Prognostic Value of Cranial Arteries Involvement on 18 F-FDG PET/CT at Giant Cell Arteritis Diagnosis in a Multicenter Cohort.

Abstract

The aim of this study was to assess the prognostic value of 18 F-FDG PET/CT-based features performed at the diagnosis of giant cell arteritis (GCA) for the risk of relapse. 18 F-FDG PET/CT was performed in patients with GCA between 2014 and 2024 before 10 days of corticosteroid therapy. Visual scores were assessed: TVS, PETVAS, "extended" PETVAS with 17 segments (PETVAS51), cephalic score "C-VAS" dedicated to the branches of the external carotid artery, and global score "G-VAS" combining all the previous territories. Maximum standardized uptake values (SUVmax) and target-to-background ratios (TBR) normalized to liver or circulating blood pool were also collected. Relapse was defined by the occurrence of suggestive clinical, biological, or imaging events. Survival analyses (Kaplan-Meier) and Cox models adjusted for age, sex, and camera type (analog or digital) were used to identify prognostic factors for relapse. Seventy-six patients were included, 41% of them relapsed after a median follow-up of 14 months. None of the conventional scores (TVS, PETVAS, PETVAS 51) or semiquantitative parameters predicted relapse. In contrast, the G-VAS >45 and C-VAS >8 were significantly associated with relapse in univariate analysis [HR = 2.64 (95% CI: 1.00; 6.98), P = 0.043 and 3.48 (1.30; 9.29), P = 0.0084, respectively]. Only a C-VAS score >8 was an independent predictor of prompt relapses in multivariate analysis [HR = 3.31 (1.16-9.38)]. In patients with GCA, increased uptake of the cephalic branches of the external carotid artery on initial 18 F-FDG PET/CT was an independent prognostic factor for relapse.