Background/Purpose: Polymyalgia rheumatica (PMR) is characterized by systemic inflammation and rapid response to glucocorticoids (GC), but its biological heterogeneity and molecular effects of early treatment remain incompletely defined.

We aimed to characterize baseline serum biomarker profiles and early kinetics after GC initiation, and associations with clinical course during the first 24 months.

Methods: We performed a longitudinal observational study including 53 patients with newly diagnosed PMR and available baseline serum sample, obtained before GC exposure or after minimal exposure. 42 patients had paired serum samples obtained at baseline and after 3 months of therapy.

GM-CSF, IFN-γ, IL-6, IL-17A, IL-17F, CXCL10, CXCL9, PDGF-AA, PDGF-AB/BB, and VEGF-A were analysed in all samples by multiplex immunoassays and run by duplicated.

Unfavourable clinical course was defined as: relapse, vasculitis, use of conventional or biological DMARD or cumulative prednisone-equivalent dose at 12 months above the cohort third quartile. Associations between baseline biomarkers and clinical course were assessed using logistic regression models adjusted by age, sex and C-reactive protein (CRP).

Biomarker kinetics were assessed using log-transformed delta values between month 3 and baseline. Linear regression models evaluated associations between biomarker changes and baseline inflammatory burden.

Results: Among 53 patients, 60% were women and mean age was 74.7 years. During follow-up, 33 patients (62.3%) relapsed, 12 (23.1%) required methotrexate, 4 (7.8%) required IL-6 receptor inhibition, and 4 had vasculitis. Median cumulative prednisone-equivalent dose was 1842.5, 2890 and 4110 mg at 6, 12 and 24 months, respectively. Patients with unfavourable clinical course had numerically higher baseline levels of GM-CSF, IL-17A, IL-17F, VEGF-A, and PDGF-AB/BB. However, none of these biomarkers reached statistical significance for the unfavourable outcomes in adjusted logistic regression models.

In the paired subgroup, median baseline CRP was 35.5 mg/L, ESR 61.5 mm/h and initial prednisone-equivalent dose 15 mg/day. After 3 months of GC therapy, IL-6, IFN-γ, CXCL9, PDGF-AA, PDGF-AB/BB and VEGF-A decreased, whereas GM-CSF and IL-17-related biomarkers showed heterogeneous kinetics. Higher baseline CRP was independently associated with greater decreases in IFN-γ (β -3.189, p=0.007), PDGF-AA (β -0.963, p=0.005) and PDGF-AB/BB (β -0.861, p=0.006). Similar associations were observed for ESR with IFN-γ (β -6.459, p=0.030), PDGF-AA (β -1.976, p=0.022) and PDGF-AB/BB (β -1.766, p=0.026).

Conclusion: In our cohort, patients with unfavourable clinical course showed numerically higher baseline of GM-CSF, IL-17A, IL-17F, VEGF-A, and PDGF-AB/BB, although none of these demonstrated independent prognostic value.

In the paired longitudinal subgroup, GC therapy was associated with early decreases in several biomarkers related to systemic inflammation, IFN-γ/CXCL9-CXCL10 signalling, and vascular remodelling, whereas GM-CSF and IL-17-related biomarkers showed more heterogeneous kinetics.

Table 1. Baseline characteristics.

Table 2. Paired samples with log-transformed delta values between month 3 and baseline.

ACR Meeting Abstracts - https://acrabstracts.org/abstract/baseline-and-early-longitudinal-serum-biomarkers-in-polymyalgia-rheumatica-kinetics-and-associations-with-clinical-course/