Continuing antidepressant therapy for 52 weeks, as opposed to stopping it at 8 weeks, was not more beneficial with regard to the primary outcome of occurrence of any mood episode.
However, a prespecified sensitivity analysis of the primary outcome and of the secondary analyses suggests that continuing antidepressant therapy for 52 weeks may prolong the time to a depressive relapse.
“Because the primary outcome is negative and the prespecified sensitivity analysis is positive and the secondary outcomes are positive, some clinicians will pick the position that they work and some that they don’t work,” lead investigator Lakshmi Yatham, MBBS, with University of British Columbia, Vancouver, told this news organization.
The study was published online in the New England Journal of Medicine.
Adjunctive antidepressant therapy – alongside mood stabilizers and/or second-generation antipsychotic medications – are often used to treat acute depressive episodes in patients with bipolar I disorder.
Currently, the Canadian Network for Mood and Anxiety Treatments (CANMAT) and International Society for Bipolar Disorders (ISBD) advise discontinuing antidepressant treatment 8 weeks after remission of depression.
Yet, the duration of antidepressant therapy for bipolar depression is “highly controversial,” due to a lack of evidence and concerns that antidepressants may induce mania, mixed states, or rapid cycling between mania and depression, Dr. Yatham said.
Dr. Yatham and colleagues assessed the safety and efficacy of continuing adjunctive antidepressant treatment (escitalopram or bupropion XL) for 52 weeks after remission, compared with discontinuing antidepressant therapy at 8 weeks after remission.
The final analysis included 177 patients (mean age 41 years, 48% men) with bipolar I disorder who had remission of depression; 90 patients continued treatment with an antidepressant for 52 weeks and 87 were switched to placebo at 8 weeks. All were taking a mood stabilizer or a second-generation antipsychotic or both.
The primary outcome, assessed in a time-to-event analysis, was any mood episode, as defined by scores on scales measuring symptoms of hypomania or mania, depression, suicidality, and mood-episode severity; additional treatment or hospitalization for mood symptoms; or attempted or completed suicide.
At 52 weeks, 28 patients (31%) in the 52-week group had experienced any mood episode (primary outcome), compared with 40 patients (46%) in the 8-week group.
The primary outcome did not reach statistical significance (hazard ratio, 0.68; 95% confidence interval, 0.43-1.10; P = .12).
The researchers note that the decision by the study team to include relapses that occurred during the first 6 weeks of the study may have affected the primary outcome.
“During the first 6 weeks, both groups were getting the same treatment, and we thought there shouldn’t be any difference in relapse, but sadly, there were more relapses in the 52-week group even though the treatments were identical,” Dr. Yatham said.
However, in a sensitivity analysis of the primary outcome after week 6, when treatment between the two groups differed, patients continuing antidepressant treatment were 40% less likely to experience a relapse of any mood event (HR, 0.60) and 59% less likely to experience a depressive episode (HR, 0.41) relative to the placebo group.
“From the point where the two groups began receiving different treatments, we see a significant benefit for patients who continued treatment with antidepressants,” Dr. Yatham said in a news release.
“Treating depression in bipolar disorder is challenging. Reducing the risk of relapse is important because it can provide patients with a great deal of stability that ultimately lets them get back to the activities they enjoy and can greatly improve their quality of life,” he added.
Although fewer patients in the 52-week group than 8-week group had a depressive episode within 52 weeks (17% vs. 40%; HR, 0.43), more had a manic or hypomanic event (12% vs. 6%; HR, 2.28).
The estimated probability of remaining free of a depressive episode at 52 weeks was 72% in the 52-week group versus 53% in the 8-week group. The estimated probability of remaining free of a manic episode at 52 weeks was 81% and 92%, respectively.
The incidence of adverse events was similar in the two groups, with a low rate of discontinuation due to adverse events and no serious adverse events. Clinically significant weight gain (≥ 7% increase in body weight) was observed in 14% of patients in the 52-week group and 7% of patients in the 8-week group.
Limitations of the trial include the fact that it was stopped early, before the planned sample size was reached, owing to slow recruitment and funding issues.
Other limitations include a lack of ethnic diversity (only 12% were White and < 1% Black) and overrepresentation of patients from India, which may limit generalizability.
In addition, the findings may not be applicable to treatment with antidepressants other than escitalopram and buproprion XL. Finally, the study population was also enriched for patients who responded to these antidepressants.