in general.

In the present study, Perlis and coauthors analyzed the occurrence of 5 SNPs near CREB₁ in more than 1800 STAR*D study patients with depression. Of 539 men without suicidal ideation at baseline, suicidal ideation developed in 54 (10%) during the 12-week course of antidepressant treat- ment. Two of the 5 SNPs were significantly more common in the men with treatment-emergent suicidality. The 2 SNPs occurred in a pattern independent of the patients' history of suicide attempts or severity of depression, which was consistent with a vulnerability to treatment-emergent effect rather than general proneness to suicide.

Perlis and associates suggest that their gender-specific finding could reflect the lower levels of anger or irritability in women in general, and as components of their suicidality. They hypothesize that antidepressant treatment might present an additional stressor in a subset of male patients with depression and a genetic predisposition for hostility and anger expression.

Although the researchers acknowledge that their findings are preliminary and that the functions of the variants are unknown and so cannot be considered causal, they anticipate that this line of investigation will have clinical implications for identifying higher-risk subgroups among depressed patients.

"The ability to identify this group could allow both patients and clinicians to proceed with greater confidence or focus more attention on the high-risk group," Perlis and colleagues wrote.

References:

References

1.

Friedman RA, Leon AC. Expanding the black box-depression, antidepressants, and the risk of suicide.

N Engl J Med

.

2007;356:2343-2346.

2.

Libby AM, Brent DA, Morrato EH, et al. Decline in treatment of pediatric depression after FDA advisory on risk of suicidality with SSRIs.

Am J Psychiatry.

2007;164:884-891.

3.

Hughes CW, Emslie GJ, Crismon ML, et al. Texas children's medication algorithm project: update from Texas consensus conference panel on medication treatment of childhood major depressive disorder.

J Am Acad Child Adolesc Psychiatry.

2007;46:667-686.

4.

Perlis RH, Purcell S, Fava M, et al. Association between treatment-emergent suicidal ideation with citalopram and polymorphisms near cyclic adenosine monophosphate response element binding protein in the STAR*D study.

Arch Gen Psychiatry.

2007;64:689-697.

5.

Perlis RH, Purcell S, Fagerness J, et al. Clinical and genetic dissection of anger expression and

CREB1

polymorphisms in major depressive disorder.

Biol Psychiatry.

2007. Feb 14; [Epub ahead of print].