Treatment with semaglutide appears effective in reducing the risk of adult-onset epilepsy or seizures relative to sodium–glucose cotransporter-2 inhibitors (SGLT2i) and other glucose-lowering drugs (GLDs) in patients with type 2 diabetes mellitus (T2DM), according to a study presented at AAN 2026.
“Semaglutide initiation was associated with lower adult-onset seizure risk, [and] signal was seen against both other GLDs and SGLT2i,” said study co-author and presenter Dr Hyunyong Koh, Department of Neurology, Baylor College of Medicine/Texas Children’s Hospital, Houston, Texas, US.
Koh and his team conducted a retrospective active-comparator, new-user target trial emulation using the NIH All of Us Controlled Tier Dataset v8. They followed adults (aged ≥18 years) with T2DM who initiated semaglutide, SGLT2i, or GLDs between December 2018 and December 2021 through December 2023.
The authors identified outcomes (ie, incident epilepsy or seizures) using the ICD-9/10 and SNOMED CT. They balanced the baseline demographics, comorbidities, medications, HbA1c, and BMI using inverse probability of treatment weighting (IPTW).
Kon and colleagues calculated hazard ratios (HRs) via Cox proportional hazards model and estimated risk differences (RD) and numbers needed to treat (NNT) through targeted maximum likelihood estimation (TMLE). They also performed counterfactual mediation to evaluate HbA1c and BMI as mediators (12-month windows over 48 months).
Lower risk
A total of 393,596 adults were identified, of whom 8,533 met the eligibility criteria for the semaglutide vs GLD comparison (2,397 vs 6,136) and 7,455 for the semaglutide vs SGLT2i comparison (1,650 vs 3,725). [Eun Y, et al, AAN 2026]
After IPTW adjustment, Koh and his team found a significant association between semaglutide use and a reduced risk of adult-onset epilepsy or seizures compared with GLDs (HR, 0.46, 95 percent confidence interval [CI], 0.25‒0.83; p=0.010) and SGLT2i (HR, 0.44, 95 percent CI, 0.22‒0.86; p=0.017).
These findings persisted in TMLE (GLDs: RD, ‒0.014; NNT, 69; p<0.001; SGLT2i: RD, ‒0.008; NNT, 129; p<0.001). Furthermore, BMI and HbA1c appeared to minimally mediate such association (Hba1c: 1.1 percent with GLDs, 3.6 percent with SGLT2i; BMI: ≤0.3 percent). These effects remained stable across follow-up of 12 to 48 months.
“[The] association was not meaningfully mediated by HbA1c or BMI and warrants prospective validation,” according to Koh.
“The effect appeared largely independent of glycaemic or weight changes, supporting possible nonglycaemic neuroprotective mechanisms of glucagon-like peptide-1 (GLP-1) receptor agonists (RAs),” the authors said.
Beyond seizure
Notably, the findings of this target trial emulation “may have implications beyond seizure risk alone and raise the possibility that semaglutide could also be relevant to broader brain health,” said another co-author Dr Yoonhyuk Jang, Department of Immunology, Harvard Medical School, Boston, Massachusetts, in a report released by AAN. [https://www.neurologyadvisor.com/reports/semaglutide-lower-risk-adult-onset-epilepsy-t2dm/]
“[H]owever, because this was a retrospective target trial emulation with a relatively small number of events, it is not yet practice-changing and should instead be viewed as a signal that supports further research into the role of GLP-1 RAs in epileptogenesis,” he added.
“Adult-onset seizure is common,” Koh said. “[It] reflects acquired brain injury burden, prognosis and quality of life are … affected, [and] disease-modifying preventive strategies remain limited.”