Giant-cell arteritis (GCA) is the most frequent vasculitis after 50 years. Corticosteroid therapy is the reference treatment for GCA. This treatment is highly effective but must be maintained for 12 to 24 months to avoid relapses, which causes the onset of numerous adverse effects in this elderly population.

Currently clinicians have no way to estimate this risk of relapse during the treatment of GCA.

Invariant T lymphocytes associated with the mucous membrane (MAIT), whose role in vasculitides has recently been shown and which produce IL-17 and IFN-γ, two key cytokines in the pathophysiology of GCA could be implicated in the pathophysiology of GCA and could constitute a predictive marker of relapse.

Our hypothesis is that blood MAIT are recruited in the artery wall in patients with GCA and that the number of circulating MAIT in the blood falls and then returns to normal if the corticoids are effective.

Given that it will be necessary to include a large number of patients to show that the persistence of a low number of circulating MAIT in patients treated with corticoids is a predictor of relapse, we propose, as the first step, to carry out a pilot study to obtain preliminary data on these new markers.

The study is classified as interventional because a lot of blood samples are taken

Inclusion Criteria:

PATIENTS

• Patients who have provided written consent
• Patients with national health insurance cover
• Age > 50 years
• Patients with Horton disease at the diagnosis, before any treatment.

Horton disease is defined by ACR criteria [7], the diagnosis is made in the presence of 3 of the following 5 criteria:

• Age at the onset of the disease of 50 years or older
• Recent-onset localized headache
• Temporal artery tenderness or decreased temporal artery pulse
• Erythrocyte sedimentation rate (ESR) greater than 50 mm in the first hour (or CRP>20 mg/L)
• Positive temporal artery biopsy (TAB) showing vasculitis with infiltration of mononuclear cells or granulomatous inflammation with or without giant cells.

Control groups:

Control Group 1: Healthy subjects Healthy subjects will be matched for age and sex with patients. They will be recruited in Dijon only. They will be healthy volunteers recruited among blood donors, voluntary hospital personnel (nurses, doctors and secretaries) and patients without an infectious or inflammatory disease, or cancer or auto-immune disease (CRP<5mg/L) recruited in the departments of investigators at the CHU of Dijon Burgundy.

Control Group 2: (PPR without ACG)

• Patient with oral consent
• Patient affiliated to a social security system
• Age > 50 years old
• Patient with PPR at diagnosis, before corticosteroid treatment
• No ACG (see protocol definition)

Control Group 3: (active infection)

• Patient with oral consent
• Patient affiliated to a social security system
• Age > 50 years old
• Inflammatory syndrome (CRP > 10 mg/L and fibrinogen> 4 g/L or CRP > 10 mg/L and VS > 30 mm/h) explained by a infectious syndrome defined by the association:
• a fever ≥ 38°C, hypothermia < 35°C or fever reported by the patient within 72 hours before inclusion
• at least 1 associated clinical and/or radiological sign to an organ infection, for example: pneumopathy, urinary infection, bacterial dermohypodermatitis, infection digestive (non-exhasutive list)
• in the absence of an explanation of the inflammatory syndrome by another cause such as an inflammatory disease or evolutive neoplastic

Exclusion Criteria:

• Adults under guardianship
• Pregnant or breast-feeding women
• Patients who have been treated with corticosteroids or immunosuppressants in the month preceding inclusion
• Patients treated with chemotherapy, immunosuppressants or biotherapy
• Contra-indication for corticosteroid therapy
• Weight <41 kg or Hg <7g/l