Corticosteroid therapy has always been the standard treatment for giant cell arteritis (GCA), with very good initial clinical efficacy but a high relapse rate when it declines.

The target population of this condition, often elderly, is particularly exposed to the numerous undesirable effects of corticosteroid therapy, and this especially as its duration lengthens with the re-increases of doses according to relapses: metabolic complications, osteo-muscular , infectious or neuropsychiatric.

Investigators propose to compare prospectively the results of a "conventional" corticosteroid regimen as recommended by European societies, to those of a "lighter and / or shorter" scheme, inspired by recent North American trials. , including the largest prospective global study in the field. Investigators hypothesize non-inferiority of the lightened regimen for relapse rate without relapse at S52, but with a decrease in treatment-related adverse events whose cumulative doses should be lower.

Investigators therefore plan to include prospectively over 3 years 150 patients, 75 for each of the two arms, with a newly diagnosed ACG. A randomization of the treatment arm will be performed and a predefined pattern of cortisone adapted to body weight will be given to the patient. Relapse rates, maintenance of remission, cumulative doses of cortisone and adverse effects of treatment will be analyzed at the 52nd week of the introduction of corticosteroid therapy. An interim analysis is planned at S28.

Treatment of giant cell arteritis (GCA) relies on the use of glucocorticoids (GC), with a very good clinical response at treatment initiation. However, relapses at GC tapering are frequent. GCA population is elderly, frequently over 80 years, and is especially affected by GC-related side effects, that increase proportionally with treatment duration. Thus, metabolic, musculo-skeletal, infectious or neuro-psychiatric complications are frequent during prolonged GC use.

After GC introduction, gradual tapering is scheduled, provided the disease remains clinically and biologically controlled. In France, guidelines recommend tapering GC on an 18-24 months timeframe, while other countries, such as the USA, usually taper GC over a shorter period, often 6-8 months. Few comparative data exist on the relapse rates or the GC-related side effects in both settings. In this prospective multicenter study, two GC-tapering schedules are planned: patients in one arm (short treatment) will be treated for 28 weeks, while patients in the second arm will be treated for 52 weeks. Each starting dose of GC and tapering doses will be adapted to body weight. The primary endpoint is to compare the remission rate without relapse at W52 between the two groups and the secondary endpoints are: 1) cumulative GC doses at W52; 2) GC-related side effects and 3) number of relapses (minor and severe) in both arms at W52.

The results of this study might considerably modify future French clinical practice if investigators confirm that a shorter GC treatment does not significantly impact the disease course while reducing GC-related side effects.

Inclusion Criteria:

• Age ≥ 50 years

• Patient with temporal arteritis giant cell match 2 of the 4 criteria of the American College of Rheumatology (ACR) that given :

  • a temporal artery biopsy compatible with a diagnosis of CAG or
  • an abdominal thoracic aortitis diagnosed by Angio CT, MR angiography or PET scanner or
  • Echo Doppler compatible with a diagnosis of CAG

• Oral corticosteroid treatment started up to 14 days, the initial dose is less or equal to 1 mg / Kg

• Patient wo has given its written consent Patient affiliated with a social security

Exclusion Criteria:

Subjects checking one of the criteria for non-inclusion may be eligible to participate in the research. These criteria may include:

• Early treatment of CAG disease with a dose> 1 mg / kg whatever the duration
• Corticosteroids already started over 14 days
• Giant arteritis cell on relapse
• dementia syndrome
• No compliant patient
• Patients who live more than 150 km from the investigation center
• Person under judicial protection, guardianship
• Hypersensitivity to prednisone or any of its excipients
• Infection requiring an systemic treatment
• Evolutive viroses (Hepatitis, Herpes, varicella-zoster virus)
• Immunization with live vaccines / mitigated during the 8 weeks preceding inclusion
• Pregnancy, breastfeeding women or women of childbearing potential not using contraception

All individual data will be analyzed in University Hospital, Caen