• infliximab
• tumour necrosis factor α
• giant cell arteritis

Corticosteroids have been the mainstay of treatment for giant cell arteritis (GCA), usually given for at least two years, with the potential for unpleasant consequences.^1,2 Infliximab, a chimeric monoclonal antibody against tumour necrosis factor α (TNFα), used successfully in the treatment of rheumatoid arthritis (RA) and other conditions,^3–5 was considered by us, on the basis of pathogenic and pathologic data of GCA, as a potentially effective treatment for this disease. Consequently, we decided to try it in two of our patients with GCA, without concomitant administration of corticosteroids, hoping that if only a few infusions of anti-TNFα cured the disease then the side effects of chronic steroid treatment might be avoided.

Our patients were both male, 85 and 80 years old, respectively, seen six months apart from each other. They had a typical clinical picture of GCA with recent onset of fatigue, weight loss, bilateral temporal headache and low grade fever, mild anaemia, erythrocyte sedimentation rate (ESR) >100 mm/1^st h and C reactive protein (CRP) >10 times normal. A temporal artery biopsy was typical for GCA. After approval by the hospital ethics committee and informed consent from the patients, 3 mg/kg of infliximab were given intravenously. Within hours the patients reported dramatic improvement, and their fever and headache disappeared. A second infusion was given after two weeks when the ESR was around 40 mm/1^st h and, and a third, one month later when the patients were in excellent condition and had a normal ESR and CRP.

The first patient did well for three months, but at that time symptoms recurred and, despite two infusions at monthly intervals, no significant improvement was noted. Consequently, a decision to stop infliximab was made, and methylprednisolone 16 mg/day was started, with excellent response. He was followed up closely.

Symptoms and a rise in the ESR recurred six weeks after the third injection in the second patient. With the experience of the first patient, we decided to stop infliximab and switch the patient to the conventional treatment, giving methylprednisolone 32 mg/day. He responded well and his ESR was 22 mm/1^st h three weeks later. He was then followed up regularly.

As described above, infliximab as monotherapy was extremely effective at the beginning of the treatment of our patients. A favourable response of GCA to this agent was not unexpected. TNFα has been found, in the pathological lesions of GCA.⁶ Furthermore, the pathology of the lesions, with the local accumulation of CD4 (+) memory cells, resembles that of RA, for which anti-TNFα is very effective. Finally, an association between GCA with different TNFα microsatellite polymorphisms has been reported.⁷

However, the initial dramatic response of our patients was not followed by a sustained improvement. It may very well be that large doses of infliximab, given monthly for a prolonged time, may be effective. However, such an approach is by no means cost effective, and should not be attempted. This trial was based on the hypothesis that, if three, or at the most, five infusions of infliximab, cured this dangerous but self limiting disease, it might be worth giving it, and avoiding the long term undesirable effects of the conventional approach. This was not found to be the case.

In a recent report infliximab was shown to be effective in three of four cases of GCA resistant to steroid.⁸ Furthermore, anti-TNFα treatment of a case of GCA resistant to steroids and immunosuppressive drugs has been reported.⁹

Consequently, we suggest that infliximab should be used in GCA only for patients who are unresponsive to, or intolerant of steroids and/or methotrexate. Probably, a larger dose than that used in RA given for relatively long time will be required. A further conclusion may be that TNFα is one of the major cytokines mediating inflammation in GCA.