Systemic inflammatory markers after GLP-1 receptor agonist versus DPP-4 inhibitor initiation in type 2 diabetes: an overlap-weighted active-comparator cohort study
- 1Department of Epidemiology, Faculty of Health Sciences, University of Debrecen
- 2Division of Metabolic Diseases, Department of Internal Medicine, Faculty of Medicine, University of Debrecen
Abstract
Background: Glucagon-like peptide-1 receptor agonists may influence systemic inflammation beyond their glucose-lowering effects, but direct comparative evidence against other incretin-based therapies remains limited.
Methods: We conducted a retrospective active-comparator, new-user cohort study using routinely collected electronic health-record data from the Clinical Center of the University of Debrecen, Hungary. Patients with type 2 diabetes initiating a glucagon-like peptide-1 receptor agonist or a dipeptidyl peptidase-4 inhibitor between January 2001 and December 2024 after a 12-month washout were included. Baseline markers were summarized during months −12 to −1 and follow-up markers during months 4–12. The follow-up window was selected to reduce the influence of very early post-initiation testing while allowing assessment of a sustained biomarker difference. Propensity-score overlap weighting was used to balance demographic, metabolic, renal, cardiovascular, medication, and healthcare-observation characteristics.
Results: The propensity-score cohort included 1,973 patients, including 1,322 glucagon-like peptide-1 receptor agonist and 651 dipeptidyl peptidase-4 inhibitor initiators; 838 had paired C-reactive protein measurements. The adjusted geometric mean ratio for follow-up C-reactive protein was 0.80 (95% confidence interval, 0.66–0.97; P = 0.024). The neutrophil-to-lymphocyte ratio was also lower (adjusted ratio, 0.92; 95% confidence interval, 0.87–0.97; P = 0.003; q=0.007), whereas platelet count was higher by 12.12×10 ⁹ /L (95% confidence interval, 4.65–19.59; P = 0.001; q=0.007). The C-reactive protein estimate remained similar after upper-tail winsorization, propensity-score re-estimation within the paired-CRP cohort, and patient-level bootstrap re-estimation of the propensity score. Restriction to patients with baseline CRP <10 mg/L yielded a directionally similar but less precise estimate, whereas stabilized inverse-probability-of-treatment weighting did not statistically support the association.
Conclusion: GLP-1 receptor agonist initiation was associated with lower follow-up CRP and NLR than DPP-4 inhibitor initiation among patients with available paired measurements in the propensity-score overlap population. However, the CRP association was attenuated in the 12-month survivor-restricted analysis and varied across other sensitivity analyses. Residual confounding by adiposity, treatment-associated weight changes, and selection related to survival and laboratory availability limit causal interpretation. The findings should be considered exploratory comparative biomarker associations rather than evidence of a direct or weight-independent anti-inflammatory effect.
Related
- Associations of mortality and sepsis after GLP-1 receptor agonist initiation versus SGLT2 inhibitor or metformin initiation in people with HIV, overweight or obesity, and type 2 diabetes: A target trial emulation
- Incidence and predictors of urinary tract infections after sodium-glucose cotransporter 2 inhibitor initiation in older adults hospitalized for heart failure
- Clinical Comparator Dependence of Inflammatory and Hematologic Biomarker Discrimination for Sepsis: A Matched Retrospective Study Using MIMIC-IV Emergency Department Data
- Divergent Eosinophilic and Paradoxical Inflammatory Adverse-Event Signatures Across Five Type 2 Asthma Biologics: A FAERS Active-Comparator Disproportionality Analysis
- GLP-1 receptor agonist initiation and newly recorded ankylosing spondylitis diagnoses in type 2 diabetes: a target trial emulation
BibTeX
@article{nagy2026systemic,
title = {{Systemic inflammatory markers after GLP-1 receptor agonist versus DPP-4 inhibitor initiation in type 2 diabetes: an overlap-weighted active-comparator cohort study}},
author = {Attila Nagy and Amr Sayed Ghanem and Battamir Ulambayar and Vivien Jendrek and Ferenc Sztanek},
journal = {Frontiers in Endocrinology},
year = {2026},
month = nov,
doi = {10.3389/fendo.2026.1959709},
url = {https://www.frontiersin.org/articles/10.3389/fendo.2026.1959709/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Systemic inflammatory markers after GLP-1 receptor agonist versus DPP-4 inhibitor initiation in type 2 diabetes: an overlap-weighted active-comparator cohort study
AU - Attila Nagy
AU - Amr Sayed Ghanem
AU - Battamir Ulambayar
AU - Vivien Jendrek
AU - Ferenc Sztanek
PY - 2026
DA - 2026/11/16
JO - Frontiers in Endocrinology
DO - 10.3389/fendo.2026.1959709
UR - https://www.frontiersin.org/articles/10.3389/fendo.2026.1959709/full
AB - Background: Glucagon-like peptide-1 receptor agonists may influence systemic inflammation beyond their glucose-lowering effects, but direct comparative evidence against other incretin-based therapies remains limited. Methods: We conducted a retrospective active-comparator, new-user cohort study using routinely collected electronic health-record data from the Clinical Center of the University of Debrecen, Hungary. Patients with type 2 diabetes initiating a glucagon-like peptide-1 receptor agonist or a dipeptidyl peptidase-4 inhibitor between January 2001 and December 2024 after a 12-month washout were included. Baseline markers were summarized during months −12 to −1 and follow-up markers during months 4–12. The follow-up window was selected to reduce the influence of very early post-initiation testing while allowing assessment of a sustained biomarker difference. Propensity-score overlap weighting was used to balance demographic, metabolic, renal, cardiovascular, medication, and healthcare-observation characteristics. Results: The propensity-score cohort included 1,973 patients, including 1,322 glucagon-like peptide-1 receptor agonist and 651 dipeptidyl peptidase-4 inhibitor initiators; 838 had paired C-reactive protein measurements. The adjusted geometric mean ratio for follow-up C-reactive protein was 0.80 (95% confidence interval, 0.66–0.97; P = 0.024). The neutrophil-to-lymphocyte ratio was also lower (adjusted ratio, 0.92; 95% confidence interval, 0.87–0.97; P = 0.003; q=0.007), whereas platelet count was higher by 12.12×10 ⁹ /L (95% confidence interval, 4.65–19.59; P = 0.001; q=0.007). The C-reactive protein estimate remained similar after upper-tail winsorization, propensity-score re-estimation within the paired-CRP cohort, and patient-level bootstrap re-estimation of the propensity score. Restriction to patients with baseline CRP <10 mg/L yielded a directionally similar but less precise estimate, whereas stabilized inverse-probability-of-treatment weighting did not statistically support the association. Conclusion: GLP-1 receptor agonist initiation was associated with lower follow-up CRP and NLR than DPP-4 inhibitor initiation among patients with available paired measurements in the propensity-score overlap population. However, the CRP association was attenuated in the 12-month survivor-restricted analysis and varied across other sensitivity analyses. Residual confounding by adiposity, treatment-associated weight changes, and selection related to survival and laboratory availability limit causal interpretation. The findings should be considered exploratory comparative biomarker associations rather than evidence of a direct or weight-independent anti-inflammatory effect.
SN - 1664-2392
ER -