AstraZeneca halts cancer drug trial in setback

Adults with serious mental illness (SMI) have a shorter lifespan than the general population, primarily driven by cardiovascular disease.

A target trial emulation of adults with SMI found that starting a GLP-1 receptor agonist was associated with significantly lower risks of death and cardiovascular events compared with starting a SGLT2 inhibitor.

Among 195,184 propensity score-matched pairs of adults with SMI, GLP-1 drug initiators had a 24% lower 4-year mortality risk compared with SGLT2 inhibitor initiators (HR 0.76, 95% CI 0.74-0.78). Death occurred in 4.91% and 6.45% of groups, respectively, equating to an absolute risk difference of -1.54 percentage points.

At year 1, all-cause mortality risk was 49% lower among GLP-1 initiators compared with SGLT2 inhibitor initiators (HR 0.51, 95% CI 0.49-0.53, P<0.001), Roger McIntyre, MD, of the University of Toronto, and colleagues reported in JAMA Psychiatry.

This association persisted across sensitivity analyses. A review of semaglutide (Ozempic, Wegovy) initiators who had SMI and type 2 diabetes indicated that the relationship was largely driven by risk reductions in cardiovascular outcomes (all P<0.001):

• 3-point major adverse cardiovascular events (MACEs): HR 0.77, 95% CI 0.76-0.79
• 5-point MACEs: HR 0.76, 95% CI 0.75-0.77
• Myocardial infarction: HR 0.71, 95% CI 0.69-0.73
• Stroke: HR 0.89, 95% CI 0.86-0.92
• Heart failure: HR 0.73, 95% CI 0.72-0.74
• Coronary artery bypass grafting: HR 0.77, 95% CI 0.70-0.85

The findings "may represent a treatment option to narrow the cardiovascular mortality gap in persons living with SMI" and support "a broad cardiometabolic benefit rather than an isolated outcome-specific signal," the authors wrote.

Life expectancy for individuals with SMI is roughly 10 to 25 years shorter than the general population, primarily due to cardiovascular disease.

"The excess cardiovascular morbidity and mortality in SMI reflects the convergence of elevated cardiometabolic risk factor burden, psychopharmacotherapy-related metabolic effects, and systemic disparities in preventive care delivery, chronic disease screening, and guideline-concordant treatment," the researchers wrote.

Underlying mechanisms driving this risk reduction may include anti-inflammatory and anti-atherogenic effects of GLP-1 drugs, alongside possible influence on motivation, behavior, cognitive organization, and mitigation of risk factors like alcohol use disorder, they suggested.

The findings didn't come as a surprise as GLP-1 drugs reduce cardiovascular mortality in the general population, McIntyre told MedPage Today.

"The impetus to conduct our study was provided by the observation that people living with serious mental illness -- such as major depressive disorder, bipolar disorder, or schizophrenia -- not only have a significantly higher rate of mortality with consequently shorter lifespan, but the cause contributing most to excess mortality in this population is cardiovascular disease," McIntyre explained.

He argued the drug class could be "potentially transformative in the psychiatric population," noting "currently no FDA-approved treatment has been shown to reduce mortality... to the extent to which we have just observed with GLP-1 receptor agonists."

The retrospective target trial emulation utilized TriNetX electronic health record data through March 2026. Using a new-user, active-comparator design, 1,528,230 adults were classified into SMI and non-SMI categories and propensity-score matched, with a mean age of approximately 60.5 years in the SMI cohort.

At 4 years, mortality was significantly reduced in type 2 diabetes patients with SMI who started semaglutide (HR 0.76) or tirzepatide (Mounjaro, Zepbound; HR 0.49). Older GLP-1 drugs -- dulaglutide (Trulicity), liraglutide (Victoza, Saxenda), exenatide (Byetta, Bydureon), lixisenatide (Adlyxin), and albiglutide (now discontinued) -- showed no consistent survival advantage over SGLT2 inhibitors.

In 10-year exploratory analyses of individuals with type 2 diabetes, semaglutide was associated with lower mortality across SMI subgroups, including major depressive disorder (HR 0.75), bipolar disorder (HR 0.76), and schizophrenia (HR 0.85, all P<.001).

The emulation could not fully control for unmeasured medication adherence, lifestyle, socioeconomic factors, or psychiatric severity, the authors acknowledged. The findings may not apply to all populations, as cost barriers and inequities in GLP-1 drug access disproportionately affect those with SMI.

"Our results provide the impetus for careful mechanistic studies to better understand how they offer this anti-mortality effect and also provide a compelling line of evidence to justify considering these agents as part of the integrated care of people with mental illness with an aim to improve health span and lifespan," McIntyre concluded.