Two controlled trials just published investigated whether infliximabmight have benefits in polymyalgia rheumatica (PR) or giant-cellarteritis (GCA), but found no significant effect. These two conditionsare rheumatic diseases with overlapping clinical features – they arefairly common and generally affect older people. Currently, thecornerstone of therapy is corticosteroid treatment, but this may needto be prolonged and many patients have significantcorticosteroid-related adverse effects. Some evidence suggestscytokines may be involved in pathogenesis, and the two studies aimed todetermine whether blockade of tumour necrosis factor (TNF) withinfliximab had any benefits.
The GCA study comparedinfliximab with placebo in newly diagnosed patients who were receivingstandard corticosteroid therapy (prednisolone or prednisone) to induceremission. Patients received corticosteroid, plus infliximab or placeboIV infusion at zero, two, and six weeks, then every eight weeksthereafter; corticosteroid dose was tapered according to a definedschedule. Planned study duration was 54 weeks and the primary outcomeswere number of patients remaining free of relapse and adverse events.
Atotal of 44 patients were randomised (infliximab n=28, placebo n=16).Interim analysis at 22 weeks found no significant difference betweenthe groups in primary efficacy outcome (remission rates infliximab 43%vs placebo 50%), neither was there any difference in time to relapse.As a result, the study was stopped early. There was no significantdifference in adverse events, although infliximab was associated with ahigher incidence of infections and infusion-site reactions. The authorsconclude that infliximab failed to increase duration of remission orreduce corticosteroid requirements in patients with newly diagnosedGCA. They comment that the study could only have demonstrated a largeeffect, and a more modest effect would have been missed: a larger trialwould be required to detect this, if present. However, a more modesteffect might not justify the risk and cost of using this drug. Thestudy also provides no evidence on use for relapsed patients or thosenot responding to corticosteroids.
The second studylooked at the use of infliximab in PR, using a similar design. Patientshad newly diagnosed PR and were initially treated with standard dosesof corticosteroid to induce remission. Infliximab or placebo was givenby IV infusion at weeks zero, two, six, 14 and 22, and thecorticosteroid dose tapered according to standard procedures. Plannedstudy duration was 52 weeks, and the primary outcome was the proportionof patients without relapse or recurrence through to week 52.
Atotal of 51 patients were randomised (infliximab n=23, placebo n=28).There was no significant difference between the groups in the primaryoutcome at the end of the study (relapse- or recurrence-free –infliximab 30% vs placebo 37%). There were also no significantdifferences in any secondary endpoints or in adverse events. Theauthors conclude that infliximab added to a standard corticosteroidregimen does not affect outcomes in patients with newly diagnosed PR.As for the GCA trial, this was too small to demonstrate a modestbenefit; again, however, a modest benefit might not justify the risksand costs of using infliximab.
An accompanying editorialdiscusses the two studies. The author notes that both were small andwould thus only demonstrate a large effect reliably. However, heconcurs that using an expensive and potentially toxic drug such asinfliximab would only be justifiable if the benefit was large. Hesuggests that the evidence for TNF as a significant pathogenic factoris limited and that other cytokines (eg, interleukin-6) may be moreimportant. Until further research has clarified the role of these,corticosteroids remain a cornerstone of treatment.
Ann Intern Med 2007;146:621-39