1164: Effect of TNF Inhibitors on the Postprandial Glucose Homeostasis Molecules: Amylin, Gastric Inhibitory Polypeptide, and Pancreatic Polypeptide in Rheumatic Patients

Background/Purpose: The effect of systemic TNF inhibition on postprandial glucose homeostasis, a parameter associated with cardiovascular health, remains unclear. This study aims to evaluate the effects of TNF inhibitors (TNFi) on hormones involved in the fine-tuning of postprandial glucose homeostasis: amylin, gastric inhibitory polypeptide (GIP), and pancreatic polypeptide (PP), in patients with rheumatoid arthritis (RA), psoriatic arthritis (PsA), and axial spondylarthritis (axSpA).

Methods: Sixty-four non-diabetic patients (RA, n=22; PsA, n=13; axSpA, n=29) undergoing TNFi therapy were prospectively evaluated at baseline and after 12, 24, and 52 weeks. Assessments included measurements of disease activity, acute-phase reactants, fasting glucose, insulin and C-peptide levels, and insulin resistance as assessed by the updated Homeostatic Model Assessment (HOMA)-2. Serum concentrations of amylin, GIP, and PP were measured at each visit. Multivariable linear mixed-effects models were used to analyze temporal changes in these molecules of glucose metabolism homeostasis.

Results: BMI remained stable throughout the one-year follow-up period. Among patients with RA, disease activity indices, including DAS28-CRP, SDAI, and CDAI, showed significant reductions at 3, 6, and 12 months. Likewise, in patients with axSpA, there was a significant decrease in disease activity, metrological, and functional indices. Finally, in patients with PsA, a significant decrease was observed in the DAPSA index at the 6-month visit, but this improvement was not maintained at the 1-year follow-up. No significant modifications were observed in lipid-related molecules, apart from triglycerides, which showed significantly lower levels after one year of treatment compared to baseline

Serum glucose levels remained consistent with baseline levels after one year of TNFi treatment. Circulating insulin and C-peptide levels remained stable during TNFi treatment. Indices of insulin sensitivity and beta-cell function, as measured by the HOMA-2 index, remained unchanged over the one-year period of TNFi treatment. The interaction analysis showed that changes in insulin resistance indices were consistent regardless of diagnosis (RA, PsA, or AS) or type of TNFi used (adalimumab versus non-adalimumab). Despite this, amylin, GIP and PP levels increased significantly shortly after treatment initiation at week 12 and maintained this significant upregulation at 12-month follow-up (Figure). Importantly, the significance of these findings persisted after adjustment for changes in CRP during the follow-up period. No significant interactions were observed between the specific TNFi used or type of rheumatic disease and changes in amylin and GIP over time. This finding suggests that changes in these hormone levels occur independently of both the TNFi used and the underlying disease.

Conclusion: Systemic TNF inhibition is associated with sustained increases in serum levels of amylin, GIP and PP independent of systemic inflammation. This modulation of postprandial glucose-regulating hormones may contribute to the cardiovascular benefits of TNFi in patients with inflammatory disease.