Semaglutide shows no higher 12-month risk of adverse neuropsychiatric outcomes than other antidiabetic drugs, study suggests

By Vijay Kumar MalesuReviewed by Danielle Ellis, B.Sc.Jul 17 2024

In a recent study published in the EClinicalMedicine, a group of researchers evaluated the 12-month neurological and psychiatric outcomes of semaglutide use in patients with type 2 diabetes mellitus (T2DM) (Chronic high blood sugar due to insulin resistance or deficiency) using a propensity-score matched cohort.

Background 

Semaglutide, a glucagon-like peptide 1 receptor agonist (GLP1-RA) approved for T2DM and obesity, is a 2023 scientific breakthrough. Healthcare spending on GLP1-RAs is set to rise with new uses and formulations. Randomized controlled trials (RCTs) show semaglutide's effectiveness in metabolic and cardiovascular outcomes. Studies indicate potential neurobiological benefits for neurological, psychiatric, and substance use disorders.

However, safety reviews by the European Medicines Agency (EMA) and the United Kingdom (UK) Medicines and Healthcare products Regulatory Agency (MHRA), triggered by mood change reports, raise concerns. Further research is needed to clarify the neuropsychiatric safety and overall brain health impact of semaglutide and GLP1-RAs in T2DM patients.

About the study 

The present study adheres to the Reporting of Studies Conducted using Observational Routinely-collected Health Data (RECORD) guidelines. Utilizing the TriNetX United States (US) Collaborative Network, a large-scale federated database, this study analyzed anonymized electronic health records (EHRs) of over 100 million patients across 62 healthcare organizations in the US. This platform includes patient demographics, diagnoses, medications, and procedures. 

Researchers identified individuals aged 18 and above with a diagnosis of T2DM and a first prescription of semaglutide between December 1, 2017, and May 31, 2021. Comparator cohorts were formed for sitagliptin, empagliflozin, and glipizide. The study matched cohorts based on 179 covariates, including demographics, socioeconomic factors, lifestyle, healthcare utilization, comorbidities, and medication history.

The study assessed the risk of 22 neurological and psychiatric outcomes within one year of the index event using International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) codes. It also estimated the risk of 15 negative control outcomes to address unmeasured confounding. Propensity-score matching and subsequent analyses were conducted using R and the TriNetX interface, with statistical significance set at a p-value of less than 0.05 and adjustments for multiple testing.

Four secondary analyses were performed: subgroup analysis by age, stratification by year of index prescription, competing risk analysis with death as a component, and exploration of outcomes at a two-year follow-up. The study's funding sources did not influence the design, data collection, analysis, interpretation, or manuscript writing. 

Study results 

Following the application of inclusion/exclusion criteria, in each cohort a total of 23,386 patients were included for the semaglutide vs. sitagliptin comparison (48.6% female, mean ± SD age 56.6 ± 12.8 years), 22,584 patients in each cohort for the semaglutide vs. empagliflozin comparison (48.9% female, mean ± SD age 57.6 ± 12.4 years), and 19,206 patients in each cohort for the semaglutide vs. glipizide comparison (49.3% female, mean ± SD age 56.3 ± 13.0 years). Propensity-score matching was achieved for every comparison and covariate, ensuring standardized mean differences (SMDs) of less than 0.1. 

Hazard ratios (HRs) below 1 indicate a lower risk with semaglutide compared to other drugs, while HRs above 1 indicate a higher risk. Semaglutide was not linked to an increased risk of any neurological or psychiatric conditions; it was associated with lower risks of cognitive deficits compared to sitagliptin (HR 0.72) and glipizide (HR 0.72) but similar to empagliflozin (HR 0.96).

Dementia (Cognitive decline affecting memory and thinking) risk was also lower with semaglutide compared to sitagliptin (HR 0.52) and glipizide (HR 0.63). Additionally, semaglutide showed a significantly lower risk of nicotine dependence compared to glipizide (HR 0.72) and empagliflozin (HR 0.77).

Other notable results include a lower risk of depression and ischemic stroke (Stroke from blocked brain blood supply) compared to sitagliptin. Negative control outcomes (NCOs) consistently showed no difference for any comparisons (HRs for composite NCOs between 0.97 and 1.03, all p-values >0.4). The risk of all-cause mortality was lower with semaglutide compared to sitagliptin (95% CI 0.50–0.66, p < 0.0001, HR 0.58), empagliflozin (p 0.035, HR 0.86, 95% CI 0.75–0.99,) and glipizide (p < 0.0001, HR 0.55, 95% CI 0.47–0.64).

Secondary analyses confirmed these findings. Age subgroup analysis showed stronger associations in older patients for cognitive deficit and dementia and younger patients for lower substance use disorders. Semaglutide's association with lower risks of death and other outcomes remained significant. Stratification by prescription time, including pre- and post- coronavirus disease 2019 (COVID-19) periods, showed consistent results. At a two-year follow-up, similar patterns were observed, with an additional potential association with a lower risk of psychosis compared to sitagliptin and glipizide.

There was no significant violation of the proportional-hazard assumption, except for a few comparisons, such as cognitive deficit between semaglutide and sitagliptin, where the HR diminished over time but remained below 1.

Conclusions 

To summarize, this large-scale study on semaglutide, a GLP1-RA approved for T2DM and obesity, found no increased risk of 22 neurological and psychiatric outcomes within 12 months compared to other antidiabetic agents, except for a higher risk of migraines with glipizide. Semaglutide was associated with possible cognitive and nicotine misuse benefits, aligning with meta-analyses suggesting GLP1-RAs' advantages for cognitive outcomes. The study's robustness stems from extensive propensity-score matching and a large sample size. These findings support semaglutide's potential to prevent cognitive deficits and substance misuse, informing regulatory reviews and public health.