Exploring Subclinical Giant Cell Arteritis in Polymyalgia Rheumatica: A Novel Phenotype in the GCA-PMR Disease Spectrum.

Abstract

Hypothesis: This study hypothesizes that patients with subclinical GCA in PMR have a distinct phenotype in the GCA-PMR disease spectrum. Aim: This study aims to define the prevalence of subclinical GCA in PMR in a newly diagnosed PMR cohort, characterise the ultrasound pattern, natural history, and biomarker profile of this cohort, and compare with prospective PMR and GCA cohorts. Objectives of the Research: 1. To define the clinical manifestations of GCA in the context of the GCA-PMR Spectrum and examine the clinical impact of a GCA fast-track clinic on an Irish GCA cohort. 2. To examine the feasibility and clinical impact of establishing a novel fast-track PMR clinic, define patient characteristics, and establish the relapse rate over 12 months. 3. To assess the prevalence of subclinical GCA with temporal and axillary ultrasound in patients with newly diagnosed PMR and describe their clinical, ultrasound, and laboratory marker patterns. 4. To define the natural history of subclinical GCA in PMR including relapse, progression to GCA, and adverse events over 1 year follow-up and compare it to PMR and GCA cohorts. 5. To compare the Halo count, Halo score, and OMERACT score in patients with subclinical GCA in PMR with GCA. 6. To determine if analysis of serum biomarkers in PMR, subclinical GCA in PMR, and GCA may help differentiate between these related conditions. Methods: This was a prospective study of consecutive patients with newly diagnosed PMR without signs or symptoms of GCA. All patients were seen in a novel PMR fast-track clinic and underwent clinical, laboratory, and ultrasound assessment of their temporal and axillary arteries. Patients with a positive vascular ultrasound, defined as the presence of a halo sign with increased intima-media thickness were considered to have subclinical GCA. Clinical and laboratory characteristics were compared with PMR patients without subclinical GCA. Ultrasound characteristics were compared with a consecutive cohort of newly diagnosed GCA patients recruited from a GCA fast-track clinic. Results: We included 73 newly diagnosed PMR patients and 24 new GCA patients. 20.5% newly diagnosed PMR patients had ultrasound evidence of subclinical vasculitis. These patients were more likely to be male (80% vs 45%) but did not differ clinically at presentation. The mean time to relapse was statistically different between the three groups; 8.2 months for subclinical GCA in PMR,9.2 months in PMR, and 11.4 months in GCA, p=0.03. PMR with subclinical GCA showed an increased propensity towards extracranial large vessel vasculitic involvement compared with classical GCA (47% vs 33%) without meeting statistical significance. While ultrasound scores steadily decreased over the first 6 months in GCA patients, there was a statistically significant increase in halo score in subclinical GCA from 3 to 12 months, p=0.02, and in OGUS score from 3 to 6 months, p<0.001. There was no statistical difference in the prevalence of cardiovascular risk factors including hypertension, hyperlipidaemia, diabetes mellitus and smoking status in patients with subclinical GCA in PMR. Subclinical GCA in PMR had a distinct biomarker profile to pure PMR. Angiopoietin-1 was significantly elevated in subclinical GCA in PMR compared with PMR patients at baseline at 25,548 [22,471.30-26468.36] versus 19,774.46 [18,051.58 -22140.71]; p<0.001. A cut-off of 21,408 pg/ml produced a sensitivity of 84.6% and a specificity of 70% for differentiating subclinical GCA in PMR from pure PMR. IL-17 levels were significantly more elevated in subclinical GCA in PMR compared with isolated PMR. ROC analysis demonstrated an AUC of 0.892, with a cut-off of 5pg/ml giving a sensitivity of 84.6%, and specificity of 79.2%. Osteopontin levels were significantly elevated in subclinical GCA in PMR compared with isolated PMR and ROC analysis found the AUC to be 0.814. A cut-off of 2766pg/ml gave a sensitivity of 84.6% and a specificity of 72.7%. GCA serum was significantly different from PMR serum at time of diagnosis, with GCA exhibiting significant elevations in angiopoietin-1 (p<0.001), angiopoietin-2 (p<0.001), CXCL-9 ((p<0.001), osteopontin (p<0.001), IL-6 (p=0.02), and IL-17 (p<0.001) compared with PMR. Conclusion: 1 in 2 patients with suspected PMR in the community did not have PMR and 1 in 5 patients with confirmed PMR had subclinical GCA. Patients with subclinical GCA in PMR had a distinct phenotype. Their ultrasound pattern showed a predilection for extracranial artery involvement and unlike GCA ultrasound scores failed to resolve at follow-up. Subclinical GCA were more likely to relapse earlier than their PMR or GCA counterparts and required significantly higher glucocorticoid treatment compared with PMR. Angiopoietin-1 IL-17, and osteopontin levels differentiate between subclinical GCA in PMR and pure PMR. Subclinical GCA in PMR is a distinct disease phenotype and biomarker signature in the GCA-PMR disease spectrum.