Higher risk profile among patients with TET2-mutated giant cell arteritis: a cluster analysis

Abstract

Objective: We aimed to assess the prevalence of clonal haematopoiesis (CH) in patients with giant cell arteritis (GCA) compared with controls and individuals with other autoimmune diseases (AIDs) and to identify high-risk clinical/genetic profiles that could influence disease outcomes.

Methods: In a prospective observational study at three hospitals, we included 49 patients diagnosed with GCA, 48 patients with other AIDs and 27 control participants. We used next-generation sequencing to detect clonal haematopoiesis (CH) among them.

Results: CH was detected in 55.1% of patients with GCA, 59.3% of controls and 18.8% of patients with other AIDs. The most commonly mutated genes in GCA and control groups were DNMT3A and TET2. No significant differences in CH prevalence were found between patients with GCA and controls or other AID when adjusted for age and sex. Cluster analysis revealed two distinct groups within the patients with GCA, one of which displayed a higher prevalence of TET2 and JAK2 variants, and was associated with worse prognosis.

Conclusions: CH is prevalent among patients with GCA but does not differ significantly from controls or other autoimmune conditions. However, specific genetic profiles, particularly mutations in TET2 and JAK2, are associated with a higher risk cluster within the GCA cohort. This observation highlights the interest of detecting CH in patients with GCA in both routine practice and clinical trials for better risk stratification. Further prospective studies are needed to determine if management tailored to the genetic profile would improve outcomes.

Keywords: epidemiology; giant cell arteritis; vasculitis.

Figure 1. Sequencing analysis of GCA, AID and control individuals. (A) Number of patients stratified by the number of detected mutations among GCA, AID and control individuals. (B) Histogram of VAF among GCA, AID and control individuals. (C) Number of variants among GCA, AID and control individuals. (D) Forest plot of the ORs for GCA diagnosis and CH prevalence versus AID or control individuals. Adjustment covariates included age and sex. Regarding the comparison between GCA and AID, no individual variant comparison was performed due to the low number of events. Regarding the comparison between GCA and control individuals, variant analyses were not adjusted due to the low number of events. The category of spliceosome variants includes SF3B1, SRSF2 and PRPF8. AID, autoimmune disease; CH, clonal haematopoiesis; GCA, giant cell arteritis; VAF, variant allele frequency.

Figure 2. Clonal landscape of patient with GCA in the blood and in TAB. (A) Circular plot showing co-occurring variant pairs among patients with GCA. (B) Co-occurrence matrix of CH variants in blood and TAB among the 12 patients with GCA with available samples, where individuals are represented by one column. No patient had CH in TAB only. CH, clonal haematopoiesis; GCA, giant cell arteritis; TAB, temporal artery biopsy.

Figure 3. Cluster analysis and survival analysis among patients with GCA. (A) The hierarchical clustering on principal components analysis of patients with GCA revealed two clusters (variables included are listed in online supplemental material). (B) Factor map showing the individuals used to generate the dendrogram. (C) Kaplan-Meier survival curves of the cumulative incidence of relapse or GC-dependency by GCA cluster p value is provided using log-rank test. GC, glucocorticoids; GCA, giant cell arteritis.