Could incretins be the EMPA-REG moment for diabetic neuropathy?
Aims: To compare the prevalence and clinical severity of diabetic peripheral neuropathy (DPN) and painful DPN among people with type 2 diabetes (T2D) treated with glucagon-like peptide-1 receptor agonists (GLP-1 RAs), sodium–glucose cotransporter-2 inhibitors (SGLT2i), or standard therapy.
Methods: This cross-sectional study included 500 individuals with T2DM receiving standard treatment (n = 208), SGLT2i (n = 152), or GLP-1 RAs (n = 140). DPN was evaluated using neurological examination, nerve conduction studies (NCS), and skin biopsy when indicated. Painful DPN was classified according to a grading system for neuropathic pain. Neuropathy severity and pain characteristics were assessed using the Toronto Clinical Neuropathy Score, Pain Numerical Rating Scale, and Neuropathic Pain Symptom Inventory. Multivariable analyses were performed to assess the associations between treatment group and neuropathy outcomes, adjusting for clinical and metabolic factors.
Results: DPN was identified in 58.6% of participants, while 26.2% had painful DPN. Participants receiving GLP-1 RAs had a lower prevalence of DPN than those receiving standard treatment (46.4% vs. 60.6%). The prevalence of painful DPN was also lower among GLP-1 RA users than among those receiving standard treatment (10.1% vs. 35.1%) or SGLT2i (28.3%). After adjustment for clinical and metabolic covariates, GLP-1 RA treatment remained associated with lower odds of DPN (odds ratio [OR] 0.60, 95% confidence interval [CI] 0.37–0.96) and painful DPN (OR 0.32, 95% CI 0.15–0.65). Participants receiving GLP-1 RAs also had lower neuropathy severity scores and pain intensity.
Conclusions: GLP-1 RA treatment was associated with a lower prevalence of DPN and painful DPN, as well as less severe neuropathy and pain and the associations remained significant after multiple adjustments for clinical and metabolic factors.
Comments. GLP-1 RAs have since their release attracted interest as potential neuroprotective agents for their anti-inflammatory and other neurobiological effects. However, clinical evidence remains conflicting, potentially reflecting differences in study designs, patient populations, treatment indications and healthcare settings, as well as confounding by indication. Interestingly, neurological sensory adverse events, particularly dysaesthesia, have also emerged in a dose-dependent manner in trials of high-dose incretin-based therapies for weight management, both in people with and without T2DM. In the STEP UP programme, dysaesthesia was reported in up to 22,9% of patients receiving 7.2 mg Semaglutide, while similar dose-related patterns have been reported for dual agonists or with the investigational triple agonist Retatrutide. Although these events do not necessarily indicate nerve damage, they highlight the complex interplay between incretin-based therapies and sensory function.
Against this background, the findings of Falco et al. are particularly interesting. In this cross-sectional study of 500 people with T2DM, GLP-1 RA use was associated with lower odds of DPN and painful DPN, alongside lower neuropathy severity and pain intensity. These findings are timely given Novo Nordisk's recent announcement that CagriSema reduced pain intensity by 2.4 points on the Numerical Rating Scale, compared with 1.3 points with placebo, in the phase 2 REIMAGINE study. Although the full dataset is not yet available for independent evaluation, these results further support investigating incretin-based therapies in painful DPN. The present study's strengths include consecutive recruitment, detailed neurological phenotyping, objective assessments including NCS and skin biopsy when indicated, and validated measures. However, its cross-sectional design precludes causal inference. Residual confounding, particularly by indication, remains a concern despite adjustment, and the relatively small GLP-1 RA group limits precision and generalisability of the findings. Furthermore, pain reduction should not be equated with neuroprotection. Prospective studies and randomised controlled trials are needed to establish whether incretin-based therapies can meaningfully reduce neuropathic pain and/or prevent DPN.
Johan Røikjer
Reference. Falco P, Di Flaviani A, Galosi E, Angotti S, De Stefano G, Di Pietro G, Di Stefano G, D'Onofrio L, Evangelisti E, Leone C, Litewczuk D, Arendt-Nielsen L, Buzzetti R, Spallone V, Truini A. Glucagon-like peptide-1 receptor agonist treatment is associated with lower frequency of diabetic peripheral neuropathy and neuropathic pain in type 2 diabetes. Pain. 2026 Nov 1;167(11):2523-2531. doi: 10.1097/j.pain.0000000000004104. Epub 2026 Aug 20. PMID: 42663556.
