Optic Ischaemic Neuropathy in Incretin-Based Therapy: A Comparative Analysis of Real-World Safety Data.
Semaglutide shows a strong safety signal for optic ischaemic neuropathy, while lixisenatide does not present any safety concerns in this regard.
Where it sits
this study against the rest of the lixisenatide corpusSummary and findings
This study evaluated the safety signals for optic ischaemic neuropathy across six incretin-based therapies using the FDA Adverse Event Reporting System. It found significant safety signals for semaglutide, tirzepatide, and liraglutide, while no signals were identified for dulaglutide, exenatide, or lixisenatide. The study highlights the need for further investigation into these findings.
Abstract
<h4>Aims</h4>Non-arteritic anterior ischemic optic neuropathy (NAION) has emerged as a safety concern with semaglutide, prompting formal regulatory review and action by the European Medicines Agency. However, its occurrence across the broader class of incretin-based therapies is insufficiently characterised. This study used the FDA Adverse Event Reporting System to evaluate and compare pharmacovigilance signals for optic ischaemic neuropathy across six incretin-based therapies.<h4>Materials and methods</h4>Adverse event reports from each drug's approval date through Q3 2025 were extracted using OpenVigil 2.1. The Medical Dictionary for Regulatory Activities term "optic ischaemic neuropathy" was used as the outcome. Disproportionality analysis was performed using reporting odds ratio (ROR) and proportional reporting ratio (PRR) with 95% confidence intervals, applying Evans' criteria, restricted to primary suspect drug reports.<h4>Results</h4>Semaglutide demonstrated a strong post-marketing safety signal for optic ischaemic neuropathy (355 reports; ROR 94.45, 95% CI: 83.02-107.45; PRR 93.77, 95% CI: 82.48-106.61). Tirzepatide and liraglutide also exhibited significant safety signals (49 reports; ROR 2.94, 95% CI: 2.20-3.93; PRR 2.94, 95% CI: 2.20-3.93; and 13 reports; ROR 4.58, 95% CI: 2.65-7.91; PRR 4.58, 95% CI: 2.65-7.91, respectively). No signal was identified for dulaglutide, exenatide, or lixisenatide. Among semaglutide-associated reports, mean age was 59 years; disability was documented in 18% and hospitalization in 11% of cases.<h4>Conclusions</h4>Significant post-marketing safety signals for optic ischaemic neuropathy were identified for semaglutide, tirzepatide, and liraglutide. The absence of signals with other glucagon-like peptide-1 receptor agonists argues against a uniform class effect. Prospective studies with neuro-ophthalmologist-confirmed diagnoses are warranted to establish causality and quantify absolute risk.
Background
This paper addresses the safety concerns associated with incretin-based therapies, particularly focusing on non-arteritic anterior ischemic optic neuropathy (NAION). Previous reports indicated a strong safety signal for semaglutide, prompting regulatory review. Understanding the occurrence of optic ischaemic neuropathy across other incretin-based therapies is crucial for evaluating their safety profiles.
Methods
The study utilized the FDA Adverse Event Reporting System to extract adverse event reports from each drug's approval date through Q3 2025. A disproportionality analysis was performed using reporting odds ratio (ROR) and proportional reporting ratio (PRR) with 95% confidence intervals, applying Evans' criteria. The analysis was restricted to primary suspect drug reports.
Results
The primary endpoint revealed that semaglutide had 355 reports associated with optic ischaemic neuropathy, yielding an ROR of 94.45 (95% CI: 83.02-107.45) and a PRR of 93.77 (95% CI: 82.48-106.61). Tirzepatide and liraglutide also showed significant safety signals with RORs of 2.94 and 4.58, respectively. No safety signal was identified for dulaglutide, exenatide, or lixisenatide.
Interpretation
The findings indicate that semaglutide, tirzepatide, and liraglutide have significant safety signals for optic ischaemic neuropathy, while lixisenatide does not. The effect size for semaglutide is notably large, but the clinical significance of the smaller signals for tirzepatide and liraglutide may require further investigation. The reliance on adverse event reports introduces confounding factors, such as underreporting and lack of clinical confirmation.
Key findings
- 355 reports for semaglutide; ROR 94.45, 95% CI: 83.02-107.45; PRR 93.77, 95% CI: 82.48-106.61.
- 49 reports for tirzepatide; ROR 2.94, 95% CI: 2.20-3.93; PRR 2.94, 95% CI: 2.20-3.93.
- 13 reports for liraglutide; ROR 4.58, 95% CI: 2.65-7.91; PRR 4.58, 95% CI: 2.65-7.91.
- No signal identified for dulaglutide, exenatide, or lixisenatide.
- Mean age of semaglutide-associated reports was 59 years; 18% documented disability and 11% hospitalization.
Limitations
- Relies on adverse event reporting data.
- Potential underreporting or misclassification of events.
- No clinical confirmation of diagnoses.
- Short follow-up period.
- Single-source data may limit generalizability.