New Studies Fuel Debate Over GLP-1 Drugs and Suicide Risk

Two new studies published in JAMA Internal Medicine do not support a previously reported association between GLP-1 receptor agonists and a signal for suicidal ideation.

In an analysis of nearly 300,000 new users of a GLP-1 agonist or SGLT2 inhibitor in Sweden and Denmark, rates of suicide death over 2.5 years were not significantly different between groups, reaching 0.23 and 0.18 events per 1,000 person-years, respectively (HR 1.25, 95% CI 0.83-1.88), reported researchers led by Peter Ueda, MD, PhD, of the Karolinska Institutet in Stockholm.

The study also found no difference between groups for new-onset depression and anxiety-related disorders, and slightly lower rates of the combined outcome of suicide death and nonfatal self-harm or self-harm alone in the GLP-1 agonist group:

• New-onset depression and anxiety-related disorders: 25.8 vs 25.4 per 1,000 person-years (HR 1.01, 95% CI 0.97-1.06)
• Suicide death and nonfatal self-harm: 1.47 vs 1.78 per 1,000 person-years (HR 0.83, 95% CI 0.70-0.97)
• Self-harm: 1.26 vs 1.62 per 1,000 person-years (HR 0.77, 95% CI 0.65-0.91)

The second study pooled data from several of the STEP trials that supported semaglutide’s (Wegovy) weight-loss indication. This post-hoc analysis showed that depression scores on the Patient Health Questionnaire-9 (PHQ-9) were lower at study end for patients on a 2.4-mg dose of semaglutide versus placebo recipients (-0.56 mean difference, 95% CI -0.81 to -0.32, P

And those on semaglutide were significantly less likely to shift to a more severe category of depression on the PHQ-9 scale (OR 0.63, 95% CI 0.50-0.79, P

Overall, 1% or fewer participants reported suicidal ideation or behavior during the study, measured by the Columbia-Suicide Severity Rating Scale, with no difference between groups.

“Results from the present study provide reassurance to healthcare professionals and their patients with overweight/obesity that semaglutide 2.4 mg does not appear to increase the risk of developing symptoms of depression or of suicidal ideation or behavior in persons who are free of such complications prior to treatment,” Wadden told MedPage Today.

In an accompanying editor’s note, JAMA Internal Medicine associate editor Timothy Anderson, MD, MAS, and deputy editor Deborah Grady, MD, MPH, agreed that the studies are reassuring, but added that “neither fully answers the question of whether these drugs are safe in those with preexisting mental health problems.”

They pointed out, for example, that the STEP trials excluded patients with current severe depression symptoms or a history of major depression, schizophrenia, or bipolar disease in the 2 years prior to enrollment.

In Line With FDA Guidance

“We know that persons with severe obesity (BMI 40-plus) have a markedly increased risk of major depression and anxiety disorders than persons of average BMI,” Wadden said. “Practitioners can expect to encounter psychological complications in such individuals and should be prepared to provide or refer patients for appropriate mental healthcare.”

“When mental health complications including depression, suicidal ideation, or behavior are observed in patients with obesity who take GLPs and other anti-obesity medications, it’s very difficult to isolate whether the complications are associated with the patient’s obesity, with life stressors, with a personal or family history of psychiatric illness, or possibly with an adverse effect of a medication for weight management,” he continued. “The key is to provide appropriate mental health support regardless of the possible cause.”

Wadden agreed with FDA guidance telling clinicians to monitor patients on GLP-1 agonists for new or worsening depression, suicidal thoughts, or any unusual changes in mood or behavior. “This can be done at routine visits using screening inventories such as the PHQ-9 or by inquiring verbally about changes in patients’ psychosocial functioning.”

Until more data are available, GLP-1 agonists should be avoided in those with a history of suicide attempts and active suicidal ideation, as per FDA guidance. “Caution suggests discontinuing the use of a weight-management medication in cases of incident suicidal ideation or suicidal behavior,” Wadden added.

The agency said this after conducting its own preliminary evaluation earlier this year of the class following some reports of suicidality and self-harm with GLP-1 agonists, but didn’t find evidence to support this link. The European Medicines Agency likewise found no evidence to support a causal link between GLP-1 receptor agonists and suicidal thoughts following its own 9-month review.

But just 2 weeks ago, a disproportionality analysis based on reports in a World Health Organization (WHO) linked semaglutide with a 45% higher odds of suicidal ideation reports compared with all other drugs in the database. Commenting on this analysis, Wadden said it’s “important to keep in mind that some analyses of post-marketing surveillance data have significant methodological limitations, including an inability to infer causality and to attribute any adverse reactions to the effects of a drug.”

STEP Trials Post Hoc Analysis

Wadden and co-authors’ post hoc analysis included 3,377 participants (69.6% female, mean age 49) from the STEP 1, 2, and 3 trials and 304 participants (77.6% female, mean age 47) from the STEP 5 trial.

Average PHQ-9 scores at baseline were 2.0 and 1.8 for the semaglutide and placebo groups, respectively. By week 68, scores were 2.0 and 2.4 (scores 0-4 indicate no/minimal depression).

Across the trials, 2.8% and 4.1% of semaglutide and placebo participants had PHQ-9 scores of 15 or greater after treatment. Scores of 15 or higher indicate moderately severe symptoms and requires evaluation by a mental health professional.

Psychiatric disorder adverse events were generally balanced between groups.

GLP-1 Receptor Agonists vs SGLT2 Inhibitors

Ueda’s group pulled nationwide register data from Sweden and Denmark for their active-comparator cohort study. It included 124,517 adults initiated a GLP-1 receptor agonist and 174,036 initiated an SGLT2 inhibitor from 2013 to 2021 primarily for type 2 diabetes. Average age was 60 and 45% were women. Liraglutide (Victoza, Saxenda; 50%) and semaglutide (41%) were the most common GLP-1 agonists.

Patients with self-harm within 3 months prior to the study were excluded for outcomes including self-harm.

A total of 77 suicide deaths in GLP-1 agonist users and 71 in SGLT2 inhibitor users were identified. Weighted incidence rates were 0.23 versus 0.18 events per 1,000 person-years for the GLP-1 agonist and SGLT2 inhibitor groups, respectively. The absolute difference also didn’t differ (0.05 events per 1,000 person-years, 95% CI -0.03 to 0.16).

Subgroup analyses continued to find no difference between the two drug classes when separated by patients with a history of psychiatric disorders, those on liraglutide, and those on semaglutide.

“While reassuring, the study could not rule out smaller absolute risk differences for suicide death, and future studies with more outcome events should be performed as data accumulate,” Ueda and co-authors wrote.

Wadden also called for additional studies delving into the safety of GLP-1 agonists and similar weight management agents among patients with more significant mental health disorders than were included in the STEP trials. “I would recommend small carefully controlled pilot studies of patients with current, well-controlled depressive disorder, bipolar disorder, and other conditions.”